Veritas NetBackup™ for

VMware Administrator's
Guide

Release 8.3
Veritas NetBackup™ for VMware Administrator's
Guide
Last updated: 2020-07-28

Legal Notice
Copyright © 2020 Veritas Technologies LLC. All rights reserved.

Veritas, the Veritas Logo, and NetBackup are trademarks or registered trademarks of Veritas
Technologies LLC or its affiliates in the U.S. and other countries. Other names may be
trademarks of their respective owners.

This product may contain third-party software for which Veritas is required to provide attribution
to the third party (“Third-party Programs”). Some of the Third-party Programs are available
under open source or free software licenses. The License Agreement accompanying the
Software does not alter any rights or obligations you may have under those open source or
free software licenses. Refer to the Third-party Legal Notices document accompanying this
Veritas product or available at:

https://www.veritas.com/about/legal/license-agreements

The product described in this document is distributed under licenses restricting its use, copying,
distribution, and decompilation/reverse engineering. No part of this document may be
reproduced in any form by any means without prior written authorization of Veritas Technologies
LLC and its licensors, if any.

THE DOCUMENTATION IS PROVIDED "AS IS" AND ALL EXPRESS OR IMPLIED

CONDITIONS, REPRESENTATIONS AND WARRANTIES, INCLUDING ANY IMPLIED
WARRANTY OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE OR
NON-INFRINGEMENT, ARE DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH
DISCLAIMERS ARE HELD TO BE LEGALLY INVALID. Veritas Technologies LLC SHALL
NOT BE LIABLE FOR INCIDENTAL OR CONSEQUENTIAL DAMAGES IN CONNECTION
WITH THE FURNISHING, PERFORMANCE, OR USE OF THIS DOCUMENTATION. THE
INFORMATION CONTAINED IN THIS DOCUMENTATION IS SUBJECT TO CHANGE
WITHOUT NOTICE.

The Licensed Software and Documentation are deemed to be commercial computer software
as defined in FAR 12.212 and subject to restricted rights as defined in FAR Section 52.227-19
"Commercial Computer Software - Restricted Rights" and DFARS 227.7202, et seq.
"Commercial Computer Software and Commercial Computer Software Documentation," as
applicable, and any successor regulations, whether delivered by Veritas as on premises or
hosted services. Any use, modification, reproduction release, performance, display or disclosure
of the Licensed Software and Documentation by the U.S. Government shall be solely in
accordance with the terms of this Agreement.

Veritas Technologies LLC

2625 Augustine Drive
Santa Clara, CA 95054
http://www.veritas.com

Technical Support
Technical Support maintains support centers globally. All support services will be delivered
in accordance with your support agreement and the then-current enterprise technical support
policies. For information about our support offerings and how to contact Technical Support,
visit our website:

https://www.veritas.com/support

You can manage your Veritas account information at the following URL:

https://my.veritas.com

If you have questions regarding an existing support agreement, please email the support
agreement administration team for your region as follows:

Worldwide (except Japan) [email protected]

Japan [email protected]

Documentation
Make sure that you have the current version of the documentation. Each document displays
the date of the last update on page 2. The latest documentation is available on the Veritas
website:

https://sort.veritas.com/documents

Documentation feedback
Your feedback is important to us. Suggest improvements or report errors or omissions to the
documentation. Include the document title, document version, chapter title, and section title
of the text on which you are reporting. Send feedback to:

[email protected]

You can also see documentation information or ask a question on the Veritas community site:

http://www.veritas.com/community/

Veritas Services and Operations Readiness Tools (SORT)

Veritas Services and Operations Readiness Tools (SORT) is a website that provides information
and tools to automate and simplify certain time-consuming administrative tasks. Depending
on the product, SORT helps you prepare for installations and upgrades, identify risks in your
datacenters, and improve operational efficiency. To see what services and tools SORT provides
for your product, see the data sheet:

https://sort.veritas.com/data/support/SORT_Data_Sheet.pdf
 Contents

Chapter 1 Introduction .......................................................................... 13

About NetBackup for VMware ......................................................... 13
About the virtual machine backups that include database data ............... 14
About the NetBackup appliance as a VMware backup host ................... 14
NetBackup for VMware components ................................................. 15
Appliance as backup host: component overview ................................. 18
Media servers as backup or discovery hosts ...................................... 18
Overview of the VMware backup process .......................................... 20
NetBackup for VMware terminology ................................................. 20

Chapter 2 Required tasks: overview ................................................ 24

Overview of VMware tasks ............................................................. 24
Overview of NetBackup tasks ......................................................... 25

Chapter 3 Notes and prerequisites .................................................. 26

NetBackup for VMware prerequisites ................................................ 26

NetBackup for VMware: notes and restrictions .................................... 27
Notes on VMware Virtual Volumes (VVols) ......................................... 33
NetBackup IPv6 parameter required for backups in VMware IPv6
environments ........................................................................ 33
NetBackup for VMware: notes on Linux virtual machines ...................... 34
Notes on the NetBackup appliance as a VMware backup host ............... 35
NetBackup for VMware support for SAN multi-pathing .......................... 35
NetBackup for VMware support for fault tolerant VMs .......................... 37
NetBackup character restrictions for the Primary VM identifier ............... 37
In the policy Query Builder, display names, resource pool names, and
vApp names are case-sensitive ................................................. 39
Notes on the hotadd transport mode ................................................. 40
Notes and limitations for tag usage in VMware Intelligent Policy queries
........................................................................................... 41
Notes and limitations for the backup and restore of VMware tag
associations .......................................................................... 42
 Contents 5

Chapter 4 Configure NetBackup communication with

VMware .......................................................................... 44

Adding the VMware backup host to NetBackup ................................... 44

Adding NetBackup credentials for VMware ........................................ 45
Using the VMware Managed Object Browser to verify the server
name ............................................................................. 50
Changing the host name of a VMware server in NetBackup .................. 50
Setting privileges for posting events to vCenter ................................... 51
Authentication token for the NetBackup vSphere plug-ins ..................... 51
Setting global limits on the use of VMware resources ........................... 52
Validating VMware virtualization server certificates in NetBackup ........... 54
ECA_TRUST_STORE_PATH for NetBackup servers and clients
..................................................................................... 58
ECA_CRL_PATH for NetBackup servers and clients ...................... 59
ECA_CRL_PATH_SYNC_HOURS for NetBackup servers and
clients ............................................................................ 60
ECA_CRL_CHECK for NetBackup servers and clients ................... 61
VIRTUALIZATION_HOSTS_CONNECT_TIMEOUT for servers
and clients ...................................................................... 62
VIRTUALIZATION_HOSTS_SECURE_CONNECT_ENABLED
for servers and clients ....................................................... 63
VIRTUALIZATION_CRL_CHECK for NetBackup servers and
clients ............................................................................ 64
VMWARE_TLS_MINIMUM_V1_2 for NetBackup servers and
clients ............................................................................ 65

Chapter 5 Configure NetBackup policies for VMware ................ 67

Configuring a VMware policy from the Policy Wizard ........................... 68
Configuring a VMware policy from the Policies utility ............................ 68
Limit jobs per policy on the Attributes tab (for VMware) ........................ 72
Backup options on the VMware tabVMware backup options .................. 73
VMware backup host ............................................................... 74
Optimizations options (VMware) ................................................ 74
Primary VM identifier options (VMware) ...................................... 76
Existing snapshot handling options (VMware) ............................... 77
Transport modes options (VMware) ............................................ 78
Application protection options (VMware) ...................................... 79
VMware - Advanced Attributes dialog ......................................... 81
About the Post vCenter events option (VMware Advanced
Attributes) ....................................................................... 85
Exclude Disks tab ......................................................................... 86
About the exclude disk options for virtual disk selection .................. 88
 Contents 6

Exclude disks from backups: an example to avoid ......................... 91

Restoring data from the backups that excluded the boot disk or
data disks ....................................................................... 91
Browse for VMware virtual machines ................................................ 91
Limiting the VMware servers that NetBackup searches when browsing
for virtual machines ................................................................. 95
Virtual machine host names and display names should be unique if
VMs are selected manually in the policy ...................................... 97
Primary VM identifier option and manual selection of virtual machines
........................................................................................... 98
About incremental backups of virtual machines ................................... 99
Configuring incremental backups ................................................... 100
Storage Foundation Volume Manager volumes in the virtual machine
.......................................................................................... 100

Chapter 6 Configure a VMware Intelligent Policy ....................... 101

About automatic virtual machine selection for NetBackup for VMware
.......................................................................................... 102
Support and use of VMware tag associations ................................... 103
The basics of a NetBackup query rule ............................................. 104
Important notes on automatic virtual machine selection ...................... 105
NetBackup requirements for automatic virtual machine selection .......... 108
Automatic virtual machine selection: Task overview ........................... 108
Options for selecting VMware virtual machines ................................. 109
Configuring automatic virtual machine selection ................................ 112
Editing an existing query in Basic Mode ........................................... 115
Using the Query Builder in Advanced Mode ..................................... 116
AND vs. OR in queries ................................................................. 117
Examples for the NetBackup Query Builder ...................................... 118
The IsSet operator in queries ........................................................ 120
About selecting virtual machines by means of multiple policies ............. 121
Order of operations in queries (precedence rules) ............................. 122
Parentheses in compound queries ................................................. 124
Query rules for resource pools ....................................................... 125
Query rules for datacenter folders (host folder) ................................. 126
Query rules for duplicate names .................................................... 127
Query rules for tags ..................................................................... 128
Query Builder field reference ......................................................... 129
Test Query screen for VMware ...................................................... 139
Test Query: Failed virtual machines ................................................ 140
Effect of Primary VM identifier parameter on Selection column in Test
Query results ....................................................................... 142
 Contents 7

Effect of Primary VM identifier parameter on VM Name column in Test

query results ........................................................................ 144
Refreshing the display of virtual environment changes in the Query
Builder ................................................................................ 144
Reducing the time required for VM discovery in a large VMware
environment ........................................................................ 146

Chapter 7 Reduce the size of backups ......................................... 147

About reducing the size of VMware backups .................................... 147

Block-level backup (BLIB): full vs incremental ................................... 148
Deleting a vSphere Client snapshot ................................................ 149

Chapter 8 Back up virtual machines .............................................. 151

Backing up virtual machines .......................................................... 151

Using the Activity Monitor to monitor virtual machine backups .............. 153
Restarting jobs individually in the Activity Monitor .............................. 155
Viewing NetBackup activity in vSphere Web Client or vSphere Client
(HTML5) ............................................................................. 155
Using OpsCenter to report on virtual machine backups ....................... 157

Chapter 9 Use Accelerator to back up virtual machines ......... 159

About the NetBackup Accelerator for virtual machines ........................ 159

Accelerator: full vs. incremental schedules ....................................... 160
How the NetBackup Accelerator works with virtual machines ............... 161
Accelerator notes and requirements for virtual machines ..................... 161
Accelerator forced rescan for virtual machines (schedule attribute)
.......................................................................................... 163
Accelerator requires the OptimizedImage attribute ............................ 164
Accelerator backups and the NetBackup catalog ............................... 164
Accelerator messages in the backup job details log ........................... 165
NetBackup logs for Accelerator with virtual machines ......................... 165
About reporting the amount of Accelerator backup data that was
transferred over the network ................................................... 166
Replacing the Accelerator image size with the network-transferred
data in NetBackup command output ......................................... 169

Chapter 10 Restore virtual machines ............................................... 173

Restore notes and restrictions ....................................................... 173
Restore notes and restrictions on Linux ........................................... 177
Restoring the full VMware virtual machine ........................................ 179
 Contents 8

Virtual Machine Recovery dialog boxes (restore to original location)

.................................................................................... 182
Virtual Machine Recovery dialogs boxes (restore to alternate
location) ........................................................................ 188
About VMware virtual machine disk restore ...................................... 196
Restoring VMware virtual machine disks by using Backup, Archive,
and Restore ........................................................................ 197
Select Virtual Disks panel ....................................................... 199
Recovery Destination panel .................................................... 200
Virtual Disk Destination panel .................................................. 201
Storage Destination panel ...................................................... 203
Recovery Summary panel ...................................................... 204
Restoring VMware virtual machine disks by using NetBackup
commands .......................................................................... 204
Determining the backup image ID from which to restore a VMware
virtual machine disk ......................................................... 205
Creating the VMware virtual machine disk restore file ................... 207
Modifying the virtual machine disk restore file ............................. 208
Validating the virtual machine disk restore file ............................. 212
Restoring the virtual machine disk or disks by using the
nbrestorevm command ................................................... 213
VMware virtual machine disk restore file .................................... 214
About restore of individual filesAbout restoring individual VMware files
.......................................................................................... 223
Restoring individual filesRestoring individual VMware virtual machine
files .................................................................................... 224
Setting up NetBackup Client Service for restore to a Windows
shared virtual machine driveSetting up NetBackup Client
Service for VMware restores to a Windows shared virtual
machine drive ................................................................ 227
Restore Marked Files dialog for restore of VMware files ................ 228
How NetBackup handles VMware tag associations at restore .............. 230
Browse and search virtual machines for restore ................................ 232
Search Virtual Machines for Restore dialog ................................ 233
Browse the vCloud or vSphere view for virtual machine clients
.................................................................................... 235
Search results dialog ............................................................. 236
If the recovery host is not at the same NetBackup release level as the
backup host ......................................................................... 236
 Contents 9

Chapter 11 Restore virtual machines with Instant Recovery

.......................................................................................... 237

About Instant Recovery for VMware ................................................ 237

Task overview for Instant Recovery for VMware ................................ 239
Performance recommendations for Instant Recovery for VMware ......... 240
Requirements for Instant Recovery for VMware ................................ 241
Notes on Instant Recovery for VMware ........................................... 242
Restarting the Client for NFS service on a Windows restore host .......... 244
Instant Recovery options on the nbrestorevm command ..................... 244
Restoring a virtual machine with Instant Recovery for VMware ............. 250
Restoring a virtual machine to a different location with Instant Recovery
for VMware .......................................................................... 254
Restoring individual files with Instant Recovery for VMware while the
current virtual machine is running ............................................. 257
Job types for Instant Recovery for VMware ...................................... 260
Reactivating a restored virtual machine with Instant Recovery for
VMware .............................................................................. 261

Chapter 12 Use NetBackup for vCloud Director ........................... 263

About NetBackup for vCloud Director .............................................. 263
Notes on creating a NetBackup policy for vCloud .............................. 264
Notes on restoring virtual machines into vCloud Director ..................... 265
Restoring virtual machines into vCloud Director ................................ 266
Reducing the time required for VM discovery in a large vCloud
environment ........................................................................ 271
Virtual machine recovery dialog boxes for vCloud Director .................. 276
Recovery Destination dialog boxRecovery Destination dialog box
(vCloud Director) ............................................................ 276
Recovery Options dialog box (vCloud Director) ........................... 277
Recovery vApp Options for vCloud Director dialog box (restore to
original location) ............................................................. 277
Virtual Machine Options dialog box for vCloud Director ................. 278
Perform Recovery dialog box for vCloud Director ........................ 278
Recovery Options dialog box for vCloud Director (restore to
alternate location) ........................................................... 279
Recovery vApp Options for vCloud Director dialog box (restore to
alternate location) ........................................................... 279
Select a vApp dialog box ........................................................ 280
Recovery Destination Options for vCloud Director dialog box
(restore to alternate location) ............................................. 281
Virtual Machine Options (restore to alternate location) .................. 281
Network connections (restore to alternate location) ...................... 281
 Contents 10

Perform Recovery dialog box for vCloud Director (restore to

alternate location) ........................................................... 281

Chapter 13 Best practices and more information ......................... 283

NetBackup for VMware best practices ............................................. 283
NetBackup for VMware with deduplication ................................. 284
Best practices for VMware tag usage .............................................. 285
Further assistance with NetBackup for VMware ................................ 286

Chapter 14 Troubleshooting ................................................................ 287

Notes on troubleshooting NetBackup for VMware .............................. 288
Beware of unsupported backup host ............................................... 289
How to determine the ESX network that NetBackup used for the backup
or restore ............................................................................ 289
NetBackup logging for VMware ...................................................... 291
Configuring VxMS logging ............................................................ 292
Format of the VxMS core.log and provider.log file names .................... 295
Configuring the VDDK logging level ................................................ 296
Preventing browsing delays caused by DNS problems ....................... 297
Changing the browsing timeout for virtual machine discovery ............... 299
Changing timeout and logging values for vSphere ............................. 300
Credentials for VMware server are not valid ..................................... 302
Snapshot error encountered (status code 156) ................................. 303
The origin of the snapshot failure: NetBackup or VMware? .................. 306
Conflict between NetBackup and VMware Storage vMotion with
vSphere 5.0 or later ............................................................... 307
Incremental backups of the VM may fail if the NTFS file system in the
guest OS is corrupted and the backup uses MSDP storage ........... 308
The restore fails when you restore individual files to a virtual machine
that has NetBackup client software ........................................... 309
Backup or restore job hangs ......................................................... 309
VMware SCSI requirements for application quiesce on Windows .......... 310
Mount point missing on a restored Windows virtual machine ................ 311
Mount points not available when restoring files from a Linux virtual
machine ............................................................................. 311
Invalid client error when you restore files using NetBackup BAR
interface installed on the virtual machine ................................... 311
VMware virtual machine does not restart after restore ........................ 312
A restored VM may not start or its file system(s) may not be accessible
.......................................................................................... 312
NetBackup job fails due to update tasks on the VMware server ............ 313
 Contents 11

The vSphere interface reports that virtual machine consolidation is

needed ............................................................................... 313
The Enable file recovery from VM backup option does not work if a
volume disk set contains a mixture of disk types .......................... 314
Linux VMs and persistent device naming ......................................... 314
For a VMware virtual machine with Windows dynamic disks, a restore
from incremental backup fails with a Windows restore host and
the hotadd transport mode ...................................................... 315
Simultaneous hotadd backups (from the same VMware backup host)
fail with status 13 .................................................................. 317
Troubleshooting VMware tag usage ................................................ 318
Ensuring that guest customizations can be restored in vCloud Director
.......................................................................................... 320
Troubleshooting vmdk restore to existing VM .................................... 321
Errors for backups of virtual machines on Virtual Volumes (VVols)
.......................................................................................... 323
Issues with the CA certificate during installation of the NetBackup client
on VMware Cloud (VMC) ........................................................ 324

Appendix A NetBackup commands to back up and restore

virtual machines ......................................................... 326
Using NetBackup commands to create a VMware policy ..................... 326
Notes on creating a VMware policy for vCloud Director ................. 330
VMware policy: bpplinfo options and keywords ........................... 331
bpplinclude options for modifying query rules in a VMware policy
.................................................................................... 333
Using the nbdiscover command to search the virtual environment
.......................................................................................... 335
Using the nbrestorevm command to restore virtual machines into
vSphere .............................................................................. 336
Using the nbrestorevm command to restore virtual machines into
vCloud Director .................................................................... 341
Using the nbrestorevm command to restore virtual machines with
Instant Recovery .................................................................. 346
Using the nbrestorevm command to restore virtual machine disks
.......................................................................................... 346
The nbrestorevm -R rename file for restoring virtual machines ........... 346
Logs for troubleshooting the nbrestorevm command .......................... 350
Query format and examples for searching virtual machine backups
.......................................................................................... 350
Allowing other servers or clients to restore virtual machines ................. 353
 Contents 12

Appendix B Configuring services for NFS on Windows .............. 355

About installing and configuring Network File System (NFS) for Granular
Recovery Technology (GRT) ................................................... 355
About configuring services for NFS on Windows 2012 or 2016
(NetBackup for VMware) ........................................................ 356
Enabling Services for Network File System (NFS) on a Windows
2012 or 2016 media server (NetBackup for VMware) ............. 356
Enabling Services for Network File System (NFS) on a Windows
2012 or 2016 restore host (NetBackup for VMware) ............... 360
About configuring services for NFS on Windows 2008 and 2008 R2
(NetBackup for VMware) ........................................................ 363
Enabling Services for Network File System (NFS) on Windows
2008 or 2008 R2 (NetBackup for VMware) ........................... 364
Disabling the Server for NFS (NetBackup for VMware) ....................... 368
Disabling the Client for NFS on the media server (NetBackup for
VMware) ............................................................................. 369
Configuring a UNIX media server and Windows backup or restore host
for Granular Recovery Technology (NetBackup for VMware) .......... 371
Configuring a different network port for NBFSD (NetBackup for
VMware) ............................................................................. 372

Appendix C The Reuse VM selection query results option

.......................................................................................... 373

About the Reuse VM selection query results option ............................ 373

The effect of virtual machine discovery on vCenter ............................ 374

Appendix D Backup of VMware raw devices (RDM) .................... 376

About VMware raw device mapping (RDM) ...................................... 376
Configurations for backing up RDMs ............................................... 377
About alternate client backup of RDMs ............................................ 377
Requirements for alternate client backup of RDMs ............................ 378
Configuring alternate client backup of RDMs .................................... 378

Index .................................................................................................................. 381

 Chapter 1
Introduction
This chapter includes the following topics:

■ About NetBackup for VMware

■ About the virtual machine backups that include database data

■ About the NetBackup appliance as a VMware backup host

■ NetBackup for VMware components

■ Appliance as backup host: component overview

■ Media servers as backup or discovery hosts

■ Overview of the VMware backup process

■ NetBackup for VMware terminology

About NetBackup for VMware

NetBackup for VMware provides backup and restore of the VMware virtual machines
that run on VMware ESX servers. NetBackup for VMware takes advantage of
VMware vStorage APIs for data protection. The backup process is off-loaded from
the ESX server to a VMware backup host.
NetBackup for VMware does the following:
■ Performs off-host backup of virtual machines (NetBackup client software is not
required on the virtual machine). Off-host backup reduces the backup processing
load on the VMware host.
■ Increases the backup speed as compared to standard file-order backup methods,
if the virtual machine is heavily populated with small files.
 Introduction 14
About the virtual machine backups that include database data

■ Automatically creates quiesced snapshots using VSS (Windows only). Creates

quiesced snapshots on Linux if snapshot quiesce is enabled in the Linux guest
OS.
■ Uses snapshot technology to keep virtual machines 100% available to users.
■ Supports VMware vSphere and vCloud Director.
■ Performs full backups and incremental backups, including block-level
incrementals.
■ Backs up the full virtual machine.
■ Backs up the virtual machines even when they are turned off.
■ Can restore selected files from the backup.

About the virtual machine backups that include

database data
When NetBackup backs up a VMware virtual machine, database data in the virtual
machine is backed up with the rest of the virtual machine. NetBackup allows the
recovery of individual database files from the backup. This feature supports Microsoft
Exchange Server, SQL Server, and SharePoint Server.
Note the following:
■ To enable restore of individual database files, a NetBackup Windows client must
be installed in the virtual machine during the backup.
■ NetBackup uses Windows Volume Shadow Copy Service (VSS) to quiesce the
database before it creates a snapshot of the virtual machine.
■ A full backup is performed of the database data with each backup job, even if
the policy schedule is incremental.
See “Application protection options (VMware)” on page 79.

About the NetBackup appliance as a VMware

backup host
The NetBackup appliance uses the VMware policy type to back up VMware virtual
machines.
The following topics contain notes on the appliance as the backup host:
■ For an overview of the appliance as backup host in a virtual environment:
See “Appliance as backup host: component overview” on page 18.
 Introduction 15
NetBackup for VMware components

■ For configuration tasks:

See “Overview of VMware tasks” on page 24.
See “Overview of NetBackup tasks” on page 25.
■ For a list of requirements and limitations:
See “Notes on the NetBackup appliance as a VMware backup host” on page 35.
■ For log files:
See “NetBackup logging for VMware” on page 291.

NetBackup for VMware components

Table 1-1 describes the components that NetBackup for VMware uses.

Table 1-1 Components of NetBackup for VMware

Component Description

Backup host NetBackup for VMware uses a special host that is called a VMware backup host (formerly called
the VMware backup proxy server). The backup host is a NetBackup client that performs backups
on behalf of the virtual machines. The backup host must have access to the datastores of the
virtual machines.

The backup host is the only host on which NetBackup client software is installed. No NetBackup
client software is required on the VMware virtual machines.

Note that the backup host is referred to as the recovery host when it performs a restore.
The backup host can be configured in any of the following ways:

■ As a NetBackup client (Windows or Linux) with a connection to separate master and media
servers. The master servers and media servers can be Windows, UNIX, or Linux.
■ As a NetBackup client that is installed on the media server (see Backup media server).
■ The NetBackup client and master and media server can all reside on the same host (Windows
or Linux).
For a list of supported platforms for the backup host, see the NetBackup Enterprise Server
and Server OS Software Compatibility List available from the following location:
NetBackup Master Compatibility List

Discovery host Used for the automatic selection of virtual machines for backup. This host discovers virtual
machines and filters them by means of the selection rules in the policy Query Builder. The
resulting list determines which virtual machines are backed up.

The discovery host can be on any platform that NetBackup supports for master or media servers.
It can also be the same host as the backup host.

You specify this host on the policy Clients tab: Click Select automatically through VMware
Intelligent Policy query, then NetBackup host to perform automatic virtual machine
selection.
 Introduction 16
NetBackup for VMware components

Table 1-1 Components of NetBackup for VMware (continued)

Component Description

Backup media server A media server that can operate as the backup host.

See “Media servers as backup or discovery hosts” on page 18.

NetBackup client Installed on the backup host.

NetBackup master Manages the backups of virtual machines, by means of the NetBackup client that is installed
server on the backup host.

NetBackup media Performs the backups to storage on behalf of the NetBackup client.
server

Virtual machine Virtual machines provide complete guest operating systems on virtualized hardware. In a
NetBackup policy, a virtual machine is configured as a NetBackup client, even though NetBackup
client software is not installed on the virtual machine.

ESX server The VMware ESX server presents a virtualized hardware environment to multiple virtual
machines; each virtual machine runs an independent operating system. Users can run
applications in the virtualized OS as if the OS was installed in its own physical computer.

vCenter Server The VMware vCenter Server (or VirtualCenter server) coordinates multiple ESX servers and
workloads. It can migrate virtual machines from one ESX server to another. It also provides the
ability to back up the virtual machines that are turned off.

The vCenter Server is optional in the NetBackup for VMware environment.

Figure 1-1 shows a NetBackup for VMware environment on a local network. The
backup host accesses the VMware datastore through the ESX servers.
 Introduction 17
NetBackup for VMware components

Figure 1-1 NetBackup for VMware: components on local network

NetBackup master server

UNIX, Linux, or Windows
LAN / WAN

vCenter
server
NetBackup media server
UNIX, Linux, or Windows

ESX
Backup host servers
NetBackup Windows or Linux
client. Can be combined with a
master or media server. VMware
Virtual machines datastore
VM VM VM

VM VM VM

Figure 1-2 shows a NetBackup for VMware environment on a SAN. The backup
host accesses the VMware datastore directly over the SAN.

Figure 1-2 NetBackup for VMware: components on SAN

NetBackup master server

UNIX, Linux, or Windows
LAN / WAN

vCenter
server
NetBackup media server
UNIX, Linux, or Windows

SAN
ESX
Backup host VMware datastore servers
NetBackup Windows or Linux (Must be available to
client. Can be combined with a VMware backup host)
master or media server. Virtual machines
VM VM VM

VM VM VM
 Introduction 18
Appliance as backup host: component overview

Appliance as backup host: component overview

As Figure 1-3 shows, the appliance can operate as the VMware backup host. A
separate Windows backup host is not required.
The appliance as backup host can also run the NetBackup media server and master
server.

Figure 1-3 NetBackup for VMware with appliance as backup host

LAN / WAN
Media
server
storage

Appliance as backup NetBackup NetBackup GUI vCenter

host uses NetBackup master server workstation server
client software to (if not on
back up VMs. appliance)
Client software and ESX
media server run on
appliance. Master servers
server can also run VMware
on appliance.
datastore
Virtual machines
Note: Windows backup
host is not required. VM VM VM

VM VM VM

The NetBackup environment can also be on a SAN:

See Figure 1-2 on page 17.
Further information is available on the appliance as backup host:
See “Notes on the NetBackup appliance as a VMware backup host” on page 35.

Media servers as backup or discovery hosts

NetBackup for VMware uses a special host that is called a VMware backup host.
The backup host is a NetBackup client that performs off-host backups of the virtual
machines. The backup host must have access to the datastores of the virtual
machines. The backup host reads the data from the datastore and sends it over
the network to the media server. The media server backs up the data to storage.
NetBackup also uses a discovery host. For the policies that automatically select
virtual machines, the discovery host filters virtual machines according to the rules
in the policy Query Builder. The discovery host returns a list of virtual machines to
be selected for backup.
 Introduction 19
Media servers as backup or discovery hosts

NetBackup can use media servers as backup hosts and as discovery hosts.
Media servers acting as backup or discovery hosts can provide the following
advantages:
■ Host redundancy: If one media server goes down, another media server takes
over.
■ Faster backup: The media server can read the data from the datastore and send
the data straight to the storage device. Without media server access to storage
devices, an ordinary backup host must send the backup data over the local
network to the media server.
Figure 1-4 shows a group of media servers that can also act as backup or discovery
hosts. The media servers can discover virtual machines for automatic selection,
and perform off-host backups and send the backup data directly to storage.

Figure 1-4 Backup media servers

NetBackup master server

UNIX, Linux, or Windows
LAN / WAN

vCenter
storage
server

ESX
Media servers servers
Each has a NetBackup client
Vmware
Virtual machines datastore
VM VM VM

VM VM VM

You can combine the flexibility of backup media servers with a standard feature of
NetBackup: storage unit groups. Create a storage unit group that contains the
storage units that your media servers can access. Any of the media servers can
operate as a backup host.
Note the following requirements:
■ To configure media servers as backup hosts: Set the policy VMware backup
host field on the VMware tab to Backup Media Server.
See “VMware backup host” on page 74.
 Introduction 20
Overview of the VMware backup process

■ To configure media servers as discovery hosts: Set the policy NetBackup host
to perform automatic virtual machine selection field on the Clients tab to
Backup Media Server.
See “Options for selecting VMware virtual machines” on page 109.

Overview of the VMware backup process

The following table describes the phases in the NetBackup backup process.

Table 1-2 NetBackup backup process

Phase Description

Phase 1 The NetBackup master server initiates the backup.

Phase 2 The NetBackup client on the VMware backup host initiates a VMware
snapshot on the virtual machine.

Phase 3 Windows: VSS synchronizes the file system on the virtual machine.

Linux: If snapshot quiesce is enabled in the Linux guest OS, the file system
is synchronized on the virtual machine. (Contact your operating system
vendor and VMware for additional information on how to enable snapshot
quiesce.)

Phase 4 The VMware server creates a snapshot on the virtual disk datastore.

Phase 5 The NetBackup client reads the snapshot from the datastores and writes
the data to the NetBackup storage unit.

NetBackup for VMware terminology

Table 1-3 lists the terminology that is used in NetBackup for VMware.
For further explanations of VMware terminology, refer to your VMware
documentation.
 Introduction 21
NetBackup for VMware terminology

Table 1-3 NetBackup for VMware terms

Term Definition

backup host The backup host is a NetBackup client that performs backups on
behalf of the virtual machines. (This host was formerly known as
the VMware backup proxy server.) The backup host is the only host
on which NetBackup client software is installed.

As an option, the backup host can also be configured as a

NetBackup master server or media server.

The backup host is referred to as the recovery host when it performs

a restore.

See “NetBackup for VMware components” on page 15.

backup media server A media server that operates as a backup host.

datastore In NetBackup for VMware, the datastore is a disk that contains the
virtual machines files.

datastore cluster A collection of datastores that can be managed as a single unit.

VMware Storage DRS manages the storage resources of the
cluster.

discovery host Discovers the virtual machines, filters them by the rules in the Query
Builder, and returns a list of virtual machines to be selected for
backup. The discovery host is used only for automatic selection of
virtual machines.

Can be the same host as the VMware backup host.

See “NetBackup for VMware components” on page 15.

guest OS An operating system that runs on a virtual machine.

hotadd The backup or restore transport mode when the backup host is
installed in a virtual machine.

hypervisor A software virtualization layer that lets different operating systems

run simultaneously on the same physical computer. The VMware
hypervisor is vSphere, which runs on the ESX server host.
 Introduction 22
NetBackup for VMware terminology

Table 1-3 NetBackup for VMware terms (continued)

Term Definition

independent disk A virtual disk that cannot be captured with a snapshot. An

independent disk can be configured as either persistent or
temporary.
Note: NetBackup for VMware cannot back up the data on an
independent disk. The backup succeeds but the backup image
contains no data for the independent disk. If you restore the disk
from the backup, the independent disk is restored to its original
allocation size but contains no data. For example, the Windows
swap file can reside on an independent disk, and is consequently
not included in the virtual machine backup.

To back up the data on an independent disk, install a NetBackup

client on the virtual machine.

See “NetBackup for VMware: notes and restrictions” on page 27.

query The combination of rules in the policy's Query Builder, by which

NetBackup selects virtual machines for backup. A query consists
of one or more rules.

Query Builder For creating filtering rules for automatic selection of virtual machines
for backup. The Query Builder is on the Clients tab of the
NetBackup policy.

query rule A single statement in a query, by which NetBackup selects virtual

machines for backup.

An example of a query rule is: Displayname Contains

"finance"

Raw device mapping Allows a virtual machine to directly access physical disks (on Fibre
(RDM) Channel or iSCSI). RDM allows the OS and applications in the
virtual machine to have pass-thru access to raw devices.

NetBackup for VMware cannot back up an RDM disk. NetBackup

client software must be installed in the virtual machine.

See “About VMware raw device mapping (RDM)” on page 376.

recovery host See backup host.

transport mode Determines how the snapshot data travels between the VMware
datastore and the backup host.

See “Transport modes options (VMware)” on page 78.

 Introduction 23
NetBackup for VMware terminology

Table 1-3 NetBackup for VMware terms (continued)

Term Definition

virtual machine An execution environment that the hypervisor creates for the
computer hardware. The resulting virtualized environment allows
different operating systems to run simultaneously on the same
physical computer.

virtual network A logical network that allows the exchange of data between virtual
machines. A virtual network uses a virtual switch (VMware vSwitch).
A virtual network can be connected to a physical network.

vmdk file In a VMware ESX server, one or more vmdk files make up the disk
image or virtual drive in a virtual machine. The .vmdk files contain
the operating system, applications, and data in the virtual machine.

VMware Tools Installed inside each VMware virtual machine. Enhances the virtual
machine performance and adds backup-related functionality.

vmx datastore Sometimes called the vmx directory or configuration datastore.

Contains the configuration files that describe the virtual machine,
such as vmx files. During a backup of a virtual machine snapshot,
vmdk writes are also cached on this datastore. Note that a separate
vmx datastore is not a VMware requirement.

vStorage VMware vStorage APIs enable data protection features for more
efficient use of storage capacity. NetBackup can use vStorage to
back up the latest vSphere environments as well as to back up
earlier VMware environments.
 Chapter 2
Required tasks: overview
This chapter includes the following topics:

■ Overview of VMware tasks

■ Overview of NetBackup tasks

Overview of VMware tasks

The VMware components including ESX servers and virtual machines must be set
up before you configure NetBackup.

Table 2-1 VMware tasks

Sequence Tasks

Phase 1 Ensure that the hardware and the SAN are configured properly. The VMware
datastore where the target virtual machine files exist must be accessible to
the VMware backup host.

A SAN connection between the backup host and the datastore is optional
if you use the NBD transfer type or NBDSSL transfer type.

To use the SAN transport type, set up the datastore on Fibre Channel or
iSCSI. In this configuration, the VMware backup host must be able to access
the datastore over the SAN.

Phase 2 Install the VMware ESX server and virtual machines.

Phase 3 Install VMware Tools on the virtual machines that you plan to back up.

VMware requires that ESX server names resolve to an IP address. It is

highly recommended that you use DNS for the naming resolution among
VMware servers.

Phase 4 Optional: install a vCenter (or VirtualCenter) server.

 Required tasks: overview 25
Overview of NetBackup tasks

Overview of NetBackup tasks

Table 2-2 lists the NetBackup configuration tasks for VMware. These tasks are
described in other NetBackup topics and guides, as indicated.

Table 2-2 NetBackup tasks

Sequence Tasks

Phase 1 Install the NetBackup master server and media server.

See the NetBackup Installation Guide.

It is recommended that the NetBackup media server and the VMware backup
host be installed on the same host.

Phase 2 Install the NetBackup Enterprise Client license on the master server, and
install NetBackup client software on the VMware backup host.

NetBackup for VMware requires an Enterprise Client license for each ESX
Server. To protect an application or database, note: an additional license
for the appropriate NetBackup package is needed for each ESX server that
hosts the application or database.

Phase 3 Add the VMware backup host to your NetBackup configuration.

See “Adding the VMware backup host to NetBackup” on page 44.

Note: This step is not required on the appliance: the backup host is already
installed on the appliance.

Phase 4 Enter NetBackup access credentials as needed: for VMware vCenter, for
vCloud Director, and for individual ESX servers.

See “Adding NetBackup credentials for VMware” on page 45.

Note: The NetBackup master server must have network access to the
VMware servers that NetBackup has credentials for.

Phase 5 Create a NetBackup policy for VMware.

See “Configuring a VMware policy from the Policies utility” on page 68.

Phase 6 Perform a backup.

See “Backing up virtual machines” on page 151.

 Chapter 3
Notes and prerequisites
This chapter includes the following topics:

■ NetBackup for VMware prerequisites

■ NetBackup for VMware: notes and restrictions

■ Notes on VMware Virtual Volumes (VVols)

■ NetBackup IPv6 parameter required for backups in VMware IPv6 environments

■ NetBackup for VMware: notes on Linux virtual machines

■ Notes on the NetBackup appliance as a VMware backup host

■ NetBackup for VMware support for SAN multi-pathing

■ NetBackup for VMware support for fault tolerant VMs

■ NetBackup character restrictions for the Primary VM identifier

■ In the policy Query Builder, display names, resource pool names, and vApp
names are case-sensitive

■ Notes on the hotadd transport mode

■ Notes and limitations for tag usage in VMware Intelligent Policy queries

■ Notes and limitations for the backup and restore of VMware tag associations

NetBackup for VMware prerequisites

The following prerequisites apply to NetBackup for VMware:
■ The VMware components including ESX servers and virtual machines must be
set up before you configure NetBackup.
See “Overview of VMware tasks” on page 24.
 Notes and prerequisites 27
NetBackup for VMware: notes and restrictions

■ The NetBackup Java Runtime Environment (JRE) must be installed on the

discovery host and the recovery host. Install the NetBackup Remote
Administration Console to install NetBackup Java.

■ The VMware backup host must run on Windows or Linux. The supported
hardware types are the same as for any NetBackup client. For the supported
platforms for the backup host, see the NetBackup Enterprise Server and Server
OS Software Compatibility List available from the following location:
NetBackup Master Compatibility List
■ To use the hotadd transfer type for backup or restore, the VMware backup or
restore host is installed in a virtual machine.
See “Notes on the hotadd transport mode” on page 40.

NetBackup for VMware: notes and restrictions

Note the following about NetBackup for VMware:
■ For information on supported VMware versions and on supported platforms for
the backup host, see the NetBackup Enterprise Server and Server OS Software
Compatibility List available from the following location:
NetBackup Master Compatibility List
For additional support information on NetBackup for VMware, see Support for
NetBackup in virtual environments:
http://www.veritas.com/docs/000006177
■ NetBackup for VMware does not support the Retain snapshot for Instant
Recovery or SLP ... option on the policy Attributes tab of the Administration
Console.
NetBackup supports these features as follows:
■ For Instant Recovery, use the nbrestorevm command.
See “About Instant Recovery for VMware” on page 237.
■ For SLP management of snapshots, use Replication Director.
For more information, see the NetBackup Replication Director Solutions
Guide.

■ NetBackup allows up to 31 snapshots per virtual machine. If the virtual machine

has more than 31 snapshots, the backup may fail with status 13. Messages
similar to the following appear in the NetBackup job details:

10/18/2012 4:56:59 PM - Critical bpbrm(pid=4604)from client

Umesh_w2k3_hypervm33: FTL - vSphere_freeze: Unable to remove
existing snapshot, too many existing snapshots (91).
10/18/2012 4:56:59 PM - Critical bpbrm(pid=4604) from client
 Notes and prerequisites 28
NetBackup for VMware: notes and restrictions

Umesh_w2k3_hypervm33: FTL - VMware_freeze: VIXAPI freeze

(VMware snapshot) failed with 26: SYM_VMC_REMOVE_SNAPSHOT_FAILED

As a reminder to consolidate or delete snapshots, the NetBackup detailed status

provides the following message when the number of snapshots exceeds 15:

Umesh_w2k3_hypervm33: WRN - vSphere_freeze: VM has 16 existing

snapshots. Snapshots may start failing if this number exceeds 32

To back up a virtual machine that has more than 31 snapshots, consolidate or

delete the existing snapshots. Then rerun the backup.

Note: Replication Director backups allow more than 31 snapshots per virtual
machine.

■ If the number of snapshot delta files for a VM's vmdk exceeds 32, snapshot
creation fails. A message similar to the following appears in the NetBackup
detailed status:

Umesh_w2k3_hypervm33: FTL - vSphere_freeze: Unable to

proceed with snapshot creation, too many existing delta files(50).

As a reminder to consolidate or delete snapshots, the NetBackup detailed status

provides the following message when a vmdk's delta files exceed 16:

Umesh_23k3_hypervm33: WRN - vSphere_freeze: VM has 17 existing

delta files for vmdk Umesh_23k3_hypervm33.vmdk. Snapshots may start
failing if this number exceeds 31

Delta files can accumulate if the VM's snapshots are not deleted or consolidated.
Consolidate or delete the existing snapshots, then rerun the backup.
■ NetBackup for VMware cannot back up the data on an independent disk, because
an independent disk cannot be captured with a snapshot. The backup succeeds
but the backup image contains no data for the independent disk.
To back up the data on an independent disk, install a NetBackup client on the
virtual machine. You can configure NetBackup to back up the virtual machine
and any independent disks as if the client was installed on a physical host. You
can restore the virtual machine and then restore the independent disk as a
separate job.
■ NetBackup for VMware does not back up standard iSCSI LUNs that are
connected to the virtual machine. If the virtual machine has an iSCSI LUN, the
backup succeeds but the drive that represents the LUN is not backed up.
Note: NetBackup for VMware supports datastores over iSCSI.
 Notes and prerequisites 29
NetBackup for VMware: notes and restrictions

■ Several notes and limitations apply to Linux virtual machines.

See “NetBackup for VMware: notes on Linux virtual machines” on page 34.
■ If a Windows virtual machine includes Veritas Storage Foundation volumes, the
Enable file recovery from VM backup option is not supported.
See “Storage Foundation Volume Manager volumes in the virtual machine”
on page 100.
■ During an incremental backup with the Enable file recovery from VM backup
option (with or without block-level incremental backup), any files that had been
moved or renamed are not backed up. Those files are not available when you
browse to restore individual files from that incremental backup.

Note: When you restore the entire VM from a block-level incremental backup,
the file metadata is updated and the moved or renamed files reflect the updated
metadata.

■ VMware virtual machine templates are for cloning virtual machines: They cannot
be turned on and used as functioning VMs. As a result, VMware has imposed
the following restrictions on backup and restore of virtual machine templates:
■ A virtual machine template cannot be captured in a snapshot. NetBackup
backs up the template to the designated storage unit.
■ Block level incremental backup (BLIB) cannot be used when backing up a
virtual machine template.

Note: As a result of this restriction, NetBackup Accelerator cannot be used

to back up VMware virtual machine templates.

■ A virtual machine template cannot be backed up over a SAN. You must

configure the NetBackup policy to use a local network transfer type, such
as nbd. Likewise, the restore of a template must be conducted over a local
network.

Note: You can use the Query Builder in the NetBackup policy to create rules
for automatic selection of virtual machine templates.

■ NetBackup supports non-ASCII characters in virtual machine objects, as follows:

■ The NetBackup master server, the backup host, and the restore host must
run NetBackup 8.1 or later. The backup host and the restore host may be
the same computer.
 Notes and prerequisites 30
NetBackup for VMware: notes and restrictions

■ Windows operating systems on the NetBackup master server, the backup

host, and the restore host do not assume non-ASCII characters in VM display
name if the system locales of Windows hosts are not set to UTF-8. Even if
the system locale is not set to UTF-8, a backup of the VMware virtual machine
whose display name contains non-ASCII character may work. However, a
restore from the VMware virtual machine backup that has non-ASCII
characters in its display name needs a Linux or UNIX restore host which
uses UTF-8 character encoding.
■ The UNIX and Linux operating systems on the NetBackup master server,
the backup host, and the restore host must use UTF-8 character encoding.
The backup host and the restore host may be the same computer.
■ The name that NetBackup uses to select a VM for backups cannot contain
non-ASCII characters. The Primary VM identifier field in the backup policy
identifies the name type that NetBackup uses to select VMs. For example,
if you specify VM display name as the Primary VM identifier, the display
name of each VM that you back up cannot contain non-ASCII characters.
See “Primary VM identifier options (VMware)” on page 76.
VM BIOS UUID and VM instance UUID names never contain non-ASCII
characters.
■ For security purposes, VM names have some restrictions.
See “NetBackup character restrictions for the Primary VM identifier”
on page 37.
Examples of objects are file and folder names, annotations, floppy image name,
parallel port or serial port file name, CD-ROM ISO name, and so on.
■ For dual-boot virtual machines, NetBackup does not support the following policy
options:
■ Enable file recovery from VM backup
■ Exclude deleted blocks
■ Exclude swap and paging files
■ Exclude boot disk
■ Exclude all data disks

■ To support the following exclude disk options, the backup host must be at release
level 8.1 or later:
■ Perform custom attribute based exclusion
Custom Attribute based disk exclusion requires that you enter in NetBackup
the credentials for the vCenter server or servers that host the VMs. ESXi
server credentials are not sufficient.
 Notes and prerequisites 31
NetBackup for VMware: notes and restrictions

See “Adding NetBackup credentials for VMware” on page 45.

■ Specific disk(s) to be excluded

■ NetBackup does not support the following exclude disks options for Replication
Director backups:
■ Exclude boot disk
■ Exclude all data disks
■ Perform custom attribute based exclusion
■ Specific disk(s) to be excluded

■ If you remove a disk from exclusion, the individual files that were last modified
before the most recent backup cannot be restored. To restore those files, restore
the entire virtual disk or the virtual machine. After the next full backup, those
files are available to restore individually.
■ NetBackup fails the backups of virtual machines that are empty. A VM may be
empty because all of its disks are empty or because disk exclusion excludes all
disks.
■ To back up a virtual machine while Storage vMotion migrates its files, NetBackup
must conduct the backup through the vCenter server.
See “Conflict between NetBackup and VMware Storage vMotion with vSphere
5.0 or later” on page 307.
■ Several notes and limitations apply to the automatic selection of virtual machines
for backup (Virtual Machine Intelligent Policy).
See “NetBackup requirements for automatic virtual machine selection”
on page 108.
■ For hotadd backup or restore: The virtual machine to back up (or restore) and
the virtual machine that contains the hotadd backup host must reside in the
same VMware data center.
See “Notes on the hotadd transport mode” on page 40.
■ NetBackup for VMware supports Windows NTFS file encryption and compression,
for backup and restore. However, NetBackup for VMware does not support
NetBackup’s compression or encryption options (in the NetBackup policy
attributes).
For Windows, UNIX, or Linux guest operating systems: NetBackup for VMware
does not support any type of compression or encryption, whether they are set
in NetBackup or in the guest OS (such as BitLocker).
 Notes and prerequisites 32
NetBackup for VMware: notes and restrictions

Note: The compressed Windows NTFS files are backed up and restored as
compressed files.

■ Storage optimization cannot be used if a snapshot exists on the virtual machine

when VMware Changed Block Tracking is turned on.
See “About reducing the size of VMware backups” on page 147.
■ If the policy's Primary VM identifier option is set to VM hostname, note:
NetBackup cannot select a VMware virtual machine for backup if it cannot obtain
an IP address for the virtual machine.
■ For Exchange, SharePoint, and SQL applications, NetBackup for VMware
supports Granular Recovery Technology (GRT) restores from full backups only.
■ See “Restore notes and restrictions” on page 173.
■ If a policy is changed from manual selection to Intelligent policy (or vice versa),
the next backup of the VM is a regular full backup, even if a backup already
exists for that VM.
For example:
■ In a new VMware policy, the Select manually option on the Clients tab is
used to select a VM. The first backup from the policy runs.
■ In the policy, an Intelligent Policy query is used to select the same VM, and
the VM is backed up a second time. Because of the switch from manual
selection to query-based selection, this second backup is a regular full
backup. Note that for a policy that uses Enable block-level incremental
backup (BLIB) or BLIB plus Accelerator, the backup processing is not limited
to changed blocks only.
For the second backup, the Detailed status log includes a message similar
to the following:

There is no complete backup image match, a regular full backup

will be performed.

The same backup behavior occurs if the policy’s VM selection is switched from
query-based to manual selection: The second backup is a regular full backup.
■ For the virtual machines that vSphere 6.5 hosts, NetBackup fails the backup if
a snapshot exists while NetBackup tries to enable VMware Change Block
Tracking. Because the VMware Change Block Tracking API behavior changed
beginning in vSphere 6.5, NetBackup fails the backup.
NetBackup enables CBT on a VM if Block Level Incremental Backups is
enabled in the backup policy and CBT is not enabled already on the VM.
 Notes and prerequisites 33
Notes on VMware Virtual Volumes (VVols)

■ In vSphere, if your virtual environment has IPv6 addresses, use only fully qualified
domain names (FQDNs) that are mapped to IPv6 addresses on the DNS server.
■ The Test query operation fails for a VMware policy in the following condition:
There is no direct connectivity between the NetBackup master server and the
ESX server (for example, ESX server in NAT environment) where a virtual
machine is to be backed up and the NetBackup host to perform automatic
virtual machine selection option is set to Backup Media Server (for example,
NAT media server).

Notes on VMware Virtual Volumes (VVols)

NetBackup supports backup and restore of the virtual machines that are configured
on Virtual Volumes (VVols).
■ Ensure that you have the required snapshot license from the array vendor.
■ Consult the storage array documentation from the vendor for space requirements.
Configuring backup and restore of virtual machines with VVols is the same as for
virtual machines without VVols, with these exceptions:
■ To restore a virtual machine with the hotadd transport mode: VMware requires
that the virtual machine and the restore host virtual machine reside on the same
VVol datastore. Otherwise, the restore must use a different transport mode (not
hotadd).

■ For a restore to standard (non-VVol) datastores, the NetBackup job creates a

vSphere snapshot of the virtual machine while NetBackup restores the data.
Note: For a restore to a VVol datastore, NetBackup restores the data to the
virtual machine without creating a vSphere snapshot.

NetBackup IPv6 parameter required for backups

in VMware IPv6 environments
For backups and restores of VMware virtual machines in an IPv6 environment, you
must configure the IPv6 support on the following NetBackup hosts:
■ The master server
■ The backup host
The Both IPv4 and IPv6 option of the IP Address Family Support host property
configures IPv6 support. The IP Address Family Support host property is on the
Network Settings host properties dialog box.
 Notes and prerequisites 34
NetBackup for VMware: notes on Linux virtual machines

If the NetBackup master server and the backup host are the same host, configure
the support on that host only. If the NetBackup master server and the backup host
are separate hosts, configure the support on each host.
For more information about the IP Address Family Support host property, see
the NetBackup Administrator's Guide, Volume I:
http://www.veritas.com/docs/DOC5332

NetBackup for VMware: notes on Linux virtual

machines
The following notes apply to virtual machines with Linux guest operating systems:
■ NetBackup cannot exclude unused or deleted blocks from the backup if the
virtual machine is configured with software RAID volumes. The policy's Exclude
deleted blocks option is not supported.
■ If snapshot quiesce is enabled in the Linux guest OS, NetBackup for VMware
synchronizes the guest OS file system before it takes the snapshot. Contact
your operating system vendor and VMware for additional information on enabling
snapshot quiesce. Note: If snapshot quiesce is not enabled, NetBackup cannot
guarantee that data in the file system is in a consistent state when the snapshot
occurs.
■ Unmounted LVM2 volumes must start with /dev
If the path of an unmounted LVM2 volume does not start with /dev, the backup
of the virtual machine fails. Note: The path of the volume is set with the dir
parameter on the LVM volume configuration file. An example of this configuration
file is /etc/lvm/lvm.conf.
■ For Linux files or directories, NetBackup for VMware has the same path name
restriction as NetBackup on a Linux physical host. Files or directories with path
names longer than 1023 characters cannot be individually backed up or restored.
Such files can be restored when you restore the entire virtual machine from a
full virtual machine backup.
■ The Linux ext4 file system includes a persistent pre-allocation feature, to
guarantee disk space for files without padding the allocated space with zeros.
When NetBackup restores a pre-allocated file (to any supported ext file system),
the file loses its preallocation and is restored as a sparse file. The restored file
is only as large as the last byte that was written to the original file. Subsequent
writes to the restored file may be non-contiguous.

Note: The restored file contains all of its original data.

 Notes and prerequisites 35
Notes on the NetBackup appliance as a VMware backup host

■ The NetBackup policy’s Enable file recovery from VM backup option is not
supported for the disks inside a Linux guest OS that are configured as follows:
■ The disks are divided into logical volumes by means of the Linux Logical
Volume Manager (LVM), and
■ The LVM volumes were created with thin-provisioning.

■ See “Restore notes and restrictions on Linux” on page 177.

Notes on the NetBackup appliance as a VMware

backup host
Note the following requirements and limitations for the appliance as the backup
host:
■ The appliance must be version 2.5 or later.
■ You must use the VMware policy type.
■ The appliance supports iSCSI connections. Refer to the NetBackup Appliance
iSCSI Guide for more information.

NetBackup for VMware support for SAN

multi-pathing
NetBackup for VMware on Windows supports dynamic multi-pathing (DMP) between
the vSphere ESXi storage and the NetBackup for VMware agent (Windows backup
host). Dynamic multi-pathing can provide SAN I/O high-availability and improved
backup throughput.
For NetBackup for VMware on a Linux backup host, the following items describe
the support for dynamic multi-pathing in a SAN environment:

NetBackup The NetBackup appliance (beginning with the 2.6.0.2 release) supports
appliance SAN dynamic multi-pathing for VMware backups. I/O is redirected
through the volume manager dynamic multi-pathing node.

For more information, see the NetBackup Appliance Administrator's

Guide for version 2.6.0.2 and later:

http://www.veritas.com/docs/DOC2792
 Notes and prerequisites 36
NetBackup for VMware support for SAN multi-pathing

Non-appliance NetBackup supports backups and restores for multi-pathing under the
Linux host following conditions:

■ The virtual disk SAN transport whitelist specifies the device node
paths that you want to use for multipathing. The virtual disk
vixDiskLib.transport.san.whitelist API function defines
the whitelist. For NetBackup purposes, Veritas recommends that
you include the DMP nodes in the whitelist; the following is an
example:

vixDiskLib.transport.san.whitelist = /dev/vx/dmp/
hitachi_usp-v0_00a0,/dev/vx/dmp/hitachi_usp-v0_00
a0s1,/dev/vx/dmp/hitachi_usp-v0_00a0s2,/dev/vx/dm
p/hitachi_usp-v0_00a0s3,/dev/vx/dmp/hitachi_usp-v
0_00a0s4,/dev/vx/dmp/hitachi_usp-v0_00a0s5

■ The virtual disk SAN transport blacklist specifies the device node
paths you want to exclude from multipathing. The virtual disk
vixDiskLib.transport.san.blacklist API function defines
the blacklist. For NetBackup purposes, Veritas recommends that
you set the blacklist to all; the following is an example:
vixDiskLib.transport.san.blacklist = all

Specify the whitelist and the blacklist in the following file on the backup
host:

/usr/openv/lib/shared/vddk/lib64/vixDiskLib.ini

For more information about the whitelist and the blacklist, see the
appropriate VMware documentation. For example, for Virtual Disk
Development Kit 6.0 the whitelist and the blacklist function descriptions
are in the VMware Virtual Disk Programming Guide.

Note: This support or limitation does not affect NetBackup support for VMware’s
Native Multipathing, which is multi-pathing between the ESXi host and storage.
This support or limitation does not affect NetBackup for VMware on Windows backup
hosts. The Windows operating system has integrated multi-pathing support. The
following Microsoft guide contains more information:
http://microsoft.com/mpio
 Notes and prerequisites 37
NetBackup for VMware support for fault tolerant VMs

NetBackup for VMware support for fault tolerant

VMs
NetBackup supports backing up and restoring primary fault tolerant virtual machines
on vSphere 6.0 and later. (NetBackup does not backup or restore the secondary
VMs.) If you choose to overwrite the VM during the restore, NetBackup deletes
both the primary and the secondary VMs during the restore process. NetBackup
then restores the primary VM.
Fault tolerance is not enabled on the restored VM. After the restore has completed,
you can turn on fault tolerance.
For information about how to turn on fault tolerance, see the VMware documentation
for your version of vSphere.

NetBackup character restrictions for the Primary

VM identifier
For VMware virtual machines in a NetBackup policy, certain characters are not
allowed in their names. The backup policy Primary VM identifier field identifies
the name type that NetBackup uses to select VMs.
See “Primary VM identifier options (VMware)” on page 76.
If the name contains disallowed characters, backups or restores may fail.
The following table describes the characters and strings that NetBackup does not
allow for backup or restore in the Primary VM identifier, except where noted.

Table 3-1 Disallowed characters and strings in the primary VM identifier

Character/string Description Notes

" Quotation mark, unicode x22.

$ Dollar sign, unicode x24.

' Apostrophe, unicode x27.

* Asterisk, unicode x2A.

, Comma, unicode x2C.

: Colon, unicode x3A.

; Semi-colon, unicode x3B.

 Notes and prerequisites 38
NetBackup character restrictions for the Primary VM identifier

Table 3-1 Disallowed characters and strings in the primary VM identifier

(continued)

Character/string Description Notes

? Question mark, unicode x3F.

@ At sign, unicode x40.

| Vertical line, unicode x7C.

` Grave accent, unicode x60.

´ Acute accent, unicode xB4.

% Percent sign, unicode x25. Disallowed in the virtual machine

display name for application-aware
VMware backups for SharePoint,
allowed elsewhere.

In VIP query and VM search

results, NetBackup converts % in
the display name of Included VMs
to the literal string %25. When you
specify a display name in a query,
replace the % character with %25.

& Ampersand sign, unicode x26.

< Less than sign, unicode x3C.

> Greater than sign, unicode x3E.

- Hyphen-minus, unicode x2D. Disallowed in the first position only.

/ Solidus, unicode x2F. Disallowed in VM display name,

allowed in other objects.

\ Reverse solidus, unicode x5C. Disallowed in VM display name,

allowed in other objects.

. Full stop (period), unicode x2E. Disallowed in VM display name

when in the last position, allowed
in other objects.
 Notes and prerequisites 39
In the policy Query Builder, display names, resource pool names, and vApp names are case-sensitive

Table 3-1 Disallowed characters and strings in the primary VM identifier

(continued)

Character/string Description Notes

Space, unicode x20. In VIP query and VM search

results, NetBackup converts a
space character in the display
name of Included VMs to the literal
string %20. When you specify a
display name in a query, replace
the space character with %20.

Unicode characters Disallowed when the backup policy

greater than x7F specifies the VM display name as
(non-ASCII) the primary VM identifier.

%2f Literal string, not a unicode Disallowed in VM display name,

character definition. allowed in other objects.

%5c Literal string, not a unicode Disallowed in VM display name,

character definition. allowed in other objects.

Additional character restrictions for VM names can be found in the NetBackup Cloud
Administrator's Guide, available from this location:
http://www.veritas.com/docs/000003214
See “NetBackup for VMware: notes and restrictions” on page 27.
See “In the policy Query Builder, display names, resource pool names, and vApp
names are case-sensitive” on page 39.

In the policy Query Builder, display names,

resource pool names, and vApp names are
case-sensitive
In VMware vSphere, virtual machine display names, resource pool names, and
vApp names are case-sensitive. For example, a virtual machine with the name
"vm1" is a different virtual machine from one that is named "VM1."
The VMware virtual machines with the name values that do not exactly match the
query string case are not returned in the result set. Backups for those virtual
machines are missed.
For example, for the following virtual machines:
 Notes and prerequisites 40
Notes on the hotadd transport mode

vmware-ted
VMware-charles
VMWARE-john
vmWARE-jason

A query specifying Displayname Contains “vmware-” returns vmware-ted but

not the other virtual machines.
See “NetBackup for VMware: notes and restrictions” on page 27.

Notes on the hotadd transport mode

NetBackup supports several transport modes for sending snapshot data between
the VMware datastore and the VMware backup host during a backup or restore.
One of those transport modes (hotadd) is used when the VMware backup host is
installed in a virtual machine.
Note the following about the hotadd transport mode:
■ The VMware backup host must be installed in a virtual machine.
■ The following is a VMware requirement: The virtual machine to back up and the
virtual machine that contains the hotadd backup host must reside in the same
VMware datacenter. The same VMware requirement applies to virtual machine
restore: The virtual machine to restore and the virtual machine that contains the
hotadd restore host must reside in the same VMware datacenter.
For hotadd backup, it is recommended at least one hotadd backup host for each
datacenter.
■ NetBackup does not support IDE disks on the virtual machine.
■ On the virtual machine to back up, no two disks should have the same name.
(Identical names can occur if the disks reside on different datastores.)
■ The ESX server (where the backup-host virtual machine resides) must have
access to the datastore of the virtual machines that you want to back up.
■ The datastore for the backup-host virtual machine must have some free space
before the hotadd backup begins. Otherwise, the backup may fail.
■ Restores that use the hotadd or SAN transport modes do not include the VM's
metadata changes in the restore. The status log of the NetBackup job contains
messages similar to the following:

07/25/2013 12:37:29 - Info tar (pid=16257) INF - Transport Type

= hotadd
07/25/2013 12:42:41 - Warning bpbrm (pid=20895) from client
 Notes and prerequisites 41
Notes and limitations for tag usage in VMware Intelligent Policy queries

<client_address>: WRN - Cannot set metadata (key:geometry.

biosSectors, value:62) when using san or hotadd transport.

Retry the restore with a different transport mode (nbd or nbdssl).

This problem is a known VMware issue..
■ Locking timeouts in the VMware VDDK may cause simultaneous hotadd backups
from the same VMware backup host to fail.
See “Simultaneous hotadd backups (from the same VMware backup host) fail
with status 13” on page 317.
■ For a list of VMware restrictions on the hotadd transport mode, refer to VMware’s
documentation.

Notes and limitations for tag usage in VMware

Intelligent Policy queries
■ NetBackup does not support the selection of virtual machines based on the
category.
■ NetBackup uses tags for virtual machine selection independent of the tag's
category. The VMware vSphere Web Client can create tags in two different
categories with the same tag name. In the example, both virtual machines are
selected if the policy is configured to include virtual machines with the “HR” tag.
Example:
■ Virtual_Machine_1 has a user-specified tag HR in the category Production

■ Virtual_Machine_2 has a user-specified tag HR in the category Test

■ NetBackup only recognizes the tags that are associated with virtual machines,
not other vCenter objects such as Datastores.
■ vCenter Server 6.0 or later is required.
■ Any Windows host with only the NetBackup Client Software installed that is also
defined in the VMware policy's Client tab as the NetBackup host to perform
automatic virtual machine selection must have NetBackup Java installed.
Install the NetBackup Remote Administration Console to install NetBackup Java.
■ Be aware of a known bug in vSphere 6.0 when the system time of the discovery
host and the vCenter Server are not synchronized. This issue is known to cause
backups to fail with a NetBackup Status Code 4263.
VMware knowledge base article: http://kb.vmware.com/kb/2125193
■ In large VMware environments, consider increasing the Maximum bearer token
lifetime from the default value. Veritas has observed issues with discovery jobs
 Notes and prerequisites 42
Notes and limitations for the backup and restore of VMware tag associations

timing out because the default value is too small. The Maximum bearer token
lifetime is a vCenter Server setting. More information on this issue is available.
See “Troubleshooting VMware tag usage” on page 318.

Notes and limitations for the backup and restore

of VMware tag associations
■ Tag associations are part of the metadata of the virtual machine. NetBackup
considers virtual machine tag association protection a best effort backup. Any
tag collection errors are shown in the Activity Monitor for the virtual machine
snapshot jobs.
If for any reason NetBackup is unable to back up tag associations, the job
completes successfully with a NetBackup Status Code 0. Any failures to retrieve
tag associations in the backup are, however, reported in the job details in the
Activity Monitor.
■ vCenter Server 6.0 or later is required.
■ VMware tag associations are only backed up when you use VMware Intelligent
Policies.
■ NetBackup backs up tag associations with virtual machines. NetBackup does
not back up tag associations with other vCenter Server objects, such as
datastores or folders.
■ Tag associations for all virtual machines are retrieved during the discovery job.
The tag associations are, however, stored in the backup image for each individual
virtual machine. Be aware there is a time difference between the discovery job
time and the backup job. If tag associations are changed between these times,
the changes are not backed up.
■ If tag associations of a virtual machine were successfully backed up, then
NetBackup attempts to recreate tag associations with the restored virtual
machine. If the recovery host version is incompatible, NetBackup completes the
recovery of the virtual machine and sets the job status to 0. The recovery job
details, however, provide information on the failure to recreate the tag
associations.
■ If tag associations of a virtual machine were not backed up successfully, then
NetBackup does not attempt to recreate tag associations with the restored Virtual
Machine. NetBackup completes the recovery of the virtual machine and sets
the job status to 0. The recovery job details provide information on the failure
to collect the tag associations that occurred at the time of backup.
■ When Replication Director for VMware is used to protect your virtual machines,
tag associations are backed up only when Application Consistent protection is
 Notes and prerequisites 43
Notes and limitations for the backup and restore of VMware tag associations

enabled. When Application Consistent protection is disabled, NetBackup does

not protect tag associations.
 Chapter 4
Configure NetBackup
communication with
VMware
This chapter includes the following topics:

■ Adding the VMware backup host to NetBackup

■ Adding NetBackup credentials for VMware

■ Changing the host name of a VMware server in NetBackup

■ Setting privileges for posting events to vCenter

■ Authentication token for the NetBackup vSphere plug-ins

■ Setting global limits on the use of VMware resources

■ Validating VMware virtualization server certificates in NetBackup

Adding the VMware backup host to NetBackup

Use this procedure to enable a NetBackup client as the backup host.

Note: This procedure is not needed for the media servers that operate as backup
hosts with the policy Backup Media Server option. With that option, NetBackup
automatically enables media servers as backup hosts.

In the NetBackup Administration Console, do the following.

 Configure NetBackup communication with VMware 45
Adding NetBackup credentials for VMware

To add the VMware backup host

1 Click Host Properties > Master Servers, double click the NetBackup master
server, and click VMware Access Hosts.
2 Click Add.
3 Enter the fully qualified domain name of the VMware backup host, and click
Add.
4 When you are finished adding hosts, click Close.
5 Click Apply and then OK.
6 If the NetBackup master server is clustered in a failover environment: repeat
this procedure to add the backup host to each master server node in the cluster.
See “Adding NetBackup credentials for VMware” on page 45.

Adding NetBackup credentials for VMware

To back up or restore virtual machines, NetBackup requires logon credentials to
the VMware ESX servers, the VMware vCenter servers, or the VMware vCloud
servers. In the NetBackup Administration Console, do the following.
To add NetBackup credentials
1 Click Media and Device Management > Credentials > Virtual Machine
Servers.
2 Click Actions > New > New Virtual Machine Server.
3 On the New Virtual Machine Server dialog, enter the name of a virtual machine
server (ESX server, vCenter server, or vCloud server).

Note: Enter the server name in the same format in which it is registered in
DNS and in the VMware server (whether short or fully-qualified).
See “Using the VMware Managed Object Browser to verify the server name”
on page 50.
 Configure NetBackup communication with VMware 46
Adding NetBackup credentials for VMware

Note: The vCenter name that you enter here must match the name that is set
on the vCenter for VimApiUrl name and Runtime name. For assistance in
setting those names on the vCenter, and for additional vCenter naming
requirements relating to the NetBackup plug-ins for vSphere: See the topic on
consistent vCenter naming in the NetBackup Plug-in for VMware vSphere Web
Client Guide:
http://www.veritas.com/docs/000003214

Note: It is recommended the use of fully qualified names. The entire name
must be properly formed without empty or null elements. For example, a
fully-qualified name must include the domain name and not end in a period (.).

Note: The ESX server name is case-sensitive. Enter the ESX server name
exactly as it is in the VMware environment. If the case is wrong, the credential
validation fails and states "...expecting <correct_name_of_server>."

NetBackup needs access to this server for either of the following reasons:
■ To browse the server's virtual machines and back them up.
■ To use the server as a target for restoring virtual machines.
Use the next dialog to identify the type of this server (vCenter, ESX, or ESX
for restore only).

4 In the Credentials pane of the New Virtual Machine Server dialog, enter the
following:
Virtual machine server type
Select the type of VMware server that NetBackup needs to access:
■ VMware Virtual Center server
Designates a vCenter (or VirtualCenter) server that manages ESX servers.
When you create a policy to back up this server's virtual machines,
NetBackup can browse this server and list its virtual machines. If the
credentials provide full access privileges to the vCenter server, you can
restore virtual machines to this server.
Note: Do not enter logon credentials for the individual ESX servers that this
vCenter server manages. NetBackup needs credentials for the vCenter
only. If you enter credentials for both an ESXi server and a vCenter that
manages it, problems such as the following may occur:
■ A VMware Intelligent Policy (VIP) may fail due to discovery of duplicate
VMs.
 Configure NetBackup communication with VMware 47
Adding NetBackup credentials for VMware

■ For manual selection of VMs (not VIP): Tags and custom attributes
(which require a vCenter) may cause intermittent problems if NetBackup
attempts to back up the VM using the ESXi server.

■ VMware ESX Server

Designates a standalone ESX server that a vCenter server does not
manage. NetBackup can browse the ESX server to present a list of its virtual
machines for backup. You can also restore virtual machines to this ESX
server. To use the server as a target for restores, enter the credentials that
provide full access privileges to the ESX server.
■ VMware Restore ESX Server
Designates an ESX server to which NetBackup can restore virtual machines.
You must enter the credentials that provide full access privileges to the
server.

Note: NetBackup accesses this type of server for restores only, not for
backups.

The restore ESX server type has the following advantages:

■ For large environments with hundreds of hosts, NetBackup may not
need full access to the vCenter server. With the restore ESX server
type, you can give NetBackup full access to a single ESX server that is
dedicated to restore.
■ SAN-based restores that go directly to a restore ESX server are faster
than restores that go through the vCenter server.
■ Allows restoring to an ESX 5.x or later server that a vCenter 5.x or later
server manages. NetBackup uses vCenter to create the virtual machine.
NetBackup then writes the .vmdk files directly to the ESX server using
the Restore ESX Server credentials to that server.

Note: VMware does not support the restore of virtual machines directly to
an ESX 5.x or later server that vCenter manages. To restore the virtual
machine, select the vCenter server as the destination. As an alternative,
you can set up an independent ESX server to be used for restores. You
must add NetBackup restore credentials for that ESX server by means of
the VMware Restore ESX Server type.

For further information on the restore ESX server, refer to the following
Veritas tech note:
http://www.veritas.com/docs/000007351
 Configure NetBackup communication with VMware 48
Adding NetBackup credentials for VMware

■ VMware vCloud Director

Designates a vCloud Director server. NetBackup can browse the vCloud
environment on this server to present a list of its virtual machines for backup.
You can also restore virtual machines to this server.

Note: The credentials must be for a system administrator account.

Note: For backup and restore to vCloud Director, both vCloud and vCenter
credentials are required (VMware Virtual Center server).

Note: If the vCloud environment uses a load-balancer to manage multiple

cells (nodes), add credentials for the load balancer, not for the cells. If
vCloud Director has multiple cells but no load balancer, add credentials for
only one of the cells, not for all of them. Note also: If the Domain Name
System (DNS) cannot resolve the name of the load balancer or cell, do the
following: Include a line in the hosts file on the VMware backup host that
specifies the IP address of the load balancer or cell.

User name
Enter the user name for the virtual machine server. Note that spaces in user
names are not allowed.
To enter a domain name with a user name, separate them with a backslash
as follows:

domain\username

Password
Enter the password for the virtual machine server.
For backup host
You can select a particular backup host for which to validate the credentials.
Note that the hosts that appear in the pull-down list can be the following: media
servers, or NetBackup clients that have been added under Host Properties,
VMware Access Hosts.
For a list of supported backup host platforms, see the NetBackup Software
Compatibility List available from the following location:
NetBackup Master Compatibility List
See “Adding the VMware backup host to NetBackup” on page 44.
 Configure NetBackup communication with VMware 49
Adding NetBackup credentials for VMware

If For backup host is set to <Any> (the default), NetBackup attempts to validate
the credentials using each backup host in turn. NetBackup continues checking
only until validation succeeds for a particular backup host or the list of backup
hosts is exhausted.
During validation: An invalid user name or password is deleted if you have
entered it for the first time and you do not select a backup host. If validation
fails for any reason, the entry is removed.
If your VMware backup host is running a version of NetBackup that is earlier
than 6.5.4, the credentials cannot be validated. An error message is displayed.
You must verify that the credentials you entered are correct for the VMware
server. If the credentials are correct, you can ignore the error message.
Connect using port number
If the default port number has not been changed on the VMware server, no
port specification is required. In that case, make sure that the Connect using
port number box is not checked.
If the VMware server has been configured to use a different port, click the
Connect using port number box and specify that port number.
5 Click OK.
6 If your site has multiple standalone ESX servers, use the Virtual Machine
Server dialog to enter credentials for each ESX server.
7 If the NetBackup master server is clustered, do the following on the other
master server nodes to verify the credentials for each backup host.
■ Fail over to the next master server node, to make it the active node.
■ On the active node, click on Media and Device Management > Credentials
> Virtual Machine Servers and double-click on the virtual machine server
credential.
■ In the For backup host field, select a particular backup host (do not select
Any).
If the backup host does not appear in the list, you must add the correct
credentials for the backup host. (Click Actions > New > New Virtual
Machine Server.)
■ Click OK.
NetBackup validates the credentials.
■ Double-click the same virtual machine server credential and validate the
credentials for another backup host (if any).
■ For other master server nodes, repeat step 7.
 Configure NetBackup communication with VMware 50
Changing the host name of a VMware server in NetBackup

See “Setting global limits on the use of VMware resources” on page 52.

Using the VMware Managed Object Browser to verify the server

name
When adding NetBackup credentials for a VMware server, enter the server name
exactly as configured in the VMware server (whether short or fully-qualified). If the
name you enter for the credentials does not match the name as defined on the
VMware server, the credential validation fails.
You can use the vSphere Managed Object Browser to verify the server's name.
To verify the server's name
1 In a web browser, open the Managed Object Browser (MOB) by entering the
fully-qualified domain name of the VMware server and /mob.
For example: https://vcenter1.acmecorp.com/mob
2 Navigate to the ManagedObjectReference:HostSystem and find the name
value for the server.
The object structure is site-dependent.
3 When you create NetBackup credentials for the server, enter the name value
exactly as it appears in the MOB.
See “Adding NetBackup credentials for VMware” on page 45.
For more information on the MOB, see your vSphere documentation.

Changing the host name of a VMware server in

NetBackup
Use the following procedure if you need to change a VMware server's host name
as it is stored in NetBackup.
Example reasons for changing the VMware server name are the following:
■ You used a short host name for the server for its NetBackup credentials, and
need to replace that name with the fully qualified host name.
■ Two hosts currently resolve to the same IP address, and one of them must be
renamed and assigned a new IP address.
To change the host name of a VMware host in NetBackup
1 Remove the current credentials that were entered for the VMware host.
Do the following:
 Configure NetBackup communication with VMware 51
Setting privileges for posting events to vCenter

■ In the NetBackup Administration Console, click on Media and Device

Management > Credentials > Virtual Machine Servers.
■ Right-click on the credentials definition for the VMware server and select
Delete (or press the Delete key).

2 In the NetBackup Administration Console, re-enter NetBackup's credentials

for the VMware server.
See “Adding NetBackup credentials for VMware” on page 45.
Make sure to enter the correct host name for the VMware server.
See “Setting global limits on the use of VMware resources” on page 52.

Setting privileges for posting events to vCenter

With the Post vCenter events option, NetBackup can send backup related events
to the vCenter server, to view in vSphere Web Client. The NetBackup plug-in for
vSphere Web Client or vSphere Client (HTML5) is not required.
Note the following requirements:
■ You must enter the credentials that give NetBackup access to the vCenter
server.
See “Adding NetBackup credentials for VMware” on page 45.
■ Make sure that the Post vCenter events option is enabled in the policy.
See “VMware - Advanced Attributes dialog” on page 81.
■ You must set the correct vCenter role privileges. Use the vSphere Web Client
to make sure that the following Global privileges are set in vCenter: Manage
custom attributes and Set custom attribute.
For assistance setting privileges, refer to the appropriate VMware vSphere
Documentation Center.

Authentication token for the NetBackup vSphere

plug-ins
With the NetBackup vSphere Web Client plug-in and the vSphere Client (HTML5)
plug-in, the VMware administrator can use the vSphere interface to recover virtual
machines. To allow the plug-in to communicate with the NetBackup master server,
you must provide an authentication token to the VMware administrator.
To create an authentication token, see the Veritas NetBackup Plug-in for VMware
vSphere Web Client Guide or the Veritas NetBackup Plug-in for VMware vSphere
Client (HTML5) Guide:
 Configure NetBackup communication with VMware 52
Setting global limits on the use of VMware resources

http://www.veritas.com/docs/DOC5332
See “Setting privileges for posting events to vCenter” on page 51.

Setting global limits on the use of VMware

resources
You can use the NetBackup Resource Limit dialog to control the number of
simultaneous backups that can be performed on a VMware resource type. The
settings apply to all NetBackup policies for the currently selected master server.
For example, to avoid overloading the ESX server, you can place a limit on the
number of concurrent backup jobs per ESX server. To control I/O overhead on the
datastore's array, you can limit the number of concurrent backups per datastore.
The following limitations apply to setting global limits on the use of VMware
resources:
■ In NetBackup 8.3, you can set global limits for VMware resources only through
the NetBackup Administration Console or the NetBackup Configuration API.
You cannot set VMware resource limits with the NetBackup web UI. For more
information about the Configuration API, see the NetBackup API Reference on
SORT.
■ New and changed resource limits may not take effect immediately. It can take
a couple of jobs before the resource limit updates are in effect.
■ The Resource Limit screen applies only to policies that use automatic selection
of virtual machines (Query Builder). If virtual machines are selected manually
on the Browse for Virtual Machines screen, the Resource Limit settings have
no effect.
■ To limit the number of simultaneous jobs per policy, use the Limit jobs per
policy setting on the policy Attributes tab. The effect of this option depends
on how the policy selects virtual machines.
See “Limit jobs per policy on the Attributes tab (for VMware)” on page 72.
To set limits on the use of VMware resources
1 In the NetBackup Administration Console, click Host Properties > Master
Servers and double-click the NetBackup master server.
2 In the Properties screen, scroll down in the left pane and click Resource Limit.
 Configure NetBackup communication with VMware 53
Setting global limits on the use of VMware resources

3 Under Application, click VMware.

4 Click in the Resource Limit column to set the maximum NetBackup usage for
a particular resource type. The settings apply to all policies.

For each resource type, the default is 0, (No Limit). A resource limit of 4 for
Datastore means that NetBackup policies can perform no more than four
simultaneous backups on any particular datastore.
Table 4-1 describes the limits:

Table 4-1 Resource types and limits

Resource type Resource limit

vCenter The maximum number of simultaneous backups per vCenter

server.

snapshot The maximum number of simultaneous snapshot operations

(create or delete) per vCenter.

Cluster The maximum number of simultaneous backups per VMware

cluster.

ESXserver The maximum number of simultaneous backups per ESX

server.
 Configure NetBackup communication with VMware 54
Validating VMware virtualization server certificates in NetBackup

Table 4-1 Resource types and limits (continued)

Resource type Resource limit

VMXDatastore Controls the maximum number of simultaneous backups per

Datastore. The Datastore is defined as the location of the VMX
file that is associated with each VM. This resource type is useful
for VMs that have VMDK files distributed across multiple
Datastores. This setting is enforced globally within a NetBackup
domain.

See “NetBackup for VMware terminology” on page 20.

Datastore The maximum number of simultaneous backups per datastore.

DatastoreFolder The maximum number of simultaneous backups per datastore

folder.

DatastoreType The maximum number of simultaneous backups per datastore

type.

VMXDatastoreNFSHost Controls the maximum number of simultaneous backups per

Datastore at the NFS host level where the Datastore type is
NFS. The VMXDatastoreNFSHost is an NFS server that
sources one or more NFS Datastores. For this resource type
the Datastore is defined as the location of the VMX file that is
associated with each VM. This resource type is useful for VMs
that have VMDK files that are distributed across multiple
Datastores. This setting is enforced globally within a NetBackup
domain.

DatastoreNFSHost The maximum number of simultaneous backups per NFS host

of the datastore.

DatastoreCluster The maximum number of simultaneous backups per datastore

cluster.

Other resource-related considerations may be of interest.

See “NetBackup for VMware best practices” on page 283.
See “Configuring a VMware policy from the Policies utility” on page 68.

Validating VMware virtualization server certificates

in NetBackup
NetBackup can now validate VMware virtualization server certificates using their
root or intermediate certificate authority (CA) certificates.
 Configure NetBackup communication with VMware 55
Validating VMware virtualization server certificates in NetBackup

The following procedure is applicable for the NetBackup master server and all
VMware access hosts.
 Configure NetBackup communication with VMware 56
Validating VMware virtualization server certificates in NetBackup

To configure secure communication between VMware virtualization server

and VMware access host
1 Configure an external CA trust store on the VMware access host.
2 Add CA certificates of the required VMware servers (VCenter, ESX, or ESXi
server) in the trust store on the access host.
In case of Windows certificate store, you need to add the CA certificate to the
Windows Trusted Root Certification Authorities.
Use the following command:
certutil.exe -addstore -f "Root" certificate filename
 Configure NetBackup communication with VMware 57
Validating VMware virtualization server certificates in NetBackup

3 Use the nbsetconfig command to configure the following NetBackup

configuration options on the access host:

ECA_TRUST_STORE_PATH Specifies the file path to the certificate bundle file

that contains all trusted root CA certificates.

This option is specific to file-based certificates. You

should not configure this option if Windows certificate
store is used.

If you have already configured this external CA

option, append the VMware CA certificates to the
existing external certificate trust store.

If you have not configured the option, add all the

required virtualization server CA certificates to the
trust store and set the option.

See “ECA_TRUST_STORE_PATH for NetBackup

servers and clients” on page 58.

ECA_CRL_PATH Specifies the path to the directory where the

certificate revocation lists (CRL) of the external CA
are located.

If the configuration option is already configured,

append the virtualization server CRLs to the CRL
cache.

If the option is not configured, add all the required

CRLs to the CRL cache and then set the option.
See “ECA_CRL_PATH for NetBackup servers and
clients” on page 59.

VIRTUALIZATION_HOSTS_ Lets you enable the validation of a virtualization

SECURE_CONNECT_ENABLED server's certificate.

See
“VIRTUALIZATION_HOSTS_SECURE_CONNECT_ENABLED
for servers and clients” on page 63.

VIRTUALIZATION_CRL_CHECK Lets you validate the revocation status of the

virtualization server certificate against the CRLs.

By default, the option is disabled.

See “VIRTUALIZATION_CRL_CHECK for

NetBackup servers and clients” on page 64.
 Configure NetBackup communication with VMware 58
Validating VMware virtualization server certificates in NetBackup

VIRTUALIZATION_HOSTS_CONNECT_TIMEOUT Lets you specify the duration (in seconds) after which
the connection between NetBackup and vCloud
Director server ends.

See
“VIRTUALIZATION_HOSTS_CONNECT_TIMEOUT
for servers and clients” on page 62.

VMWARE_TLS_MINIMUM_V1_2 Lets you specify the Transport Layer Security (TLS)

version to be used for communication between
NetBackup and VMware servers.

See “VMWARE_TLS_MINIMUM_V1_2 for

NetBackup servers and clients” on page 65.

For more information on external CA support in NetBackup, refer to the

NetBackup Security and Encryption Guide.

ECA_TRUST_STORE_PATH for NetBackup servers and clients

The ECA_TRUST_STORE_PATH option specifies the file path to the certificate bundle
file that contains all trusted root CA certificates.
This certificate file should have one or more certificates in PEM format.
Do not specify the ECA_TRUST_STORE_PATH option if you use the Windows certificate
store.
The trust store supports certificates in the following formats:
■ PKCS #7 or P7B file having certificates of the trusted root certificate authorities
that are bundled together. This file may either be PEM or DER encoded.
■ A file containing the PEM encoded certificates of the trusted root certificate
authorities that are concatenated together.
This option is mandatory for file-based certificates.

Table 4-2 ECA_TRUST_STORE_PATH information

Usage Description

Where to use On NetBackup servers or clients.

If certificate validation is required for VMware, RHV servers, or
Nutanix AHV, this option must be set on the NetBackup master
server and respective access hosts, irrespective of the certificate
authority that NetBackup uses for host communication
(NetBackup CA or external CA).
 Configure NetBackup communication with VMware 59
Validating VMware virtualization server certificates in NetBackup

Table 4-2 ECA_TRUST_STORE_PATH information (continued)

Usage Description

How to use Use the nbgetconfig and the nbsetconfig commands to

view, add, or change the option.

For information about these commands, see the NetBackup

Commands Reference Guide.

Use the following format:

ECA_TRUST_STORE_PATH = Path to the external CA

certificate

For example: c:\rootCA.pem

If you use this option on a Flex Appliance application instance,

the path must be /mnt/nbdata/hostcert/.

Equivalent Administration No equivalent exists in the NetBackup Administration Console

Console property host properties.

ECA_CRL_PATH for NetBackup servers and clients

The ECA_CRL_PATH option specifies the path to the directory where the Certificate
Revocation Lists (CRL) of the external certificate authority (CA) are located.
These CRLs are copied to NetBackup CRL cache. Revocation status of the external
certificate is validated against the CRLs from the CRL cache.
CRLs in the CRL cache are periodically updated with the CRLs in the directory that
is specified for ECA_CRL_PATH based on the ECA_CRL_PATH_SYNC_HOURS option.
See “ECA_CRL_PATH_SYNC_HOURS for NetBackup servers and clients”
on page 60.
If the ECA_CRL_CHECK option is not set to DISABLE (or 0) and the ECA_CRL_PATH
option is not specified, NetBackup downloads the CRLs from the URLs that are
specified in the CRL distribution point (CDP) and uses them to verify revocation
status of the peer host's certificate.
See “ECA_CRL_CHECK for NetBackup servers and clients” on page 61.

Note: For validating the revocation status of a virtualization server certificate, the
VIRTUALIZATION_CRL_CHECK option is used.

See “VIRTUALIZATION_CRL_CHECK for NetBackup servers and clients”

on page 64.
 Configure NetBackup communication with VMware 60
Validating VMware virtualization server certificates in NetBackup

Table 4-3 ECA_CRL_PATH information

Usage Description

Where to use On NetBackup servers or clients.

If certificate validation is required for VMware, RHV servers,
or Nutanix AHV, this option must be set on the NetBackup
master server and respective access hosts, irrespective of
the certificate authority that NetBackup uses for host
communication (NetBackup CA or external CA).

How to use Use the nbgetconfig and the nbsetconfig commands

to view, add, or change the option.

For information about these commands, see the NetBackup

Commands Reference Guide.

Use the following format to specify a path to the CRL directory:

ECA_CRL_PATH = Path to the CRL directory

If you use this option on a Flex Appliance application instance,

the path must be /mnt/nbdata/hostcert/crl.

Equivalent Administration No equivalent exists in the NetBackup Administration

Console property Console host properties.

ECA_CRL_PATH_SYNC_HOURS for NetBackup servers and clients

The ECA_CRL_PATH_SYNC_HOURS option specifies the time interval in hours to update
the Certificate revocation lists (CRL) in the NetBackup CRL cache with the CRLs
in the directory specified for the ECA_CRL_PATH configuration option.
See “ECA_CRL_PATH for NetBackup servers and clients” on page 59.
The ECA_CRL_PATH_SYNC_HOURS option is not applicable if CDP is used for CRLs.
By default, CRLs in the cache are updated every one hour.
During host communication, revocation status of the external certificate is validated
against the CRLs from the CRL cache.

Table 4-4 ECA_CRL_PATH_SYNC_HOURS information

Usage Description

Where to use On NetBackup servers or clients.

 Configure NetBackup communication with VMware 61
Validating VMware virtualization server certificates in NetBackup

Table 4-4 ECA_CRL_PATH_SYNC_HOURS information (continued)

Usage Description

How to use Use the nbgetconfig and the nbsetconfig commands to

view, add, or change the option.

For information about these commands, see the NetBackup

Commands Reference Guide.

Use the following format:

ECA_CRL_PATH_SYNC_HOURS = Number of hours

Minimum number of hours that you can specify - 1 hour

Maximum number of hours that you can specify - 720 hour

The default value is one hour.

Equivalent Administration No equivalent exists in the NetBackup Administration Console

Console property host properties.

ECA_CRL_CHECK for NetBackup servers and clients

The ECA_CRL_CHECK option lets you specify the revocation check level for external
certificates of the host. It also lets you disable the revocation check for the external
certificates. Based on the check, revocation status of the certificate is validated
against the Certificate Revocation List (CRL) during host communication.
You can choose to use the CRLs from the directory that is specified for the
ECA_CRL_PATH configuration option in the configuration file (bp.conf on UNIX or
Windows registry) or the CRL Distribution Point (CDP).
See “ECA_CRL_PATH for NetBackup servers and clients” on page 59.

Table 4-5 ECA_CRL_CHECK information

Usage Description

Where to use On NetBackup servers or clients.

 Configure NetBackup communication with VMware 62
Validating VMware virtualization server certificates in NetBackup

Table 4-5 ECA_CRL_CHECK information (continued)

Usage Description

How to use Use the nbgetconfig and the nbsetconfig commands to

view, add, or change the option.

For information about these commands, see the NetBackup

Commands Reference Guide.

Use the following format:

ECA_CRL_CHECK = CRL check

You can specify one of the following:

■ DISABLE (or 0) - Revocation check is disabled. Revocation

status of the certificate is not validated against the CRL during
host communication.
■ LEAF (or 1) - Revocation status of the leaf certificate is
validated against the CRL. This is the default value.
■ CHAIN (or 2) - Revocation status of all certificates from the
certificate chain are validated against the CRL.

Equivalent Administration No equivalent exists in the NetBackup Administration Console

Console property host properties.

VIRTUALIZATION_HOSTS_CONNECT_TIMEOUT for servers and

clients
The VIRTUALIZATION_HOSTS_CONNECT_TIMEOUT option lets you specify the duration
(in seconds) after which the connection between NetBackup and vCloud Director
server ends.

Table 4-6
Usage Description

Where to use On NetBackup master server or all access hosts.

 Configure NetBackup communication with VMware 63
Validating VMware virtualization server certificates in NetBackup

Table 4-6 (continued)

Usage Description

How to use Use the nbgetconfig and the nbsetconfig commands to view,
add, or change the option.

For information about these commands, see the NetBackup

Commands Reference Guide.

By default, the VIRTUALIZATION_HOSTS_CONNECT_TIMEOUT option

is set to 60 seconds.

Use the following format to specify the time-out value:

VIRTUALIZATION_HOSTS_CONNECT_TIMEOUT = Time-out
value in seconds

Equivalent No equivalent exists in the NetBackup Administration Console

Administration host properties
Console property

VIRTUALIZATION_HOSTS_SECURE_CONNECT_ENABLED for
servers and clients
The VIRTUALIZATION_HOSTS_SECURE_CONNECT_ENABLED option enables the
validation of virtualization server certificates using its root or intermediate certificate
authority (CA) certificates.
Before you enable the option, review the steps from the 'Validating VMware
virtualization server certificates in NetBackup ' section in the NetBackup for VMware
Administrator's Guide.

Table 4-7 VIRTUALIZATION_HOSTS_SECURE_CONNECT_ENABLED

information

Usage Description

Where to use On NetBackup master server or all access hosts.

 Configure NetBackup communication with VMware 64
Validating VMware virtualization server certificates in NetBackup

Table 4-7 VIRTUALIZATION_HOSTS_SECURE_CONNECT_ENABLED

information (continued)

Usage Description

How to use Use the nbgetconfig and the nbsetconfig commands to view,
add, or change the option.

For information about these commands, see the NetBackup

Commands Reference Guide.

By default, the
VIRTUALIZATION_HOSTS_SECURE_CONNECT_ENABLED is set to
UNDEFINED, but the security certificate validation is enabled for RHV
and Nutanix AHV servers and disabled for VMware servers.

Use the following format to enable certificate validation for the RHV,
VMware, or Nutanix AHV servers:

VIRTUALIZATION_HOSTS_SECURE_CONNECT_ENABLED = YES

Equivalent No equivalent exists in the NetBackup Administration Console

Administration host properties.
Console property

VIRTUALIZATION_CRL_CHECK for NetBackup servers and clients

The VIRTUALIZATION_CRL_CHECK option lets you specify the revocation check level
for external certificates of the virtualization server. Based on the check, revocation
status of the virtualization server certificate is validated against the certificate
revocation list (CRL) during host communication.
By default, the VIRTUALIZATION_CRL_CHECK option is disabled. If you want to validate
the revocation status of the virtualization server certificate against certificate
revocation list (CRL), set the option to a different value.
You can choose to use the CRLs from the directory that is specified for the
ECA_CRL_PATH configuration option or the CRL distribution point (CDP).

See “ECA_CRL_PATH for NetBackup servers and clients” on page 59.

Table 4-8 VIRTUALIZATION_CRL_CHECK information

Usage Description

Where to use On NetBackup master server or all access hosts.

 Configure NetBackup communication with VMware 65
Validating VMware virtualization server certificates in NetBackup

Table 4-8 VIRTUALIZATION_CRL_CHECK information (continued)

Usage Description

How to use Use the nbgetconfig and the nbsetconfig commands to

view, add, or change the option.

For information about these commands, see the NetBackup

Commands Reference Guide.

Use the following format:

VIRTUALIZATION_CRL_CHECK = CRL check

You can specify one of the following:

■ DISABLE (or 0) - Revocation check is disabled. Revocation

status of the certificate is not validated against the CRL during
host communication. This is the default value.
■ LEAF (or 1) - Revocation status of the leaf certificate is
validated against the CRL.
■ CHAIN (or 2) - Revocation status of all certificates from the
certificate chain are validated against the CRL.

Equivalent Administration No equivalent exists in the NetBackup Administration Console

Console property host properties.

VMWARE_TLS_MINIMUM_V1_2 for NetBackup servers and clients

The VMWARE_TLS_MINIMUM_V1_2 option lets you specify the Transport Layer Security
(TLS) version to be used for communication between NetBackup and VMware
servers.

Table 4-9
Usage Description

Where to use On NetBackup master server or all access hosts.

 Configure NetBackup communication with VMware 66
Validating VMware virtualization server certificates in NetBackup

Table 4-9 (continued)

Usage Description

How to use Use the nbgetconfig and the nbsetconfig commands to view, add,
or change the option.

For information about these commands, see the NetBackup Commands

Reference Guide.

Use the following format:

VMWARE_TLS_MINIMUM_V1_2 = YES | NO

By default, the VMWARE_TLS_MINIMUM_V1_2 option is set to YES.

If the option is set to YES, TLS 1.2 version and the following cipher suites
are used for communication with VMware servers:

TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384,
TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384,
TLS_DHE_RSA_WITH_AES_256_GCM_SHA384,
TLS_DHE_DSS_WITH_AES_256_GCM_SHA384,
TLS_ECDHE_ECDSA_WITH_AES_128_GCM_SHA256,
TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256,
TLS_DHE_RSA_WITH_AES_128_GCM_SHA256,
TLS_DHE_DSS_WITH_AES_128_GCM_SHA256,
TLS_RSA_WITH_AES_256_GCM_SHA384,
TLS_RSA_WITH_AES_128_GCM_SHA256

If the option is set to NO, the TLS 1.2 or earlier version with the default
cipher suite is used for communication between NetBackup and VMware
server. The cipher suite is used based on the TLS and cipher suite
configuration that are set on the respective VMware server.

Equivalent No equivalent exists in the NetBackup Administration Console host

Administration properties.
Console property
 Chapter 5
Configure NetBackup
policies for VMware
This chapter includes the following topics:

■ Configuring a VMware policy from the Policy Wizard

■ Configuring a VMware policy from the Policies utility

■ Limit jobs per policy on the Attributes tab (for VMware)

■ Backup options on the VMware tabVMware backup options

■ Exclude Disks tab

■ Browse for VMware virtual machines

■ Limiting the VMware servers that NetBackup searches when browsing for virtual
machines

■ Virtual machine host names and display names should be unique if VMs are
selected manually in the policy

■ Primary VM identifier option and manual selection of virtual machines

■ About incremental backups of virtual machines

■ Configuring incremental backups

■ Storage Foundation Volume Manager volumes in the virtual machine

 Configure NetBackup policies for VMware 68
Configuring a VMware policy from the Policy Wizard

Configuring a VMware policy from the Policy

Wizard
You can create a NetBackup for VMware policy by means of the Policy Configuration
Wizard.
Use the following procedure.
To create a backup policy by means of the Policy Configuration Wizard
1 In the NetBackup Administration Console (on the NetBackup master server),
click on the name of the master server.
2 Click Create a Policy.
3 Under Create a backup policy for, click VMware and Hyper-V, and click
Next.
4 Enter a name for the policy.
5 Click VMware as the virtual machine type.
6 Click Next, and follow the remaining panels in the wizard.
The wizard creates the policy according to your selections.
See “Configuring a VMware policy from the Policies utility” on page 68.

Configuring a VMware policy from the Policies

utility
You can create a NetBackup policy (full or incremental) to back up the virtual
machine in either of two ways:
■ Use the Policies utility of the NetBackup Administration Console (described in
this topic).
■ Use the Policy Configuration Wizard.
See “Configuring a VMware policy from the Policy Wizard ” on page 68.

Note: To configure a policy for Replication Director, see the NetBackup Replication
Director Solutions Guide.
 Configure NetBackup policies for VMware 69
Configuring a VMware policy from the Policies utility

To configure a policy to back up the virtual machine

1 In the NetBackup Administration Console, click Policies and select Actions
> New > Policy.
2 Enter a name for the policy.
3 For Policy type, select VMware.
VMware backup options are available on the VMware tab (described later in
this procedure).
4 Select a policy storage unit or storage unit group.
Storage unit groups are described in the NetBackup Administrator’s Guide,
Volume I.
5 In most cases, you can leave the Disable client-side deduplication option
at the default (unchecked).
The Disable client-side deduplication option is described in the NetBackup
Administrator's Guide, Volume I.
6 To enable the Accelerator, click Use Accelerator.
When you click Use Accelerator, the Enable block-level incremental backup
option on the VMware tab is selected and grayed out.
See “About the NetBackup Accelerator for virtual machines” on page 159.
 Configure NetBackup policies for VMware 70
Configuring a VMware policy from the Policies utility

7 Click the VMware tab.

Use this dialog to set VMware-related options.

See “Backup options on the VMware tabVMware backup options” on page 73.
 Configure NetBackup policies for VMware 71
Configuring a VMware policy from the Policies utility

8 Click the Exclude Disks tab.

Use this tab to exclude disks from the backups.

See “Exclude Disks tab” on page 86.
9 To define a schedule, click the Schedules tab and click New.
For assistance with the Accelerator forced rescan option:
See “Accelerator forced rescan for virtual machines (schedule attribute)”
on page 163.
On the Schedules Attributes tab, you can select Full backup, Differential
Incremental Backup, or Cumulative Incremental Backup.
Note that incremental backups require one of the following selections on the
policy VMware tab:
■ Enable file recovery from VM backup, or
■ Enable block-level incremental backup.
Enable block-level incremental backup requires an ESX server 4.0 and
a virtual machine at vmx-07 or later.
 Configure NetBackup policies for VMware 72
Limit jobs per policy on the Attributes tab (for VMware)

The following topic contains additional information.

See “About incremental backups of virtual machines” on page 99.

10 Click the Clients tab to select the virtual machines to back up.
The options for selecting virtual machines are as follows:

Select manually, and The Browse for VMware Virtual Machines screen appears. You can type the host name in
click New. the Enter the VMware hostname field, or click Browse and select virtual machine.

Note: The Backup Selections tab is set to ALL_LOCAL_DRIVES. Individual drives cannot
be specified.

See “Browse for VMware virtual machines” on page 91.

Select automatically With this option, NetBackup can automatically select virtual machines for backup based on
through VMware the filtering criteria that you enter. The following topics explain how to specify the criteria:
Intelligent Policy
See “About automatic virtual machine selection for NetBackup for VMware” on page 102.
query
See “Configuring automatic virtual machine selection” on page 112.
Note: The Backup Selections tab is filled in with the filtering rules from the query. NetBackup
uses the ALL_LOCAL_DRIVES directive to back up the virtual machines.

Enable vCloud Enables backup of the virtual machines that reside in a vCloud environment. Requires the
Director integration automatic selection of virtual machines.

With this option, the policy selects for backup only the virtual machines that vCloud manages:
it skips the virtual machines that are not in vCloud.

See “About NetBackup for vCloud Director” on page 263.

NetBackup host to This option appears when you click Select automatically through VMware Intelligent
perform automatic Policy query.
virtual machine
This host discovers virtual machines and automatically selects them for backup based on
selection
the query rules.

See “About automatic virtual machine selection for NetBackup for VMware” on page 102.

11 When you click OK, a validation process checks the policy and reports any
errors. If you click Cancel, no validation is performed.

Limit jobs per policy on the Attributes tab (for

VMware)
The Limit jobs per policy option operates as follows, depending on how the policy
selects virtual machines.
 Configure NetBackup policies for VMware 73
Backup options on the VMware tabVMware backup options

For the policies that select virtual machines automatically

(Query Builder)
The Limit jobs per policy option controls the number of parent (discovery) jobs
that run simultaneously for the policy. This option does not limit the number of
snapshot jobs and backup (bpbkar) jobs that the parent job launches. For example,
if this option is set to 1 and you begin a backup of a policy that discovers 100 virtual
machines: all the snapshot jobs and backup jobs for each of the 100 virtual machines
are allowed to execute simultaneously. Only the initial discovery job counts against
Limit jobs per policy. If you begin a second backup of the policy, its discovery job
cannot start until all the child jobs from the first backup are complete.

For the policies that use manual selection of virtual

machines
Limit jobs per policy controls the number of virtual machines that the policy can
back up simultaneously. Because no discovery job is needed, each virtual machine
backup begins with a snapshot job. Each snapshot counts against the Limit jobs
per policy setting. If this option is set to 1: the backup of the next virtual machine
that is specified in the policy cannot begin until the first snapshot job and its backup
are complete.
See “Setting global limits on the use of VMware resources” on page 52.

Backup options on the VMware tabVMware backup

options
In the Administration Console, the VMware tab appears when you select VMware
as the policy type.
The following topics describe the VMware backup options.
See “VMware backup host” on page 74.
See “Optimizations options (VMware)” on page 74.
See “Primary VM identifier options (VMware)” on page 76.
See “Existing snapshot handling options (VMware)” on page 77.
See “Transport modes options (VMware)” on page 78.
See “Application protection options (VMware)” on page 79.
See “VMware - Advanced Attributes dialog” on page 81.
For more information, see the NetBackup for VMware Guide:
HTTP://www.veritas.com/docs/DOC5332
 Configure NetBackup policies for VMware 74
Backup options on the VMware tabVMware backup options

VMware backup host

The VMware backup host is a NetBackup client that performs backups on behalf
of the virtual machines.

Table 5-1 VMware backup host selection

Option Description

Backup Media Server This option allows a media server that is selected in the policy to operate as the backup
host. (The storage unit determines the selection of the media server.) To operate as the
backup host, the media server must contain NetBackup client software.
Note: The storage unit that is specified in the policy must be unique to the media servers
that NetBackup supports as VMware backup hosts. If the storage unit is available on a
media server that is not a supported VMware backup host, the snapshot may not succeed
(status 20). For a list of supported platforms for the VMware backup host, see the NetBackup
Software Compatibility List (SCL) available from the following location:

NetBackup Master Compatibility List

Note: When the Backup Media Server option is selected, NetBackup cannot determine
a host to perform policy validation. To validate the policy, temporarily select one of the
possible media servers as the backup host (do not select Backup Media Server). When
the policy validates successfully, reset the backup host to Backup Media Server.

See “Media servers as backup or discovery hosts” on page 18.

backup_host_name Select a backup host to perform the backup.

The pull-down list contains any media servers that are supported as backup host. It also
contains any NetBackup clients that have been added in the Administration Console under
Host Properties, VMware Access Hosts.

For a list of supported platforms for the backup host, see the NetBackup Software
Compatibility List available from the following location:

NetBackup Master Compatibility List

See “Adding the VMware backup host to NetBackup” on page 44.

See “Backup options on the VMware tabVMware backup options” on page 73.

Optimizations options (VMware)

The following options set the type and scope of the VMware virtual machine backup.
 Configure NetBackup policies for VMware 75
Backup options on the VMware tabVMware backup options

Table 5-2 Optimizations

Option Description

Enable file recovery This option allows restore of individual files from the backup. With or without this option,
from VM backup you can restore the entire virtual machine.

You can also use this option for incremental backups: in the policy schedule, select
differential incremental backup or cumulative incremental backup.

To perform a VMware backup to a deduplication storage unit, select this option. This option
provides the best deduplication rates.

To back up a virtual machine that contains Veritas Storage Foundation Volume Manager
volumes, disable this option. Also make sure that the Exclude deleted blocks option is
disabled.
Note: For a Linux virtual machine, the name of an LVM volume can include any of the
following special characters: . (period), _ (underscore), - (hyphen). No other special
characters are supported. If other special characters are in the volume name, the Enable
file recovery from VM backup option does not work. As a result, you cannot restore
individual files from that volume.

Note: During an incremental backup, any files that had been moved or renamed are not
backed up. Those files are not available when you browse to restore individual files from
the incremental backup. However, when you restore the entire VM from a block-level
incremental backup, note: the file metadata is updated and the moved or renamed files in
the restored VM reflect the updated metadata.

Enable block-level For block-level backups of the virtual machine. This option reduces the size of the backup
incremental backup image.

For the Exchange, SQL, and SharePoint Agents, this option is selected and grayed out if
you enable Use Accelerator.

On the Attributes tab, Perform block level incremental backups is automatically selected
and grayed out.

See “Block-level backup (BLIB): full vs incremental” on page 148.

Exclude deleted blocks Reduces the size of the backup image by excluding any unused or deleted blocks within
the file system on the virtual machine. This option supports the following file systems:
Windows NTFS, and Linux ext2, ext3, and ext4.

This option uses proprietary mapping technology to identify vacant sectors (allocated but
empty) within the file system.

See “About reducing the size of VMware backups” on page 147.

To back up a virtual machine that contains Veritas Storage Foundation Volume Manager
volumes, disable this option. Also make sure that the Enable file recovery from VM
backup option is disabled.
 Configure NetBackup policies for VMware 76
Backup options on the VMware tabVMware backup options

Table 5-2 Optimizations (continued)

Option Description

Exclude swap and Reduces the size of the backup image by excluding the data in the guest OS system paging
paging files file (Windows) or the swap file (Linux).
Note: This option does not exclude the swapping and paging files from the backup: it only
excludes the data in those files. If the files are restored, they are restored as empty files.

Note: For a Linux virtual machine, this option disables the swap file when you restore the
virtual machine. You must reconfigure the swap file after the virtual machine is restored.
To allow the virtual machine to be restored with its swap file enabled, do not select Exclude
swap and paging files.

See “Backup options on the VMware tabVMware backup options” on page 73.

Primary VM identifier options (VMware)

This setting specifies the type of name by which NetBackup recognizes virtual
machines when it selects them for backup.
The name you use for the Primary VM identifier may have restrictions.
See “NetBackup character restrictions for the Primary VM identifier” on page 37.

Table 5-3 Primary VM identifier

Option Description

VM hostname The network host name for the virtual machine. (This option is the default.) NetBackup
obtains the host name by means of a reverse lookup on the virtual machine's IP address.
Note: For NetBackup to look up the IP address, the virtual machine must already be
turned on.

If no host name can be found, the IPv4 address is used as the host name. In case of
an IPv6 address, the field remains blank.

See “Preventing browsing delays caused by DNS problems” on page 297.

 Configure NetBackup policies for VMware 77
Backup options on the VMware tabVMware backup options

Table 5-3 Primary VM identifier (continued)

Option Description

VM display name The name of the virtual machine as displayed in the VMware interface. A display name
is assigned to the virtual machine when the virtual machine is created.

When virtual machines are included in a NetBackup policy, restrictions apply to the
characters that are allowed in the virtual machine display name.

See “NetBackup character restrictions for the Primary VM identifier” on page 37.
Note: The restrictions also apply to other vSphere objects, such as floppy image
name, parallel port or serial port file name, and CD-ROM ISO name.

Each display name must be unique in your VMware environment.

See “NetBackup for VMware: notes and restrictions” on page 27.

See “Restoring individual filesRestoring individual VMware virtual machine files”

on page 224.

VM BIOS UUID The ID assigned to the virtual machine when the virtual machine is created. This ID
may or may not be unique, depending on whether the virtual machine has been
duplicated. This option is included for compatibility with the policies that use the older
VM UUID identifier.

VM DNS Name The VMware DNS Name of the virtual machine. In vSphere Client, this name appears
on the virtual machine's Summary tab.
Note: This name may or may not be associated with the virtual machine’s IP address.
VMware Tools obtains this name from the host name that is configured in the virtual
machine. For further information on this name, refer to the documentation for the guest
operating system.

VM instance UUID The globally unique ID assigned to the virtual machine when the virtual machine is
created. This ID uniquely identifies the virtual machine within a vCenter server. Even
if the virtual machine has been duplicated (such as within a vCloud), only the original
virtual machine retains this instance ID. (The virtual machine duplicates are assigned
different instance UUIDs.)

This option is recommended instead of the VM BIOS UUID option.

See “Backup options on the VMware tabVMware backup options” on page 73.

Existing snapshot handling options (VMware)

This option specifies the action that NetBackup takes when a snapshot is discovered
before NetBackup creates a new snapshot for the virtual machine backup. After it
creates a snapshot, NetBackup usually deletes the snapshot when the backup
 Configure NetBackup policies for VMware 78
Backup options on the VMware tabVMware backup options

completes. If snapshots are not automatically deleted (whether created by

NetBackup or not), the performance of the virtual machine may eventually decline.
Undeleted snapshots can cause restore failures due to lack of disk space. If the
virtual machine was configured on multiple datastores and a leftover snapshot
existed on the virtual machine when it was backed up, note: NetBackup tries to
restore all .vmdk files to the snapshot datastore. As a result, the datastore may not
have enough space for the .vmdk files, and the restore fails. (For a successful
restore, you can restore the virtual machine to an alternate location. Use the Storage
Destination dialog in the Virtual Machine Recovery dialogs to select a datastore
for the .vmdk files.)

Table 5-4 Existing snapshot handling: Options

Option Description

Continue backup NetBackup ignores any existing virtual machine snapshots (including snapshots
previously created by NetBackup) and proceeds with snapshot creation and the
backup.

Abort if any snapshot(s) exist If any snapshot exists on the virtual machine, NetBackup aborts the job for that
virtual machine only.

Remove NetBackup snapshot(s) If a virtual machine snapshot exists that a NetBackup backup previously created:
and continue backup NetBackup removes the old snapshot, creates an updated snapshot, and proceeds
with the virtual machine backup.

Abort if NetBackup snapshot(s) If a virtual machine snapshot exists that a NetBackup backup previously created,
exist NetBackup aborts the job for that virtual machine only.
(This option is the default.)

See “Backup options on the VMware tabVMware backup options” on page 73.

Transport modes options (VMware)

The transport modes determine how the snapshot data travels from the VMware
datastore to the VMware backup host. The appropriate mode depends in part on
the type of network that connects the VMware datastore to the VMware backup
host.
By default, all modes are selected. NetBackup tries each transport mode in order,
from top to bottom. It uses the first mode that succeeds for all disks in the virtual
machine.
 Configure NetBackup policies for VMware 79
Backup options on the VMware tabVMware backup options

Table 5-5 Transport Modes

Mode Description

san For unencrypted transfer over Fibre Channel (SAN) or iSCSI.

Note: This mode is not supported for the virtual machines that use VMware Virtual
Volumes (VVols).

See “Notes on VMware Virtual Volumes (VVols)” on page 33.

hotadd Lets you run the VMware backup host in a virtual machine.

See “Notes on the hotadd transport mode” on page 40.

Note: For the virtual machines that use VVols, the virtual machine and the backup host
(hotadd) virtual machine must reside on same VVol datastore.

For instructions on this transport mode and on installing the backup host in a VMware
virtual machine, refer to your VMware documentation.

nbd For unencrypted transfer over a local network that uses the Network Block Device (NBD)
driver protocol. This mode of transfer is usually slower than Fibre Channel.

nbdssl For encrypted transfer (SSL) over a local network that uses the Network Block Device
(NBD) driver protocol. This mode of transfer is usually slower than Fibre Channel.

Move Up, Move Down Use these options to change the order in which NetBackup tries each selected mode.
Highlight a mode and click Move Up or Move Down.

For example: assume that all four transport modes are selected, and the order is SAN,
hotadd, nbd, and nbdssl. If one of the virtual disks cannot be accessed using SAN, the
SAN transport mode is not used for any of the virtual machine’s disks. NetBackup then
tries to use the hotadd mode for all the disks. NetBackup continues to try each mode until
it finds one that succeeds for all the disks.

See “Backup options on the VMware tabVMware backup options” on page 73.

Application protection options (VMware)

To enable file-level recovery of database data that resides in the virtual machine,
select from the following options. These options apply to full backups of the virtual
machine; they do not apply to incremental backups.
 Configure NetBackup policies for VMware 80
Backup options on the VMware tabVMware backup options

Table 5-6 Application Protection options

Database option Description

Enable Exchange This option enables recovery of the Exchange databases or mailbox messages from the
Recovery virtual machine backups. If this option is disabled, you can recover the entire virtual machine
from the backup, but you cannot recover the databases or mailbox messages individually.

To use the Enable Exchange Recovery option, you must enable the Enable file recovery
from VM backup option. If you choose to exclude disks on the Exclude disks tab, ensure
that the disks do not contain Exchange Server data.

For more information, refer to the NetBackup for Exchange Server Administrator's Guide.

Truncate logs For Exchange, this option truncates the transaction logs when the snapshot of the virtual
machine is complete. Truncating the logs frees up disk space in the virtual machine.
Note: To use this option, install the Veritas VSS Provider on the virtual machine. For
instructions on how to install this provider, refer to the NetBackup for Exchange Server
Administrator's Guide.

If you do not select Truncate logs, the logs are not truncated.

Enable SQL Server This option enables recovery of individual files from SQL Server data in the virtual machine
Recovery backup. If this option is disabled, you can recover the entire virtual machine from the backup,
but you cannot recover the SQL Server files individually.

To use the Enable SQL Server Recovery option, you must enable the Enable file recovery
from VM backup option. If you choose to exclude disks on the Exclude disks tab, ensure
that the disks do not contain SQL Server data.

For more information, refer to the NetBackup for SQL Server Administrator's Guide.

Truncate logs For SQL Server, this option truncates the transaction logs when the backup occurs. Truncating
the logs frees up disk space in the virtual machine.
Note: To use this option, install the Veritas VSS Provider on the virtual machine. For
instructions on how to install this provider, refer to the NetBackup for SQL Server
Administrator's Guide.

If you do not select Truncate logs, the logs are not truncated.

Enable SharePoint This option enables recovery of SharePoint objects from the virtual machine backup. If this
Recovery option is disabled, you can recover the entire virtual machine from the backup, but you cannot
recover the SharePoint objects individually.

To use the Enable SharePoint Recovery option, you must enable the Enable file recovery
from VM backup option. If you choose to exclude disks on the Exclude disks tab, ensure
that the disks do not contain SharePoint Server data.

For more information, refer to the NetBackup for SharePoint Server Administrator's Guide.

See “Backup options on the VMware tabVMware backup options” on page 73.
 Configure NetBackup policies for VMware 81
Backup options on the VMware tabVMware backup options

VMware - Advanced Attributes dialog

This dialog appears when you click Advanced... on the VMware policy tab.
You can use the VMware Advanced Attributes dialog to set the following additional
parameters for VMware backup. In most situations, the best settings are the defaults.

Table 5-7 VMware Advanced Attributes

Configuration Description
parameter

Virtual machine This option is enabled by default. In the great majority of cases, you should accept the default.
quiesce
I/O on the virtual machine is quiesced before NetBackup creates the snapshot. Without quiescing
file activity, data consistency in the snapshot cannot be guaranteed. If not consistent, the
backed-up data may be of little or no value.

If this option is disabled, the snapshot is created without quiescing I/O on the virtual machine. In
this case, you must perform your own analysis for data consistency in the backup data.

Caution: Veritas does not recommend that you disable quiesce. In most cases, this option should
be enabled.
Note: To use this option, VMware Tools must be installed on the virtual machine.

Note: To use this option with Linux virtual machines, snapshot quiesce must be enabled in the
Linux guest OS.

Ignore diskless If this option is enabled:

VMs
NetBackup does not back up a replicated (passive) VM in a vCenter Site Recovery Manager
(SRM) environment if that VM has no vmdk files. NetBackup skips that VM and backs up the
corresponding active VM, which has vmdk files.

Note that virtual machines without vmdk files can occur in a vCenter SRM environment. If a
replicated virtual machine has never been active, it is in passive mode and may have no vmdk
file(s).
Note: If this option is enabled and NetBackup does not have access to the vCenter where the
active virtual machine runs: the Query Builder policies run without error; no attempt is made to
back up the virtual machine. For the policies that use manual selection of virtual machines,
backups fail with status 156, because the virtual machine cannot be located.

If this option is disabled:

NetBackup attempts to back up a virtual machine regardless of whether it has vmdk files. If the
virtual machine has no vmdk files, the backup fails with status 156.
 Configure NetBackup policies for VMware 82
Backup options on the VMware tabVMware backup options

Table 5-7 VMware Advanced Attributes (continued)

Configuration Description
parameter

Post vCenter Enables NetBackup to send backup related events to the vCenter server. The events appear in
events vSphere Client under Home > Inventory > Hosts and Clusters, Tasks & Events tab.

See “Viewing NetBackup activity in vSphere Web Client or vSphere Client (HTML5)” on page 155.
The following options are available:

■ All Events: NetBackup posts an event to the vCenter server on each backup success or
failure. This setting is the default.
■ No Events: Disables the Post vCenter events option. NetBackup does not post any events
to the vCenter server.
■ Error Events: NetBackup posts an event to the vCenter server only for backup failures.

Further information on Post vCenter events is available:

See “About the Post vCenter events option (VMware Advanced Attributes)” on page 85.

Multiple This option is disabled by default. If it is enabled, the query rules can select virtual machines from
organizations per different vCloud Director organizations and back them up to the same storage unit.
policy
If you do not want backups of virtual machines from different organizations to be stored on the
same drive, leave this option disabled.

Ignore Instant If this option is enabled (the default): NetBackup skips any virtual machine that was restored with
Recovery VMs Instant Recovery for VMware if the virtual machine is running from a NetBackup NFS datastore.

When the virtual machine data files have been migrated to the production datastore, the virtual
machine can be backed up.
NetBackup identifies Instant Recovery virtual machines according to the following criteria:

■ The virtual machine has a snapshot that is named NBU_IR_SNAPSHOT,

■ And the virtual machine is running from a datastore and the name of the datastore begins
with NBU_IR_.

If the virtual machine meets all these criteria, it is not backed up if this option is enabled.

If this option is disabled: NetBackup backs up the virtual machine even if it is running from the
NetBackup NFS datastore.

Snapshot retry Sets the number of times the snapshot is retried. The default is 10. The range is 0 to 100.
count
This option and the snapshot time-out and snapshot creation interval provide flexibility in the
creation of snapshots. For most environments, the default values are usually best. In special
circumstances, it may be helpful to adjust these settings. Example considerations are the size
of the virtual machine and the processing load on the VMware server.
 Configure NetBackup policies for VMware 83
Backup options on the VMware tabVMware backup options

Table 5-7 VMware Advanced Attributes (continued)

Configuration Description
parameter

Snapshot timeout Sets a time-out period (in minutes) for completion of the snapshot. The default is 0, which means
(minutes) no time-out.

If snapshots do not complete, set this option to a specific period to force a time-out. Consider
using the snapshot creation interval to retry the snapshot at a later time.

Snapshot Determines the wait time (in seconds) before the snapshot is retried. The default is 10 seconds.
creation interval The range is 0 to 3600.
(seconds)

VMware server Specifies a colon-delimited list of virtual machine servers that NetBackup communicates with for
list this policy. In large virtual environments, you can use this list to improve backup performance:
NetBackup communicates only with the servers that are in this list. For example, exclude from
the list any vCenter or vCloud servers that do not contain virtual machines to be backed up by
this policy. Also, if a duplicate of the virtual machine exists on a different server: the duplicate is
not backed up if it resides on a server that is not included in this list.

Important: IPv6 addresses are not supported in the VMware server list field. Use fully qualified
domain names or host names.
Note: Each host name must match exactly the name as configured in the NetBackup credentials.

Note: Separate the names with a colon (:) not a comma (,).

This option does not affect either of the following: the list of all possible values in the Query Builder
for automatic selection of virtual machines, or browsing of virtual machines for manual selection.

If the list is blank (the default), NetBackup communicates with any servers in the virtual
environment.
 Configure NetBackup policies for VMware 84
Backup options on the VMware tabVMware backup options

Table 5-7 VMware Advanced Attributes (continued)

Configuration Description
parameter

Treat Tags as Tags were introduced with VMware vCenter Version 5.1. The APIs to interface with the tagging
unset if unable to service were not released until VMware vCenter Version 6.0. NetBackup for VMware supports
evaluate the tags that are assigned to virtual machine objects starting with VMware vCenter Version 6.0.

If you have a mixed vCenter environment, such as 5.1, 5.5, and 6.0: you can use this configuration
parameter to modify how NetBackup treats tags for vCenter Version 5.1/5.5. This configuration
parameter also applies to all versions of ESXi hosts whose credentials were added to the Virtual
Machine Servers list.

This option is disabled by default. If you use the tag field in the VMware Intelligent Policy query
and your policy searches for virtual machines across a mixed vCenter environment, note:
NetBackup reports the virtual machines that are discovered from vCenter Server 5.1/5.5 and
ESXi hosts as failed if it needs to evaluate the tag portion of the query to make an include or
exclude decision.

When you enable this option NetBackup treats tags as unset. If you use the tag field in the
VMware Intelligent Policy query and your policy searches for virtual machines across a mixed
vCenter environment, note: NetBackup evaluates the tag part of the query as if no tags were set
on the virtual machines that are discovered from vCenter Server 5.1, 5.5, and ESXi.

Continue VIP
Note: This option applies to VMware Intelligent Policies (VIP) only.
discovery if one
vSphere login When this option is set to Yes: For a VIP policy's discovery job, NetBackup ignores a failed logon
fails to a vCenter and attempts to log on and discover VMs on other vCenters. On any vCenter that
NetBackup can log on to, the VMs that match the VIP policy's query are backed up.

When this option is set to No (the default): If the attempt to log on to a vCenter fails, the discovery
job fails and no VMs are backed up for any vCenters.

More information is available on the types of NetBackup jobs for VMware:

See “Using the Activity Monitor to monitor virtual machine backups” on page 153.

Perform snapshot This option is disabled by default. If it is enabled and a quiesced snapshot cannot be created,
without quiescing the snapshot is created without quiescing I/O on the virtual machine. The resulting snapshot is
if quiesced referred to as crash consistent. In this case, you must perform your own analysis for data
snapshots fail consistency in the backed-up data.
Warning: In most cases, Veritas does not recommend enabling this option. NetBackup cannot
guarantee that all required data has been flushed to disk when the snapshot occurs. The data
that is captured in the snapshot may be incomplete.

If this option is disabled, the backup fails if a quiesced snapshot cannot be created.
 Configure NetBackup policies for VMware 85
Backup options on the VMware tabVMware backup options

Table 5-7 VMware Advanced Attributes (continued)

Configuration Description
parameter

VMDK Used to specify the VMDK compression method. When this option is set to none (default)
Compression compression is not used during the backup job.

NetBackup uses the preferred VMDK compression method when it opens VMDKs. Backup jobs
automatically set the compression method to none if NetBackup is unable to read the VMDK
with the preferred method.

See “About the exclude disk options for virtual disk selection” on page 88.
See “Backup options on the VMware tabVMware backup options” on page 73.
See “Configuring a VMware policy from the Policies utility” on page 68.

About the Post vCenter events option (VMware Advanced Attributes)

The Post vCenter events option is available on the VMware - Advanced Attributes
dialog of a NetBackup policy. (On the VMware tab of the policy, click Advanced....)
Post vCenter events enables NetBackup to send backup related events to the
vCenter server.

Note: To post events to vCenter, NetBackup must perform the backup through a
vCenter server. If NetBackup accesses the ESX server directly, the backup
information cannot be displayed in vSphere Client.

Note: You must set the required permissions in vCenter:

See “Setting privileges for posting events to vCenter” on page 51.

Note: If a vSphere administrator created an attribute named NB_LAST_BACKUP

of type Global, NetBackup cannot post backup events to that attribute. You must
remove the NB_LAST_BACKUP attribute from vSphere. Make sure that Post
vCenter events is set to All Events or Error Events. At the next backup, NetBackup
creates a NB_LAST_BACKUP attribute of type Virtual Machine and posts events
to that attribute.

Post vCenter events also records the date and time of the last successful backup
of the virtual machine:
 Configure NetBackup policies for VMware 86
Exclude Disks tab

■ The date and time appear in vSphere Client on the Summary tab as a custom
attribute under Annotations. The attribute is labeled NB_LAST_BACKUP.
■ The date and time appear in vSphere Web Client on the Virtual Machines
display.
The events can also be viewed with the NetBackup vSphere Web Client plug-in
and the NetBackup vSphere Client (HTML5) plug-in:
See “Viewing NetBackup activity in vSphere Web Client or vSphere Client (HTML5)”
on page 155.
For instructions on installing and using the NetBackup plug-in, see the following
guides:
NetBackup Plug-in for VMware vSphere Web Client Guide
NetBackup Plug-in for VMware vSphere Client (HTML5) Guide
http://www.veritas.com/docs/DOC5332

Exclude Disks tab

In the Change Policy dialog box, the Exclude Disks tab appears for policies of
the VMware policy type. These options determine the kind of disks on the virtual
machine that are excluded from the backup. These options can reduce the size of
the backup, but should be used with care. These options are intended only for the
virtual machines that have multiple virtual disks.
The following options appear on the Exclude Disks tab.

Table 5-8 Options on the Exclude Disks tab of the policy

Option Description

No disks Backs up all virtual disks that are configured for the virtual machine.
excluded

Exclude boot disk The virtual machine's boot disk (for example the C drive) is not included
in the backup. Any other disks (such as D) are backed up. Consider
this option if you have another means of recreating the boot disk, such
as a virtual machine template for boot drives.

See “About the exclude disk options for virtual disk selection”
on page 88.
Note: A virtual machine that is restored from this backup cannot start.
Data files are available in the restored data disks.
 Configure NetBackup policies for VMware 87
Exclude Disks tab

Table 5-8 Options on the Exclude Disks tab of the policy (continued)

Option Description

Exclude all data The virtual machine's data disks (for example the D drive) are not
disks included in the backup for this policy. Only the boot disk is backed up.
Consider this option only if you have a different policy that backs up
the data disks.

See “About the exclude disk options for virtual disk selection”
on page 88.
Note: When the virtual machine is restored from the backup, the virtual
machine data for the data disk may be missing or incomplete.

Perform custom Exclude disks by a VMware Custom Attribute that is applied to a virtual
attribute based machine. The VMware Custom Attribute identifies the disks that you
exclusion want to exclude from backups. If you select this option, also enter the
name of the Custom Attribute. NetBackup then excludes the disks that
are defined in that attribute. The attribute must have comma separated
values of device controllers for the disks to be excluded. For example:

scsi0-0,ide0-0,sata0-0,nvme0-0

The default value is NB_DISK_EXCLUDE_LIST. You can use this value

as the custom attribute or choose your own value.
Note: Custom Attribute based disk exclusion requires that you enter
in NetBackup the credentials for the vCenter server or servers that host
the VMs. ESXi server credentials are not sufficient.

See “Adding NetBackup credentials for VMware” on page 45.

Your VMware administrator must use a VMware interface to apply the

custom attribute to the disks that you want to exclude from the virtual
server. The Virtual Disk Exclusion wizard of the NetBackup plug-ins
for vSphere provides a method to add a Custom Attribute to a virtual
machine or virtual machines. For more information, see the NetBackup
Plug-in for VMware vSphere Web Client Guide or the NetBackup Plug-in
for VMware vSphere Client (HTML5) Guide available through the
following URL:

http://www.veritas.com/docs/DOC5332
 Configure NetBackup policies for VMware 88
Exclude Disks tab

Table 5-8 Options on the Exclude Disks tab of the policy (continued)

Option Description

Specific disks to Exclude a specific disk by selecting the disk controller type and device
be excluded numbers that represent the virtual device node of the disk. Then click
Add. NetBackup adds the controller ID to the list of nodes to be
excluded. Repeat for each disk that you want to exclude.

To delete a disk from the list of disks to exclude, select the disk controller
type and device numbers, and then click Delete. The list is a
comma-separated collection of controller IDs.

Alternatively, you can add or delete specific disks by editing the contents
of the text box in which the excluded disks appear. Wildcard characters
are not supported.

Note: NetBackup does not support the exclude disks options for Replication Director
backups.

See “Exclude disks from backups: an example to avoid” on page 91.

See “Restoring data from the backups that excluded the boot disk or data disks”
on page 91.
For more information about VMware, see the NetBackup for VMware Guide.

About the exclude disk options for virtual disk selection

The backup policy Exclude Disks tab has options to exclude virtual disks from a
backup. By default, no disks are excluded. You should use this setting in most
cases.
If you want to exclude disks from a backup, the other options are Exclude boot
disk and Exclude all data disks. These options are intended for the virtual
machines that have multiple virtual disks. You should use these options with care.
To exclude a boot disk or data disk, note the following requirements:
■ The virtual machine must have more than one disk.
■ NetBackup must be able to identify the boot disk.
■ The boot disk must not be part of a managed volume (Windows LDM or Linux
LVM). The boot disk must be fully contained on a single disk.
The boot disk must include the following:
■ The boot partition.
■ The system directory (Windows system directory or Linux boot directory).
 Configure NetBackup policies for VMware 89
Exclude Disks tab

Important! The exclude disk options are meant only for the following cases:
■ Exclude boot disk: Consider this option if you have another means of recreating
the boot disk, such as a virtual machine template for boot drives. If Exclude
boot disk is enabled, the policy does not back up the boot disk.

Note: When the virtual machine is restored from the backup, the virtual machine
data for the boot disk may be missing or incomplete.

Note the following about Exclude boot disk:

■ If the virtual machine has a boot disk but has no other disks, the boot disk
is backed up. It is not excluded.
■ If the virtual machine's boot disk is an independent disk, but the virtual
machine has no other disks, the boot drive is backed up. The restored boot
drive however contains no data, because NetBackup cannot back up the
data in an independent disk.
■ If the virtual machine has a boot drive and an independent drive, the boot
drive is not backed up. Only the independent drive is included in the backup.
Since NetBackup cannot back up the data in an independent disk, the
restored independent disk contains no data.
Refer to the explanation on independent disks in the following topic:
See “NetBackup for VMware terminology” on page 20.
■ Adding a virtual disk and changing this option before the next backup can
have unexpected results.
See “Exclude disks from backups: an example to avoid” on page 91.

■ Exclude all data disks: Consider this option if you have a different policy or
other backup program that backs up the data disks. If Exclude all data disks
is enabled in a policy, that policy does not back up the data disks.
Note the following about excluding data disks:
■ If the virtual machine has only one disk (such as C:), that drive is backed
up. It is not excluded.
■ If the virtual machine's boot disk is an independent disk, and the virtual
machine has a separate data disk, the boot disk is backed up. The restored
boot disk however contains no data, because NetBackup cannot back up
the data in an independent disk.

Note: When the virtual machine is restored from the backup, the virtual machine
data for the data disk may be missing or incomplete.
 Configure NetBackup policies for VMware 90
Exclude Disks tab

■ Perform custom attribute based exclusion: If this option is enabled, NetBackup

excludes the disks that have a custom attribute from the backup. The default
value for this attribute is NB_DISK_EXCLUDE_DISK. You can use this default
attribute or change the attribute name on the Exclude Disks tab of the backup
policy. Note that this option gives the VMware administrator control over which
disks are excluded.

Note: When a virtual machine is restored from the backup, the virtual machine
data for the excluded disk may be missing or incomplete.

■ The attribute on the virtual machine must be populated with comma separated
values of controller IDs for the disks to be excluded.
■ If the custom attribute is not populated or does not exist on the virtual
machine, none of the disks (except independent disks) is excluded.
■ If you remove disks from the custom attribute value between the differential
backups, only those files that changed since the last backup are available
to restore individually. You can restore the entire virtual disk or the VM, in
which case all files are restored including those you cannot restore
individually. After the next full backup, you can restore any of the files
individually.
■ If you add disks to the custom attribute value between the differential
backups, those disks are excluded from the next backup.

■ Specific disk(s) to be excluded: If this option is enabled, NetBackup excludes

the disks that you specify. Note that this option gives the NetBackup administrator
control over which disks are excluded from backups.

Note: When a virtual machine is restored from the backup, the virtual machine
data for the excluded disk may be missing or incomplete.

■ You must include comma-separated values of controller IDs on the Exclude

Disks tab of the backup policy.
■ If the disks do not exist on the specified controller and device IDs, none of
the disks (except independent disks) is excluded.
■ If you remove controllers from the exclusion list between the differential
backups, only those files that changed since the last backup are available
to restore. All files are available to restore after the next full backup.
■ If you add controllers to the exclusion list between the differential backups,
their disks are excluded from the next backup.
 Configure NetBackup policies for VMware 91
Browse for VMware virtual machines

Caution: Use of the exclude disk options can have unintended consequences if
these rules are not followed.

The following topics contain important guidance on the exclude disk options:
See “Restoring data from the backups that excluded the boot disk or data disks”
on page 91.

Exclude disks from backups: an example to avoid

You should use the options on the Exclude Disks tab of the Change Policy dialog
box with care. For example, if you add a disk to the virtual machine and change
the settings that exclude disks, note: The next backup may not capture the virtual
machine in the state that you intended. You should back up the entire virtual machine
(that is, do not exclude any disks) before you exclude a disk from future backups.
See “VMware - Advanced Attributes dialog” on page 81.

Restoring data from the backups that excluded the boot disk or data
disks
If the policy's Excludes Disks option excluded the boot disk or data disks, you can
restore the backed-up data as follows:
■ If Enable file recovery from VM backup was enabled on the backup policy:
You can restore individual files from those portions of the virtual machine that
the Virtual disk selection option did not exclude.
See “Restoring individual filesRestoring individual VMware virtual machine files”
on page 224.
■ If the Excludes Disks option was set to Exclude boot disk: You can restore
the virtual machine and move the restored data disk(s) to another virtual machine.

Browse for VMware virtual machines

When you configure a NetBackup policy, you can use the Browse for Virtual
Machines screen to select virtual machines for backup. You can enter the virtual
machine's host name manually or browse and select it from a list.
As an alternative, NetBackup can automatically select virtual machines based on
a range of criteria.
See “About automatic virtual machine selection for NetBackup for VMware”
on page 102.
 Configure NetBackup policies for VMware 92
Browse for VMware virtual machines

■ Enter the VM display name (or VM hostname, VM BIOS UUID, VM DNS

Name, VM instance UUID)
Click this option if you want to manually enter the name or other identifier of a
virtual machine. The format of the name depends on your system. It may be the
fully qualified name or another name, depending on your network configuration
and how the name is defined in the guest OS. If NetBackup cannot find the
name you enter, policy validation fails.

Note: The type of name to enter depends on the Primary VM identifier setting
on the VMware tab of the policy.

To enter a name, make sure that Browse for Virtual Machines is not selected.
■ Browse for Virtual Machines
Click this option to discover vCenter servers, clusters, data centers, vApps, and
ESX servers (shown in the left pane). You can select virtual machines from a
list (in the right pane).
NetBackup uses the following icons for vSphere objects:

vCenter server

Datacenter

Cluster
 Configure NetBackup policies for VMware 93
Browse for VMware virtual machines

vApp

ESX server

Virtual machine (powered off)

Virtual machine (powered on)

Virtual machine template

The virtual machines that are contained in a VMware vApp are displayed in the
right pane along with other virtual machines.
The virtual machine names that are listed may be derived from a cache file. Use
of the cache file is faster than rediscovering the virtual machines on the network
if your site has a large number of virtual machines.
If NetBackup cannot obtain the IP address of the virtual machine, the host name
and IP address are displayed as NONE.
Note that virtual machine host names or display names must be unique within
a master server's policies:
See “Virtual machine host names and display names should be unique if VMs
are selected manually in the policy” on page 97.
■ Last Update
To update the cache file and re-display virtual machines, click the refresh icon
to the right of the Last Update field. This field shows the date and time of the
most recent cache file that contains the names of virtual machines.
For NetBackup to access the virtual machines, note the following:
■ The NetBackup master server must have credentials for the VMware vCenter
or ESX server(s).
See “Adding NetBackup credentials for VMware” on page 45.
■ DNS problems may prevent or slow down discovery of the virtual machines.
See “Preventing browsing delays caused by DNS problems” on page 297.
■ To limit the search to particular vCenter or ESX servers, you can create a
BACKUP registry entry as an exclude list. Excluding unneeded servers can
dramatically speed up the search for virtual machines.
 Configure NetBackup policies for VMware 94
Browse for VMware virtual machines

See “Limiting the VMware servers that NetBackup searches when browsing
for virtual machines” on page 95.
■ The browsing time out value must not be set too low.
See “Changing the browsing timeout for virtual machine discovery”
on page 299.

■ VM hostname, Display Name, DNS Name, Instance UUID, etc.

When you select an ESX server in the left pane, information about its virtual
machines appears in columns in the right pane.
You can change the arrangement of these columns by right-clicking in the right
pane:

The columns are the following:

VM hostname The host name of the virtual machine.

Display Name The display name of the virtual machine.

Instance UUID The UUID of the virtual machine.

BIOS UUID The BIOS UUID of the virtual machine.

Dns Name The VMware "DNS Name" of the virtual machine. In

vSphere Client, this name appears on the virtual machine's
Summary tab.

See “Primary VM identifier options (VMware)” on page 76.

IP Address The IP address of the virtual machine.

Operating System The guest OS system as defined when the virtual machine
was created.
 Configure NetBackup policies for VMware 95
Limiting the VMware servers that NetBackup searches when browsing for virtual machines

DataStore The datastore where the virtual machine configuration files

are stored. If the datastore is in a datastore cluster,
NetBackup displays the name of the datastore (not the
name of the cluster).

Powered The power on/off status of the virtual machine when it was
last discovered.

RDM Status Shows whether the virtual machine has a physical disk in
raw device mapping mode (RDM) or has an independent
drive.

Change Tracking Support Shows whether the virtual machine supports the VMware
Changed Block Tracking feature (for BLIB). Changed
Block Tracking requires ESX server 4.0 and a virtual
machine at vmx-07 or later.

Virtual Center Name The vCenter server that manages the ESX server (also
appears in the left pane).

Template Shows whether the virtual machine is a template.

The virtual machines you selected are listed in the Clients tab.

Limiting the VMware servers that NetBackup

searches when browsing for virtual machines
As part of creating a NetBackup policy, you must specify which virtual machines to
back up. One approach is to let NetBackup search the network and list all available
virtual machines. However, if your VMware environment contains many VMware
servers and virtual machines, it may take too long to search and list all of them. For
example, consider an environment with ten vCenter servers. To back up the virtual
 Configure NetBackup policies for VMware 96
Limiting the VMware servers that NetBackup searches when browsing for virtual machines

machines on one of the ten vCenter servers, browsing virtual machines on all ten
servers is unnecessary.
To speed up browsing, you can exclude particular VMware servers from the search.
When it probes the network for virtual machines, NetBackup queries only the
VMware servers that are not named in the exclude list for the backup-host.
Use one of the following procedures, depending on the platform of the backup host
(Windows or Linux).
To limit the VMware servers that NetBackup discovers, per Windows backup
host
1 On the Windows desktop of the backup host, click Start > Run and enter
regedit.

2 To be on the safe side, make a backup of the current registry (File > Export).
3 Go to HKEY_LOCAL_MACHINE > SOFTWARE > Veritas > NetBackup >
CurrentVersion > Config and create a key called BACKUP.
4 Right-click in the right pane and click New > String Value. Enter
excludeVMservers as the name.

5 Right-click the excludeVMservers name and click Modify.

6 In the Edit String dialog, enter a comma-delimited list of the VMware servers
that are NOT to be queried when NetBackup browses the network. Do not
enter spaces. You can enter vCenter servers and individual ESX servers.

Note: The exclude list is used on the next backup. If any bpfis processes are
running, the exclude list has no effect on them.
The exclude list applies only to this backup host. The servers are not queried
when NetBackup browses for virtual machines to list on the policy's Browse
for Virtual Machines dialog of the NetBackup Administration Console.
 Configure NetBackup policies for VMware 97
Virtual machine host names and display names should be unique if VMs are selected manually in the policy

To limit the VMware servers that NetBackup discovers, per Linux backup host
1 On the Linux backup host, create (or open) the following file:
/usr/openv/netbackup/virtualization.conf

2 Add the following to the file:

[BACKUP]
"excludeVMservers"="server1,server2,server3"

where server1,server2,server3 is a comma-delimited list of the VMware servers

that are NOT to be queried when NetBackup browses the network. Do not
enter spaces. You can enter vCenter servers and individual ESX servers.
Note: If the file already contains a [BACKUP] line, do not add another [BACKUP]
line. Any other lines that already exist under [BACKUP] should remain as they
are.
3 Save the file.
Note: The exclude list is used on the next backup. If any bpfis processes are
running, the exclude list has no effect on them.
The exclude list applies only to this backup host. The servers are not queried
when NetBackup browses for virtual machines to list on the policy's Browse
for Virtual Machines dialog of the NetBackup Administration Console.
See “Browse for VMware virtual machines” on page 91.
As an alternative, NetBackup can automatically select virtual machines based on
a range of criteria.
See “About automatic virtual machine selection for NetBackup for VMware”
on page 102.

Virtual machine host names and display names

should be unique if VMs are selected manually in
the policy
This topic applies to NetBackup for VMware.
Certain VMware environments do not require unique names for virtual machines.
For instance, virtual machines within a vCenter server can have the same host or
display names as virtual machines in another vCenter server. The same is true of
datacenters, which are logical groupings of virtual resources within a vCenter server.
Virtual machine host names or display names must be unique within a datacenter.
They do not need to be unique between two datacenters on the same vCenter. A
 Configure NetBackup policies for VMware 98
Primary VM identifier option and manual selection of virtual machines

virtual machine named VM1 can exist in datacenter A. Another virtual machine
(also named VM1) can exist in datacenter B, on the same vCenter server.
Identically named virtual machines can present a problem for any policies that are
configured as follows:
■ The master server's policies use the Select manually option on the Clients tab
to select the VMs for backup.
■ The Primary VM identifier option on the VMware tab identifies VMs by their
host names or display names.
These policies may back up a different but identically named VM, instead of the
VM that you selected. In that case, the VM that you selected is not backed up. For
these policies to work, the virtual machines' display names or host names must be
unique.
Consider the following options:
■ For manual policies that identify VMs by display name or host name, change
the VM names so that each VM has a unique host name or display name.
■ As an alternative, configure the policies' Primary VM identifier option to identify
the VMs by their UUIDs instead of by host name or display name. Use the type
of UUID that is appropriate for your virtual machine environment.
See “Primary VM identifier options (VMware)” on page 76.
■ Instead of policies with manual-selection, use VMware Intelligent policies to
select the VMs through a query. Even if the Primary VM identifier option is set
to host name or display name, NetBackup identifies each VM by its UUID.
See “Browse for VMware virtual machines” on page 91.

Primary VM identifier option and manual selection

of virtual machines
This topic applies to the manual selection of virtual machines on the NetBackup
Browse for Virtual Machines screen.
A change to the policy's Primary VM identifier value can affect backups. If you
change this option, you may have to delete the virtual machine selections on the
Clients tab and re-enter them. Otherwise, NetBackup may no longer be able to
identify the virtual machines to back up.
For example, the host names in the Clients tab cannot be used and the virtual
machines are not backed up in the following case:
■ If you change the Primary VM identifier from VM hostname to VM display
name, and
 Configure NetBackup policies for VMware 99
About incremental backups of virtual machines

■ The display names of the virtual machines are different from the host names.
In this case, delete the host name entries on the Clients tab and browse the network
to select the virtual machines by display name.
See “Browse for VMware virtual machines” on page 91.

Note: When creating virtual machines, use the same name for both host name and
display name. If the Primary VM identifier is changed, the existing entries on the
Clients tab still work.

If the policy selects virtual machines automatically, refer to the following topic:
See “Effect of Primary VM identifier parameter on Selection column in Test Query
results” on page 142.

About incremental backups of virtual machines

NetBackup enables full virtual machine and file-level incrementals in the same
backup (the Enable file recovery from VM backup option).
Better support for incremental backup is available when you use BLIB (Enable
block-level incremental backup). BLIB requires ESX 4.x and virtual machines at
vmx-07 or later.

Note: Individual file recovery is supported from full backups and from incremental
backups, as long as the Enable file recovery from VM backup policy option is
enabled.

Note: Make sure that the virtual machines to back up are time synchronized with
the backup host. Otherwise, some changed data may not be included in the backup,
depending on the clock differential between the backup host and the virtual machine.

Note: For incremental backups, you do not have to configure the client on the
VMware backup host for timestamps. The VMware policies automatically default
to the use of timestamps.

See “Configuring incremental backups” on page 100.

 Configure NetBackup policies for VMware 100
Configuring incremental backups

Configuring incremental backups

Use the following procedure for virtual machine backup.
More information is available on creating a policy.
See “Configuring a VMware policy from the Policies utility” on page 68.
To configure incremental backup of a virtual machine
1 In the NetBackup policy Attributes tab, select the VMware policy type.
2 On the VMware tab, select the VMware backup host.
3 Optional: for BLIB, select Enable block-level incremental backup.
4 Select Enable file recovery from VM backup.
More information is available on the VMware options.
See “Backup options on the VMware tabVMware backup options” on page 73.
5 Click OK.
6 On the Schedules tab, select Differential Incremental Backup or Cumulative
Incremental Backup.
7 Fill in the Clients tab.
See “About incremental backups of virtual machines” on page 99.

Storage Foundation Volume Manager volumes in

the virtual machine
To back up a virtual machine that contains Veritas Storage Foundation Volume
Manager volumes, make sure the following options on the policy's VMware tab are
disabled:
Enable file recovery from VM backup
Exclude deleted blocks.

Note: Restore of selected files from a backup of the full virtual machine is not
supported if the virtual machine contains Storage Foundation Volume Manager
volumes.

See “NetBackup for VMware: notes and restrictions” on page 27.

See “Configuring a VMware policy from the Policies utility” on page 68.
 Chapter 6
Configure a VMware
Intelligent Policy
This chapter includes the following topics:

■ About automatic virtual machine selection for NetBackup for VMware

■ Support and use of VMware tag associations

■ The basics of a NetBackup query rule

■ Important notes on automatic virtual machine selection

■ NetBackup requirements for automatic virtual machine selection

■ Automatic virtual machine selection: Task overview

■ Options for selecting VMware virtual machines

■ Configuring automatic virtual machine selection

■ Editing an existing query in Basic Mode

■ Using the Query Builder in Advanced Mode

■ AND vs. OR in queries

■ Examples for the NetBackup Query Builder

■ The IsSet operator in queries

■ About selecting virtual machines by means of multiple policies

■ Order of operations in queries (precedence rules)

■ Parentheses in compound queries

 Configure a VMware Intelligent Policy 102
About automatic virtual machine selection for NetBackup for VMware

■ Query rules for resource pools

■ Query rules for datacenter folders (host folder)

■ Query rules for duplicate names

■ Query rules for tags

■ Query Builder field reference

■ Test Query screen for VMware

■ Test Query: Failed virtual machines

■ Effect of Primary VM identifier parameter on Selection column in Test Query

results

■ Effect of Primary VM identifier parameter on VM Name column in Test query

results

■ Refreshing the display of virtual environment changes in the Query Builder

■ Reducing the time required for VM discovery in a large VMware environment

About automatic virtual machine selection for

NetBackup for VMware
Instead of manually selecting the virtual machines for backup, you can configure
NetBackup to automatically select virtual machines based on a range of criteria.
You specify the criteria (rules) in the Query Builder on the NetBackup policy Clients
tab. NetBackup creates a list of the virtual machines that currently meet the rules
and adds those virtual machines to the backup.
This feature is called the VMware Intelligent Policy.
Automatic selection of virtual machines has the following advantages:
■ Simplifies the policy configuration for sites with large virtual environments.
You do not need to manually select virtual machines from a long list of hosts:
NetBackup selects all the virtual machines that meet the selection rules in the
policy's Query Builder.
■ Allows the backup list to stay up-to-date with changes in the virtual environment.
Eliminates the need to revise the backup list whenever a virtual machine is
added or removed.
■ Virtual machine selection takes place dynamically at the time of the backup.
Examples of automatic virtual machine selection are the following:
 Configure a VMware Intelligent Policy 103
Support and use of VMware tag associations

Table 6-1 Examples for automatic virtual machine selection

Example Description

Add new virtual machines At the next backup, the policy can automatically
discover the virtual machines that have recently
been added to the environment. If the virtual
machines match the query rules that you configure
in the policy, they are automatically backed up.

Limit the backup list to the virtual If some of your virtual machines are occasionally
machines that are currently turned on turned off, NetBackup can be configured to
automatically exclude those from the backup list.
Among the virtual machines it discovers,
NetBackup backs up only the virtual machines that
are turned on.

Back up virtual machines based on Examples of physical boundaries are vCenter

physical boundaries servers, ESX servers, datastores, and clusters.
For example, a query rule can select all the virtual
machines in a particular ESX server, so the policy
backs up only those virtual machines.

Back up virtual machines based on Examples of logical boundaries are folders, vApps,
logical boundaries templates, and resource pools. For example, a
query rule can select all the virtual machines in a
particular folder, so the policy backs up only those
virtual machines.

Back up virtual machines based on NetBackup can include or exclude virtual machines
VMware tags based on the user assigned tags.

See “The basics of a NetBackup query rule” on page 104.

Support and use of VMware tag associations

NetBackup supports using VMware tags for virtual machine selection. You use this
feature when you configure VMware Intelligent Policies to protect virtual machines.
More information about this feature is available.
See “Notes and limitations for tag usage in VMware Intelligent Policy queries”
on page 41.
See “Query rules for tags” on page 128.
NetBackup also supports the backup and restore of VMware tag associations with
virtual machines when you use VMware Intelligent Policies. The tag association
metadata for all tags that are associated with a virtual machine are backed up with
 Configure a VMware Intelligent Policy 104
The basics of a NetBackup query rule

that virtual machine. If those tags exist on the vCenter Server, they are recreated
when the virtual machine is restored. More information about this feature is available.
See “Notes and limitations for the backup and restore of VMware tag associations”
on page 42.
See “How NetBackup handles VMware tag associations at restore” on page 230.

The basics of a NetBackup query rule

For automatic virtual machine selection, NetBackup uses query rules to determine
which VMware virtual machines to select for backup. You create the rules in the
Query Builder, on the Clients tab of the policy.
A query rule consists of the following:
■ A keyword, such as Displayname or Datacenter (many keywords are available).
For example: For automatic selection of the virtual machines with the display
names that contain certain characters, you need the Displayname keyword in
the rule.
■ An operator, such as Contains, StartsWith, or Equal.
The operator describes how NetBackup analyzes the keyword. For example:
Displayname StartsWith tells NetBackup to look for the display names that
start with particular characters.
■ Values for the keyword.
For the Displayname keyword, a value might be "prod". In that case, NetBackup
looks for the virtual machines that have the display names that include the
characters prod.
■ An optional joining element (AND, AND NOT, OR, OR NOT) to refine or expand
the query.
The policy uses these elements to discover and select virtual machines for backup.
Table 6-2 contains the examples of rules.

Table 6-2 Examples of rules

Rule Description

Displayname Contains "vm" NetBackup selects the virtual machines that have the
characters vm anywhere in their display names.

Displayname EndsWith "vm" NetBackup selects the virtual machines that have the
characters vm at the end of their display names.
 Configure a VMware Intelligent Policy 105
Important notes on automatic virtual machine selection

Table 6-2 Examples of rules (continued)

Rule Description

Datacenter AnyOf NetBackup selects the virtual machines that use

"datacenter1","datacenter2" datacenter1 or datacenter2.

Powerstate Equal poweredOn NetBackup selects only the virtual machines that are
currently turned on.

Powerstate Equal poweredOn NetBackup selects only virtual machines that are
AND Tag Equal "Production" currently powered on with the "Production" tag.

Many topics are available on automatic selection of VMware virtual machines. To

begin, consider the following topics.
See “Important notes on automatic virtual machine selection” on page 105.
See “Automatic virtual machine selection: Task overview” on page 108.
See “Configuring automatic virtual machine selection” on page 112.
See “AND vs. OR in queries” on page 117.
See “Query Builder field reference” on page 129.

Important notes on automatic virtual machine

selection
The Virtual Machine Intelligent Policy feature in NetBackup is a different approach
to VMware virtual machine selection in the policy. It represents a paradigm shift in
the way you select virtual machines for backup. As with all major changes, the
effective use of this feature requires forethought, preparation, and care.
 Configure a VMware Intelligent Policy 106
Important notes on automatic virtual machine selection

Table 6-3 Important notes on automatic virtual machine selection!

Note! Explanation

Create rules carefully.... Instead of manually selecting virtual machines for backup, you create guidelines for
automatic selection of virtual machines. The guidelines are called rules; you enter the
rules in the policy's Query Builder.

You make the rules, and NetBackup follows them.

If the rules state: Back up all virtual machines with a host name that contains "prod",
NetBackup does that. Any virtual machine that is added to the environment with a host
name containing "prod" is automatically selected and backed up when the policy runs.
Virtual machines with the names that do not contain "prod" are not backed up. To have
other virtual machines automatically backed up, you must change the query rules (or
create additional policies).

Changes to the virtual If many virtual machines are temporarily added to your environment and happen to fall
environment can affect within the scope of the query rules, they are backed up. The backups can therefore
backup times. run much longer than expected.

Test the query rules. Test the query rules ahead of time. The policy includes a Test Query function for that
purpose. It's important to verify that your query operates as expected. Otherwise, the
query may inadvertently select too many or too few virtual machines.

As an alternative, you can use the nbdiscover command to test a query. Refer to
the NetBackup Commands Reference Guide.

Note also: The policy's Primary VM identifier parameter can affect the automatic
selection process.

See “Effect of Primary VM identifier parameter on Selection column in Test Query

results” on page 142.
 Configure a VMware Intelligent Policy 107
Important notes on automatic virtual machine selection

Table 6-3 Important notes on automatic virtual machine selection!

(continued)

Note! Explanation

A query test does not create The automatic selection process is dynamic. Changes in the virtual environment may
the backup list. NetBackup affect which virtual machines the query rules choose when the backup runs.
creates the backup list when
Note: If virtual machine changes occur, the virtual machines that are selected for
the backup runs.
backup may not be identical to those listed in your query test results.

Backup
list

Query test

Backup
list
Backup
execution

The policy does not display a If you select virtual machines manually (with the Browse for Virtual machines screen),
list of the virtual machines the selected virtual machines are listed on the policy Clients tab. But when you use
that are to be backed up. the Query Builder for automatic selection, the selected virtual machines are not listed
on the Clients tab.
Use the Activity Monitor or
OpsCenter. For a list of the backed up virtual machines, use the NetBackup Activity Monitor or the
OpsCenter web interface.

See “Using the Activity Monitor to monitor virtual machine backups” on page 153.

See “Using OpsCenter to report on virtual machine backups” on page 157.

 Configure a VMware Intelligent Policy 108
NetBackup requirements for automatic virtual machine selection

Table 6-3 Important notes on automatic virtual machine selection!

(continued)

Note! Explanation

When you save the policy, the When you save a policy, policy validation does not consult the query rules and select
query rules are not validated. virtual machines for backup. Because of the potential for changes in the virtual
environment, virtual machine selection must wait until the backup runs. As a result,
when you save the policy, NetBackup does not check the policy attributes against a
backup list. If the query rules select the virtual machines that are incompatible with a
policy attribute, policy validation cannot flag that fact. The incompatibility becomes
apparent when NetBackup determines the backup list at the time of the backup.
Policy Take for example a policy that is configured for Enable block-level incremental
backup (BLIB). BLIB works only with ESX 4.0 virtual machines at version vmx-07 or
later. If the query rules select a virtual machine at a version earlier than vmx-07, the
policy cannot back up that virtual machine. The mismatch between the policy and the
virtual machine is revealed when the backup runs, not when the policy is validated.
The Activity Monitor's job details log indicates which virtual machines can or cannot
be backed up.

See special requirements Certain requirements apply.

See “NetBackup requirements for automatic virtual machine selection” on page 108.

NetBackup requirements for automatic virtual

machine selection
Note the following requirements for automatic selection of VMware virtual machines:
■ The system where the NetBackup Administration Console runs must have access
to the vCenter server.
■ Automatic virtual machine selection requires no additional license beyond the
NetBackup Enterprise Client license.
■ Automatic virtual machine selection is required for backups in vCloud Director
or for Replication Director for VMware.
See “The basics of a NetBackup query rule” on page 104.

Automatic virtual machine selection: Task

overview
This topic is a high-level overview of how to set up a NetBackup policy for automatic
selection of VMware virtual machines. Follow the links in the table for more details.
 Configure a VMware Intelligent Policy 109
Options for selecting VMware virtual machines

Table 6-4 Automatic selection of virtual machines: overview of the tasks

Steps to configure automatic Description and notes

selection

Configure a VMware policy Use the policy Attributes tab.

See “Configuring a VMware policy from the Policies utility” on page 68.

Set rules for virtual machine selection On the policy Clients tab, click Select automatically through VMware
in the policy Query Builder Intelligent Policy query.

Choose a host for virtual machine selection (default is the VMware backup
host).

To add rules, use the Query Builder drop-down fields.

See “Configuring automatic virtual machine selection” on page 112.

See “Options for selecting VMware virtual machines” on page 109.

Test the rules Click Test Query in the Query Builder on the Clients tab. Virtual machines
are labeled as included or excluded, based on the rules.
Note: The list of virtual machines is not saved in the Clients tab.

Note: The query rules are also displayed in the Backup Selections tab. The
backup selections are pre-set to All_LOCAL_DRIVES (not displayed).

As an alternative, you can use the nbdiscover command to test a query.

Refer to the NetBackup Commands Reference Guide.

Execute a backup When the policy executes, NetBackup consults the rules in the Query Builder,
creates a list of virtual machines, and backs them up.

Check the backup To see which virtual machines were backed up, use the Activity Monitor, or
run a Virtual Client Summary report in OpsCenter.

See “Using the Activity Monitor to monitor virtual machine backups” on page 153.

See “Using OpsCenter to report on virtual machine backups” on page 157.

Options for selecting VMware virtual machines

This topic describes the options on the policy Clients tab, or in the Backup Policy
Wizard.
You can use these options to manually select virtual machines, or to configure
NetBackup to select virtual machines automatically. For automatic selection, you
specify the selection criteria (rules) in the policy's Query Builder. When the backup
job runs, NetBackup discovers the virtual machines that currently meet the criteria
and backs up those virtual machines.
 Configure a VMware Intelligent Policy 110
Options for selecting VMware virtual machines

A procedure is available.
See “Configuring automatic virtual machine selection” on page 112.

Table 6-5 Virtual machine selection

Option Description

Select manually Click this option and click New to manually enter virtual machines names, or to browse and
select them from a list.

See “Browse for VMware virtual machines” on page 91.

Note: The rest of the fields and options are for automatic selection of virtual machines.

Select automatically Click this option to allow NetBackup to automatically select virtual machines for backup based
through VMware on the rules that you enter in the Query Builder.
Intelligent Policy
query

Enable vCloud Enables backup of the virtual machines that reside in a vCloud environment. Requires the
Director integration automatic selection of virtual machines.

With this option, the policy selects for backup only the virtual machines that vCloud manages:
it skips the virtual machines that are not in vCloud.

NetBackup host to This host discovers virtual machines and automatically selects them for backup based on
perform automatic your query rules. The resulting list determines which virtual machines are backed up.
virtual machine
To designate your media servers as discovery hosts, select Backup Media Server from the
selection
pull-down.

Table 6-6 Query Builder

Option Description

Query Builder (Join, Use these pull-down fields to define rules for automatic selection of virtual machines. From
Field, Operator, left to right, each pull-down further defines the rule.
Values)
Click the plus sign to add the rule to the Query pane.

Click the reset icon (curved arrow) to blank out the pull-down fields.

See “Query Builder field reference” on page 129.

See “Examples for the NetBackup Query Builder” on page 118.

Advanced Mode Places the Query Builder in Advanced Mode for manual entry of rules.

See “Using the Query Builder in Advanced Mode” on page 116.

See “Query Builder field reference” on page 129.

See “Examples for the NetBackup Query Builder” on page 118.

 Configure a VMware Intelligent Policy 111
Options for selecting VMware virtual machines

Table 6-6 Query Builder (continued)

Option Description

Basic Mode Returns the Query Builder from Advanced Mode to Basic Mode.

See “Query Builder field reference” on page 129.

Edit Use this option to change an existing query rule when in Basic Mode, as follows:

■ Click the rule and then click Edit.

■ Make new selections in the Query Builder pull-down fields.
■ Click the save option (diskette icon).

Remove Deletes a query rule when in Basic Mode. Click on the rule and then click Remove.

Test Query... Click this option to test which virtual machines NetBackup selects based on the rules in the
Query Builder.
Note: This test option does not create the backup list for the policy. When the next backup
runs from this policy, NetBackup re-discovers virtual machines and consults the query rules.
At that time, NetBackup backs up the virtual machines that match the rules.

See “Test Query screen for VMware” on page 139.

As an alternative, you can use the nbdiscover command to test a query. Refer to the
NetBackup Commands Reference Guide.

Reuse VM selection Sets the refresh rate of an internal cache of the query results. NetBackup uses the cache to
query results for: determine which virtual machines to select at the time of the backup. The cache speeds up
the selection of virtual machines without burdening the vCenter server at each scheduled
backup.

A faster cache refresh rate synchronizes the cache with the changes recorded in vCenter,
such as the addition or removal of virtual machines. However, each cache refresh consumes
vCenter resources.

With a slower refresh rate, new virtual machines may not be included immediately in the next
backup. New or changed virtual machines are included when the cache is refreshed. Note
that fewer vCenter resources are consumed with a slower refresh rate.

The default is 8 hours. For 8 hours, NetBackup uses the cache and does not attempt to
rediscover virtual machines. Changes to the virtual environment do not affect the cache
during that period. After 8 hours, the next execution of the policy causes NetBackup to
rediscover virtual machines. If any changes match a rule in the query, the list of selected
virtual machines is modified accordingly.
Note: The cache is refreshed before the next scheduled backup whenever the policy is
changed and saved.

More information is available on the Reuse VM selection query results for: option:

See “About the Reuse VM selection query results option” on page 373.
 Configure a VMware Intelligent Policy 112
Configuring automatic virtual machine selection

Configuring automatic virtual machine selection

This topic applies to policy configuration for NetBackup for VMware.
NetBackup can automatically select VMware virtual machines for backup based on
the criteria that you enter. You specify the criteria (rules) in the Query Builder on
the NetBackup policy Clients tab. You can set up rules to include certain virtual
machines for backup, or to exclude virtual machines.
When the backup job runs, NetBackup creates a list of the virtual machines that
currently meet the query rules and backs them up.
See “About automatic virtual machine selection for NetBackup for VMware”
on page 102.
The following is the policy Clients tab with Select automatically through VMware
Intelligent Policy query selected.

Figure 6-1 Policy Clients tab for automatic selection of virtual machines

The Query Builder can operate in Basic Mode or in Advanced Mode.

 Configure a VMware Intelligent Policy 113
Configuring automatic virtual machine selection

To configure automatic virtual machine selection in Basic Mode

1 On the policy Attributes tab, select VMware for the policy type.
2 On the policy VMware tab, select a VMware backup host.
Review the other options on the VMware tab.
See “Backup options on the VMware tabVMware backup options” on page 73.
3 Make other policy selections as needed (for example, create a Schedule).
4 Click the Clients tab, and click Select automatically through VMware
Intelligent Policy query.
If you selected virtual machines manually in the Browse for Virtual Machines
dialog, those virtual machines are removed from the policy.
5 To back up virtual machines in vCloud Director, click Enable vCloud Director
integration.
More information is available on backups and restores into vCloud.
See “About NetBackup for vCloud Director” on page 263.

Note: Enable vCloud Director integration makes several vCloud Director

keywords available in the policy Query Builder Field, for rule-based selection
of virtual machines. If this option is not selected, NetBackup cannot use the
vCloud keywords to locate virtual machines in vCloud Director, and the backup
fails.
 Configure a VMware Intelligent Policy 114
Configuring automatic virtual machine selection

6 To create a rule, make selections from the pull-down menus.

For the first rule, you can start with the Field pull-down, depending on the type
of rule. (For the first rule, the only selections available for the Join field are
blank (none), or NOT.)

Then make a selection for Operator:

For the Value(s) field: Click the folder icon to browse for values, enter the value
manually, or in some cases use the Value(s) drop-down (as in this example).
The characters you enter manually in the Values field must be enclosed in
single quotes or double quotes. Note that browsing for values may take some
time in large virtual environments.
See “Query Builder field reference” on page 129.

7 Click the plus sign to add the rule to the Query pane.
 Configure a VMware Intelligent Policy 115
Editing an existing query in Basic Mode

8 Create more rules as needed.

See “Query Builder field reference” on page 129.
See “Examples for the NetBackup Query Builder” on page 118.
9 To see which virtual machines NetBackup currently selects based on your
query, click Test Query.
On the Test Query screen, the Virtual machines in your current environment
that match the rules for selection in the policy are labeled INCLUDED. Note
however that the Test Query option does not create the backup list for the
policy. When the next backup runs from this policy, NetBackup re-discovers
virtual machines and consults the query rules. At that time, NetBackup backs
up the virtual machines that match the query rules.
The list of virtual machines is saved but the virtual machines are not displayed
in the policy's Clients tab.
See “Test Query screen for VMware” on page 139.
10 You can specify how long NetBackup uses the latest query results as the
backup list for future executions of the policy. Set the time period in Reuse VM
selection query results for.
The default is 8 hours.
More information is available on this setting:
See “About the Reuse VM selection query results option” on page 373.
11 To create queries manually instead of using the pull-down menus, click
Advanced Mode.
See “Using the Query Builder in Advanced Mode” on page 116.

Editing an existing query in Basic Mode

This topic applies to policy configuration for NetBackup for VMware. You can use
the Query Builder to enter rules for the automatic selection of VMware virtual
machines for backup.
To edit an existing query in Basic Mode
1 Click on the query rule you want to change and click Edit.
2 Make selections in the pull-down menus.
3 Click the save option (diskette icon).
4 To delete a rule, click on it and click Remove.
See “Configuring automatic virtual machine selection” on page 112.
 Configure a VMware Intelligent Policy 116
Using the Query Builder in Advanced Mode

See “Using the Query Builder in Advanced Mode” on page 116.

Using the Query Builder in Advanced Mode

This topic applies to policy configuration for NetBackup for VMware. You can use
the Query Builder to enter rules for the automatic selection of VMware virtual
machines for backup.
The Query Builder's Advanced Mode provides more flexibility in crafting rules for
virtual machine selection, including the use of parentheses for grouping.
To use the Query Builder in Advanced Mode
1 Set up a VMware policy and specify a VMware backup host or backup media
server.
For assistance, you can refer to the first few steps of the following procedure:
See “Configuring automatic virtual machine selection” on page 112.
2 Click the Clients tab.
3 Click Select automatically through VMware Intelligent Policy query.
4 Beneath the Query Builder pane, click Advanced Mode.
5 You can use the Query Builder drop-down menus to add query rules. You can
also type in rules manually.
Here are a few example queries:
VMFolder Contains "mango"

Datastore StartsWith "Acc" OR Datastore StartsWith "Prod"

vCenter Contains "ROS" AND ESXserver Equal "VM_test1" AND

Powerstate Equal poweredOn
 Configure a VMware Intelligent Policy 117
AND vs. OR in queries

6 To insert a rule between existing rules, place the cursor where you want the
new rule to start and type it in.
When you create a rule with the drown-down menus, it appears at the end of
the query. You can cut and paste it into the proper location.
7 To establish the proper order of evaluation in compound queries, use
parentheses to group rules as needed. Compound queries contain two or more
rules, joined by AND, AND NOT, OR, or OR NOT.
More information is available on the use of parentheses and on the order of
precedence.
See “AND vs. OR in queries” on page 117.
See “Order of operations in queries (precedence rules)” on page 122.
See “Parentheses in compound queries” on page 124.

AND vs. OR in queries

This topic applies to policy configuration for NetBackup for VMware. You can use
the Query Builder to enter rules for the automatic selection of VMware virtual
machines for backup.
The Join field in the Query Builder provides connectors for joining rules (AND, AND
NOT, OR, OR NOT). The effect of AND versus OR in the Query Builder may not
be obvious at first glance.
In essence, AND and OR work in this way:
■ AND limits or restricts the scope of the query.
■ OR opens up the query to an additional possibility, expanding the scope of the
query.
Note: Do not use AND to join the rules that are intended to include additional virtual
machines in the backup list. For instance, AND cannot be used to mean "include
virtual machine X AND virtual machine Y."
For example: To include the virtual machines that have either "vm1" or "vm2" in
their names, use OR to join the rules:

Displayname Contains "vm1"

OR Displayname Contains "vm2"

If you use AND to join these rules:

Displayname Contains "vm1"

AND Displayname Contains "vm2"
 Configure a VMware Intelligent Policy 118
Examples for the NetBackup Query Builder

the result is different: the backup list includes only the virtual machines that have
both vm1 and vm2 in their names (such as "acmevm1vm2"). A virtual machine with
the name "acmevm1" is not included in the backup.
Table 6-7 provides the examples with AND and OR.

Table 6-7 Queries with AND, OR

Query Description

Displayname Contains "vm1" This query selects any virtual machine that has either vm1 or vm2 in its
display name. For example, this query selects both "seabizvm1" and
OR Displayname Contains "vm2"
"seabizvm2" for backup.

vCenter Equal "vCenterServer_1" This query is very specific. Virtual machines with the host names that contain
"manu" are included in the backup only if: they reside in vCenter server
AND Datacenter Equal "dc_A"
"vCenterServer_1", datacenter "dc_A", and ESX server "prod".
AND ESXserver Equal "prod"
The virtual machines that do not reside in that hierarchy are not included.
AND VMHostName Contains "manu" For example: if a virtual machine resides in "vCenterServer_1" and datacenter
"dc_A", but not in ESX server "prod", that virtual machine is not included.

vCenter Equal "vCenterServer_1" This query uses the same keywords and values, but combines them with
OR. The result may be a much larger list of virtual machines.
OR Datacenter Equal "dc_A"
A virtual machine that meets any of these rules is included:
OR ESXserver Equal "prod"
■ Any virtual machines in vCenter "vCenterServer_1". Their host names,
OR VMHostName Contains "manu"
datacenter, or ESX server do not matter.
■ Any virtual machines in datacenter "dc_A". Their host names or server
do not matter.
■ Any virtual machines in ESXserver "prod". Their host names, datacenter,
or vCenter server do not matter.
■ Any virtual machines with a host name that contains "manu". Their server
or datacenter do not matter.

See “Order of operations in queries (precedence rules)” on page 122.

See “Parentheses in compound queries” on page 124.
See “Examples for the NetBackup Query Builder” on page 118.

Examples for the NetBackup Query Builder

This topic applies to policy configuration for NetBackup for VMware. You can use
the Query Builder to enter rules for the automatic selection of VMware virtual
machines for backup.
The following table provides example query rules.
 Configure a VMware Intelligent Policy 119
Examples for the NetBackup Query Builder

To use the Query Builder, you must click Select automatically through VMware
Intelligent Policy query on the Clients tab.
Another topic is available on the difference between AND and OR in a query.
See “AND vs. OR in queries” on page 117.

Table 6-8 Query Builder examples

Example query Query result when backup job executes

No query rules specified (Query pane is empty) All virtual machines are added to the backup list. Exceptions are
those that do not have a host name, or that have invalid characters
in the display name.

See “Effect of Primary VM identifier parameter on Selection column

in Test Query results” on page 142.

Displayname Contains "prod" All virtual machines with the display names that contain the string
"prod" are added to the backup list.

See “Effect of Primary VM identifier parameter on Selection column

in Test Query results” on page 142.

powerstate Equal "poweredOn" Any virtual machine that is turned on is added to the backup list.

VMGuestOS Equal "windows7Guest" All virtual machines with a guest OS of Windows 7 are added to the
backup list.

DisplayName AnyOf "grayfox7”,"grayfox9" The virtual machines named "grayfox7”and "grayfox9" are added to
the backup list. (Note that each value must be enclosed in its own
quotes, with a comma in between.)

powerstate Equal "poweredOn" In datastore Storage_1: any virtual machine that is turned on and
has a guest OS of Red Hat Linux 4 is added to the backup list.
AND Datastore Equal "Storage_1"

AND VMGuestOS Equal "rhel4Guest"

vCenter Equal "vCenterServer_1" In the vCenter server vCenterServer_1, virtual machines that are in
ESX servers with names containing "prod" are added to the backup
AND ESXserver Contains "prod"
list.

Cluster Equal "VMcluster_1" In cluster VMcluster_1, all virtual machines with the host names that
contain "Finance", in ESX servers ESX_1, ESX_2, ESX_3, are added
AND ESXserver AnyOf
to the backup list.
"ESX_1","ESX_2","ESX_3"

AND VMHostName Contains "Finance"

VMFolder StartsWith "Prod" For any folder whose name starts with "Prod" or whose name is not
"VM_test", add its virtual machines to the backup list.
OR VMFolder NotEqual "VM_test"
 Configure a VMware Intelligent Policy 120
The IsSet operator in queries

Table 6-8 Query Builder examples (continued)

Example query Query result when backup job executes

Examples with IsSet See “The IsSet operator in queries” on page 120.

Datacenter Contains "prod" AND Tag Equal This query selects any virtual machine where the Datacenter contains
"Finance" "prod" and the user-specified tag is "Finance".

Datacenter Equal "prod" AND NOT Tag Equal Selects virtual machines where Datacenter is "prod" but excludes
"Test" any virtual machines that have the user-specified tag "Test".

Click Advanced to see the query rule in Advanced Mode. Only Advanced Mode
supports the use of parentheses for grouping sets of rules.
See “Using the Query Builder in Advanced Mode” on page 116.
See “Configuring automatic virtual machine selection” on page 112.

The IsSet operator in queries

This topic applies to policy configuration for NetBackup for VMware. You can use
the Query Builder to enter rules for the automatic selection of VMware virtual
machines for backup.
In a query, you can use the IsSet operator to ensure that certain virtual machines
are included or excluded from the backup.
For example: if the Primary VM identifier parameter is set to VM hostname,
NetBackup is unable to identify virtual machines for backup that do not have a host
name. You can use IsSet to exclude such virtual machines from the backup list.

Table 6-9 Examples of queries with the IsSet operator

Query rules with IsSet operator Effect of the query on virtual machine selection

Cluster Contains "dev" INCLUDED: Any virtual machine in a cluster that has a name that contains
the string "dev" if the virtual machine also has a VMware DNS name.
AND VMDNSName IsSet
EXCLUDED: Any virtual machines that do not have a VMware DNS Name.

Without VMDNSName IsSet in this query, virtual machines without a DNS

name cannot be excluded. They would be listed as FAILED.
 Configure a VMware Intelligent Policy 121
About selecting virtual machines by means of multiple policies

Table 6-9 Examples of queries with the IsSet operator (continued)

Query rules with IsSet operator Effect of the query on virtual machine selection

Displayname Contains "prod" INCLUDED: Any virtual machine with a display name that contains the string
"prod" if the virtual machine also has a host name.
AND VMHostName IsSet
EXCLUDED: Any virtual machines that do not have host names.

Without VMHostName IsSet in this query, virtual machines without a host

name cannot be excluded. They would be listed as FAILED.

The policy's Primary VM identifier parameter has an important effect on which

virtual machines NetBackup can back up. This parameter affects the test query
results.
See “Effect of Primary VM identifier parameter on Selection column in Test Query
results” on page 142.

About selecting virtual machines by means of

multiple policies
This topic applies to policy configuration for NetBackup for VMware. You can use
the Query Builder to enter rules for the automatic selection of VMware virtual
machines for backup.
If your virtual environment has many virtual machines with inconsistent naming
conventions, you may need multiple policies working in tandem. It may be difficult
to create a single policy that automatically selects all the virtual machines that you
want to back up.
For this situation, configure several policies such that each policy backs up a portion
of the environment. One policy backs up a particular set or group of virtual machines,
such as those that have host names. A second policy backs up a different group
of virtual machines that were not backed up by the first policy, and so forth. When
all the policies have run, all the virtual machines are backed up.
The following table describes the policies that are designed to back up the virtual
environment in three phases. Note that each policy relies on a different setting for
the Primary VM identifier parameter.
 Configure a VMware Intelligent Policy 122
Order of operations in queries (precedence rules)

Table 6-10 Three policies that back up virtual machines in phases

Policy Query Builder rules Backup result

First policy VMHostName IsSet This policy backs up all virtual machines that have a
host name. Any virtual machines that do not have a
Primary VM identifier
host name are excluded from the backup.
parameter: VM hostname

Second policy NOT VMHostName IsSet This policy backs up all virtual machines that do not
have a host name but that do have a valid display name.
Primary VM identifier AND VMHasVDSName Equal
Any virtual machines that do not have a host name or
parameter: VM display name 'TRUE'
a valid display name are excluded from the backup.

See “NetBackup character restrictions for the Primary

VM identifier” on page 37.

Third policy NOT VMHostName IsSet This policy backs up the virtual machines that were not
backed up by the first two policies. This policy selects
Primary VM identifier AND NOT VMHasVDSName
the virtual machines that do not have a host name or a
parameter: VM UUID Equal 'TRUE'
valid display name, but that do have a UUID.

More information is available on the Primary VM identifier parameter and its effect
on virtual machine selection.
See “Effect of Primary VM identifier parameter on Selection column in Test Query
results” on page 142.
See “The basics of a NetBackup query rule” on page 104.

Order of operations in queries (precedence rules)

This topic applies to policy configuration for NetBackup for VMware. You can use
the Query Builder to enter rules for the automatic selection of VMware virtual
machines for backup.
The information in this topic is for advanced users who understand precedence in
programming languages. In the Query Builder, the order in which operations occur
can determine which virtual machines are selected and backed up.
The following table lists the order of operations, or precedence, from highest to
lowest (7 is the highest). For example, an operation with a precedence of 6 (such
as Contains) is evaluated before an operation with a precedence of 5 (such as
Greater).
 Configure a VMware Intelligent Policy 123
Order of operations in queries (precedence rules)

Table 6-11 Order of operations

Operation Description Precedence

!x Produces the value 0 if x is true (nonzero) and 7

the value 1 if x is false (0).

x Contains y Does y exist somewhere in x 6

x StartsWith y Does x start with y 6

x EndsWith y Does x end with y 6

x AnyOf list Does x appear in list 6

x Greater y Is x greater than y 5

x GreaterEqual y Is x greater than or equal to y 5

x Less y Is x less than y 5

x LessEqual y Is x less than or equal to y 5

x Equal y Is x equal to y 4

x NotEqual y Is x not equal to y 4

Not x operator produces the value 0 if x is true 3

(nonzero) and the value 1 if x is false (0).

x And y True if both x and y are true 2

x OR y True if either x or y are true 1

Note the following:

■ AND has a higher precedence than OR.
In the Query Builder's Advanced Mode, you can use parentheses to change the
order of evaluation in the rules that use AND or OR.
See “Parentheses in compound queries” on page 124.
■ In the Query Builder's Advanced Mode, you can combine two or more operations
in a single rule without AND or OR to join them. Precedence determines the
order in which the operations are evaluated within the rule.
Example of a rule that includes three operations:

Displayname StartsWith “L” NotEqual Displayname contains “x”

This rule selects the following virtual machines:

Virtual machines with the names that start with L but do not contain x.
Virtual machines with the names that do not start with L but that do contain x.
 Configure a VMware Intelligent Policy 124
Parentheses in compound queries

Explanation: The StartsWith and Contains operations have a precedence of 6,

whereas NotEqual has a lower precedence of 3. Starting on the left, the
StartsWith operation is evaluated first and the Contains operation is evaluated
next. The last operation to be evaluated is Not Equal.
See “Using the Query Builder in Advanced Mode” on page 116.
See “The basics of a NetBackup query rule” on page 104.

Parentheses in compound queries

This topic applies to policy configuration for NetBackup for VMware. You can use
the Query Builder to enter rules for the automatic selection of VMware virtual
machines for backup.
You can use the Query Builder to make precise queries containing as many rules
as necessary to identify the appropriate virtual machines. In a query such as
powerstate Equal "poweredOn", the result of the query is easy to predict: only the
virtual machines that are turned on are included in the backup. But if several rules
are combined with AND and OR, the result may not be obvious. This kind of query
is called a compound query. Compound queries contain two or more rules, joined
by AND, AND NOT, OR, or OR NOT.
The order in which the Query Builder evaluates compound rules affects the outcome
of the query. Grouping the rules with parentheses can change the order of evaluation
and thus the outcome of the query.
The examples in the following table demonstrate how the Query Builder evaluates
compound queries with and without parentheses.

Note: Only the Query Builder's Advanced Mode supports the use of parentheses.

Table 6-12 Examples of compound queries with and without parentheses

Example query The following virtual machines are selected

ESXServer Equal "ESX001" OR Folder All virtual machines under ESX001 (regardless of power state), and virtual
Equal "FolderEngA" AND powerstate machines under FolderEngA that are turned on
Equal ON
To select only the virtual machines that are turned on in the ESX server and
in the folder, use parentheses (see next example).

(ESXServer Equal "ESX001" OR Folder All the virtual machines that are turned on in ESX001 and in FolderEngA.
Equal "FolderEngA") AND powerstate
Equal ON

See “AND vs. OR in queries” on page 117.

 Configure a VMware Intelligent Policy 125
Query rules for resource pools

See “Order of operations in queries (precedence rules)” on page 122.

Query rules for resource pools

This topic applies to policy configuration for NetBackup for VMware. You can use
the Query Builder to enter rules for the automatic selection of VMware virtual
machines for backup.
If the resource pool that you query is nested, the choice of Operator determines
which virtual machines in the resource pool hierarchy are discovered.
For example, assume the following hierarchy of resource pools that contain virtual
machines:

Res/ResourcePool_1
VM1
VM2
/ResourcePool_2
VM3
VM4
/ResourcePool_3
VM5
VM6

where ResourcePool_1 contains virtual machines VM1 and VM2, and so forth.
The following table shows the query results with the Contains, Equal, StartsWith,
and EndsWith operators. (Other operators can be used.)

Note: If you want the query to include all virtual machines in a hierarchy of nested
resource pools, do not use Equal as the Operator.

Table 6-13 Example rules for nested resource pools

Query rule Included virtual machines

Resourcepool Contains "Res/ResourcePool_1" Includes all the virtual machines in the three resource
pools (VM1 through VM6).

Resourcepool Equal "Res/ResourcePool_1" Includes only the virtual machines that are in
ResourcePool_1 (VM1, VM2). Virtual machines in the
sub-pools are not included (VM3 through VM6).

Resourcepool Equal "Res/ResourcePool_1/ResourcePool_2" Includes only the virtual machines that are in
ResourcePool_2 (VM3, VM4).
 Configure a VMware Intelligent Policy 126
Query rules for datacenter folders (host folder)

Table 6-13 Example rules for nested resource pools (continued)

Query rule Included virtual machines

Resourcepool StartsWith "Res/ResourcePool" Includes all the virtual machines in the three resource
pools (VM1 through VM6).

Resourcepool StartsWith Includes only the virtual machines that are in

"Res/ResourcePool_1/ResourcePool_2" ResourcePool_2 and 3. Virtual machines in
ResourcePool_1 are not included.

Resourcepool EndsWith "ResourcePool_2" Includes the virtual machines in ResourcePool_2 (VM3,

VM4) but not in ResourcePool_1 or 3.

These examples also apply to host folders.

See “Query rules for datacenter folders (host folder)” on page 126.
See “Test Query screen for VMware” on page 139.

Query rules for datacenter folders (host folder)

This topic applies to policy configuration for NetBackup for VMware. You can use
the Query Builder to enter rules for the automatic selection of VMware virtual
machines for backup.
In NetBackup terminology, a host folder is a folder that has been defined within a
VMware datacenter. A host folder can contain ESX servers or clusters, as well as
other folders. For example:

Folder_1
ESX1
ESX2
subfolder_A
ESX3
ESX4
subfolder_B
ESX_5
ESX_6

If you want NetBackup to select all the virtual machines within the top-level folder
and any subfolders, use the Contains or StartsWith operator. For example:
HostFolder Contains "Folder_1"
 Configure a VMware Intelligent Policy 127
Query rules for duplicate names

Note: If you want the query to include all virtual machines in the hierarchy of folders,
do not use Equal as the Operator.

If you want NetBackup to select the virtual machines within a subfolder only (such
as subfolder_A), use the Contains or Equal operator. For example:
HostFolder Equal "Folder_1/subfolder_A"
In this case, NetBackup includes only the virtual machines that reside on servers
ESX3 and ESX4.
For host folders, these operators work the same as they do for resource pools. For
further query builder examples, refer to the following topic (substitute host folder
for resource pool in the examples):
See “Query rules for resource pools” on page 125.

Note: If an ESX cluster is not contained within a folder and you click the browse
for values button: The ESX cluster name appears in the Select From List dialog.

See “The basics of a NetBackup query rule” on page 104.

See “Configuring automatic virtual machine selection” on page 112.

Query rules for duplicate names

If you have clusters, datastores, or virtual machine display names that have
duplicates elsewhere in your virtual environment, note: The query rules must specify
the parent datacenter or host folder to avoid conflicts during discovery. (A host
folder is one that has been defined within a datacenter.)
Take the following example of duplicate virtual machine names:

Folder_1
ESXi_prod
VM_1
Folder_2
ESXi_mrkt
VM_1

To back up ESXi_prod/VM_1 but not ESXi_mrkt/VM_1, use the following query:

HostFolder Equal "Folder_1"

AND ESXserver Equal "ESXi_prod"
AND Displayname Equal "VM_1"
 Configure a VMware Intelligent Policy 128
Query rules for tags

To back up only ESXi_mrkt/VM_1, use the following query:

HostFolder Equal "Folder_2"

AND ESXserver Equal "ESXi_mrkt"
AND Displayname Equal "VM_1"

Note: To back up both of these virtual machines from the same policy, include both
of these rules in the policy, as follows:

Instead of Equal, you can use other field values, such as Contains.
See “Query rules for datacenter folders (host folder)” on page 126.

Query rules for tags

Be aware of the following rules when you use tags in your queries:
■ If you use VMware tags, you can base your backup selections on these tags.
Be aware these tags are case-sensitive, so UNIX is different from unix.
■ When NetBackup uses tags to select virtual machines, the selection is based
only on tag names. The selection is independent of the category.
Example:
■ Virtual_machine_1 has a user-specified tag HR from the category
production.

■ Virtual_machine_2 has a user-specified tag HR from the category test.

Queries that select virtual machines with the tag HR select both virtual machines.
■ NetBackup uses a different VMware interface to access tag information from
vCenter Servers than is used for other query fields. Therefore, tag related calls
to the vCenter Server execute only if they are required. Calls are skipped if the
query is satisfied without the tag information. NetBackup only collects this
information once per vCenter Server. NetBackup collects tag metadata as part
of virtual machine backup, but it collects tag metadata only from those vCenter
Servers for which a virtual machine is selected. If no virtual machine is selected
 Configure a VMware Intelligent Policy 129
Query Builder field reference

from a vCenter Server and tags are not used in the query then tag metadata is
not collected from that vCenter Server.
■ Example 1: Tag Equal "Production" OR Powerstate Equal poweredOn
■ Example 2: Powerstate Equal poweredOn OR Tag Equal "Production"
In Example 1, NetBackup retrieves virtual machine data as well as tag data from
each virtual server for which it has credentials.
In Example 2, NetBackup retrieves virtual machine data for each virtual server
for which it has credentials. But NetBackup only needs to retrieve tag data for
virtual machines where the Powerstate is not equal to poweredOn.

Query Builder field reference

This topic applies to policy configuration for NetBackup for VMware. You can use
the Query Builder to enter rules for the automatic selection of VMware virtual
machines for backup.
Table 6-14 describes the drop-down fields and options for creating rules in the
Query Builder.

Table 6-14 Query Builder drop-down options: Join, Field, Operator, Value(s)

Query Builder Description

drop-down
fields

Join Selects a connector to join rules.

For the first rule, choices are blank (none) or NOT. After you add a rule,
the available connectors are AND, AND NOT, OR, OR NOT.

Field Selects a parameter on which to build the rule. Select one from the list
(scroll down for additional parameters).

You can type the first character to speed up selection. For example,
on entering “d”, the list moves to first entry starting with "d". Another
entry of "d" moves through the list to the next entry starting with "d".
The selected entry is automatically filled in.

See Table 6-15 on page 130.

Operator Selects an operator. The available operators depend on the parameter

that is selected for Field.

See Table 6-16 on page 137.

 Configure a VMware Intelligent Policy 130
Query Builder field reference

Table 6-14 Query Builder drop-down options: Join, Field, Operator, Value(s)
(continued)

Query Builder Description

drop-down
fields

Value(s) Specifies a value for the Field parameter.

The Value field allows manual entry. It may also be a drop-down,

depending on the selections that are made in the other fields.

For manual entry, you can specify multiple comma-separated values.

See Table 6-17 on page 138.

Allows browsing for specific values, depending on the selections that

are made in the other drop-down fields.

Adds the current drop-down selections to the Query pane as a new

rule.

Blanks out the drop-down fields.

Field (keywords)
Table 6-15 describes the keywords available in the Field drop-down. The table also
indicates whether the values for each keyword (in the Values field) are
case-sensitive.
Note that the inclusion or exclusion of virtual machines is not determined by the
Field keyword alone, but depends on the rule you construct: the combination of
Join, Field, Operator, and Value(s).

Table 6-15 Keywords in the Field drop-down

Field keyword Data type Description

Annotation Alphanumeric The text that is added to virtual machine annotations in vSphere Client.
string
Values are case-sensitive.
 Configure a VMware Intelligent Policy 131
Query Builder field reference

Table 6-15 Keywords in the Field drop-down (continued)

Field keyword Data type Description

Cluster Alphanumeric The name of the cluster (a group of ESX servers) that the virtual
string machine is configured in.

Values are not case-sensitive.

Note: A virtual machine may be assigned to an ESX server that is
not in a cluster.

Note also that in VMware, a cluster name need only be unique within
a datacenter path.

Datacenter Alphanumeric The name of the VMware datacenter.

string
Values are not case-sensitive.

DatacenterPath Alphanumeric The folder structure that defines the path to a datacenter. Use this
string option if the datacenter name that you want to filter on is not unique
in your environment.

Values are case-sensitive.

Datastore Alphanumeric The name of the datastore.

string
Values are case-sensitive.
Note: Multiple ESX servers can share access to the same datastore.
Also, a datastore name can be duplicated between multiple ESX
servers. Use DatacenterPath or ESXserver to uniquely identify the
datacenter.

DatastoreCluster Alphanumeric The name of the datastore cluster that contains the datastores.
string
Values are not case-sensitive.

DatastoreFolder Alphanumeric The name of the folder that contains the datastores.
string
Values are not case-sensitive.

DatastoreNFSHost Alphanumeric The name of the datastore's NFS host.

string
Values are not case-sensitive.

DatastoreNFSPath Alphanumeric The folder structure that defines the path to an NFS datastore. Use
string this option if the NFS host name of the datastore that you want to filter
on is not unique in your environment.

Values are not case-sensitive.

 Configure a VMware Intelligent Policy 132
Query Builder field reference

Table 6-15 Keywords in the Field drop-down (continued)

Field keyword Data type Description

DatastoreType Alphanumeric The type of the datastore. Values are NFS, NFS41, VMFS, vsan, and
string VVOL.

Values are not case-sensitive.

Displayname Alphanumeric The virtual machine's display name.

string
Values are case-sensitive.

ESXserver Alphanumeric The name of the ESX server.

string
Values are not case-sensitive.

The ESX host name must match the name as defined in the vCenter
server.

HostFolder Alphanumeric The folder path between the datacenter level and a cluster, ESX hosts,
string or a subfolder.

If an ESX cluster is not contained within a folder and you click the
browse for values button, note: The ESX cluster name appears in the
Select From List dialog.

Values are not case-sensitive.

See “Query rules for datacenter folders (host folder)” on page 126.

InstanceUUID Alphanumeric The virtual machine's instance UUID.

string
Example query:
InstanceUUID Equal
"501b13c3-52de-9a06-cd9a-ecb25aa975d1"

Values are not case-sensitive.

Network Alphanumeric The name of the network switch (on an ESX server) or distributed
string switch.

Values are not case-sensitive.

NetworkFolder Alphanumeric The name of the folder that contains the network.
string
Values are not case-sensitive.

Powerstate Alphabetic The power state of the virtual machine.

Values are poweredOff, poweredOn, suspended.

 Configure a VMware Intelligent Policy 133
Query Builder field reference

Table 6-15 Keywords in the Field drop-down (continued)

Field keyword Data type Description

Resourcepool Alphanumeric The name of the resource pool. (A resource pool is similar to a vApp.)
string
Values are case-sensitive.

If a resource pool contains other resource pools (sub-pools), the choice

of Operator determines whether virtual machines in the sub-pools are
included.

See “Query rules for resource pools” on page 125.

Tag Alphanumeric The name of the tag.

string
Values are case-sensitive.

When NetBackup uses tags to select virtual machines, the selection

is based only on tag names. The selection is independent of the
category.

See “Query rules for tags” on page 128.

Template Boolean TRUE if the virtual machine is a virtual machine template.

vApp Alphanumeric The name of the vApp.

string
Values are case-sensitive.

A vApp is a collection of virtual machines. vApps can also contain

resource pools and other vApps. vApps are components of standalone
ESX servers or of clusters.
Like vSphere Client, NetBackup refers only to the top level of a vApp
that contains sub vApps.

For the following rule:

vApp Equal "vapp1"

if vapp1 has a sub vApp named "vapp2", any virtual machines in vapp1
or vapp2 are included. You cannot make a rule that refers specifically
to vapp2.

vCDCatalog Alphanumeric The name of the vCloud Director catalog.

string
Values are not case-sensitive.

vCDIsExpired Alphabetic Expired if the vCloud Director vApp or vApp template is expired.

Possible values are Expired, Not Expired, and Unknown. Unknown

indicates that an error occurred between vCloud Director and the
vSphere environment. Examples are errors in provisioning or in a
deletion operation.
 Configure a VMware Intelligent Policy 134
Query Builder field reference

Table 6-15 Keywords in the Field drop-down (continued)

Field keyword Data type Description

vCDIsvAppTemplate Boolean TRUE if the vCloud Director vApp is a template.

vCDOrg Alphanumeric The name of the vCloud Director organization.

string
Values are not case-sensitive.

vCDOrgvDC Alphanumeric The name of the organization virtual datacenter in vCloud Director.
string
Values are not case-sensitive.

vCDServer Alphanumeric The name of the vCloud Director server.

string
Values are not case-sensitive.

vCDvApp Alphanumeric The name of the vCloud Director vApp.

string
Values are not case-sensitive.

vCenter Alphanumeric The name of the vCenter server.

string
Values are not case-sensitive.

The vCenter name that is specified in the Query Builder must match
the name as entered in the Administration Console for NetBackup
credentials. (Media and Device Management > Credentials > Virtual
Machine Servers.)

Note that a fully qualified domain name is recommended.

vCenterVersion Alphanumeric The version of the vCenter Server. For example: 5.1.0, 5.5.0, 6.0.0
string
The possible values of this field are automatically updated and
populated based on the environment. Only the versions of the vCenter
servers that are registered with NetBackup are shown.

VMDNSName Alphanumeric The virtual machine DNS name in vSphere Client.

string
Values are not case-sensitive.
 Configure a VMware Intelligent Policy 135
Query Builder field reference

Table 6-15 Keywords in the Field drop-down (continued)

Field keyword Data type Description

VMFolder Alphanumeric The name of the VM folder (within a datacenter), including the path to
string the folder that contains the VMs.

Values are not case-sensitive.

For example, assume the following VM folders containing a total of 65

VMs:

vm\VM_backup_prod1 (contains 5 VMs)

vm\VM_backup_prod1\cluster1 (contains 10 VMs)
vm\VM_backup_prod2 (contains 50 VMs)

To include the VMs in vm\VM_backup_prod1 but not the VMs in

cluster1 or in any other folder:

VMFolder Equal “vm\VM_backup_prod1”

To include the VMs in vm\VM_backup_prod1 and in its subfolder

cluster1:

VMFolder Equal “vm\VM_backup_prod1”

OR VMFolder StartsWith “vm\VM_backup_prod1\\”

Note: The first backslash is an escape character that causes the

following backslash to be interpreted as a literal character.

To include all 65 VMs:

VMFolder StartsWith “vm\VM_backup_prod”
Note: Any VM that is in a path that begins with vm\VM_backup_prod
is included.

VMGuestOS Alphanumeric The virtual machine guest OS type that is recorded in vSphere Client.
string
Values are case-sensitive.

VMHasIDE Boolean TRUE if the virtual machine has IDE drives.

VMHasMDS Boolean TRUE if the virtual machine has multiple datastores.

You can use this keyword to select any virtual machine that is
configured to use more than one datastore.

VMHasNVME Boolean TRUE if the virtual machine has NVMe drives.

VMHasRDM Boolean TRUE if the virtual machine uses Raw Device Mapping (RDM).

VMHasSATA Boolean TRUE if the virtual machine has SATA drives.

 Configure a VMware Intelligent Policy 136
Query Builder field reference

Table 6-15 Keywords in the Field drop-down (continued)

Field keyword Data type Description

VMHasSnap Boolean TRUE if a VMware snapshot of the virtual machine is currently active.

VMHasVDSName Boolean TRUE if the virtual machine has a display name that is valid for use
as the host name.

VMHostName Alphanumeric The virtual machine name that is derived from a reverse lookup of its
string IP address.

Values are not case-sensitive.

VMIsConn Boolean TRUE if the virtual machine is connected and available.

For example: If a virtual machine's ESX server is down, that virtual

machine is not connected.

VMVersion Alphanumeric The VMware version of the virtual machine.

string
Values are case-sensitive.

For example: vmx-04, vmx-07, vmx-08.

VMXDatastore Alphanumeric The name of the vmx datastore (sometimes called the vmx directory
string or configuration datastore).

Values are case-sensitive.

More information on the vmx datastore is available.

See “NetBackup for VMware terminology” on page 20.

VMXDatastoreFolder Alphanumeric The name of the folder that contains the vmx datastores.
string
Values are not case-sensitive.

VMXDatastoreNFSHost Alphanumeric The name of the vmx datastore's NFS host.

string
Values are not case-sensitive.

VMXDatastoreNFSPath Alphanumeric The folder structure that defines the path to a vmx NFS datastore. Use
string this option if the NFS host name of the datastore that you want to filter
on is not unique in your environment.

Values are not case-sensitive.

VMXDatastoreType Alphanumeric The type of the vmx datastore. Values are NFS or VMFS.
string
Values are not case-sensitive.
 Configure a VMware Intelligent Policy 137
Query Builder field reference

Table 6-15 Keywords in the Field drop-down (continued)

Field keyword Data type Description

[vSphere custom attributes] Alphanumeric The value of a custom attribute that is set in vSphere Client for one or
string more virtual machines.

Note: in vSphere Client, the attribute must have a value for at least
one virtual machine. The attribute type must be Virtual Machine.

The values are case-sensitive.

Custom Attribute based disk exclusion requires that you enter in

NetBackup the credentials for the vCenter server or servers that host
the VMs. ESXi server credentials are not sufficient.

See “Adding NetBackup credentials for VMware” on page 45.

Operators
Table 6-16 describes the operators available in the Operator drop-down.

Table 6-16 Operators in the Operator drop-down

Operator Description

AnyOf Matches any of the specified values in the Value(s) field.

For example: If the ESX servers in the Value(s) field are "ESX01","ESX02,"ESX03", AnyOf
matches any ESX server that has one of those names. If the names of your servers are not
identical to any of the specified values, no match occurs. A server named ESX01A is not a match.

Contains Matches the value in the Value(s) field wherever that value occurs in the string.

For example: If the Value(s) entry is "dev", Contains matches strings such as "01dev", "01dev99",
"devOP", and "Development_machine".

EndsWith Matches the value in the Value(s) field when it occurs at the end of a string.

For example: If the Value(s) entry is "dev", EndsWith matches the string "01dev" but not "01dev99",
"devOP", or "Development_machine".

Equal Matches only the value that is specified in the Value(s) field.

For example: If the display name to search for is "VMtest27", Equal matches virtual machine
names such as VMTest27 or vmtest27 or vmTEST27, and so forth. The name VMtest28 is not
matched.

Greater Matches any value that is greater than the specified Value(s), according to the UTF-8 collating
sequence.

GreaterEqual Matches any value that is greater than or equal to the specified Value(s), according to the UTF-8
collating sequence.
 Configure a VMware Intelligent Policy 138
Query Builder field reference

Table 6-16 Operators in the Operator drop-down (continued)

Operator Description

IsSet Determines whether a value is returned for the Field keyword. Use IsSet with another rule as a
condition, to ensure that the query selects the appropriate virtual machines.

Note that you do not make an entry under Value(s) for a rule that uses IsSet.

See “The IsSet operator in queries” on page 120.

See “Effect of Primary VM identifier parameter on Selection column in Test Query results”
on page 142.

See “Test Query: Failed virtual machines” on page 140.

Less Matches any value that is less than the specified Value(s), according to the UTF-8 collating
sequence.

LessEqual Matches any value that is less than or equal to the specified Value(s), according to the UTF-8
collating sequence.

NotEqual Matches any value that is not equal to the value in the Value(s) field.

StartsWith Matches the value in the Value(s) field when it occurs at the start of a string.

For example: If the Value(s) entry is "box", StartsWith matches the string "box_car" but not
"flatbox".

Value(s)
Table 6-17 describes the characters that can be entered in the Value(s) field. The
Field keyword determines case sensitivity.

Note: The character string you enter in the Value(s) field must be enclosed in single
quotes or double quotes.

Table 6-17 Characters you can enter for Value(s)

Character types String characters allowed

Alphanumerics A to Z, a to z, 0 to 9, - (minus sign), and special characters.

Note: Decimal numbers only.

Wildcards * (asterisk) matches everything.

For example: "*prod*" matches the string "prod" preceded or followed by any characters.

? (question mark) matches any single character.

For example: "prod??" matches the string "prod" followed by any two characters.
 Configure a VMware Intelligent Policy 139
Test Query screen for VMware

Table 6-17 Characters you can enter for Value(s) (continued)

Character types String characters allowed

Escape character \ (backslash) escapes the wildcard or meta-character that follows it.

For example: To search for a string that contains an asterisk (such as test*), enter "test\*"

Quotation marks Note: The characters you enter in Value(s) must be enclosed in single or double quotes.

To search for a string that contains quotation marks, either escape each quote (\") or enclose
the entire string in the opposite type of quotes.

For example: To search for a string that includes double quotes (such as "name"), enter
'"name"' (enclosing it in single quotes) or "\"name\""

See “The basics of a NetBackup query rule” on page 104.

Test Query screen for VMware

This topic applies to policy configuration for NetBackup for VMware. You can use
the Query Builder to enter rules for the automatic selection of VMware virtual
machines for backup. Use the Test Query screen to test the rules.
This screen lists the virtual machines that NetBackup discovered in your virtual
environment when you clicked Test Query. Later changes in the virtual environment
may affect which virtual machines match the query rules. For example: if virtual
machines are added, the test results may not be identical to the virtual machines
that are selected for backup when the backup runs.
When the next backup runs from this policy, the following occur: NetBackup
re-discovers virtual machines, consults the query rules, and backs up the virtual
machines that match the rules.
The list of backed up virtual machines is saved but the virtual machines are not
displayed in the policy's Clients tab. You can use the Activity Monitor to view the
virtual machine jobs, or you can run a Virtual Client Summary report in OpsCenter.

Note: An alternative to the Test Query screen is the nbdiscover command. For
more information, see the NetBackup Commands Reference Guide.

See “Using the Activity Monitor to monitor virtual machine backups” on page 153.
See “Using OpsCenter to report on virtual machine backups” on page 157.
The Test Query function runs in the background. You can continue to configure
the policy while the test runs. Any changes you make in the Query Builder however
 Configure a VMware Intelligent Policy 140
Test Query: Failed virtual machines

are not included in the currently running test. You must re-initiate the test to see
the results of your Query Builder changes.

Table 6-18
Field Description

Test query for Lists the rules in the Query Builder that were used in this test. The rules
policy are specified in the Query Builder on the policy Clients tab.

Test Query VM Name: Shows the display name of all discovered virtual machines.
Results Selection: Lists the virtual machines that were discovered, as follows:

■ INCLUDED: The virtual machine matches the rules in the query.

■ EXCLUDED: The virtual machine does not match the rules in the
query.
■ FAILED: The virtual machine cannot be selected for backup because
of a host name problem or other error. Also, the query cannot exclude
the virtual machine. An explanation appears at the bottom of the Test
Query screen. For example:

VM does not have a host name to use as a

client name, display name =

See “Test Query: Failed virtual machines” on page 140.

The operator IsSet can be used to filter out such virtual machines.
More information is available on IsSet.
See Table 6-16 on page 137.
See “The IsSet operator in queries” on page 120.

Included: Gives a tally of how many virtual machines were included, excluded, or
failed in the test.
Excluded:

Failed:

See “Using the Activity Monitor to monitor virtual machine backups” on page 153.
See “About automatic virtual machine selection for NetBackup for VMware”
on page 102.
See “Configuring automatic virtual machine selection” on page 112.

Test Query: Failed virtual machines

This topic applies to policy configuration for NetBackup for VMware. You can use
the Query Builder to enter rules for the automatic selection of VMware virtual
machines for backup. Use the Test Query screen to test the rules.
 Configure a VMware Intelligent Policy 141
Test Query: Failed virtual machines

If the query rules cannot exclude a virtual machine, and that virtual machine cannot
be selected for backup, it is marked as FAILED. The virtual machine is listed as
not run in the job details log.
For example: the virtual machine does not have the type of name specified by the
Primary VM identifier parameter (such as host name or display name). Or the
virtual machine name contains invalid characters. In any case, a virtual machine
that is listed as FAILED should be investigated: it may be one that you want to back
up.
To see the reason for the failure, click on the virtual machine in the Test Query
Results. An explanation appears at the bottom of the screen.
For example:

Explanation: The virtual machine that is listed as Ubuntu-vS in the example does
not have a VMware host name. In the NetBackup policy, on the VMware tab, the
Primary VM identifier parameter may be set to VM hostname. In that case,
NetBackup cannot refer to the virtual machine by its host name and thus cannot
back it up.
You can fix this problem in a couple of ways:
■ Use vSphere Client to configure a host name for the virtual machine.
■ To exclude the virtual machines that have no host name, construct a query with
the IsSet operator.
 Configure a VMware Intelligent Policy 142
Effect of Primary VM identifier parameter on Selection column in Test Query results

See “The IsSet operator in queries” on page 120.

See “Effect of Primary VM identifier parameter on Selection column in Test Query
results” on page 142.
See “Test Query screen for VMware” on page 139.

Effect of Primary VM identifier parameter on

Selection column in Test Query results
This topic applies to policy configuration for NetBackup for VMware. You can use
the Query Builder to enter rules for the automatic selection of VMware virtual
machines for backup.
The NetBackup policy's Primary VM identifier parameter tells NetBackup how to
identify virtual machines. For example, if the parameter is set to VM hostname,
NetBackup identifies virtual machines by their host names. If they do not have a
host name, the policy cannot back them up.
The Primary VM identifier parameter has a direct effect on the query test results.
Note that for each virtual machine, the query test result is one of three possibilities:
INCLUDED, EXCLUDED, or FAILED.
If NetBackup cannot identify a virtual machine according to the Primary VM
identifier parameter, one of two test results can occur:
■ If the virtual machine is filtered out by the query rules, it is listed as EXCLUDED.
■ If the virtual machine is not filtered out by the query rules, it is listed as FAILED.
The following table gives the test query results from example combinations of
the Primary VM identifier parameter and a query rule.

Table 6-19 Effect of Primary VM identifier parameter and query rules on test
query results

Primary VM Query rule in Query Test query result

identifier setting on Builder
VMware policy tab

VM hostname VMHostName Contains "VM" INCLUDED: Any virtual machines with a host name that
contains "VM". Since the Primary VM identifier parameter tells
NetBackup to select the virtual machine by host name, it can
back up the virtual machines.

EXCLUDED: All other virtual machines.

 Configure a VMware Intelligent Policy 143
Effect of Primary VM identifier parameter on Selection column in Test Query results

Table 6-19 Effect of Primary VM identifier parameter and query rules on test
query results (continued)

Primary VM Query rule in Query Test query result

identifier setting on Builder
VMware policy tab

VM hostname Displayname Contains "VM" INCLUDED: Any virtual machines that have a host name and
that have a display name that contains "VM".

EXCLUDED: Any virtual machines that have a host name, but

that do not have a display name containing "VM".

FAILED: Any virtual machines that do not have a host name.

Since the Primary VM identifier parameter is set to VM
hostname, NetBackup cannot select the virtual machine for
backup.

VM hostname Displayname Contains "VM" INCLUDED: Any virtual machines that have a host name and
that have a display name that contains "VM".
AND VMHostName IsSet
EXCLUDED: All other virtual machines. The IsSet rule means
that if a virtual machine does not have a host name, it is
excluded.

VM hostname Displayname Contains "VM" INCLUDED:

AND VMHostName IsSet ■ Any virtual machines that have a host name and that have
a display name that contains "VM".
OR Annotation Contains "test"
■ Any virtual machines without a host name that have an
AND NOT VMHostName annotation that contains "test".
IsSet
EXCLUDED: All other virtual machines.

VM display name Displayname Contains "VM" INCLUDED: Any virtual machines with the display names that
contain "VM". Since the Primary VM identifier parameter tells
NetBackup to select the virtual machine by display name, it
can back up the virtual machines.

EXCLUDED: All other virtual machines.

VM display name VMHostName Contains "VM" INCLUDED: Any virtual machines that have a display name
and that have a host name that contains "VM".

EXCLUDED: Any virtual machines that have a display name,

but that do not have a host name containing "VM".

FAILED: Any virtual machines that do not have a display name.

Since the Primary VM identifier parameter is set to VM
display name, NetBackup cannot select those virtual machines
for backup.
 Configure a VMware Intelligent Policy 144
Effect of Primary VM identifier parameter on VM Name column in Test query results

See “Test Query: Failed virtual machines” on page 140.

See “The IsSet operator in queries” on page 120.
See “Effect of Primary VM identifier parameter on VM Name column in Test query
results” on page 144.

Effect of Primary VM identifier parameter on VM

Name column in Test query results
This topic applies to policy configuration for NetBackup for VMware. You can use
the Query Builder to enter rules for the automatic selection of VMware virtual
machines for backup. You can use the Test Query screen to test the rules.
The policy's Primary VM identifier parameter affects the type of virtual machine
name that appears in the VM Name column of the Test Query screen, as follows:
■ If a virtual machine is EXCLUDED or FAILED, it is listed according to its virtual
machine display name. The Primary VM identifier parameter does not matter.
■ But if a virtual machine is listed as INCLUDED, note: The name that appears
under VM Name is the type of name that is specified on the Primary VM
identifier parameter.
For example: If the Primary VM identifier parameter is VM hostname, the
included virtual machine is listed according to its host name. Even if the query
rule specified Display name (such as Displayname Equal "vm1"), the virtual
machine appears on the Test Query screen by its host name.
See “Primary VM identifier options (VMware)” on page 76.
See “Effect of Primary VM identifier parameter on Selection column in Test Query
results” on page 142.

Refreshing the display of virtual environment

changes in the Query Builder
By default, NetBackup waits one hour before the policy Query Builder detects
changes in the virtual environment. Until one hour has passed, the Query Builder
does not detect the changes when you click the "Load values" folder icon next to
the Value(s) field. To make the changes immediately available to the Value(s) field,
use the following procedure to refresh the display.
 Configure a VMware Intelligent Policy 145
Refreshing the display of virtual environment changes in the Query Builder

Note: The Query Builder's Reuse VM selection query results for option does not
affect the display of virtual environment changes in the Query Builder. The reuse
option determines how long NetBackup reuses the current backup list for future
executions of the policy.

To refresh the Query Builder's view of the virtual environment (Windows):

1 On the Windows desktop of the local host, click Start > Run and enter regedit.
2 To be on the safe side, make a backup of the current registry (File > Export).
3 Go to HKEY_LOCAL_MACHINE > SOFTWARE > Veritas > NetBackup >
CurrentVersion > Config and create a key that is called BACKUP.
4 Under BACKUP, create a new DWORD that is called xmlCacheLimit.
5 Set this DWORD to the number of seconds for the refresh.
A value of 15 allows the Query Builder to be refreshed after 15 seconds.
6 If the policy editor is open in the NetBackup Administration Console, close
it and reopen it.
To refresh the Query Builder's view of the virtual environment (Linux):
1 On the Linux desktop of the local host, create (or open) the following file:
/usr/openv/netbackup/virtualization.conf

2 Enter a new dword line under [BACKUP] to set the number of seconds for the
refresh. For example:

[BACKUP]
"xmlCacheLimit"=dword:15

This example allows the Query Builder to be refreshed after 15 seconds.

Note: If the file already contains a [BACKUP] line, do not add another [BACKUP]
line. Any other lines that already exist under [BACKUP] should remain as they
are.

3 Save the file.

4 If the policy editor is open in the NetBackup Administration Console, close it
and reopen it.
 Configure a VMware Intelligent Policy 146
Reducing the time required for VM discovery in a large VMware environment

Reducing the time required for VM discovery in a

large VMware environment
NetBackup VMware Intelligent Policies use query rules to automatically search and
filter the vSphere environment. By default, the query rules search all the VMware
servers in your environment. If the environment contains many VMware servers
with many virtual machines, VM discovery may take a long time. You can speed
up VM discovery by limiting the search to specific VMware servers or virtual
machines.
The following is an example of a policy Query Builder rule that searches all VMware
servers and all virtual machines:
vmware:/?filter=Displayname Contains "vm1"

To limit the search to particular servers or virtual machines, insert an additional

expression in the Query Builder rule as explained in the following procedure.
To use the Administration Console to reduce the time required for VM
discovery in a VMware environment
1 In the NetBackup Administration Console, open the VMware Intelligent Policy.
2 On the Clients tab of the policy, make sure Select automatically through
VMware Intelligent Policy query is selected.
3 In the Query Builder, click Advanced.
4 Under Query (Advanced Mode), create one or more rules to search for VMs
in specific VMware servers or for specific virtual machines.
Be aware that to create two or more rules, you must be in Advanced Mode not
Basic Mode. Additionally, each query rule must begin on its own line.
You can use the types of rules shown:
■ To search for VMs in a particular VMware server
vmware://VMware_server?filter=filter

■ To search for a specific virtual machine on a particular VMware server

vmware://VMware_server/vm/virtual_machine_instance_uuid
 Chapter 7
Reduce the size of
backups
This chapter includes the following topics:

■ About reducing the size of VMware backups

■ Block-level backup (BLIB): full vs incremental

■ Deleting a vSphere Client snapshot

About reducing the size of VMware backups

This topic applies to policy configuration for NetBackup for VMware.
NetBackup provides the following options for reducing the backup size for a VMware
virtual machine:
 Reduce the size of backups 148
Block-level backup (BLIB): full vs incremental

Table 7-1 Options for reducing the virtual machine backup size

Option Description

Block level BLIB reduces the size of backups (full and incremental) by tracking block-level changes. Only
incremental backup the blocks that have changed since the last full or incremental are included in the backup. For
(BLIB) incremental backups, this option applies to cumulative and to differential backups.

BLIB works with VMware’s Changed Block Tracking in vSphere to track block-level changes
in the virtual machine.

The Enable block-level incremental backup option is enabled by default on the NetBackup
policy VMware tab. NetBackup uses BLIB for storage optimization when the backup runs.
Note: The first backup you run with BLIB must be a full backup.

See “Block-level backup (BLIB): full vs incremental” on page 148.

Note: BLIB works only with ESX 4.0 or later virtual machines at version vmx-07 or later.

Note: If you used vSphere Client to manually create a VM snapshot and that snapshot is
currently active, you may have to delete the snapshot.

See “Deleting a vSphere Client snapshot” on page 149.

Exclusion of deleted Reduces the size of virtual machine backups by excluding any deleted sectors in the file system
blocks. on the virtual machine.

To enable this option, click Exclude deleted blocks on the policy VMware tab.

Refer to Exclude deleted blocks in the following topic:

See “Optimizations options (VMware)” on page 74.

Block-level backup (BLIB): full vs incremental

This topic applies to policy configuration for NetBackup for VMware.
When you use the Enable block-level incremental backup option in the policy,
NetBackup uses VMware's Changed Block Tracking feature (CBT) to reduce the
backup size.
This option reduces the size of full backups as well as the size of incremental
backups, as follows.
 Reduce the size of backups 149
Deleting a vSphere Client snapshot

Table 7-2 Block-level backup of the virtual machine: full vs incremental

schedule

Type of backup Optimization that is used in backup

Backup of entire virtual Backs up only the blocks that have changed since the .vmdk was
machine, with full created. Note that the blocks that are not initialized are excluded
schedule from the backup.

Backup of entire virtual Backs up only the blocks that have changed since the last backup,
machine, with as follows:
incremental schedule
■ For cumulative incrementals, BLIB backs up only the blocks
that changed since the last full backup.
■ For differential incrementals, BLIB backs up only the blocks
that changed since the previous backup of any kind.

Note: On incremental backups, the Enable block-level

incremental backup option backs up the changed files as well
as their metadata.

Note: VMware CBT may occasionally reset tracking of file

changes in the virtual machine, such as after a power failure or
hard shutdown. For the next backup, NetBackup reads all the
data from the vmdk files and the backup takes longer than
expected. If deduplication is enabled, the deduplication rate is
lower than expected.

The following VMware article contains more information on CBT:

Changed Block Tracking (CBT) on virtual machines (1020128)

See “About reducing the size of VMware backups” on page 147.

Deleting a vSphere Client snapshot

This topic applies to the use of BLIB (Perform block level incremental backup)
with NetBackup for VMware.
To use BLIB with NetBackup for VMware, you must delete an existing vSphere
Client snapshot if both of the following are true:
■ You used the vSphere Client interface to manually create a snapshot of the
virtual machine and that snapshot is currently active.
■ A NetBackup policy with the Enable block-level incremental backup feature
had never been used to back up the virtual machine before you started the
vSphere Client snapshot.
 Reduce the size of backups 150
Deleting a vSphere Client snapshot

In this case, NetBackup cannot enable BLIB for the virtual machine. You must
delete the vSphere Client snapshot. Then, when the NetBackup policy runs, BLIB
is enabled.
More information is available on NetBackup block-level backups:
See “About reducing the size of VMware backups” on page 147.
See “Block-level backup (BLIB): full vs incremental” on page 148.
To delete a vSphere Client snapshot
1 In the vSphere Client interface, right-click on the virtual machine and select
Snapshot > Snapshot Manager.
2 Select the snapshot and click Delete.
3 To back up the virtual machine with BLIB, you can now run a NetBackup policy
that uses Enable block-level incremental backup.
NetBackup backups from this policy continue to use storage optimization, even
if you manually create a snapshot of the virtual machine using vSphere Client.
See “Configuring a VMware policy from the Policies utility” on page 68.
 Chapter 8
Back up virtual machines
This chapter includes the following topics:

■ Backing up virtual machines

■ Using the Activity Monitor to monitor virtual machine backups

■ Restarting jobs individually in the Activity Monitor

■ Viewing NetBackup activity in vSphere Web Client or vSphere Client (HTML5)

■ Using OpsCenter to report on virtual machine backups

Backing up virtual machines

This topic applies to NetBackup for VMware.
Virtual machine backups initiate from a NetBackup policy. You can start the backup
manually from a policy, or have it run automatically according to a schedule that is
defined in the policy.
To create the policy, you can use the Policies utility in the NetBackup Administration
Console, or the Policy Configuration Wizard.
See “Configuring a VMware policy from the Policies utility” on page 68.
See “Configuring a VMware policy from the Policy Wizard ” on page 68.
For further information on policies and backup schedules, see the "Creating backup
policies" chapter of the NetBackup Administrator's Guide, Volume I.
 Back up virtual machines 152
Backing up virtual machines

To back up a virtual machine manually from an existing policy

1 Click on Policies in the NetBackup Administration Console, right-click on the
policy, and click Manual Backup....

The Manual Backup dialog appears.

2 Select the type of schedule for the backup.

 Back up virtual machines 153
Using the Activity Monitor to monitor virtual machine backups

3 Select the clients (virtual machines) to back up.

If the policy was configured for automatic selection of virtual machines, the
Clients pane shows the VMware backup host rather than the virtual machines.
4 Click OK to start the backup.
5 To see the job progress, click Activity Monitor in the NetBackup Administration
Console.
It may take a few moments for the job to appear in the Activity Monitor. Do not
start another backup if a job does not immediately appear. If the job does not
show up after a few moments, close and restart the NetBackup Administration
Console.
Note that your VMware backup request launches more than one job. The first
job automatically creates and deletes snapshots. This job has a dash (-) in the
Schedule column. The second job backs up the virtual machine files from the
snapshot.

Using the Activity Monitor to monitor virtual

machine backups
This topic applies to NetBackup for VMware.
You can use the NetBackup Activity Monitor to keep track of the VMware virtual
machines that a policy backs up.
To monitor virtual machine backups
1 Click Activity Monitor in the left pane of the Administration Console.
2 Note each job as it appears by row.
If the policy selects virtual machines automatically (based on a query), the
backup consists of three generations of jobs:
■ The first job discovers the virtual machines. This job is labeled Backup.
(This job is unique to policies that use a query to select virtual machines.)
■ The discovery (Backup) job starts a child job to take a VMware snapshot
of the virtual machine. A snapshot job is started for each virtual machine.
Each of these jobs is labeled Snapshot.
■ Each snapshot job starts a child job to back up the virtual machine. A backup
job is started for each virtual machine. Each of these jobs is labeled Backup.
The job flow is as follows:

discovery job --> snapshot job --> backup job

 Back up virtual machines 154
Using the Activity Monitor to monitor virtual machine backups

3 To trace the discovery job to the virtual machine backup jobs, note the Job ID
and the Parent Job ID columns.

In this example, the discovery job is 119. Job 119 started job 120 (Snapshot).
Job 120 started job 121 (Backup).
4 Right-click on the job row and click Details.
On the Job Details screen, click the Detailed Status tab for more information
on the job.
5 If a column is not displayed in the Activity Monitor, or to re-arrange the columns,
do the following:
■ Right-click anywhere in the Jobs pane and click Columns > Layout.
■ Click the row for the heading that you want to display, hide, or re-arrange
(such as Parent Job ID).

■ Use Move Up, Move Down, Show, and Hide as needed.

■ Click OK. The columns should now be visible in the Activity Monitor.

See “Limit jobs per policy on the Attributes tab (for VMware)” on page 72.
See “Restarting jobs individually in the Activity Monitor” on page 155.
 Back up virtual machines 155
Restarting jobs individually in the Activity Monitor

See “Backing up virtual machines” on page 151.

See “Using OpsCenter to report on virtual machine backups” on page 157.

Restarting jobs individually in the Activity Monitor

This topic is about NetBackup policies that select virtual machines automatically (a
Virtual Machine Intelligent Policy).
If the policy automatically selects virtual machines for backup, you can restart the
virtual machine jobs individually. This feature is handy if the policy backs up a large
number of virtual machines: you can restart one or more of the jobs individually
rather than re-running the entire policy.
To restart jobs individually in the Activity Monitor
◆ In the Activity Monitor, right-click on the job and select Restart Job.
In some cases, to restart a child job you may have to restart its parent job.
See “Using the Activity Monitor to monitor virtual machine backups” on page 153.

Viewing NetBackup activity in vSphere Web Client

or vSphere Client (HTML5)
In VMware vCenter, NetBackup can record the backup activity for virtual machines.
You can view the events in vSphere Web Client or vSphere Client (HTML5) at the
level of any parent object (such as folder, datacenter, cluster, or host). You can
also view the events for the virtual machine.

Note: The NetBackup plug-in is not required.

Make sure that the policy's Post vCenter events option is enabled:
See “VMware - Advanced Attributes dialog” on page 81.
To view backup events and the last backup time in vSphere Web Client
1 Open vSphere Web Client or vSphere Client (HTML5).
2 Go to Home > Events.
 Back up virtual machines 156
Viewing NetBackup activity in vSphere Web Client or vSphere Client (HTML5)

3 In the Event Console, enter NetBackup in the search field.

Each NetBackup event includes the following details:

Description Shows the NetBackup operation that succeeded or failed, and

includes policy details and duration. When you click on the row
of the event, the Event Details pane shows the same
information.

Type The types are Information for a successful operation, and Error
for a failed operation.

Date Time The date and time of the event.

Task Not used.

Target The virtual machine that was backed up.

Click on the virtual machine link to see the Summary tab.

Note: On the Summary tab in vSphere 6.5, under Custom
Attributes, the timestamp of the virtual machine's last backup
is listed on the NB_LAST_BACKUP attribute.

User The user that ran the backup.

4 To view backup events for a particular object (such as ESX host or VM), select
the object and do the following:
■ Click the Monitor tab.
 Back up virtual machines 157
Using OpsCenter to report on virtual machine backups

■ Click Task & Events.

■ Click the Events view.
■ Enter NetBackup in the search field.

Using OpsCenter to report on virtual machine

backups
This topic applies to NetBackup for VMware.
To use OpsCenter to obtain a backup report on VMware virtual machines
1 Start the OpsCenter console in a web browser.
On a system that has a network connection to the OpsCenter server, enter the
following in a browser:

http://host.domain/opscenter

where host.domain is the fully qualified domain name or IP address of the

OpsCenter server.
2 Click the Reports tab.
3 Click Client Reports > Virtual Client Summary.
The Virtual Client Summary report is described in the OpsCenter Reporting
Guide.
 Back up virtual machines 158
Using OpsCenter to report on virtual machine backups

See “Using the Activity Monitor to monitor virtual machine backups” on page 153.
See “Backing up virtual machines” on page 151.
 Chapter 9
Use Accelerator to back
up virtual machines
This chapter includes the following topics:

■ About the NetBackup Accelerator for virtual machines

■ Accelerator: full vs. incremental schedules

■ How the NetBackup Accelerator works with virtual machines

■ Accelerator notes and requirements for virtual machines

■ Accelerator forced rescan for virtual machines (schedule attribute)

■ Accelerator requires the OptimizedImage attribute

■ Accelerator backups and the NetBackup catalog

■ Accelerator messages in the backup job details log

■ NetBackup logs for Accelerator with virtual machines

■ About reporting the amount of Accelerator backup data that was transferred
over the network

■ Replacing the Accelerator image size with the network-transferred data in

NetBackup command output

About the NetBackup Accelerator for virtual

machines
NetBackup Accelerator reduces the backup time for VMware backups. NetBackup
uses VMware Changed Block Tracking (CBT) to identify the changes that were
 Use Accelerator to back up virtual machines 160
Accelerator: full vs. incremental schedules

made within a virtual machine. Only the changed data blocks are sent to the
NetBackup media server, to significantly reduce the I/O and backup time. The media
server combines the new data with previous backup data and produces a traditional
full NetBackup image that includes the complete virtual machine files.

Note: Accelerator is most appropriate for virtual machine data that does not
experience a high rate of change.

Accelerator has the following benefits:

■ Performs the full backups faster than traditional backup. Creates a compact
backup stream that uses less network bandwidth between the backup host and
the server.
Accelerator sends only changed data blocks for the backup. NetBackup then
creates a full traditional NetBackup image that includes the changed block data.
■ Accelerator backups support Granular Recovery Technology (GRT) for restoring
Exchange, SQL, and SharePoint applications (using a full schedule only).
■ Accelerator backups (full and incremental) support instant recovery of virtual
machines.
■ If the Enable file recovery from VM backup option on the policy VMware tab
is enabled, you can restore individual files from the backup (full or incremental).
■ Reduces the I/O on the backup host.
■ Reduces the CPU load on the backup host.
See “How the NetBackup Accelerator works with virtual machines” on page 161.

Accelerator: full vs. incremental schedules

NetBackup Accelerator supports full and incremental backups.
Note: After an initial full backup, Accelerator backups with a full schedule have
about the same effect on I/O and performance as traditional incremental backups.
The NetBackup catalog however includes all catalog references that would be made
if the backup was a traditional (non-Accelerator) full.
For virtual machine restore, note the following about full vs. incremental backups
with Accelerator:
■ For applications (Exchange, SQL, and SharePoint), NetBackup Accelerator
supports Granular Recovery Technology (GRT) restores from full backups only.
■ For any other kind of virtual machine restore, the Accelerator supports full
backups and incremental backups.
 Use Accelerator to back up virtual machines 161
How the NetBackup Accelerator works with virtual machines

See “How the NetBackup Accelerator works with virtual machines” on page 161.

How the NetBackup Accelerator works with virtual

machines
To enable acceleration of virtual machine backups, click Use Accelerator on the
policy Attributes tab.
See “Configuring a VMware policy from the Policies utility” on page 68.
The NetBackup Accelerator creates the backup stream and backup image for each
virtual machine as follows:
■ If the virtual machine has no previous backup, NetBackup performs a full backup
and uses VMware Changed Block Tracking to track the data in use for each
VMDK.
■ At the next backup, NetBackup identifies data that has changed since the
previous backup. Only changed blocks and the header information are included
in the backup, to create a full virtual disk backup.
■ The backup host sends to the media server a tar backup stream that consists
of the following: The virtual machine's changed blocks, and the previous backup
ID and data extents (block offset and size) of the unchanged blocks.
■ The media server reads the virtual machine's changed blocks, the backup ID,
and information about the data extents of the unchanged blocks. From the
backup ID and data extents, the media server locates the rest of the virtual
machine's data in existing backups.
■ The media server directs the storage server to create a new full image that
consists of the following: The newly changed blocks, and the existing unchanged
blocks that reside on the storage server. The storage server may not write the
existing blocks but rather link them to the image.
See “Accelerator notes and requirements for virtual machines” on page 161.
See “Accelerator forced rescan for virtual machines (schedule attribute)” on page 163.

Accelerator notes and requirements for virtual

machines
Note the following about Accelerator for virtual machines:
■ Accelerator for virtual machines uses VMware Changed Block Tracking (CBT)
to identify the changes that were made within a virtual machine.
 Use Accelerator to back up virtual machines 162
Accelerator notes and requirements for virtual machines

VMware CBT may occasionally reset tracking of file changes, such as after a
power failure or hard shutdown. In that case, for the next backup NetBackup
reads all the data from the vmdk files and the backup takes longer than expected.
If deduplication is enabled, the deduplication rate is lower than expected.
For more information on CBT, see the following VMware article:
Changed Block Tracking (CBT) on virtual machines (1020128)

■ Supports the disk storage units that have the following storage destinations:
■ Cloud storage. Storage that a supported cloud storage vendor provides.
■ NetBackup Media Server Deduplication Pool. In addition to NetBackup media
servers, NetBackup 5200 series appliances support Media Server
Deduplication Pool storage.
■ Qualified third-party OpenStorage devices.
To verify that your storage unit supports Accelerator, refer to the NetBackup
hardware compatibility list for the currently supported OST vendors:
NetBackup Master Compatibility List
■ It is recommended that you not enable Expire after copy retention for any
storage units that are used with storage lifecycle policies (SLP) in combination
with Accelerator. The Expire after copy retention can cause images to expire
while the backup runs. To synthesize a new full backup, the SLP backup needs
the previous backup image. If the previous image expires during the backup,
the backup fails.
■ Update the NetBackup device mapping files if needed.
The NetBackup device mapping files contain all storage device types that
NetBackup can use. To add support for the new devices or upgraded devices
that support Accelerator, download the current device mapping files from the
website.
See the NetBackup Administrator's Guide Volume I for information on the device
mapping files and how to download them.
■ Storage unit groups are supported only if the storage unit selection in the group
is Failover.
■ Supports the full backups and incremental backups. Every Accelerator backup
(from a full schedule or incremental schedule) results in a complete image of
the virtual machine.
■ You can use incremental backups (cumulative or differential) as follows: To
reduce the file-mapping overhead and to reduce the number of files that are
recorded in the NetBackup catalog. Cumulative backups may involve more
file-mapping because they do not use the random indexing method to determine
 Use Accelerator to back up virtual machines 163
Accelerator forced rescan for virtual machines (schedule attribute)

which files have changed. In some cases, differential backups may be faster
than cumulative backups.
■ If a backup of the virtual machine does not exist, NetBackup performs a full
backup. On the backup host it also accesses the VMware CBT information. This
initial backup occurs at the speed of a normal (non-accelerated) full backup.
Subsequent Accelerator backups of the virtual machine use VMware Changed
Block Tracking to accelerate the backup.

Note: When you first enable a VMware policy to use Accelerator, the next backup
(whether full or incremental) is in effect a full backup: It backs up all the virtual
machine files that are selected in the policy. If that backup is an incremental, it
may not complete within the backup window. When you upgrade policies to
NetBackup 7.6 or later and enable Accelerator, make sure that the next full
backup completes in the time that is required.

■ If the storage unit that is associated with the policy cannot be validated when
you create the policy, note: The storage unit is validated later when the backup
job begins. If Accelerator does not support the storage unit, the backup fails. In
the bpbrm log, a message appears that is similar to one of the following:

Storage server %s, type %s, doesn't support image include.

Storage server type %s, doesn't support accelerator backup.

■ Accelerator requires the storage to have the OptimizedImage attribute enabled.

See “Accelerator requires the OptimizedImage attribute” on page 164.
■ Because of a VMware restriction, BLIB is not supported for VMware templates.
As a result, NetBackup Accelerator cannot be used to back up VMware virtual
machine templates.

Accelerator forced rescan for virtual machines

(schedule attribute)
Accelerator for virtual machines uses Changed Block Tracking (CBT) technology
from VMware to identify changed blocks. NetBackup requires the changed blocks
when it creates a full virtual machine (synthesized) image. NetBackup is therefore
dependent on VMware CBT for correctly identifying changed blocks. To protect
against any potential omissions by underlying VMware CBT using timestamps, the
Accelerator forced rescan option conducts the backup by collecting all in-use
blocks as reported by VMware CBT.
 Use Accelerator to back up virtual machines 164
Accelerator requires the OptimizedImage attribute

When Accelerator forced rescan is used, all the data on the virtual machine is
backed up. This backup is similar to the first Accelerator backup for a policy. For
the forced rescan job, the optimization percentage for Accelerator is 0. The duration
of the backup is similar to a non-Accelerator full backup.

Note: Under normal operations, an Accelerator forced rescan schedule is not

necessary for VMware backups. It can be used to enforce a new baseline backup
in case VMware CBT issues are discovered. Engaging with Veritas support is
recommended in such situations.
Accelerator forced rescan is unavailable if the Use Accelerator option on the
Attributes tab is not selected.

See “Accelerator messages in the backup job details log” on page 165.

Accelerator requires the OptimizedImage attribute

Accelerator requires that the storage has the OptimizedImage attribute enabled.
To ensure that your storage is configured properly, see the documentation for your
storage option:
■ NetBackup Media Server Deduplication Pool.
The OptimizedImage attribute is enabled by default beginning with the
NetBackup 7.1 release. If you created the storage servers and pools in an earlier
release, you must configure them for OptimizedImage.
See the NetBackup Deduplication Guide.
■ Backups to a third-party disk appliance.
The storage device must support the OptimizedImage attribute.
See the NetBackup OpenStorage Solutions Guide for Disk.
■ Cloud storage that NetBackup supports.
See the NetBackup Cloud Administrator's Guide.

Accelerator backups and the NetBackup catalog

Use of Accelerator does not affect the size of the NetBackup catalog. A full backup
with Accelerator generates the same catalog size as a full backup of the same data
without Accelerator. The same is true of incremental backups: use of Accelerator
does not require more catalog space than the same backup without Accelerator.
A potential catalog effect does exist, depending on how often you use Accelerator
with full backups. A full backup with Accelerator completes faster than a normal
 Use Accelerator to back up virtual machines 165
Accelerator messages in the backup job details log

full. It may therefore be tempting to replace your incremental backups with

Accelerator full backups. Note: Since a full backup requires more catalog space
than an incremental, replacing incrementals with fulls increases the catalog size.
When changing your incrementals to fulls, you must weigh the advantage of
Accelerator fulls against the greater catalog space that fulls require compared to
incrementals.

Accelerator messages in the backup job details

log
When a virtual machine is first backed up, Accelerator is not used for that backup.
The following messages appear in the job details log:

7/25/2012 4:45:35 PM - Info bpbrm(pid=6192) There is no complete

backup image match with track journal, a regular full backup will
be performed

...

7/25/2012 4:53:22 PM - Info bpbkar32(pid=5624) accelerator sent

5844728320 bytes out of 5844726784 bytes to server, optimization 0.0%

When subsequent backups of the virtual machine use Accelerator, the following
messages appear in the job details log:

7/27/2012 4:40:01 AM - Info bpbrm(pid=412) accelerator enabled

...

7/27/2012 4:43:07 AM - Info bpbkar32(pid=4636) accelerator sent

74764288 bytes out of 5953504256 bytes to server, optimization 98.7%

This message is a key trace for Accelerator. In this example Accelerator was
successful at reducing the backup data by 98.7%.

NetBackup logs for Accelerator with virtual

machines
Accelerator does not require its own log directory. For log messages about
Accelerator, see the following standard NetBackup log directories.
 Use Accelerator to back up virtual machines 166
About reporting the amount of Accelerator backup data that was transferred over the network

Table 9-1 NetBackup logs that may contain Accelerator information

Log directory Resides on

UNIX: /usr/openv/netbackup/logs/bpbrm NetBackup master or media

server
Windows: install_path\NetBackup\logs\bpbrm

UNIX: /usr/openv/netbackup/logs/bptm NetBackup media server

Windows: install_path\NetBackup\logs\bptm

/usr/openv/netbackup/logs/bpbkar Backup host

Windows: install_path\NetBackup\logs\bpbkar

/usr/openv/netbackup/logs/bpfis Backup host

Windows: install_path\NetBackup\logs\bpfis

VxMS logs See “NetBackup logging for

VMware” on page 291.

To create the log directories, run the following command on the NetBackup servers
and backup host:
On Windows:
install_path\NetBackup\logs\mklogdir.bat

On UNIX/Linux:
/usr/openv/netbackup/logs/mklogdir

About reporting the amount of Accelerator backup

data that was transferred over the network
For Accelerator backup reporting, several NetBackup commands can report the
amount of data that is transferred over the network for each Accelerator backup.
The amount of transferred data is often much less than the size of the Accelerator
backup image.
For each Accelerator backup, NetBackup combines the client's (or VM's) changed
blocks with the unchanged data from previous backups to synthesize a backup
image. However, NetBackup sends only the changed data over the network when
the backup occurs. The resulting backup image may be much larger than the amount
of backup data that travels the network. For backup reporting, it may be important
to distinguish between the backup image size and the amount of data that was
transferred over the network.
 Use Accelerator to back up virtual machines 167
About reporting the amount of Accelerator backup data that was transferred over the network

For Accelerator backups, the network-transferred data can appear in the output of
the following NetBackup commands: bpdbjobs, bpimagelist, and bpclimagelist.
Table 9-2 lists the default location of these commands.

Table 9-2 Default location of bpdbjobs, bpimagelist, and bpclimagelist

Command Default location

bpdbjobs, Windows:
bpimagelist
install_path\NetBackup\bin\admincmd\

UNIX, Linux

/usr/openv/netbackup/bin/admincmd/

bpclimagelist Windows:

install_path\NetBackup\bin\

UNIX, Linux

/usr/openv/netbackup/bin/

The following example uses the bpimagelist command to show the results of a
backup of acmevm2:
bpimagelist -backupid acmevm2

Example output:

In this example, the backup image size in kilobytes is 7799632, and the amount of
data that was transferred over the network is 225792.
You can use the following commands to show the amount of data that was
transferred over the network for an Accelerator backup.
 Use Accelerator to back up virtual machines 168
About reporting the amount of Accelerator backup data that was transferred over the network

bpimagelist
bpimagelist -backupid backup_id [-l | -L | -json | -json_compact]

Brackets [ ] indicate optional elements, and the vertical bars | indicate that you can
choose only one of the options within the brackets.
Table 9-3 describes how the network-transferred data field appears in the
bpimagelist output.

Table 9-3 The bpimagelist options that show the amount of

network-transferred data for Accelerator backups

bpimagelist How the network-transferred data field appears

option

No option The field is unlabeled. For example: 225792

See the bpimagelist example output earlier in this topic.

-l The field is unlabeled (same as no option). For example: 225792

-L The field is labeled. For example:

Kilobytes Data Transferred: 225792

-json The field is labeled. For example:

"kilobytes_data_transferred": 225792,

-json_compact The field is labeled. For example:

"kilobytes_data_transferred":225792,

bpdbjobs
bpdbjobs –jobid job_id -report –most_columns

or
bpdbjobs –jobid job_id -report –all_columns

The network-transferred data field appears at the end of the output.

bpclimagelist
bpclimagelist -client client_name

This command can only show the network-transferred data in the field that normally
shows the Accelerator backup image size. To show the network-transferred data
with this command, you must configure a NetBackup setting:
See “Replacing the Accelerator image size with the network-transferred data in
NetBackup command output” on page 169.
 Use Accelerator to back up virtual machines 169
Replacing the Accelerator image size with the network-transferred data in NetBackup command output

Additional details on these commands are available in the NetBackup Commands

Reference Guide or in the NetBackup man pages.

Replacing the Accelerator image size with the

network-transferred data in NetBackup command
output
You can configure the output of bpimagelist, bpdbjobs, and bpclimagelist to
show the amount of Accelerator backup data that was transferred over the network
instead of the backup image size.
The following is the default bpimagelist output that shows the Accelerator image
size (see the circled value 7799632). The amount of network-transferred data
appears farther down in the output (225792):

You can configure NetBackup command output to show the network-transferred

data in the image size field. In the output, the image size value is replaced with the
network-transferred data value (see the following example). A script that reads the
image size from the command output now reads the amount of network-transferred
data.
In the following bpimagelist output, the image size field shows the
network-transferred data (225792):
 Use Accelerator to back up virtual machines 170
Replacing the Accelerator image size with the network-transferred data in NetBackup command output

Note: The same change occurs in the labeled output of the commands (such as
with the -L option of bpimagelist). For example, the Kilobytes field shows the
transferred data value (225792 in the example) rather than the Accelerator backup
image size.

To enable the reporting of network-transferred data in the Accelerator image

size field of bpimagelist, bpdbjobs, and bpclimagelist
◆ Use the bpsetconfig command to enable the output change.
To enable this change for the bpclimagelist command, enter the bpsetconfig
command on the master server. To enable this change for bpimagelist or
bpdbjobs, enter the bpsetconfig command on the server where you intend
to run bpimagelist or bpdbjobs.
Refer to Table 9-4 for the bpsetconfig command to use based on the type of
Accelerator backup that you want to report on.
 Use Accelerator to back up virtual machines 171
Replacing the Accelerator image size with the network-transferred data in NetBackup command output

Table 9-4 To enable the reporting of network-transferred data in the

Accelerator image size field of bpimagelist, bpclimagelist, or
bpdbjobs output

Type of backup Enter this command

to report on

Incremental Windows
VMware
Accelerator echo REPLACE_IMAGE_SIZE_WITH_DATA_TRANSFERRED = REPLACE_IMAGE_SIZE_
backups FOR_ACCL_INC_VMWARE | install_path\NetBackup\bin\admincmd\bpsetconfig

UNIX, Linux

echo "REPLACE_IMAGE_SIZE_WITH_DATA_TRANSFERRED = REPLACE_IMAGE_SIZE_

FOR_ACCL_INC_VMWARE" | /usr/openv/netbackup/bin/admincmd/bpsetconfig

All VMware Windows

Accelerator
backups (full and echo REPLACE_IMAGE_SIZE_WITH_DATA_TRANSFERRED = REPLACE_IMAGE_SIZE_
incremental) FOR_ACCL_ALL_VMWARE | install_path\NetBackup\bin\admincmd\bpsetconfig

UNIX, Linux

echo "REPLACE_IMAGE_SIZE_WITH_DATA_TRANSFERRED = REPLACE_IMAGE_SIZE_

FOR_ACCL_ALL_VMWARE" | /usr/openv/netbackup/bin/admincmd/bpsetconfig

Incremental Windows
Accelerator virtual
machine backups echo REPLACE_IMAGE_SIZE_WITH_DATA_TRANSFERRED = REPLACE_IMAGE_SIZE_
(VMware and FOR_ACCL_INC_VIRTUAL | install_path\NetBackup\bin\admincmd\bpsetconfig
Hyper-V)
UNIX, Linux

echo "REPLACE_IMAGE_SIZE_WITH_DATA_TRANSFERRED = REPLACE_IMAGE_SIZE_

FOR_ACCL_INC_VIRTUAL" | /usr/openv/netbackup/bin/admincmd/bpsetconfig

All Accelerator Windows

virtual machine
backups (VMware echo REPLACE_IMAGE_SIZE_WITH_DATA_TRANSFERRED = REPLACE_IMAGE_SIZE_
and Hyper-V, full FOR_ACCL_ALL_VIRTUAL | install_path\NetBackup\bin\admincmd\bpsetconfig
and incremental)
UNIX, Linux

echo "REPLACE_IMAGE_SIZE_WITH_DATA_TRANSFERRED = REPLACE_IMAGE_SIZE_

FOR_ACCL_ALL_VIRTUAL" | /usr/openv/netbackup/bin/admincmd/bpsetconfig
 Use Accelerator to back up virtual machines 172
Replacing the Accelerator image size with the network-transferred data in NetBackup command output

Table 9-4 To enable the reporting of network-transferred data in the

Accelerator image size field of bpimagelist, bpclimagelist, or
bpdbjobs output (continued)

Type of backup Enter this command

to report on

All incremental Windows

Accelerator
backups (physical echo REPLACE_IMAGE_SIZE_WITH_DATA_TRANSFERRED = REPLACE_IMAGE_SIZE_
clients and virtual FOR_ACCL_INC_ALL | install_path\NetBackup\bin\admincmd\bpsetconfig
machines)
UNIX, Linux

echo "REPLACE_IMAGE_SIZE_WITH_DATA_TRANSFERRED = REPLACE_IMAGE_SIZE_

FOR_ACCL_INC_ALL" | /usr/openv/netbackup/bin/admincmd/bpsetconfig

All Accelerator Windows

backups (full and
incremental, echo REPLACE_IMAGE_SIZE_WITH_DATA_TRANSFERRED = REPLACE_IMAGE_SIZE_
physical clients FOR_ACCL_ALL_ALL | install_path\NetBackup\bin\admincmd\bpsetconfig
and virtual
UNIX, Linux
machines)
echo "REPLACE_IMAGE_SIZE_WITH_DATA_TRANSFERRED = REPLACE_IMAGE_SIZE_
FOR_ACCL_ALL_ALL" | /usr/openv/netbackup/bin/admincmd/bpsetconfig

To reset the command output to the default setting

◆ To disable the reporting of network-transferred data in the Accelerator image
size field (return to default), enter the following:
Windows
echo REPLACE_IMAGE_SIZE_WITH_DATA_TRANSFERRED =
REPLACE_IMAGE_SIZE_DISABLED |
install_path\NetBackup\bin\admincmd\bpsetconfig

UNIX, Linux
echo "REPLACE_IMAGE_SIZE_WITH_DATA_TRANSFERRED =
REPLACE_IMAGE_SIZE_DISABLED" |
/usr/openv/netbackup/bin/admincmd/bpsetconfig

The following topic provides background on the Accelerator image size vs. the
amount of data that was transferred over the network:
See “About reporting the amount of Accelerator backup data that was transferred
over the network” on page 166.
 Chapter 10
Restore virtual machines
This chapter includes the following topics:

■ Restore notes and restrictions

■ Restore notes and restrictions on Linux

■ Restoring the full VMware virtual machine

■ About VMware virtual machine disk restore

■ Restoring VMware virtual machine disks by using Backup, Archive, and Restore

■ Restoring VMware virtual machine disks by using NetBackup commands

■ About restore of individual filesAbout restoring individual VMware files

■ Restoring individual filesRestoring individual VMware virtual machine files

■ How NetBackup handles VMware tag associations at restore

■ Browse and search virtual machines for restore

■ If the recovery host is not at the same NetBackup release level as the backup
host

Restore notes and restrictions

This topic is about restores from a NetBackup backup of a VMware virtual machine.
Before you begin the restore, note the following:
■ Cross-platform restore of individual files is not supported. You can restore
Windows files to Windows guest operating systems but not to Linux. You can
restore Linux files to supported Linux guest operating systems but not to
Windows. In other words, the restore host must be the same platform as the
files that you want to restore.
 Restore virtual machines 174
Restore notes and restrictions

See “About restore of individual filesAbout restoring individual VMware files”

on page 223.
■ Unless a NetBackup client is installed on the virtual machine, you must do the
restore from the NetBackup master server.
■ To restore files to the original virtual machine location, the destination must be
specified as the virtual machine's host name (not display name or UUID).
■ To restore directly to an ESX server, the name that is specified for the restore
must match the ESX server’s official host name. The name must be in the same
format in which it is registered in DNS and in the VMware server (whether short
or fully-qualified).
See “Adding NetBackup credentials for VMware” on page 45.
■ If the VM’s display name was changed after the VM was backed up, the
pre-recovery check may fail when you click Start Recovery:

VM exists overwrite -Failed. Vmxdir for VM exists

You can ignore the error and click Start Recovery, but note: The restore may
succeed but the folder that contains the vmx file for the newly restored VM has
a different name than the vmx folder of the existing VM. VMware does not
rename this folder when the VM is renamed, but continues to use the existing
folder.
As an alternative, restore the VM to a different location.

■ A virtual machine template cannot be restored to a standalone ESX server.

Because templates are a feature of vCenter servers, you must restore the
template through vCenter. If you restore a template to a standalone ESX server,
the template is converted to a normal virtual machine and is no longer a template.

■ NetBackup supports backup and recovery of VMware NVRAM files and the
vTPM devices that are associated with virtual machines.
■ A NetBackup 8.3 or later backup or recovery host is required for NVRAM
and vTPM protection. Supported recovery methods include Full VM recovery
and VMware Instant Recovery.
■ NetBackup does not support the backup or restore of NVRAM and vTPM
for the virtual machines whose display names begin with a period ('.'). An
existing VMware limitation prevents downloading or uploading data store
files beginning with a period ('.') to a virtual machine's working directory as
these appear as hidden files.

■ If the virtual machine was backed up by its display name or UUID, and the
display name differs from the host name, note: You must specify the correct
destination client for the restore. Use the Specify NetBackup Machines and
Policy Type dialog in the NetBackup Backup, Archive, and Restore interface.
 Restore virtual machines 175
Restore notes and restrictions

See “Invalid client error when you restore files using NetBackup BAR interface
installed on the virtual machine” on page 311.
■ Restore of individual files from a backup of the full virtual machine is not
supported if the virtual machine contains Storage Foundation Volume Manager
volumes.
■ To restore Windows NTFS-encrypted files individually, you must install a
NetBackup client on the virtual machine.
See “NetBackup for VMware best practices” on page 283.
■ If the attempt to restore a full virtual machine fails while using the SAN transport
type, try the NBD transport type instead.
■ Restoring a virtual machine with a transport mode of NBD or NBDSSL may be
slow in the following cases:
■ The virtual machine had many small data extents due to heavy fragmentation.
(A file system extent is a contiguous storage area defined by block offset
and size.)
■ The restore is from a block-level incremental backup and the changed blocks
on the disk were heavily fragmented when the incremental backup occurred.
For faster restores in either of these cases, use the hotadd transport mode
instead of NBD or NBDSSL.
■ VMware does not support the restore of virtual machines directly to an ESX 5.x
server that vCenter manages. To restore the virtual machine, select the vCenter
server as the destination.
As an alternative, you can set up an independent ESX server to be used for
restores. You must add NetBackup restore credentials for that ESX server by
means of the VMware restore ESX server server type.
See “Adding NetBackup credentials for VMware” on page 45.
■ For the SAN transport mode, the job may be slow when you restore to a vCenter
Server. For greater speed, designate a VMware restore ESX server as the
destination for the restore.
See “Adding NetBackup credentials for VMware” on page 45.
■ For the SAN transport mode and a restore host on Windows 2008 R2 or later,
the restore fails if the datastore’s LUN is offline. The detailed status log contains
messages similar to the following:

5/22/2013 4:10:12 AM - Info tar32(pid=5832) done. status: 24:

socket write failed
5/22/2013 4:10:12 AM - Error bpbrm(pid=5792) client restore EXIT
STATUS 24: socket write failed
 Restore virtual machines 176
Restore notes and restrictions

■ Make sure the status of the SAN disk on the restore host is online (not offline).
Disk status can be checked or changed using the Windows diskpart.exe
utility or the Disk Management utility (diskmgmt.msc). When the disk status
reads online, retry the restore.
■ If multipathing is enabled, make sure all the paths are online.

■ A restore by means of the SAN transport mode may be slow in other

circumstances. The following Veritas article provides details:
VMware Transport Modes: Best practices and troubleshooting
■ The APIs in VMware's Virtual Disk Development Kit (VDDK) contain the following
limitation: The maximum write speed during virtual machine restore is roughly
one third of the hardware's maximum speed. The following Veritas tech note
contains further information:
Two causes of slow NetBackup for VMware restore performance
http://www.veritas.com/docs/TECH169860
■ If a virtual machine had vmdk files in different directories in the same datastore,
note: When the virtual machine is restored to the original location its vmdk files
are restored to a single directory, not to the original directories. (This behavior
follows current VMware design.)
As a workaround, do the following: Remove the vmdk files from the restored
virtual machine, move the files to their respective directories, then re-attach the
moved files to the virtual machine.
■ If the original VM contains encrypted vmdk files, after restoring the full VMware
virtual machine to the original location or after an in-place disk restore, the
restored disks may not be compliant to the VM encryption policy. Therefore, the
restored VM must be reconfigured manually to comply with the policy. Otherwise,
the virtual disks of the restored VM might be left in an unencrypted state.
■ When restoring large files, make sure that no snapshots are active on the
destination virtual machine. Otherwise, the files are restored to the VMware
configuration datastore, which may be too small to contain the files you want to
restore. In that case, the restore fails.
The configuration datastore (sometimes called the vmx directory) contains the
configuration files that describe the virtual machine, such as *.vmx files. Note
that active snapshots of vmdk files are also stored on the configuration datastore.
■ If you cancel the virtual machine restore before it completes, the not-fully-restored
virtual machine remains at the target location. NetBackup does not delete the
incomplete virtual machine when the restore job is canceled. You must manually
remove the incomplete virtual machine.
 Restore virtual machines 177
Restore notes and restrictions on Linux

■ If the virtual machine display name contains unsupported characters, the backup
may succeed but the restore fail. To restore the virtual machine, you must change
the display name to contain supported characters only and retry the restore.
See “NetBackup character restrictions for the Primary VM identifier” on page 37.
■ NetBackup for VMware does not support individual file restore by means of
client-direct restore.

■ On a restore, NetBackup recreates the linking between a hard link and its original
file only in this case: The link file and its target file are restored in the same job.
If each file is restored individually in separate restore jobs, they are restored as
separate files and the link is not re-established.
■ If you restore a VM in vCloud to an expired vApp, the vApp is automatically
renewed and added back into the vCloud organization. If the expired vApp
contained other VMs, all those VMs are also removed from the expired list and
added to the organization.
Note that in vCloud Director, an expired vApp must be renewed before you can
import a VM into that vApp.
■ With a remote connection from a Windows Java GUI that uses the English locale,
the restore of files that have non-ASCII characters may fail.
See the following tech note for further information on how to restore the files:
http://www.veritas.com/docs/TECH75745
■ In VMware vSphere 6.0 U1b and later, a full restore of a virtual machine may
trigger an alarm if the original VM was not deleted. The alarm is a VM MAC
address conflict alarm. This VMware alarm behavior is by design. If there is a
MAC address conflict, VMware eventually changes the MAC address of the new
VM. If you do not want to receive alarms, disable the VM MAC address conflict
alarms in vCenter.
■ See “NetBackup for VMware: notes and restrictions” on page 27.
See “Restore notes and restrictions on Linux” on page 177.
See “If the recovery host is not at the same NetBackup release level as the backup
host” on page 236.

Restore notes and restrictions on Linux

This topic relates to the restore of files from a NetBackup backup of a VMware
virtual machine that runs Linux.
Before you begin the restore, note the following:
■ Cross-platform restore of individual files is not supported. You can restore Linux
files to supported Linux guest operating systems but not to Windows.
 Restore virtual machines 178
Restore notes and restrictions on Linux

■ To migrate an ext2 or ext3 file system to ext4, note: Make sure to follow the
instructions under "Converting an ext3 file system to ext4" on the following page
of the Ext4 wiki:
https://ext4.wiki.kernel.org/index.php/UpgradeToExt4
If you do not follow these instructions, data in a newly created ext4 file is not
promptly flushed from memory to disk. As a result, NetBackup cannot back up
the data of recently created files in the ext4 file system. (The NetBackup snapshot
captures the file as zero length.)
As a workaround for the file systems that were not correctly migrated, do one
of the following:
■ Run the Linux sync command on the ext4 file system before starting each
backup.
■ Make sure that snapshot quiesce is enabled in the Linux guest OS. Contact
your operating system vendor and VMware for additional information.

■ For Linux virtual machines, NetBackup cannot restore individual files from
software RAID volumes. The files are restored when you restore the entire virtual
machine.
■ The Linux ext4 file system includes a persistent pre-allocation feature, to
guarantee disk space for files without padding the allocated space with zeros.
When NetBackup restores a pre-allocated file (to any supported ext file system),
the file loses its preallocation and is restored as a sparse file. The restored
sparse file is only as large as the last byte that was written to the original file.
Note also that subsequent writes to the sparse file may be non-contiguous.
■ NetBackup supports backup and restore of Linux LVM2 volumes, including
individual file restore from an LVM2 volume. Note however that NetBackup does
not support individual file restore from a snapshot that was created by means
of the snapshot feature in LVM2. If an LVM2 snapshot exists at the time of the
backup, the data in the snapshot is captured in the backup. The data can be
restored along with the rest of the virtual machine data when you recover the
entire virtual machine.
■ NetBackup supports backup of Linux FIFO files and socket files. NetBackup
does not support restoring FIFO files and socket files individually. FIFO files
and socket files can be restored along with the rest of the virtual machine data
when you recover the entire virtual machine.
■ For Linux virtual machines, the ext2, ext3, and ext4 file systems are supported
for individual file restore. If a partition uses some other file system, the backup
succeeds but files in that partition cannot be individually restored. To allow
NetBackup to present mount points for restore in the Backup, Archive, and
Restore interface, note: The "/" (root) partition must be formatted with ext2, ext3,
or ext4.
 Restore virtual machines 179
Restoring the full VMware virtual machine

See “Mount points not available when restoring files from a Linux virtual machine”
on page 311.
■ When you restore Linux files individually to an NFS-shared device on a Linux
virtual machine, NetBackup can only restore the file data and attributes. The
extended attributes cannot be restored to NFS-shared devices.
■ For a virtual machine that is running a Linux guest operating system: When you
restore a virtual machine, the ESX server may assign the virtual machine a new
(virtual) MAC address. After you restart the virtual machine, you may have to
configure its MAC address. For instance, the original MAC address of the virtual
machine may be in a configuration file that has to be updated.
Refer to your VMware documentation for more details.
■ For Linux, additional notes apply.
See “NetBackup for VMware: notes on Linux virtual machines” on page 34.
See “Restore notes and restrictions” on page 173.
See “Restoring the full VMware virtual machine” on page 179.
See “If the recovery host is not at the same NetBackup release level as the backup
host” on page 236.

Restoring the full VMware virtual machine

This topic describes how to r19546022estore the VMware virtual machines that
were backed up with NetBackup for VMware.
To restore the full virtual machine
1 In the NetBackup Administration Console, click Backup, Archive, and Restore.
2 Click Actions > Specify NetBackup Machines and Policy Type.
 Restore virtual machines 180
Restoring the full VMware virtual machine

3 Specify the following:

Server to use for The NetBackup master server that directed the backup
backup and
restores

Source client for The VMware virtual machine that was backed up.
restores
Click Search VM Clients to search or browse for a virtual
machine. Use this option to locate a virtual machine in a large,
multi-layered virtual environment.

See “Browse and search virtual machines for restore”

on page 232.

Or enter the type of name that was selected for the Primary VM
identifier option on the policy VMware tab. For example, if the
Primary VM identifier option is set to VM host name, enter the
virtual machine's host name.

Destination client This field is ignored. The destination for the restore is specified
for restores in a different screen as explained in a later procedure.

Policy type for The type of policy that made the backup (VMware).
restores

4 Click OK.
5 For the Restore type, select Virtual Machine Backups.
6 You can use the Start date and End date icons to select the date range within
which the backup occurred.
 Restore virtual machines 181
Restoring the full VMware virtual machine

7 Under Directory Structure, click on the backup you want to restore.

8 Click Restore.
The type of dialog that appears depends on whether or not the backup was
enabled for vCloud.

vSphere (not vCloud) See “Virtual Machine Recovery dialog boxes (restore to original location)” on page 182.

See “Virtual Machine Recovery dialogs boxes (restore to alternate location)” on page 188.

vCloud See “Virtual machine recovery dialog boxes for vCloud Director” on page 276.

See “About VMware virtual machine disk restore” on page 196.

See “About restore of individual filesAbout restoring individual VMware files”
on page 223.
 Restore virtual machines 182
Restoring the full VMware virtual machine

Virtual Machine Recovery dialog boxes (restore to original location)

This topic applies to NetBackup for VMware. It describes how to restore a VMware
virtual machine to its original location.
The following topics describe the Backup, Archive, and Restore screens for restoring
a virtual machine to the original location. These screens do not apply to vCloud
Director.
See “Virtual machine recovery dialog boxes for vCloud Director” on page 276.
See “Recovery Destination dialog box” on page 182.
See “Recovery Options dialog box” on page 182.
See “Virtual Machine Options dialog box” on page 184.
See “Perform Recovery dialog box” on page 187.

Recovery Destination dialog box

This topic applies to NetBackup for VMware. It describes how to restore a VMware
virtual machine.
You can select the type of destination for the virtual machine recovery: the original
location or alternate location.
This dialog includes the following:
■ Original settings of the virtual machine at the backup
Shows the configuration of the virtual machine at the time of the backup.
When you restore the virtual machine to its original location, these configuration
settings are used and cannot be changed.
When you restore to an alternate location (not to the original location), you can
change these settings after you click Next.
■ Recover virtual machine to
Select Original Location or Alternate Location.
For restore to the original location:
See “Recovery Options dialog box” on page 182.
For restore to an alternate location:
See “Restoring the full VMware virtual machine” on page 179.

Recovery Options dialog box

This topic applies to NetBackup for VMware. It describes how to restore a VMware
virtual machine.
 Restore virtual machines 183
Restoring the full VMware virtual machine

Select the recovery host and transport mode, as follows:

■ Destination settings
Shows the configuration of the virtual machine at the time of backup. These
settings are applied to the virtual machine when it is restored.

Note: This information is included for recovery to vSphere only (not for vCloud).

■ NetBackup Recovery Host

The host that performs the recovery. The default is the host that performed the
backup (the backup host).
In most cases, you can use the host that performed the backup. Selecting a
different host as the recovery host can result in a slow restore or a failed restore,
depending on the transport mode.
For example, the restore fails if you select the SAN transport mode in the
following case:
■ The host that performed the backup used a SAN connection to access the
datastore.
■ The host you select as recovery host does not have SAN access to the
datastore.

■ Media Server
You can use this option to select a media server that has access to the storage
unit that contains the backup image. An example of such an environment is a
Media Server Deduplication Pool (MSDP) with multiple media servers.

Note: If the storage unit that contains the backup image is not shared with
multiple media servers, this option is grayed out.

■ Transport modes
Determines how the restore data is sent from the recovery host to the VMware
datastore. The appropriate option depends in part on the type of network that
connects the VMware datastore to the recovery host.
The default is the type that the backup used.

san For unencrypted transfer over Fibre Channel (SAN)

or iSCSI.
 Restore virtual machines 184
Restoring the full VMware virtual machine

nbd For unencrypted transfer over a local network that

uses the Network Block Device (NBD) driver protocol.
This mode of transfer is usually slower than Fibre
Channel.

nbdssl For encrypted transfer (SSL) over a local network that

uses the Network Block Device (NBD) driver protocol.
This mode of transfer is usually slower than Fibre
Channel.

hotadd Requires the recovery host in a virtual machine.

For instructions on this transport mode and on

installing the recovery host in a VMware virtual
machine, refer to your VMware documentation.

Move Up, Move Down Use these options to change the order in which
NetBackup tries each selected mode. Highlight a mode
and click Move Up or Move Down.

For example: Assume that all four transport modes

are selected, and the order is SAN, hotadd, nbd, and
nbdssl. If one of the virtual disks cannot be accessed
using SAN, the SAN transport mode is not used for
any of the virtual machine’s disks. NetBackup then
tries to use the hotadd mode for all the disks.
NetBackup continues to try each mode until it finds
one that succeeds for all the disks.

See “Perform Recovery dialog box” on page 187.

Virtual Machine Options dialog box

Virtual Machine Options:
■ Restore BIOS UUID xxx instead of creating a new UUID
Retains the UUID of the original virtual machine (note that the UUID is a globally
unique identifier). The virtual machine is restored with the same UUID that it
had before the restore.
Note the following:
■ If a virtual machine with the same display name but with a different UUID
exists at the target restore location, the restore fails. You must either delete
the existing virtual machine and run the restore, or keep the existing virtual
machine and abandon the restore.
 Restore virtual machines 185
Restoring the full VMware virtual machine

■ If you do not want to keep the existing virtual machine, you can do one of
the following: Remove the existing virtual machine, or log into the ESX server
and remove the directory where the virtual machine resides.

■ Restore Instance UUID xxx instead of creating a new UUID

Retains the Instance UUID of the original virtual machine (note that the Instance
UUID is a vCenter specific unique identifier of a virtual machine). The virtual
machine is restored with the same Instance UUID that it had when it was backed
up.
If the restore of the virtual machine is to a standalone ESXi host, this option is
disabled.
If a virtual machine with the same Instance UUID exists at the target restore
location, a message appears that the UUID is used already. In this case, the
original instance UUID is not restored and a new UUID is assigned for the virtual
machine.
■ Overwrite the existing virtual machine
If a virtual machine with the same display name exists at the destination, that
virtual machine must be deleted before the restore begins. Otherwise, the restore
fails. Select this option to have the virtual machine deleted.
■ Remove backing information for devices like DVD/CD-ROM drives, serial
or parallel ports
If a virtual machine had an ISO file that was mounted on a CD when the virtual
machine was backed up, note: The ISO file must be available on the target ESX
server when you restore the virtual machine. If the ISO file is not available on
the ESX server, the restore fails.
This option restores the virtual machine without restoring an ISO file that was
mounted when the virtual machine was backed up.
■ Remove network interfaces
Removes any network interfaces from the restored virtual machine.
Select this option if:
■ The network connections on the destination virtual machine have changed
since the backup was made.
■ Or if the original virtual machine still exists and a duplicate may cause
conflicts.

■ Remove tag associations

When this option is selected, NetBackup does not attempt to restore tag
associations when it restores the virtual machine. If the box is not checked,
NetBackup attempts to restore all tag associations from the backup. If NetBackup
cannot restore one or more of the tag associations, the restore exits with a
NetBackup Status Code 1.
 Restore virtual machines 186
Restoring the full VMware virtual machine

■ Power on virtual machine after recovery

Select this option to have the recovered virtual machine automatically turned
on when the recovery is complete.
■ Override default job priority
Place a check mark in this box to override the default priority.
■ Job Priority
Use this field to set a higher priority.
■ Retain original hardware version
This option restores the virtual machine with its original hardware version (such
as 4). It retains the original version even if the target ESX server by default uses
a different hardware version (such as 7 or 8). If the target ESX server does not
support the virtual machine’s hardware version, the restore may fail.
If this option is disabled, the restored virtual machine is converted to the default
hardware version that the ESX server uses.
Format of restored virtual disks:
■ Original provisioning
Restores the virtual machine's virtual disks with their original provisioning.

Note: Regarding restore into vCloud Director: For Original provisioning,

NetBackup restores the virtual machine's virtual disks with the format they had
when the virtual machine was backed up. That format may be different from
either of the following: The default format of the vCloud organization when the
virtual machine was originally created, or the format for the destination
organization.

■ Thick provisioning Lazy Zeroed

Configures the restored virtual disks in the thick format. The virtual disk space
is allocated when the disk is created. This option restores the populated blocks,
but initializes vacant blocks with zeros later, on demand.
■ Thick provisioning Eager Zeroed
Configures the restored virtual disks in the thick format. Restores the populated
blocks and immediately initializes vacant blocks with zeros (eager zeroed).
Creation of the virtual disks may take more time with this option. However, if
the restore occurs over a SAN, the eager zeroed feature may speed up the
restore by reducing network communication with the vCenter server.
■ Thin provisioning
Configures the restored virtual disks in the thin format. Restores the populated
blocks but does not initialize vacant blocks or commit them.
 Restore virtual machines 187
Restoring the full VMware virtual machine

Thin provisioning saves disk space through dynamic growth of the vmdk file.
The vmdk files are no larger than the space that the data on the virtual machine
requires. The virtual disks automatically increase in size as needed.

Note: If the vmdk is completely written, VMware automatically converts a thin

disk to Thick provisioning Eager Zeroed.

For more information on virtual disk provisioning, refer to VMware's

documentation.

Perform Recovery dialog box

This topic applies to NetBackup for VMware. It describes how to restore a VMware
virtual machine.
Review the settings to be applied to the restored virtual machine.
■ Recovery settings
Lists the settings that are used for the recovery.
■ Run Pre-Recovery Check
Click this option to verify the following: The credentials, the available space on
the datastore, and other important requirements. For a description of the checks,
see Table 10-1.
■ Start Recovery
Starts the recovery of the virtual machine.

Table 10-1 Pre-recovery checks

Type of check Description

Verify vSphere server Verifies that NetBackup has the correct credentials to access the vSphere server for the
credential restore.

Verify datacenter path Verifies that NetBackup can access the datacenter that is designated for the restore.

Verify folder path Verifies that NetBackup can access the VM folder that is designated for the restore.

Unsupported non-ASCII Checks for unsupported characters in the VM display name, datastore name, or network
characters name.
 Restore virtual machines 188
Restoring the full VMware virtual machine

Table 10-1 Pre-recovery checks (continued)

Type of check Description

VM exists overwrite If the original VM exists at the restore destination, verifies that the Overwrite the existing
virtual machine option is selected. If the VM exists and the overwrite option is selected,
the VM is overwritten with the restored VM.

If the original VM does not exist at the restore destination, verifies that the original VM's
vmx configuration file directory does not exist. If the original VM does not exist but the vmx
directory exists, specify a different display name for the restored the VM.

Adequate space on Verifies that the datastore (or datastore cluster) for the restore has enough space for the
datastore (or datastore restored VM's vmx configuration file.
cluster) for configuration
file

VMware connectivity test Verifies that the NetBackup restore host can communicate with the vCenter server.

Datastore/Datastore Verifies that the datastore (or datastore cluster) for the restore has enough space for the
Cluster space available restored VM files.

SAN datastore Verifies that the selected datastore has a compatible block size.
compatibility
For example, this check fails if it detects the following problem: The VM’s vmdk files are
not a multiple of the VMFS block size. In that case, the last partial-block write may fail
during the restore, and the restore job fails with status 2820. The NetBackup job details
log contains messages similar to the following:

12/12/2011 3:12:28 AM - Critical bpbrm(pid=3560) from client

io.acme.com: FTL - Virtual machine restore: file write failed

In this case, use the NBD or the NBDSSL transport mode when you restore the VM.

Verify unique BIOS UUID If only the SAN transport mode is selected for the restore, determines if a VM with the same
BIOS UUID exists on the specified vCenter.

See “Virtual Machine Recovery dialogs boxes (restore to alternate location)”

on page 188.

Virtual Machine Recovery dialogs boxes (restore to alternate location)

This topic applies to NetBackup for VMware. It describes how to restore a VMware
virtual machine.
The following topics describe the Backup, Archive, and Restore screens for restoring
a virtual machine to an alternate location. These screens do not apply to vCloud
Director.
See “Virtual machine recovery dialog boxes for vCloud Director” on page 276.
 Restore virtual machines 189
Restoring the full VMware virtual machine

See “Recovery Destination dialog box (restore to alternate location)” on page 189.
See “Recovery Options dialog box (restore to alternate location)” on page 189.

Recovery Destination dialog box (restore to alternate

location)
This topic applies to NetBackup for VMware. It describes how to restore a VMware
virtual machine to an alternate location.
■ Original settings of the virtual machine when it was backed up
Shows the configuration of the virtual machine at the time of the backup. When
restoring to an alternate location (not to the original location), you can change
these settings after you click Next.
■ Recover virtual machine to
Select Alternate location.
See “Recovery Options dialog box (restore to alternate location)” on page 189.

Recovery Options dialog box (restore to alternate location)

This topic applies to NetBackup for VMware. It describes how to restore a VMware
virtual machine to an alternate location.
The Recovery Options dialog initially displays the original virtual machine settings
that were recorded in the NetBackup catalog when the virtual machine was backed
up. Before accepting any default values, you may have to verify that the original
hosts or servers are still available.
The dialog contains the following fields:
■ NetBackup Recovery Host
The recovery host transfers the virtual machine files to the VMware destination
that is specified in other fields on this dialog. The default for the recovery host
is the backup host that backed up the virtual machine.
Consider the following when selecting a recovery host:
■ To use the original VMware backup host for the recovery, verify that the
backup host still exists. The original backup host may have been removed
or reconfigured after the backup occurred.
■ The current processing load on the host that you want to use may be a factor.
The recovery host undergoes significant I/O activity when it restores a virtual
machine.
■ The network connection between the recovery host and the destination
datastore may limit the type and speed of data transmission. For example,
 Restore virtual machines 190
Restoring the full VMware virtual machine

to use the SAN transport mode, the recovery host must have access to the
appropriate LUNs in the datastore.

■ Media Server
You can use this option to select a media server that has access to the storage
unit that contains the backup image. An example of such an environment is a
Media Server Deduplication Pool (MSDP) with multiple media servers.

Note: If the storage unit that contains the backup image is not shared with
multiple media servers, this option is grayed out.

■ Transport modes
Determines how the data is sent from the recovery host to the VMware datastore.
The appropriate mode depends on the type of connection between the
NetBackup recovery host and the VMware datastore.
More information is available about these options.
See “Recovery Options dialog box” on page 182.
■ vCenter Server
Specifies the vCenter server for the restored virtual machine. To use the original
vCenter Server (the default), verify that the original vCenter server still exists.
To use a different vCenter Server, consider whether it is compatible with the
virtual machine you want to restore. If you attempt to restore to an earlier vSphere
server that does not support the virtual machine’s hardware version, note: The
VMware Tools component converts the virtual machine to an earlier hardware
version. But the restore may fail if the virtual machine uses devices that only
the later hardware version supports.
In other cases, you can use the Retain original hardware version option on
the Virtual Machine Options dialog.
The restore may finish faster if you use the SAN transport mode with a restore
ESX server (bypassing any vCenter servers).
When choosing a target host for the restore, note the following:
■ If NetBackup has credentials to a vCenter server but does not have
credentials to a restore ESX server, you can perform the restore. NetBackup
uses the vCenter credentials to access the ESX server but does not send
the virtual machine data through the vCenter.

Note: If you use the SAN transport mode for the restore, the restore job
may take longer than expected.
 Restore virtual machines 191
Restoring the full VMware virtual machine

The credentials must give NetBackup full access privileges to the vCenter
server.
See “Adding NetBackup credentials for VMware” on page 45.
■ If NetBackup has credentials to a vCenter server as well as to a VMware
restore ESX server, the restore may complete faster. In this case, by default
NetBackup communicates with the restore ESX server during data transfer.
You can select a vCenter folder in which to place the restored virtual machine.
The credentials must give NetBackup full access privileges to the restore
ESXi server.
Select the vCenter Server, and click Change to select the VMware restore
ESXi server.

Note: NetBackup uses the selected vCenter to create the virtual machine.
NetBackup then writes the .vmdk files directly to the ESX server using the
Restore ESX Server credentials to that server.

■ If NetBackup has credentials to an ESX server, you can designate that server
as the target for the restore. The credentials must give NetBackup full access
privileges to the ESX server.
Select None for the vCenter Server, and click Change to select the ESX
server.

Note: To restore to an ESX server, the ESX must be standalone, not

managed by vCenter. To restore to an ESX under vCenter, NetBackup must
have VMware restore ESX server credentials for the ESX. Select the vCenter,
and click Change to select the VMware restore ESX server.

■ ESX Server/Cluster
Specifies the ESX server or cluster on which the restored virtual machine is to
reside. To use the original ESX server or cluster (the default), verify that the
original ESX server or cluster still exists.
Click Change to browse for a different ESX server or cluster.
See “Select ESX server dialog box (restore to alternate location)” on page 193.
■ Datacenter
Shows the VMware data center containing the selected ESX server.
■ Folder
A folder is a logical entity within the vCenter that contains VMware objects, such
as datastores, networks, and virtual machines.
 Restore virtual machines 192
Restoring the full VMware virtual machine

By default, this field shows the VMware folder that contained the virtual machine
when the virtual machine was backed up. If you select a different ESX server
to which to restore the virtual machine, this field changes to None. You must
then click Change to select a folder within the data center for the restored virtual
machine.
See “Select Folder dialog box (restore to alternate location)” on page 193.
■ Display Name
Specifies the VMware display name for the restored virtual machine. The default
is the display name that the virtual machine had when it was backed up.
The display name must be unique for the vCenter Server where the virtual
machine is restored.

Note: If a virtual machine with this display name already exists at this location
(or at the original location), you are prompted to click Overwrite the existing
virtual machine. You cannot restore the virtual machine if the result is two
virtual machines with the same display name on the same vCenter server.

■ Resource Pool/vApp
Use this option to have the restored virtual machine assigned to either a VMware
resource pool or to a vApp. Resource pools manage the host's CPU and memory.
vApps are logical containers for virtual machines, and also share some
functionality with virtual machines.
■ Datastore/Datastore Cluster
Specifies the VMware datastore or datastore cluster that contains the virtual
machine configuration files. Click Change to make a selection.
This datastore (sometimes called the vmx directory) contains the configuration
files that describe the virtual machine, such as *.vmx files. Active snapshots of
vmdk files are also stored on this datastore.

Note: The Datastore field shows the name of the datastore that contained the
virtual machine data when the virtual machine was backed up. Even if the
datastore was in a datastore cluster, the field shows the name of the datastore,
not the datastore cluster. When the virtual machine is restored, NetBackup
determines how the datastore is currently configured (in a cluster or not) and
configures the virtual machine accordingly.

■ Use the same datastore/datastore cluster for all VMDKs

This option assigns all the restored virtual machines to the datastore or datastore
cluster that is selected in the Datastore/Datastore Cluster field.
 Restore virtual machines 193
Restoring the full VMware virtual machine

If this option is not selected, you assign the VMDKs to specific datastores or
datastore clusters in a later dialog (the Storage Destination dialog).
See “Select ESX server dialog box (restore to alternate location)” on page 193.
See “Select Folder dialog box (restore to alternate location)” on page 193.
See “Storage Destination dialog box (restore to alternate location)” on page 194.

Select ESX server dialog box (restore to alternate location)

Use this dialog to enter the name of the ESX server or cluster, or drill down in the
list to select the server or cluster.
■ Specify ESX server
Enter the name of the ESX server or cluster in this field. As you type a name,
the list of available servers or clusters automatically expands and the nearest
match is highlighted.
■ Search
If necessary, click Search to locate the server name or cluster name that you
typed.

Note: Some older ESX servers may not support the version of the virtual machine
that you want to restore. Verify that the ESX server is compatible with the virtual
machine.

See “Select Folder dialog box (restore to alternate location)” on page 193.
See “Storage Destination dialog box (restore to alternate location)” on page 194.

Select Folder dialog box (restore to alternate location)

This topic applies to NetBackup for VMware. It describes how to restore a VMware
virtual machine to an alternate location.
The Select Folder dialog shows the vCenter folders that exist on the datacenter
that is shown in the Datacenter field.
Pick a folder from the list of available folders. The restored virtual machine is placed
in this folder.
If two datacenters of the same name exist in the vCenter server, all folders for the
identically named datacenters are displayed. After you select a folder and click OK,
verify that the folder on the Recovery Options dialog contains the destination ESX
server. If the selected folder is on the wrong datacenter, the restore fails.
See “Storage Destination dialog box (restore to alternate location)” on page 194.
 Restore virtual machines 194
Restoring the full VMware virtual machine

Select Datastore or Datastore Cluster dialog box (restore

to alternate location)
Use this screen to select a datastore or datastore cluster for the restored virtual
machine. You can use the search field, or drill down manually into the list.

Note: Only datastores and clusters visible to the ESX server are displayed.

Specify Datastore
Type a name and click Search. The list of available datastores and datastore
clusters automatically expands and the nearest match is highlighted.
See “Recovery Options dialog box (restore to alternate location)” on page 189.

Virtual Machine Options dialog box (restore to alternate

location)
This dialog is described in a previous topic:
See “Virtual Machine Options dialog box” on page 184.

Storage Destination dialog box (restore to alternate

location)
This topic applies to NetBackup for VMware. It describes how to restore a VMware
virtual machine to an alternate location.
Select the datastores where the .vmdk files are to be restored.
■ Source VMDK File Name
Shows the names of the .vmdk files that were backed up.
■ Datastore/Datastore Cluster
Shows the VMware datastore or datastore cluster where the .vmdk files are to
be restored. Click in this field to select another datastore.
See “Select Datastore or Datastore Cluster dialog box (restore to alternate
location)” on page 194.
If the datastore is in a datastore cluster, NetBackup displays the name of the
datastore (not the name of the cluster).
■ Datastore/Datastore Cluster Status
Indicates whether the datastore or datastore cluster has enough space to contain
the selected .vmdk file.

OK The datastore or datastore cluster has enough space to

contain the selected .vmdk file.
 Restore virtual machines 195
Restoring the full VMware virtual machine

Insufficient space The datastore or datastore cluster does not have enough
space to contain the selected .vmdk file.

See “Network Connections and Other Recovery Options dialog box (restore to
alternate location)” on page 195.

Network Connections and Other Recovery Options dialog

box (restore to alternate location)
This topic applies to NetBackup for VMware. It describes how to restore a VMware
virtual machine to an alternate location.
For restore into vSphere, this dialog displays all of the networks that the restored
virtual machine can connect to.
For restore into vCloud Director, this dialog displays the networks that the virtual
machine was connected to (if any) at the time of the backup.
For vCloud Director: If the virtual machine was not connected to any networks during
backup, or if none of those networks are currently available, the dialog states: "No
networks available for selected virtual machine." In that case, if a network connection
is required, create it after the restore completes.
■ Retain original network configuration
If possible, this option retains the original MAC address and other configuration
settings of the virtual machine. Note that vCenter assigns a new MAC address
if the original exists. This option configures the virtual machine with its original
network labels, even if those networks are not available at the restore location.
When you select this option, any network names that are shown in this dialog
are ignored.
If this option is unselected, the virtual machine's original network settings are
discarded. You can select available networks under Network Names.
■ Network Names
The available virtual networks are displayed in addition to physical networks.
Select the networks that you want the restored virtual machine to be connected
to. This option creates new network adaptors for the selected networks.

Note: On the restored virtual machine, the network adapters are not configured.

Note: Click the Network Names check box to select or unselect all the networks
in the list.
 Restore virtual machines 196
About VMware virtual machine disk restore

Use Move Up and Move Down to reorder the networks. The order in the list
determines the order of the corresponding virtual network adapters (NIC cards)
for the virtual machine.
When the virtual machine is backed up, its network connections (physical or
virtual) are recorded in the NetBackup catalog.
For vSphere, NetBackup determines what networks currently exist at the restore
location and lists the network names as follows:
■ If a network that was recorded in the backup catalog for the virtual machine
still exists, it is listed on this screen and automatically selected.
■ If a network is available that was not listed in the NetBackup catalog, it is
listed on this screen but not selected.
■ If a network was recorded in the backup catalog but is not currently available,
it is not listed.

See “Perform Recovery dialog box (restore to alternate location)” on page 196.

Perform Recovery dialog box (restore to alternate location)

This topic applies to NetBackup for VMware. It describes how to restore a VMware
virtual machine to an alternate location.
■ Recovery settings
Lists the settings that are used for the recovery.
■ Run Pre-Recovery Check
Verifies the credentials and appropriate paths and connectivity, determines
whether the datastore or datastore cluster has available space, and reviews
other requirements. You must run this check at least once. You can proceed
with the restore even if the check fails. More details are available on the
Pre-Recovery Check.
See “Perform Recovery dialog box” on page 187.
■ Start Recovery
Starts the recovery of the virtual machine.
See “Restoring the full VMware virtual machine” on page 179.
See “About restore of individual filesAbout restoring individual VMware files”
on page 223.

About VMware virtual machine disk restore

NetBackup supports the restore of individual VMware virtual machine disks to the
following destinations:
 Restore virtual machines 197
Restoring VMware virtual machine disks by using Backup, Archive, and Restore

To the original VM You can restore the disks to the same VM from which the disks were
backed up. You can either overwrite the original disks or attach the
virtual disks without overwriting the original disks.

NetBackup creates a temporary VM to which it restores the virtual disks.

Then, NetBackup attaches the virtual disks to the existing, target VM.
Finally, NetBackup deletes the temporary VM after the disk or disks
are attached successfully.

A special case, called in-place disk restore, replaces all disks of an

existing VM with the data in its backup. Raw devices (RDMs) and
independent disks are not replaced or deleted. For In-place Disk
Restore, the disks are restored to the same disk controller configuration
acquired at the time of backup.

To a different VM You can restore the disks to a different VM.

NetBackup creates a temporary VM to which it restores the virtual disks.

Then, NetBackup attaches the virtual disks to the existing, target VM.
Finally, NetBackup deletes the temporary VM after the disk or disks
are attached successfully.

You can also perform an in-place disk restore to a different VM.

To a new VM NetBackup creates a new virtual machine and restores the specified
disks to the new VM. The new VM is intended to be a container for the
restored disks. It does not have enough resources to run most operating
systems. After the restore, you should attach the restored virtual disks
to a VM that can support them and then delete the restore VM.

The following are the general support requirements for virtual disk restore.
■ Sufficient storage must exist for the restore.
■ NetBackup does not support the following virtual machine disk restores:
■ From NetBackup Replication Director for VMware backups.
■ To templates. However, virtual disks from a backup of a VM template can
be restored to a virtual machine.

Restoring VMware virtual machine disks by using

Backup, Archive, and Restore
This restore method uses the Backup, Archive, and Restore interface to select the
parameters for the restore and to restore the disks. NetBackup 8.1 or later is
required.
See “About VMware virtual machine disk restore” on page 196.
 Restore virtual machines 198
Restoring VMware virtual machine disks by using Backup, Archive, and Restore

To restore virtual machine disks

1 In the NetBackup Administration Console, select Backup, Archive, and
Restore.

Note: The NetBackup Administration Console does not support the in-place
disk restore option.

2 Select the Restore Files tab.

3 Click Actions > Specify NetBackup Machines and Policy Type.
4 Specify the following:

Server to use for The NetBackup master server that directed the backup
backup and
restores

Source client for The VMware virtual machine that was backed up.
restores
Click Search VM Clients to search or browse for a virtual
machine. Use this option to locate a virtual machine in a large,
multi-layered virtual environment.

See “Browse and search virtual machines for restore”

on page 232.

Or enter the type of name that was selected for the Primary VM
identifier option on the policy VMware tab. For example, if the
Primary VM identifier option is set to VM host name, enter the
virtual machine's host name.

Destination client This field defaults to the same client selected in the Source
for restores client for restores field.

You select the destination client in the wizard, so you do not

have to change this field.

Policy type for Select VMware.

restores

If any backups for the client are within the specified date range, NetBackup
populates the Directory Structure pane and the Contents of... pane with the
information about the most recent backup.
5 In the Restore type drop-down menu on the Restore Files tab, select Virtual
Disk Restore.
The Directory Structure pane changes to show the client name.
 Restore virtual machines 199
Restoring VMware virtual machine disks by using Backup, Archive, and Restore

6 In the Directory Structure pane, select the client.

7 To restore disks from a backup other than the most recent:

a. On the Actions menu, select Set Date Range > Backup History.

b. In the Backup History dialog box, select the wanted backup and click OK.

8 Click Restore.
The Select Virtual Disks panel of the Restore Virtual Machine Disks wizard
appears.
9 Complete the wizard panels to restore the virtual disks.
See “Select Virtual Disks panel” on page 199.
See “Recovery Destination panel” on page 200.
See “Virtual Disk Destination panel” on page 201.
See “Storage Destination panel” on page 203.
See “Recovery Summary panel” on page 204.

Select Virtual Disks panel

The Select Virtual Disks wizard panel shows all of the virtual disks that were in
the VM at back-up time, even those that were excluded from the backup. By default,
the wizard displays the virtual disks. To select file systems, click File Systems.
 Restore virtual machines 200
Restoring VMware virtual machine disks by using Backup, Archive, and Restore

■ Virtual Disks
Select the disks that you want to restore.
If you use the check box in the column header to select all, the wizard selects
only those disks that were included in the backup.
■ File Systems
Select the wanted file systems. When you select a file system, the wizard selects
the virtual disks on which the file system resides.
If you use the check box in the column header to select all, the wizard selects
only those file systems that were included in the backup.
See “Restoring VMware virtual machine disks by using Backup, Archive, and
Restore” on page 197.

Recovery Destination panel

On the Recovery Destination panel, you can configure the transport mode and
the recovery destination parameters.
■ NetBackup Recovery Host
The host that performs the restore. If not specified, NetBackup uses the backup
host value from the backup image.
■ Media Server
You can use this option to select a media server that has access to the storage
unit that contains the backup image. An example of such an environment is a
Media Server Deduplication Pool (MSDP) with multiple media servers. Note: If
the storage unit that contains the backup image is not shared with multiple media
servers, this option is grayed out.
■ Transport modes
The transport modes to use for the restore. Select the wanted transport modes
and move them so they are in the wanted order. If all methods fail, the restore
fails.
By default, the wizard displays the transport mode combination and the order
that were used for the backup.
■ Recover virtual disks to
The destination VM for the restore, as follows:
■ Original Virtual Machine. NetBackup restores the selected disks to the VM
from which they were backed up.
You can select whether to overwrite the existing virtual disks on the Storage
Destination panel.
See “Storage Destination panel” on page 203.
 Restore virtual machines 201
Restoring VMware virtual machine disks by using Backup, Archive, and Restore

■ Alternate Virtual Machine. NetBackup restores the selected disks to a

different VM than the original.
■ New (Temporary) Virtual Machine. NetBackup creates a new virtual
machine and restores the selected disks to the new VM. The new VM is
intended to be a container for the restored disks. It does not have enough
resources to run most operating systems. After the restore, you should attach
the restored virtual disks to a VM that can support them and then delete the
temporary VM.

■ Advanced Restore Options

The following items describe the advanced restore options:
■ Delete restored staging VM on error. Whether to delete the temporary VM
if the disk attach operation fails. If you select No and the disks are not
successfully attached to the target VM, you can access the data on the
temporary VM.
■ Wait time in seconds for VM shutdown. The restore process shuts down
the target virtual machine before it attaches the disk or disks. The duration
of the shutdown operation depends on the VMware workload. Use this
parameter to specify how long the restore process should wait for shutdown
before giving up on restore.

Virtual Disk Destination panel

The contents of the Virtual Disk Destination wizard panel depend on the recovery
destination, as follows:
■ Original Virtual Machine
The panel displays Destination details of the original VM.
You can select the power state of the VM after the restore. You also can refresh
Virtual Machine Server details.
See Power on virtual machine after restore and Last update in Table 10-2.
■ Alternate Virtual Machine
The panel displays Destination details of the original VM.
To select a destination VM:

a. Optionally, select a different Virtual Machine Server.

b. Optionally, enter a term in the Search box to find a VM on the Virtual Machine
Server and click Search.
 Restore virtual machines 202
Restoring VMware virtual machine disks by using Backup, Archive, and Restore

c. Select the destination VM for the restored virtual disks.

When you select a VM, the wizard populates the Destination details with the
details of that VM.

You can select the power state of the VM after the restore. You also can refresh
Virtual Machine Server details.
See Power on virtual machine after restore and Last update in Table 10-2.
■ New (Temporary) Virtual Machine
You must select the destination details.
See the field descriptions in Table 10-2.

Table 10-2 Virtual disk destination field descriptions

Field Description

Virtual Machine Server The name of the VMware host on which the destination VM
resides or should be created.

ESX Server Restore to a new VM only.

The name of the ESX host to which NetBackup should restore

the virtual machine disks. By default, NetBackup uses the
value from the backup. To use a different host, click Browse
to select or search for a host.

Data Center Restore to a new VM only.

The data center to which to restore the virtual disks.

Folder Restore to a new VM only.

The pathname of the VM folder to which NetBackup should

restore the virtual machine disks. By default, NetBackup uses
the value from the backup. Click Browse to select a different
folder.

Virtual Machine Restore to a new VM only.

The display name of the temporary virtual machine for the

restored disk or disks. The wizard populates this field with
the name of the original VM client and appends a timestamp
to it. The timestamp is the decimal number of seconds since
January 1, 1970.

The name must be unique.

 Restore virtual machines 203
Restoring VMware virtual machine disks by using Backup, Archive, and Restore

Table 10-2 Virtual disk destination field descriptions (continued)

Field Description

Resource Pool/vApp Restore to a new VM only.

The resource pool or vApp to which NetBackup should restore

the virtual machine disks. By default, NetBackup uses the
value from the backup.

Power on virtual machine Restore to the original or to an alternate VM only.

after restore
Select to start the virtual machine after the restore completes
successfully.

Last update Restore to the original or to an alternate VM only.

The date and time on which the Virtual Machine Server

provided the information for this wizard panel. Click the
refresh option to update the Virtual Machine Server details.

See “Restoring VMware virtual machine disks by using Backup, Archive, and
Restore” on page 197.

Storage Destination panel

The Storage Destination wizard panel shows the virtual disks to restore. They are
either the disks you selected or the disks on which the file systems you selected
for restore reside.
Settings for all virtual disks region
To change the values that apply to all of the virtual disks, use the fields in the
Settings for all virtual disks region of the wizard panel, as follows:
■ Restored virtual disks provisioning
The default disk provisioning for all of the disks to be restored: Thin, Thick Lazy
Zeroed, or Thick Eager Zeroed.
■ Datastore
The name of the Datastore or the datastore cluster that is the destination for
the restore.
Click Browse to select a different datastore or datastore cluster.
■ Overwrite all virtual disks
Whether to overwrite the existing virtual disk or disks on the target VM. If Yes,
overwrite the original virtual disk and retain the disk UUID. If No, restore the
virtual disk to the target VM as a new disk; VMware assigns a new UUID to the
disk.
 Restore virtual machines 204
Restoring VMware virtual machine disks by using NetBackup commands

Original Virtual Disk list

To override the global settings, change the values for individual virtual disks in the
Original Virtual Disk list.
See “Restoring VMware virtual machine disks by using Backup, Archive, and
Restore” on page 197.

Recovery Summary panel

The Recovery Summary wizard panel displays the parameters of the recovery.
Verify the settings for the recovery. If they are correct, click Start Recovery.
Otherwise, click Back to return and change the settings.
See “Restoring VMware virtual machine disks by using Backup, Archive, and
Restore” on page 197.

Restoring VMware virtual machine disks by using

NetBackup commands
This restore method requires that you describe in a text file the parameters of the
restore and then use a NetBackup command to restore the disks. The file in which
you describe the restore includes the following information:
■ The backup image from which you want to restore.
■ The target VM if restoring to an existing VM.
■ The virtual machine disks to restore.
■ Other parameters that control aspects of the restore.
The NetBackup nbrestorevm command includes an option that you can use to
create the required restore parameters file from a backup image. The NetBackup
nbrestorevm command then reads that file to accomplish the restore.

The restore parameters file uses the JavaScript Object Notation (JSON) format to
describe the restore parameters.
The following table identifies the tasks in the process to restore an individual VMware
virtual machine disk or disks. It also includes links to the individual task topics.
 Restore virtual machines 205
Restoring VMware virtual machine disks by using NetBackup commands

Table 10-3 Restore VMware virtual machine disk process

Step Description Link

Step 1 Optionally, determine the ID of the See “Determining the backup image ID
backup image from which to create the from which to restore a VMware virtual
restore parameters file. If you want to use machine disk” on page 205.
a date range to specify the backup image,
you do not have to perform this task.

Step 2 Create the file that describes the See “Creating the VMware virtual
parameters of the restore. You use the machine disk restore file” on page 207.
file as input for the command that
restores the virtual disks.

Step 3 Modify the parameters of the virtual See “Modifying the virtual machine disk
machine disk restore file. You must restore file” on page 208.
modify the file to describe your specific
restore instance.

Step 4 Validate the virtual machine disk restore See “Validating the virtual machine disk
file to ensure that the restore occurs restore file” on page 212.
correctly.

Step 5 Restore the disk or disks. See “Restoring the virtual machine disk
or disks by using the nbrestorevm
command” on page 213.

See “Troubleshooting vmdk restore to existing VM” on page 321.

Determining the backup image ID from which to restore a VMware

virtual machine disk
This topic describes how to determine the ID of the backup image from which to
restore a VMware virtual machine disk. You then use the backup image ID to specify
the backup from which to create the restore parameters file. Two methods exist to
determine the backup ID, as follows:

NetBackup Administration See To determine the backup image ID by using

Console NetBackup Administration Console.

bpclimagelist command See To determine the backup image ID by using the

bpclimagelist command.

Note: If you want to use a date range to specify the backup from which to create
the restore parameters file, you can skip this task.
 Restore virtual machines 206
Restoring VMware virtual machine disks by using NetBackup commands

This task is part of a larger process that describes how to restore an individual
VMware virtual machine disk or disks.
See “Restoring VMware virtual machine disks by using NetBackup commands”
on page 204.
To determine the backup image ID by using NetBackup Administration Console
1 Use the NetBackup Administration Console Catalog utility to search for the
backup image of the client from which to restore. The Search Results pane
of the Catalog utility shows the backup ID:

2 Continue to the next task in the individual virtual machine disk restore process.
See “Creating the VMware virtual machine disk restore file” on page 207.
To determine the backup image ID by using the bpclimagelist command
1 Invoke the NetBackup bpclimagelist command to display the backups for
the client, as follows:
On UNIX: /usr/openv/netbackup/bin/bpclimagelist -client client_name
-Likelydate -policy policy_name

On Windows: install_path\NetBackup\bin\bpclimagelist -client

client_name -Likelydate -policy policy_name

2 Construct the backup ID by combining the client name, an underscore character

(_), and the backup time notation (the decimal number of seconds since January
1, 1970). For the following example output, vm1.example.com_1454006694 is
the backup ID:

bpclimagelist -client vm1.example.com -Likelydate -policy backups

Client: vm1.example.com
Policy: backups
Keyword: NONE
Backup Time: Thu Jan 28 12:44:54 2016 (1454006694)
Schedule Type: FULL

3 Continue to the next task in the individual virtual machine disk restore process.
See “Creating the VMware virtual machine disk restore file” on page 207.
 Restore virtual machines 207
Restoring VMware virtual machine disks by using NetBackup commands

Creating the VMware virtual machine disk restore file

This topic describes how to use the NetBackup nbrestorevm command to create
the text file that describes the virtual machine disk restore parameters. The file uses
the JavaScript Object Notation (JSON) format to describe the restore parameters.
Two methods exist to specify the backup that is used to create the restore
parameters file, as follows:

Specify a backup See To create the VMware virtual machine disk restore file by specifying
image ID a backup ID .

A separate procedure describes how to determine the backup ID.

See “Determining the backup image ID from which to restore a VMware

virtual machine disk” on page 205.

Specify a date If you specify a date range, NetBackup uses the most recent backup
range in that date range.

See To create the VMware virtual machine disk restore file by specifying
a date range.

This task is part of a larger process that describes how to restore an individual
VMware virtual machine disk or disks.
See “Restoring VMware virtual machine disks by using NetBackup commands”
on page 204.
To create the VMware virtual machine disk restore file by specifying a backup
ID
1 Create the file that describes the virtual machine disk or disks that you want
to restore by specifying a backup ID, as follows:
On Windows: install_path\NetBackup\bin\nbrestorevm -vmw -C
vm_client -S master_server -backupid value -restorespecout
filename

On UNIX: /usr/openv/netbackup/bin/brestorevm -vmw -C vm_client -S

master_server -backupid value -restorespecout filename

■ The -backupid value option and argument specifies the backup image
from which to create the file.
■ The -restorespecout filename option and argument specifies the name
of the file.
 Restore virtual machines 208
Restoring VMware virtual machine disks by using NetBackup commands

By default, nbrestorevm creates the file in the current working directory.

To create the file in a different directory, specify a pathname for the
filename.

2 Continue to the next task in the individual virtual machine disk restore process.
See “Modifying the virtual machine disk restore file” on page 208.
To create the VMware virtual machine disk restore file by specifying a date
range
1 Create the file that describes the virtual machine disk or disks that you want
to restore by specifying a date range, as follows:
On Windows: install_path\NetBackup\bin\nbrestorevm -vmw -C
vm_client -S master_server -s mm/dd/yyyy hh:mm:ss -e mm/dd/yyyy
hh:mm:ss -restorespecout filename

On UNIX: /usr/openv/netbackup/bin/nbrestorevm -vmw -C vm_client

-S master_server -s mm/dd/yyyy hh:mm:ss -e mm/dd/yyyy hh:mm:ss
-restorespecout filename

■ The hh:mm:ss for the start date and the end date are optional.
■ The -restorespecout filename option and argument specifies the name
of the file.
By default, nbrestorevm creates the file in the current working directory.
To create the file in a different directory, specify a pathname for the
filename.

2 Continue to the next task in the individual virtual machine disk restore process.
See “Modifying the virtual machine disk restore file” on page 208.

Modifying the virtual machine disk restore file

This topic describes how to modify the file that describes the virtual machine disk
restore parameters. You must modify the file to describe your specific restore
instance.
Two methods exist to modify the restore file, as follows:

A text editor With a text editor, you can modify any or all of the fields.

See To modify a VMware virtual machine disk restore file by using a

text editor.
 Restore virtual machines 209
Restoring VMware virtual machine disks by using NetBackup commands

A Veritas provided With the restore_spec_utility.pl Perl script, you can perform
Perl script the following operations:

■ –modify, to change the value of a field, one field at a time. Use

this option to modify the fields that do not describe
VMWareVirtualDiskDestination parameters.
■ -keep_disks, to specify the disks to retain in the restore
parameters file. Any other disks are removed from the file.
■ -delete_disks, to specify the disks to delete from the restore
parameters file. The disks that remain are restored.

Examine the script for a description of the options and example usage
statements.

Perl must be installed on the host on which you modify the restore
parameters file. Examine the script to determine which modules are
required; some may require separate installation.

See To modify a VMware virtual machine disk restore file by using the
restore_spec_utility.pl script.

The following table describes some restore scenarios and the values of the fields
that are required to accommodate those scenarios.

Table 10-4 Restore parameters file items to edit

Item Description

Disks Edit the VMwareVirtualDiskDestination section of the file so that

it contains only the virtual machine disk or disks that you want to restore.
To ensure data integrity, be aware of the following:

■ If a file system volume spans multiple virtual machine disks, restore

all of the virtual disks for that volume.
■ If a volume manager manages the volumes, restore all of the virtual
disks in that volume set.

Restore to a For a restore of virtual disks to a datastore cluster, specify the name of
datastore cluster the datastore cluster in the Datastore field of the
VMwareVirtualDiskDestination section. Do not specify the cluster
in the Path field.
Note: If the restore is to an existing VM, the VMDKs are not located
in the existing datastore path for the VM. They reside in a datastore
path based on the name of the temporary VM used in the restore. The
name of the temporary VM is available in the details for the restore job
in NetBackup.
 Restore virtual machines 210
Restoring VMware virtual machine disks by using NetBackup commands

Table 10-4 Restore parameters file items to edit (continued)

Item Description

Restore to an By default, the restore parameters file that the nbrestorevm

existing VM -restorespecout command creates is configured to restore to a
new VM. To restore the virtual disks to an existing VM, do the following:

■ Specify the name of the VM in the VMName field. The VM must exist
in the target vCenter or ESXi serve; if it does not exist, the restore
fails with status code 2820.
■ Change the value of the AttachDisksToExistingVM field from
No to Yes or, in the case of in-place disk restores,
DeleteAllDisksAndReplace.
■ If a virtual disk to restore exists on the target VM already, configure
the OverwriteExistingDisk field, as follows:
■ To overwrite the original virtual disk and retain the disk UUID,
specify Yes.
■ To restore the virtual disk to the target VM as a new disk, specify
No. VMware assigns a new UUID to the disk. The default is No.

Disk UUIDs are always preserved with in-place disk restores.

■ The restore process shuts down the target virtual machine before
it attaches the disk or disks. By default, NetBackup does not turn
on the target VM. To turn on the target VM, specify Yes for the
PowerOn option.

The VMwareAdvancedRestoreOptions section of the restore

parameters file includes other parameters that are related to restoring
to an existing VM.

The restore job details show the datastore path of the virtual disk when
it was backed up and after it is restored.

Standalone ESXi To restore to a standalone ESXi hypervisor when the backup was
hypervisor through a vCenter, edit the following fields so that their values are None,
as follows:

■ "vCenterServer": "None"
■ "Folder": "None"
■ "Datacenter" : "None"

The VM name The nbrestorevm -restorespecout command adds a timestamp

to the name of the original VM client when it populates the VMName field
in the restore parameters file. The timestamp is the decimal number of
seconds since January 1, 1970.

Depending on the requirements of the restore instance, you may need

to edit the VMName field.
 Restore virtual machines 211
Restoring VMware virtual machine disks by using NetBackup commands

Table 10-4 Restore parameters file items to edit (continued)

Item Description

Other fields Edit any of the other fields so that the values that describe the restore
meet your needs.

See “VMware virtual machine disk restore file” on page 214.

This task is part of a larger process that describes how to restore an individual
VMware virtual machine disk or disks.
See “Restoring VMware virtual machine disks by using NetBackup commands”
on page 204.
To modify a VMware virtual machine disk restore file by using a text editor
1 Open the restore file in a text editor.
2 Modify the parameters as necessary, save the file, and then close the editor.
See “VMware virtual machine disk restore file” on page 214.
3 Continue to the next task in the individual virtual machine disk restore process.
See “Validating the virtual machine disk restore file” on page 212.
To modify a VMware virtual machine disk restore file by using the
restore_spec_utility.pl script

1 Invoke the Perl script to modify the restore parameters file as necessary, one
field at a time. Examples of three supported -req_type operations follow:
■ The following is an example of a modify operation to change the VMName
field:
Windows: perl.exe
C:\Progra~1\Veritas\NetBackup\bin\goodies\restore_spec_utility.pl
-req_type modify -file_path restore_filename -field VMName
-value new_vm_name
UNIX: perl
/usr/openv/netbackup/bin/goodies/restore_spec_utility.pl
-req_type modify -file_path restore_filename -field VMName
-value new_vm_name

■ The following is an example of a keep_disks operation to retain a subset

of the disks that are described in a restore parameters file. It retains the
scsi0-1 and scsi0-2 virtual disks and deletes all of the other defined disks
from the file:
Windows: perl.exe
C:\Progra~1\Veritas\NetBackup\bin\goodies\restore_spec_utility.pl
 Restore virtual machines 212
Restoring VMware virtual machine disks by using NetBackup commands

-req_type keep_disks -file_path restore_filename -controller

scsi0-1,scsi0-2
UNIX: perl
/usr/openv/netbackup/bin/goodies/restore_spec_utility.pl
-req_type keep_disks -file_path restore_filename -controller
scsi0-1,scsi0-2

■ The following is an example of a delete_disks operation to delete a subset

of the disks that are described in a restore parameters file. It deletes the
scsi0-1 and scsi0-2 virtual disks and retains the other defined disks in
the file:
Windows: perl.exe
C:\Progra~1\Veritas\NetBackup\bin\goodies\restore_spec_utility.pl
-req_type delete_disks -file_path restore_filename -controller
scsi0-0,scsi0-1
UNIX: perl
/usr/openv/netbackup/bin/goodies/restore_spec_utility.pl
-req_type delete_disks -file_path restore_filename -controller
scsi0-0,scsi0-1

■ The following is an example of modifying a restore specification for In-place

disk restore:
Windows: perl.exe
C:\Progra~1\Veritas\NetBackup\bin\goodies\restore_spec_utility.pl
-req_type modify -file_path restore_filename -field
AttachDisksToExistingVM -value DeleteAllDisksAndReplace
UNIX: perl
/usr/openv/netbackup/bin/goodies/restore_spec_utility.pl
-req_type modify -file_path restore_filename -field
AttachDisksToExistingVM -value DeleteAllDisksAndReplace

2 Repeat step 1 until you have finished modifying the file.

3 Continue to the next task in the individual virtual machine disk restore process.
See “Validating the virtual machine disk restore file” on page 212.

Validating the virtual machine disk restore file

This topic describes how to use the NetBackup nbrestorevm command to validate
the file that describes the virtual machine disk restore parameters.
This task is part of a larger process that describes how to restore an individual
VMware virtual machine disk or disks.
 Restore virtual machines 213
Restoring VMware virtual machine disks by using NetBackup commands

See “Restoring VMware virtual machine disks by using NetBackup commands”

on page 204.
To validate the virtual machine disk restore file
1 Invoke the following NetBackup command:
On Windows: install_path\NetBackup\bin\nbrestorevm –validate
-restorespec filename

On UNIX: /usr/openv/netbackup/bin/nbrestorevm –validate

-restorespec filename

2 Examine the command output. If the output indicates problems, edit the file to
fix the problems.
3 Repeat steps 1 and 2 until the file is valid.
4 Continue to the next task in the individual virtual machine disk restore process.
See “Restoring the virtual machine disk or disks by using the nbrestorevm
command” on page 213.

Restoring the virtual machine disk or disks by using the nbrestorevm

command
This topic describes how to use the NetBackup nbrestorevm command to restore
one or more virtual machine disks to a new VMware virtual machine.
This task is part of a larger process that describes how to restore an individual
VMware virtual machine disk or disks.
See “Restoring VMware virtual machine disks by using NetBackup commands”
on page 204.
To restore a VMware virtual machine disk
1 If the file is valid, restore the virtual machine by invoking the following
NetBackup command:
On Windows: install_path\NetBackup\bin\nbrestorevm -restorespec
filename [-L progress_log] [-w [hh:mm:ss]]

On UNIX: /usr/openv/netbackup/bin/nbrestorevm -restorespec filename

[-L progress_log] [-w [hh:mm:ss]]

■ The -restorespec filename option specifies the name of the file that
contains the parameters of the restore.
 Restore virtual machines 214
Restoring VMware virtual machine disks by using NetBackup commands

■ The -L progress_log option specifies the name of an existing file in which

to write progress information. Only default paths are allowed for this option;
the following are the default paths:
UNIX systems: /usr/openv/netbackup/logs/user_ops/proglog
Windows systems: install_path\NetBackup\logs\user_ops\proglog
■ The -w [hh:mm:ss] option causes NetBackup to wait for a completion
status from the server before it returns you to the system prompt.
The required date and time values format in NetBackup commands varies
according to your locale. The /usr/openv/msg/.conf file (UNIX) and the
install_path\VERITAS\msg\LC.CONF file (Windows) contain information
such as the date-time format for each supported locale. The files contain
specific instructions on how to add or modify the list of supported locales
and formats.
You can optionally specify a wait time in hours, minutes, and seconds. The
maximum wait time you can specify is 23:59:59. If the wait time expires
before the restore is complete, the command exits with a time-out status.
The restore, however, still completes on the server.
If you specify 0 or do not specify a time, the wait time is indefinite for the
completion status.

2 Monitor the restore in the NetBackup Administration Console Activity

Monitor.
See “Troubleshooting vmdk restore to existing VM” on page 321.
3 After the restore completes, do the following as necessary:
■ For the virtual disks that are restored to a new VM, add the disks to a VM
that can support them after the restore succeeds. Optionally, delete the VM
that NetBackup created for the restore.
■ For the virtual disks that are restored to an existing VM, assign a drive letter
(Windows) or mount the disk (Linux) so you can access the data.
How to add disks, delete virtual machines, and mount drives is beyond the
scope of the NetBackup documentation. See your VMware documentation.

VMware virtual machine disk restore file

This topic describes the parameters that NetBackup uses to restore a virtual machine
disk or disks. NetBackup requires that the parameters be described in a text file
that uses the JavaScript Object Notation (JSON) format. The NetBackup
nbrestorevm command reads the file to obtain the necessary information about
the virtual machine disks. NetBackup has no requirements for the file name other
than those imposed by the operating system.
 Restore virtual machines 215
Restoring VMware virtual machine disks by using NetBackup commands

You can use the nbrestorevm command to generate the file from an existing backup.
How to do so is described in a different topic.
See “Creating the VMware virtual machine disk restore file” on page 207.
The following is an example template of the required file:

{
"ClientType": "VMware",
"ClientName": "VM-client-name",
"RestoreType": "SelectiveDiskRestore",
"BackupImageSelection": {
"MasterServer": "Master-server-name",
"StartDate": "mm/dd/yy hh:mm:ss",
"EndDate": "mm/dd/yy hh:mm:ss",
"BackupId": "clientname_timestamp"
},
"VMwareRestoreParameters": {
"vCenterServer": "vCenter-name-for-restore",
"VMwareRecoveryHost": "Recovery-host-name",
"DefaultDiskProvisioning": "thin",
"TransportMode": "san:hotadd:nbd",
"vmdk_compression": "none",
"VMwareVirtualMachineDestination": {
"VMName": "Restore-vm-name",
"AttachDisksToExistingVM": "No",
"PowerOn": "No",
"Datacenter": "Path-of-Datacenter-for-destination-vm",
"ESX": "Hostname-of-the-ESX-host",
"Folder": "Path-to-destination-VM-folder",
"ResourcePool/Vapp": "Path-of-vApp-or-resource-pool-destination",
"VmxDatastore": ""
},
"VMwareVirtualDiskDestination": [
{
"VirtualDisk" : "/DS1/BackedupVM/BackedupVM.vmdk",
"OverwriteExistingDisk": "No",
"Datastore": "[Datastore-name]",
"Path": "”,
"Provisioning": "thin"
"Controller": "scsi0-0"
},
{
"VirtualDisk": "/DS2/BackedupVM/BackedupVM_1.vmdk",
 Restore virtual machines 216
Restoring VMware virtual machine disks by using NetBackup commands

"OverwriteExistingDisk": "No",
"Datastore": "",
"Path": "[datastore_name] MyVm/MyVM_1.vmdk”,
"Provisioning": "eagerzeroed"
"Controller": "scsi0-1"
}
]
"VMwareAdvancedRestoreOptions": {
"DeleteRestoredVMOnError": "No",
"VMShutdownWaitSeconds": 900
}
}
}

The following is an example of the required file for In-place disk restore:

{
"BackupImageSelection": {
"StartDate": "mm/dd/yy hh:mm:ss",
"BackupId": "clientname_timestamp"
"EndDate": "mm/dd/yy hh:mm:ss",
"MasterServer": "Master-server-name",
},
"ClientName": "VM-client-name",
"VMwareRestoreParameters": {
"vmdk_compression": "none",
"VMwareAdvancedRestoreOptions": {
"VMShutdownWaitSeconds": 900,
"DeleteRestoredVMOnError": "No"
},
"VMwareRecoveryHost": "VM-recovery-host-name",
"VMwareVirtualMachineDestination": {
"ResourcePool/Vapp": "Path-of-vApp-or-resource-pool-destination",
"VmxDatastore": "Vmx-datastore-name",
"Datacenter": "Path-of-Datacenter-for-destination-vm",
"AttachDisksToExistingVM": "DeleteAllDisksAndReplace",
"ESX": "Hostname-of-the-ESX-host",
"VMName": "Restore-vm-name",
"Folder": "Path-to-destination-VM-folder",
"PowerOn": "Yes"
},
"DefaultDiskProvisioning": "unknown",
"TransportMode": "nbdssl",
"VMwareVirtualDiskDestination": [],
 Restore virtual machines 217
Restoring VMware virtual machine disks by using NetBackup commands

"vCenterServer": "vCenter-name-for-restore"
},
"ClientType": "VMware",
"RestoreType": "SelectiveDiskRestore"
}

The following subsections describe the parameters in the file. The optional sections
or optional fields that you do not want to use must be omitted from the file.
See “About VMware virtual machine disk restore” on page 196.

Global fields
The first section of the file specifies the required information about the client that
contains the disks that you want to restore. The following table describes the fields
in the first, global section:

Table 10-5 Global fields (required)

Field name Description Required?

ClientType The client type as configured in the backup policy. For Vmware virtual machine Required.
disk restore, use VMware.

ClientName The client name as configured in the backup policy. Required.

RestoreType The type of restore. For Vmware virtual machine disk restore, use Required.
SelectiveDiskRestore.

BackupImageSelection
The BackupImageSelection section of the file specifies the information that
NetBackup requires to identify the backup image to restore. If this section is not
specified, NetBackup restores the most recent backup. The following table describes
the fields in the BackupImageSelection section:

Table 10-6 The BackupImageSelection section (optional)

Field name Description Required?

MasterServer The fully-qualified domain name of the NetBackup master server to use to query Optional.
the VM details.

If not specified, NetBackup uses the master server that backs up the VM.

StartDate The start date to look for backup images, in mm/dd/yy hh:mm:ss format. If more Optional.
than one backup image exits in the date range, NetBackup selects the most recent
backup.

If not specified, the start date is 6 months earlier than the current date.
 Restore virtual machines 218
Restoring VMware virtual machine disks by using NetBackup commands

Table 10-6 The BackupImageSelection section (optional) (continued)

Field name Description Required?

EndDate The end date to look for backup images, in mm/dd/yy hh:mm:ss format. If more Optional.
than one backup image exits in the date range, NetBackup selects the most recent
backup.

If not specified, NetBackup uses the current date.

BackupId The ID of the backup image to use for the restore, in clientname_backuptime Optional.
format. The backuptime is the decimal number of seconds since January 1,
1970.

If not specified, NetBackup uses the most recent backup image. If you specify a
StartDate, EndDate, and a valid BackupId, NetBackup restores from the
BackupId image.

VMwareRestoreParameters
The VMwareRestoreParameters section of the file contains the following:
■ The restore parameters of the VM. All of the fields in this section are optional;
however, the section is required because it also contains two required
subsections. See Table 10-7.
■ A VMwareVirtualMachineDestination subsection, which specifies the
destination parameters for the restore.
See the section called “VMwareVirtualMachineDestination” on page 219.
■ A VMwareVirtualDiskDestination subsection, which specifies the disks to
restore and the restore parameters for those disks.
See the section called “VMwareVirtualDiskDestination” on page 221.
■ A VMwareAdvancedRestoreOptions subsection, which specifies parameters to
restore to an existing VM.
See the section called “VMwareAdvancedRestoreOptions” on page 223.

Table 10-7 The VMwareRestoreParameters section (required)

Field name Description Required?

vCenterServer The host name of the destination vCenter for the restore, in the same Optional.
format as specified in the NetBackup Virtual Server credentials.

To restore to a standalone ESXi hypervisor when the backup was

through a vCenter, the value of this field must be None.
 Restore virtual machines 219
Restoring VMware virtual machine disks by using NetBackup commands

Table 10-7 The VMwareRestoreParameters section (required) (continued)

Field name Description Required?

VMwareRecoveryHost The host that performs the restore. Optional.

If not specified, NetBackup uses the backup host value from the
backup image.

DefaultDiskProvisioning The default disk provisioning for all of the disks to be restored: thin, Optional.
thick, eagerzeroed, or unknown.

For each disk, you can override this default by specifying a different
Provisioning value in the VMwareVirtualDiskDestination
section of the file.

If neither DefaultDiskProvisioning or Provisioning are

specified, NetBackup uses the provisioning specified in the backup.

TransportMode The transport mode combination to use for the restore as specified Optional.
in lowercase, colon separated values: hotadd:nbd:nbdssl:san.
The order of the specification is significant; NetBackup attempts
each method in turn until the restore succeeds. If all methods fail,
the restore fails.

If not specified, NetBackup uses the transport mode combination

that was used for the backup.

vmdk_compression The vmdk compression type to use for full and selective vmdk Optional.
restores. Vmdk compression types are: zlib, skipz, and fastlz.

If not specified, NetBackup uses none as the value.

VMwareVirtualMachineDestination
The VMwareVirtualMachineDestination section of the file specifies the destination
parameters for the restore. The following table describes the fields in the
VMwareVirtualMachineDestination section. This section is subordinate to the
VMwareRestoreParameters section.
 Restore virtual machines 220
Restoring VMware virtual machine disks by using NetBackup commands

Table 10-8 The VMwareVirtualMachineDestination section (required)

Field name Description Required?

VMName The unique display name of the new virtual machine for the Required.
restored disk or disks. The nbrestorevm command adds a
timestamp to the name of the original VM client when it populates
this field. The timestamp is the decimal number of seconds since
January 1, 1970.

NetBackup restores the virtual machine disks to a new VM.

Therefore, if this name conflicts with an existing display name, the
restore fails.

AttachDisksToExistingVM Whether to restore the selected VMDKs to an existing VM or to a Required.

new VM, as follows:

■ If the value is Yes, the VM specified in the VMName field must

exist in the target vCenter or ESX server. If it does not exist,
the restore fails with status code 2820.
■ If the value is No, the VM specified in the VMName field must
not exit in the target vCenter or ESX server. If it exists, the
restore fails with status code 2820.
■ If the value is DeleteAllDisksAndReplace, the VM
specified in the VMName field must exist in the target vCenter
or ESX server. If it does not exist, the restore fails with status
code 2820.

The default value is No.

PowerOn Whether to turn on the target VM after the restore, as follows: Required.

■ If the value is Yes, the target VM is powered ON at the end of

a successful restore.
■ If the value is No, the target VM is not turned on after the
restore.

If the restore is to an existing VM, the VM is turned off before the

virtual disks are attached to it during the restore.

The default value is No.

Datacenter The name of the VMware Datacenter for the virtual disk, in Optional.
pathname format.

To restore to a standalone ESXi hypervisor when the backup was

through a vCenter, the value of this field must be None.

If not specified, NetBackup uses the value from the backup.

 Restore virtual machines 221
Restoring VMware virtual machine disks by using NetBackup commands

Table 10-8 The VMwareVirtualMachineDestination section (required)

(continued)

Field name Description Required?

ESX The name of the ESX host to which NetBackup should restore the Optional.
virtual machine disks.

If not specified, NetBackup uses the value from the backup.

Folder The pathname of the VM folder to which NetBackup should restore Optional.
the virtual machine disks.

To restore to a standalone ESXi hypervisor when the backup was

through a vCenter, the value of this field must be None.

If not specified, NetBackup uses the value from the backup.

ResourcePool/Vapp The pathname of the resource pool to which NetBackup should Optional.
restore the virtual machine disks. If the restore is to a vApp, specify
the path of the vApp.

If not specified, NetBackup uses the value from the backup.

VmxDatastore The name of the Datastore to which NetBackup should restore Optional.
the .vmx configuration file and other VM configuration files. You
may enclose the name in square brackets but do not have to.

If not specified, NetBackup uses the value from the backup.

DefaultDiskDatastore The name of the Datastore to which NetBackup should restore Optional.
all the virtual disks for in-place disk restore. If not specified,
NetBackup uses the value from the backup. This option is only
used for In-place disk restore and is ignored otherwise.

VMwareVirtualDiskDestination
The VMwareVirtualDiskDestination section of the file is a JSON array that
specifies the disks to restore and the restore parameters for those disks. It can
contain one or more sets of the fields that are described in the following table, one
set per virtual machine disk. A comma must separate fields in a set, and a comma
must separate sets. This section is subordinate to the VMwareRestoreParameters
section.

Note: This section is not applicable for in-place disk restores.

 Restore virtual machines 222
Restoring VMware virtual machine disks by using NetBackup commands

Table 10-9 The VMwareVirtualDiskDestination section (required)

Field name Description Required?

VirtualDisk The full pathname of the virtual disk to restore. This path must match Required.
exactly the path of the .vmdk file when it was backed up.

OverwriteExistingDisk Whether to overwrite the existing virtual disk or disks on the target VM, Required.
as follows:

■ If the value is Yes, overwrite the original virtual disk and retain the
disk UUID.
■ If the value is No, restore the virtual disk to the target VM as a new
disk. VMware assigns a new UUID to the disk.

The default value is No.

Datastore The name of the Datastore that is the destination for the restore. You Optional.
may enclose the name in square brackets but do not have to. (VMware
generates the Datastore pathname using the naming conventions
for the VM.)

For a restore of virtual disks to a datastore cluster, specify the name

of the datastore cluster in this field.

If not specified, NetBackup uses the value that is specified in the Path
field. If neither Datastore or Path are specified, NetBackup uses the
Datastore from the backup image.

Path The full pathname to the restore destination for the virtual disk, in the Optional.
following format:

[datastore_name] MyVM/MyVM.vmdk

The nbrestorevm -restorespecout option does not populate the

Path field.

If you specify a Path and it is not available or a disk already exists at

that path, the restore fails. If neither Datastore or Path are specified,
NetBackup uses the Datastore from the backup image.

If a specific vmdk name is not required, Veritas recommends that you

leave this field empty.

Provisioning The disk provisioning for this specific disk: thin, thick, eagerzeroed, Optional.
or unknown.

If not specified, the NetBackup uses the DefaultDiskProvisioning

value.
 Restore virtual machines 223
About restore of individual filesAbout restoring individual VMware files

Table 10-9 The VMwareVirtualDiskDestination section (required)

(continued)

Field name Description Required?

Controller The virtual disk controller to which the disk is attached in the original Optional
VM.

This field is informational only to help you determine which virtual disk
or disks to restore. The value is not used during a restore.

VMwareAdvancedRestoreOptions
The VMwareAdvancedRestoreOptions section of the file specifies parameters to
restore to an existing VM. This section is subordinate to the
VMwareRestoreParameters section.

Table 10-10 The VMwareAdvancedRestoreOptions section (optional)

Field name Description Required?

DeleteRestoredVMOnError NetBackup creates a temporary VM during the restore process Optional.

and then attaches the virtual disks to the target VM. The value of
this parameter determines whether NetBackup deletes the
temporary VM if the disk attach operation fails, as follows:

■ If the value is Yes, delete the temporary VM.

■ If the value is No, do not delete the temporary VM. If the disks
are not successfully attached to the target VM, you can access
the data on the temporary VM.

The default value is No.

VMShutdownWaitSeconds For restores to an existing VM, the restore process shuts down Optional.
the target virtual machine before it attaches the disk or disks. The
duration of the shutdown operation depends on the VMware
workload. Use this parameter to specify how long the restore
process should wait for shutdown before giving up on restore.

The default value is 900 seconds (15 minutes).

About restore of individual filesAbout restoring

individual VMware files
This topic applies to NetBackup for VMware.
You can use either of the following procedures to restore files individually.
 Restore virtual machines 224
Restoring individual filesRestoring individual VMware virtual machine files

One-stage process Restore individual files directly to a virtual machine in one of the
following ways:

■ Restore the files to a virtual machine on which NetBackup client

software is installed.
■ Windows only: Restore the files to virtual machine drives that
have been mapped to a host on which NetBackup client software
is installed.

See “Setting up NetBackup Client Service for restore to a Windows

shared virtual machine driveSetting up NetBackup Client Service for
VMware restores to a Windows shared virtual machine drive”
on page 227.

Two-stage process Restore individual files to a host on which the NetBackup client
software is installed (not to the virtual machine). See the NetBackup
Backup, Archive, and Restore Getting Started Guide on how to
restore to different locations. Then manually copy the restored files
to the virtual machine. (NetBackup does not perform this step.)

Note: Individual file recovery is supported from full backups and from incremental
backups, as long as the Enable file recovery from VM backup policy option is
enabled.

See “Restoring individual filesRestoring individual VMware virtual machine files”

on page 224.
See “Restoring VMware virtual machine disks by using Backup, Archive, and
Restore” on page 197.
See “Restoring the full VMware virtual machine” on page 179.

Restoring individual filesRestoring individual

VMware virtual machine files
This topic describes how to restore files from a NetBackup backup of a VMware
virtual machine.
If the VMware backup was made with the Enable file recovery from VM backup
option, you can restore individual files.
See “Optimizations options (VMware)” on page 74.
 Restore virtual machines 225
Restoring individual filesRestoring individual VMware virtual machine files

To restore individual files

1 In the NetBackup Administration Console, click Backup, Archive, and Restore
2 Click Actions > Specify NetBackup Machines and Policy Type....
3 Enter the following:

Server to use for The NetBackup master server that directed the backup.
backup and
restores

Source client for The VMware virtual machine that was backed up.
restores
Click Search VM Clients to search or browse for a virtual
machine. This option helps to locate a virtual machine in a large,
multi-layered virtual environment.

See “Browse and search virtual machines for restore”

on page 232.

Or, enter the type of name that was selected for the Primary
VM identifier option on the policy VMware tab. For example, if
the Primary VM identifier option is set to VM host name, enter
the virtual machine's host name.

Destination client Enter one of the following:

for restores
■ The host name of the VMware virtual machine that was
backed up, if NetBackup client software has been installed
on the virtual machine.
Enter the host name even if the policy identified the virtual
machine by its VM display name, UUID, or DNS name. Do
not enter the VM display name, UUID, or DNS name.
■ Any NetBackup client that has the same operating system
and file system of the VM that was backed up.

Note: The NetBackup Client Service must be logged on as

Administrator.

See “Setting up NetBackup Client Service for restore to a

Windows shared virtual machine driveSetting up NetBackup
Client Service for VMware restores to a Windows shared virtual
machine drive” on page 227.

Policy type for The type of policy that made the backup (VMware).
restores

Date / time range The time period within which to search for backups.

4 Click OK.
 Restore virtual machines 226
Restoring individual filesRestoring individual VMware virtual machine files

5 For the Restore type, select Normal Backups.

6 You can use the Start date and End date icons to select the date range within
which the backup occurred.
7 Under Directory Structure, drill into the folders and click the folders and files
you want to restore.

8 Click Restore.
9 On the Restore Marked Files dialog: If NetBackup client software is not
installed on the destination virtual machine, select one of the following:
■ Restore everything to its original location
■ Restore everything to a different location (maintaining existing
structure)
■ Restore individual directories and files to different locations
Destinations must be entered as UNC path names that refer to shared
drives on the virtual machine. For example, to restore the file
E:\folder1\file1 on virtual machine vm1, enter the following destination:

\\vm1\e$\folder1\file1

The NetBackup Client Service must be logged on under an account that

has Administrator privileges.
See “Setting up NetBackup Client Service for restore to a Windows shared
virtual machine driveSetting up NetBackup Client Service for VMware restores
to a Windows shared virtual machine drive” on page 227.
 Restore virtual machines 227
Restoring individual filesRestoring individual VMware virtual machine files

See “About restore of individual filesAbout restoring individual VMware files”

on page 223.
See “Restoring the full VMware virtual machine” on page 179.
See “About VMware virtual machine disk restore” on page 196.
See “Restore notes and restrictions” on page 173.

Setting up NetBackup Client Service for restore to a Windows shared

virtual machine driveSetting up NetBackup Client Service for VMware
restores to a Windows shared virtual machine drive
This topic relates to the restore of files from a NetBackup backup of a VMware
virtual machine.
To restore individual files to a Windows virtual machine that has a shared drive,
note: the NetBackup Client Service must be logged on under an account that has
Administrator privileges (not as the Local System account). An account with
Administrator privileges lets NetBackup write to the directories on the virtual machine
to which the data is restored.
If you try to restore files while the NetBackup Client Service is logged on as the
Local System account, the restore fails.
To log on the NetBackup Client Service as Administrator
1 In Windows Services on the VMware recovery host, double-click the NetBackup
Client Service.
2 Check the Log On tab: if the service is not logged on under an account that
has Administrator privileges, stop the service.
3 Change the logon to the Administrator account, or to an account that has
Administrator privileges.
The account must have Administrator privileges in the domain in which both
the virtual machine and the VMware backup host reside.
4 Restart the service.
5 Retry the restore.
See “Restoring individual filesRestoring individual VMware virtual machine files”
on page 224.
See “Restore notes and restrictions” on page 173.
 Restore virtual machines 228
Restoring individual filesRestoring individual VMware virtual machine files

Restore Marked Files dialog for restore of VMware files

This topic describes the Restore Marked Files dialog for restoring individual folders
and files from a VMware virtual machine backup.

Table 10-11 Options for individual file restore on the Restore Marked Files
dialog box

Option Description

Destination Select from the following options.

 Restore virtual machines 229
Restoring individual filesRestoring individual VMware virtual machine files

Table 10-11 Options for individual file restore on the Restore Marked Files
dialog box (continued)

Option Description

Restore everything to Restores the folders and files to the location where they resided
its original location when the backup occurred.

Restore everything to Restores the folders and files with their original hierarchy, but to a
a different location different location.
(maintaining existing
Use the Destination field to enter the restore location.
structure)
Click Browse to browse to the restore location.

Restore individual Restores the folders and files to individually designated locations.
directories and files To designate a restore destination for each source folder, double
to different locations click on its row. Use the following dialog to enter or browse for the
destination:

To restore to a Windows mounted drive: Destinations must be

entered as UNC path names that refer to shared drives on the
virtual machine. For example, to restore the file
E:\folder1\file1 on virtual machine vm1, enter the following
destination:

\\vm1\e$\folder1\file1

See “Setting up NetBackup Client Service for restore to a Windows

shared virtual machine driveSetting up NetBackup Client Service
for VMware restores to a Windows shared virtual machine drive”
on page 227.

Create and restore to This option is not implemented.

a new virtual hard
disk file

Options For a description of the restore options, see the NetBackup help
for the Backup, Archive, and Restore interface.
 Restore virtual machines 230
How NetBackup handles VMware tag associations at restore

Table 10-11 Options for individual file restore on the Restore Marked Files
dialog box (continued)

Option Description

Media Server You can use this option to select a media server that has access
to the storage unit that contains the backup image. An example of
such an environment is a Media Server Deduplication Pool (MSDP)
with multiple media servers.
Note: If the storage unit that contains the backup image is not
shared with multiple media servers, this option is grayed out.

Override default Determines the restore job's priority for restore resources. A higher
priority priority means that NetBackup assigns the first available drive to
the first restore job with the highest priority. Enter a number
(maximum 99999). The default for all restore jobs is 0, the lowest
priority possible. Any restore job with a priority greater than zero
has priority over the default setting.

See “Restoring the full VMware virtual machine” on page 179.

See “About restore of individual filesAbout restoring individual VMware files”
on page 223.
See “If the recovery host is not at the same NetBackup release level as the backup
host” on page 236.

How NetBackup handles VMware tag associations

at restore
To better understand the NetBackup restore behavior, it's important to understand
some of the internal mechanics of how VMware handles tag associations. For each
tag that is created in VMware, there is a corresponding internal identifier you cannot
see or edit. This feature allows VMware to function correctly without the need to
account for variation in naming conventions. Tags are replicated throughout the
environment through the Platform Services Controller (PSC).
NetBackup recognizes and uses tag names when they are part of the VIP query.
Backups of the virtual machine store the tag name and VMware internal identifier
for all tags that are associated with that virtual machine. During the restore, however,
NetBackup only creates the tag associations based on the VMware internal identifiers
that are defined on the target vCenter server.
 Restore virtual machines 231
How NetBackup handles VMware tag associations at restore

Figure 10-1 Sample VMware environment

Tag A (B) Tag A (B)

Identifier: 135 Identifier: 135

vCenter1 vCenter2

Platform Services
Controller

VM1
Tag A (B)
Identifier: 135

Tag A
Identifier: 246

vCenter3 VM2
Tag A
Identifier: 246

Figure 10-1 shows a sample VMware environment with multiple vCenter servers,
virtual machines, and tags. Assume that a backup of all virtual machines completes
successfully with a NetBackup status code 0.
 Restore virtual machines 232
Browse and search virtual machines for restore

■ If you restore VM1 to either vCenter1 or vCenter2, it is restored with tag A and
the restore exits with a NetBackup status code 0. This behavior is true both for
a restore to the same name as well as an alternate client restore. This behavior
is the result of VMware replicating tags across all vCenter servers that are
attached to a single PSC.
■ If you restore VM1 to vCenter3, it is restored without any tags. The restore exits
with a NetBackup status code 1. This behavior is because VMware uses the
internal identifier. While there is a tag name A in vCenter3, the internal identifier
for tag A does not match the internal identifier that is restored. This behavior is
true both for a restore to the same name as well as an alternate client restore.
■ After the NetBackup backup, if tag A is renamed to B, when VM1 is restored to
either vCenter1 or vCenter2 it is restored with tag B. The restore exits with a
NetBackup status code 0. This behavior is because VMware uses the internal
identifier, and now associates this identifier with the tag name B.
■ After the NetBackup backup, if tag A is deleted, when VM1 is restored to either
vCenter1 or vCenter2 it is restored without any tag associations. The restore
exits with a NetBackup status code 1.
If for any reason the backup of the virtual machines in Figure 10-1 did not
successfully capture the tag associations, the backup exits with NetBackup status
code 0. The reasons for failing to capture tag associations appear in the Activity
Monitor. Any restores based on this backup exit with NetBackup status code 0, but
no tag information is restored. Depending on the backup error, more information
regarding tag associations may appear in the Activity Monitor.

Browse and search virtual machines for restore

To browse and search virtual machines for restore
1 In the Backup, Archive, and Restore interface, click Actions > Specify
NetBackup Machines and Policy Type....
2 Under Policy type for restores, select VMware.
3 Click Search VM Clients.
The Browse and Search Virtual Machines for Restore dialog appears. Use
this dialog to find a virtual machine.

Note: This dialog finds the virtual machines that were backed up with the VMware
policy type. It does not search for other types of backups.
 Restore virtual machines 233
Browse and search virtual machines for restore

Note: If the backup was made before NetBackup 7.6, the backup image must be
imported and verified before it can be browsed or searched.

Table 10-12 Browse and Search Virtual Machines for Restore dialog box

Dialog box item Description

Search virtual Search for virtual machines by name, IP address, UUID, or by other
machines criteria.

See “Search Virtual Machines for Restore dialog” on page 233.

Browse virtual Browse for virtual machines by drilling down through the virtual
machines objects.
Select the type of view for browsing:

■ vCloud view
Shows the virtual objects as they appear in vCloud Director.
■ vSphere view
Shows the virtual objects as they appear in vSphere Client.

See “Browse the vCloud or vSphere view for virtual machine clients”
on page 235.

Note: Although VMware allows up to 128 characters for an object name, the
NetBackup database restricts searches to the first 126 characters only.

Search Virtual Machines for Restore dialog

Use this screen to search for virtual machines by name or other identifier, or to use
other filter criteria.
Enter Display Name, Host Name, IP Address, UUID, DNS Name or vApp to
search for:
Enter the name or number of the virtual machine to search for and click Next. Any
matching virtual machines appear in the Search results dialog.
See “Search results dialog” on page 236.
Advanced Search:
Click to create a rule-based search query. See Table 10-13.
 Restore virtual machines 234
Browse and search virtual machines for restore

Table 10-13 Advanced Search: filter parameters

Search item Description

Available when you click the plus sign to add a rule to the
search.

AND: Limits or restricts the scope of the search.

OR: Opens the search to an additional possibility,

expanding the scope of the search.

Select a parameter on which to search.

You can type the first character of the parameter to select

it.

Select an operator.

You can type the first character of the operator to select it.

The available operators depend on the parameter that was

selected.

Enter a value for the parameter. The search looks for the
virtual machines that match the value.

Choose a date and time.

Refine or expand the search (add another rule).

 Restore virtual machines 235
Browse and search virtual machines for restore

Table 10-13 Advanced Search: filter parameters (continued)

Search item Description

Remove the rule from the search.

Basic Search Search by display name, host name, IP address, UUID,

DNS name, or vApp.

Next Run the search. The matching virtual machines appear in

the Search results dialog.

See “Search results dialog” on page 236.

Note the following about the search behavior:

■ Although VMware allows up to 128 characters for an object name, the NetBackup
database restricts searches to the first 126 characters only.
■ To search for a vSphere vApp by name without specifying its path: Use the
search operator Contains or EndsWith (do not use Equal). To use Equal, you
must specify the entire path of the vApp.
■ To search by IP address: If the virtual machine has two or more IP addresses
(such as virtual addresses), use Contains as the search operator instead of
Equal.
■ You must include a leading forward slash (/) when searching for a data center,
VM folder, or resource pool. If you use the Contains operator, a leading slash
is not required.
■ Regarding wildcards: * and ? are not supported for Greater, GreaterEqual,
Less, and LessEqual operators. The * and ? characters are treated as literals,
not wildcards.
For the rest of the operators, * matches everything, and ? matches any single
character.
Note: SQL has a limitation that wildcards * and ? (SQL equivalents are % and
_) do not match null values. If a virtual machine does not have an IP address,
the query “IPAddress Equal *” does not find the virtual machine.

Browse the vCloud or vSphere view for virtual machine clients

The Browse virtual machines using the vCloud (or vSphere) view dialog shows
your virtual environment starting with the vCenter server. The objects appear
according to the view you chose (vCloud or vSphere).
 Restore virtual machines 236
If the recovery host is not at the same NetBackup release level as the backup host

To select a virtual machine, click on an ESX server or vApp. Highlight the virtual
machine you want and then click Select.
To switch the view to vCloud or vSphere, click on the link in the upper right.

Note: Although VMware allows up to 128 characters for an object name, the
NetBackup database restricts searches to the first 126 characters only.

See “Browse and search virtual machines for restore” on page 232.

Search results dialog

The virtual machines that match an advanced search (from the Search Virtual
Machines for Restore dialog) are listed in the Search results dialog. The search
criteria are listed at the top.
Click on the virtual machine that you want to restore and click Select. Click Refresh
on the Specify NetBackup Machines and Policy Type screen.
If no virtual machines appear, click Back to revise your search criteria.
See “Browse and search virtual machines for restore” on page 232.

If the recovery host is not at the same NetBackup

release level as the backup host
Note the following about mixed-level backups and restores:
■ To restore from a NetBackup 7.x backup
To restore a virtual machine from a backup that a NetBackup 7.x backup host
performed, a NetBackup 7.x recovery host is required. You cannot use a
NetBackup 6.5.x recovery host to restore a virtual machine that was backed up
by a NetBackup 7.x backup host.
■ To restore from a NetBackup 6.5.x backup
To restore a virtual machine (or selected files) from a NetBackup 6.5.x backup,
the VMware Converter must be installed on the NetBackup recovery host. The
recovery host can run NetBackup 6.5.x or 7.x. A staging area is required.
See “Restoring the full VMware virtual machine” on page 179.
See “Restoring individual filesRestoring individual VMware virtual machine files”
on page 224.
 Chapter 11
Restore virtual machines
with Instant Recovery
This chapter includes the following topics:

■ About Instant Recovery for VMware

■ Task overview for Instant Recovery for VMware

■ Performance recommendations for Instant Recovery for VMware

■ Requirements for Instant Recovery for VMware

■ Notes on Instant Recovery for VMware

■ Restarting the Client for NFS service on a Windows restore host

■ Instant Recovery options on the nbrestorevm command

■ Restoring a virtual machine with Instant Recovery for VMware

■ Restoring a virtual machine to a different location with Instant Recovery for

VMware

■ Restoring individual files with Instant Recovery for VMware while the current
virtual machine is running

■ Job types for Instant Recovery for VMware

■ Reactivating a restored virtual machine with Instant Recovery for VMware

About Instant Recovery for VMware

NetBackup can recover a virtual machine almost instantly, without waiting to transfer
the virtual machine's data from the backup. NetBackup starts the virtual machine
 Restore virtual machines with Instant Recovery 238
About Instant Recovery for VMware

directly from the backup image and makes it accessible to users on the target ESX
host immediately. You can copy files (including vmdk files) without restoring the
entire virtual machine. To restore the virtual machine, use VMware Storage vMotion
to migrate the virtual machine data files from the backup image to the ESX host.
Some example uses for instant recovery:
■ Access and restore individual files and folders from any type of OS and then
delete the virtual machine. (Note for Windows or Linux: Instead of instant
recovery, you can use the policy Enable file recovery from VM backup option
and restore individual files with the Backup, Archive, and Restore interface.)
■ Test a patch on a restored virtual machine before you apply the patch to
production systems.
■ Troubleshoot a virtual machine or host, such as when the production ESX host
is down. You can start the virtual machine from its backup and use it until the
production system is back online.
■ Permanently recover the virtual machine by means of Storage vMotion.
■ Verify the backup image.
■ Copy a vmdk file and then delete the virtual machine.
■ Verify an application.
In any case, the virtual machine is started directly from the backup image and is
available in seconds or minutes. The startup time depends on the network speed
and storage speed, not on the size of the virtual machine.
Table 11-1 describes the steps in a virtual machine instant recovery.

Table 11-1 How Instant Recovery for VMware works

Sequence Actions

Step 1 Run the nbrestorevm command* to access the virtual machine from its backup
image. The NetBackup File System Service (NBFSD) on the media server
accesses the backup image file system and mounts the image as an NFS
datastore. The datastore becomes accessible to the ESX host where the
virtual machine is to be restored.

On the same command, select a temporary datastore that is accessible to

the ESX host.

Step 2 NetBackup creates a virtual machine on the ESX host and configures the
virtual machine with write access to a temporary (local) datastore.

Step 3 NetBackup creates a snapshot of the virtual machine. Any new write requests
in the virtual machine use the temporary datastore. The virtual machine uses
the NFS datastore as read-only.
 Restore virtual machines with Instant Recovery 239
Task overview for Instant Recovery for VMware

Table 11-1 How Instant Recovery for VMware works (continued)

Sequence Actions

Step 4 NetBackup starts up the virtual machine on the ESX host.

Step 5 To keep the restored VM: Use Storage vMotion to copy the virtual machine
data from the NFS datastore to the temporary datastore.

Step 6 When vMotion is complete, use nbrestorevm to unmount the NFS datastore.

*NetBackup provides a command-line interface for instant recovery of virtual

machines (nbrestorevm). A graphical user interface (the Instant Recovery Wizard)
is available in the NetBackup vSphere Web Client plug-in and the NetBackup
vSphere Client (HTML5) plug-in. For details, see the NetBackup Plug-in for VMware
vSphere Web Client Guide and the NetBackup Plug-in for VMware vSphere Client
(HTML5) Guide, available from this location:
http://www.veritas.com/docs/000003214
For additional information on Instant Recovery for VMware:
See “Task overview for Instant Recovery for VMware” on page 239.
See “Notes on Instant Recovery for VMware” on page 242.
See “Instant Recovery options on the nbrestorevm command” on page 244.
See “Restoring a virtual machine with Instant Recovery for VMware” on page 250.

Task overview for Instant Recovery for VMware

Table 11-2 describes the tasks for Instant Recovery for VMware.

Table 11-2 Instant Recovery tasks

Step Description Reference topic

Step 1 Review the performance See “Performance recommendations

recommendations for Instant Recovery for VMware”
on page 240.

Step 2 Review the notes and requirements See “Requirements for Instant
Recovery for VMware” on page 241.

See “Notes on Instant Recovery for

VMware” on page 242.
 Restore virtual machines with Instant Recovery 240
Performance recommendations for Instant Recovery for VMware

Table 11-2 Instant Recovery tasks (continued)

Step Description Reference topic

Step 3 Restart the Client for NFS service on See “Restarting the Client for NFS
the restore host service on a Windows restore host”
on page 244.

Step 4 Review the Instant Recovery options See “Instant Recovery options on the
on the nbrestorevm command nbrestorevm command” on page 244.

Step 5 Use the nbrestorevm command to See “Restoring a virtual machine with
perform Instant Recovery Instant Recovery for VMware”
on page 250.

See “Restoring individual files with

Instant Recovery for VMware while
the current virtual machine is running”
on page 257.

Performance recommendations for Instant

Recovery for VMware
After Instant Recovery, the virtual machine is on an NFS-attached datastore that
is presented by the NetBackup media server. The performance of the virtual machine
and of Storage vMotion depends on the following: the network speed and latency
between the ESXi host and the media server, and the speed of NetBackup storage
that the backup is recovered from.
It is recommended the following:
■ A SAN connection from the NetBackup media server to its disk storage unit.
■ For Fibre Channel SAN, a minimum speed of 4 gigabits per second.
■ For iSCSI SAN, a minimum speed of 1 gigabit per second.
■ When you use Storage vMotion to migrate a restored virtual machine, migrate
one virtual machine at a time per media server. The migration may be slow if
you simultaneously migrate multiple virtual machines per media server.
■ For disaster recovery testing, it is recommended that you restore no more than
three or four virtual machines per media server. The number to restore depends
on the I/O load on the media server. It is recommended restoring each VM
one-by-one, not simultaneously.
 Restore virtual machines with Instant Recovery 241
Requirements for Instant Recovery for VMware

Note: For large-scale recovery of multiple virtual machines, use the virtual machine
restore feature in the Backup, Archive, and Restore interface. Do not use Instant
Recovery for VMware.

Requirements for Instant Recovery for VMware

For virtual machine instant recovery, your environment must meet the following
requirements:
■ The virtual machine to restore must have been backed up from a VMware policy.
■ The target ESX server for the restore must be at vSphere 5.0 or later.
■ The restore host can be on Windows or Linux.

Note: For the VMware virtual machines that have non-ASCII characters in their
paths, NetBackup does not support Instant Recovery using Windows restore
hosts and media servers. You must use a Linux restore host and media server
for Instant Recovery of such virtual machines.
The requirements and limitations for non-ASCII character support are described
in a different topic.
See “NetBackup for VMware: notes and restrictions” on page 27.

■ For a restore host that is separate from the NetBackup master server or media
server: You must add the restore host to the list of servers that can access the
master server.
In the NetBackup Administration Console, click Host Properties > Master
Servers, double-click the NetBackup master server, and click Servers. On the
Additional Servers tab, click Add to add the restore host.
See “Allowing other servers or clients to restore virtual machines” on page 353.

■ The NFS Client service must be enabled on the ESXi host.

■ The Network File System (NFS) must be installed on the Linux media server
and restore host and the portmap service must be active.
For information about how to install NFS, see the media server host operating
system documentation.
■ The Services for Network File System (NFS) must be installed on the Windows
media server and restore host.
See “About configuring services for NFS on Windows 2012 or 2016 (NetBackup
for VMware)” on page 356.
 Restore virtual machines with Instant Recovery 242
Notes on Instant Recovery for VMware

See “About configuring services for NFS on Windows 2008 and 2008 R2
(NetBackup for VMware)” on page 363.
The NetBackup media server platform must support Granular Recovery
Technology. See the NetBackup Enterprise Server and Server OS Software
Compatibility List:
NetBackup Master Compatibility List
■ The Client for NFS service may have to be restarted on a NetBackup Windows
restore host.
See “Restarting the Client for NFS service on a Windows restore host”
on page 244.
■ The media server must use IPv4 or have a dual stack configuration if the vCenter
server has a dual stack configuration.
■ NetBackup requires logon credentials for the vCenter server and the restore
host.
See “Adding NetBackup credentials for VMware” on page 45.

Notes on Instant Recovery for VMware

Note the following about instant recovery of VMware virtual machines:
■ Supports the following storage unit types (disk only): BasicDisk, AdvancedDisk,
Media Server Deduplication Pool (MSDP), and qualified third-party OpenStorage
devices.
Note: Snapshot-only backups are not supported.
■ Does not support a virtual machine that had the disks that were excluded from
the backup. The policy Virtual disk selection option must have been set to
include all disks.
■ Does not support a virtual machine that has a disk in raw device mapping mode
(RDM) or that has a disk in Persistent mode.
■ Supports the following policy schedule types: Full backups, and the incremental
backups that include the Use Accelerator option with a disk-based storage unit.
Incrementals without the Use Accelerator policy option are not supported.
■ Does not support virtual machine templates.
■ If the virtual machine contains an IDE drive, the restored virtual machine may
not start. This issue is not unique to instant recovery.
See “VMware virtual machine does not restart after restore” on page 312.
■ To avoid host name or IP address conflicts between the current virtual machine
and the virtual machine version you want to restore: Shut down the virtual
machine in your production environment before you start the recovery. Then
 Restore virtual machines with Instant Recovery 243
Notes on Instant Recovery for VMware

change the display name of the current virtual machine, or use the -R option on
nbrestorevm to rename the restored virtual machine.
■ For a virtual machine that is running under a high load, migration of the virtual
machine may take longer than expected. For this reason, NetBackup changes
the virtual machine's fsr.maxSwitchoverSeconds property to 900.
For example, this increase may be necessary when the virtual machine is
restored from a deduplication storage unit.
The following VMware Knowledge Base article contains more information on
the fsr.maxSwitchoverSeconds property:
Using Storage vMotion to migrate a virtual machine with many disks timeout
■ Note the following about the virtual machine's datastore name:
■ If the name of the datastore includes spaces, the name should be enclosed
in double quotes ("").
■ A virtual machine restore may fail if the name of the datastore (that was used
at the time of the backup) ended with a period.
The following tech note contains additional information.
http://www.veritas.com/docs/TECH199771

■ Instant recovery cannot restore a vCloud virtual machine into vCloud. The virtual
machine is restored into vSphere. You can copy or import the restored virtual
machine into vCloud by means of the Copy option in vCloud.
Note that the vCloud Move option does not work with a virtual machine that
runs from a NetBackup datastore.
■ Storage lifecycle policies (SLPs) can use Auto Image Replication to replicate a
virtual machine backup image to another NetBackup domain. To restore the
virtual machine from the replicated image, you must include the -vmproxy option
on the nbrestorevm command. Use the -vmproxy option to specify the backup
host (access host) that is in the domain where the virtual machine was replicated.
Without the -vmproxy option, nbrestorevm defaults to the backup host in the
original domain and the restore fails.
■ Supports recovery of virtual machines containing independent disks. The
independent disks that are associated with the virtual machine are recovered
to the virtual machine working directory on the temporary datastore. This
functionality requires a NetBackup 8.3 or later recovery host.
 Restore virtual machines with Instant Recovery 244
Restarting the Client for NFS service on a Windows restore host

Restarting the Client for NFS service on a

Windows restore host
It may be necessary to stop and restart the NFS Client service. If you use the
Microsoft services snap-in (Services.msc) to restart it, the service does not start
until you restart the server.
To restart the Client for NFS service without a server restart
◆ From the Windows command prompt, run the following commands:

net stop nfsclnt

net stop nfsrdr
net start nfsrdr
net start nfsclnt

The Client for NFS service should restart without a restart of the server.

Instant Recovery options on the nbrestorevm

command
NetBackup provides a command-line interface for instant recovery of virtual
machines: the nbrestorevm command.
The following is a list of the available nbrestorevm options for performing Instant
Recovery of a VMware virtual machine. The options are listed in man page format.

Note: Although the nbrestorevm command has additional options, only the options
that are described in this topic apply to Instant Recovery.

To initiate Instant Recovery (activate the virtual machine)

Options without brackets are required.

nbrestorevm -vmw -ir_activate -C vm_client

-temp_location temp_location_for_writes
[-S master_server] [-vmpo] [-vmInstanceId] [-vmsn] [-vmst]
[-vmserver vm_server] [-vmproxy vm_proxy] [-vmkeephv] [-vmid]
[-vmnewdiskuuid] [-s mm/dd/yyyy [HH:MM:SS]]
[-e mm/dd/yyyy [HH:MM:SS]]
[-R absolute_path_to_rename_file]
[-disk_media_server media_server]
 Restore virtual machines with Instant Recovery 245
Instant Recovery options on the nbrestorevm command

Note: Only -vmw, -ir_activate, -C, and -temp_location are required. If the other
options are not specified, NetBackup automatically supplies values for those options
from the backup. In most cases, if you do not restore the virtual machine to a different
location, you can omit the bracketed options.

See “To restore a virtual machine with instant recovery” on page 251.
See “Restoring individual files with Instant Recovery for VMware while the current
virtual machine is running” on page 257.

To list details about the activated virtual machine

nbrestorevm -ir_listvm

To deactivate or delete the virtual machine

nbrestorevm -ir_deactivate instant_recovery_identifier [-force]

To complete the VM instant recovery job after the data is

migrated
nbrestorevm -ir_done instant_recovery_identifier

To reactivate a virtual machine that was interrupted during

recovery
nbrestorevm -ir_reactivate instant_recovery_identifier [-force]
nbrestorevm -ir_reactivate_all -vmhost vm_host -media_server
media_server_activate_vm [-force]

See “Reactivating a restored virtual machine with Instant Recovery for VMware”
on page 261.
Table 11-3 describes the nbrestorevm options for Instant Recovery of virtual
machines. These options are also described in the NetBackup Commands Reference
Guide, and in the man page for nbrestorevm.

Table 11-3 Options on the nbrestorevm command for Instant Recovery

Option Description

-vmw Indicates the type of virtual machine to restore (VMware).

 Restore virtual machines with Instant Recovery 246
Instant Recovery options on the nbrestorevm command

Table 11-3 Options on the nbrestorevm command for Instant Recovery

(continued)

Option Description

-C virtual_machine The name of the virtual machine as identified in the backup. For example, if the policy
backed up the virtual machine by its host name, specify that host name.
Note: If a virtual machine with the same name already exists on the target ESX host, the
command fails. You can change the display name of the virtual machine in vCenter. As an
alternative, use the -R option on nbrestorevm to change the display name or location of
the restored virtual machine.

-ir_activate Starts the restore by mounting the backup image of the virtual machine as an NFS datastore.
The datastore becomes accessible to the ESX host where the virtual machine is to be
restored.

The following procedure has further details:

See “To restore a virtual machine with instant recovery” on page 251.

-temp_location A temporary datastore on the ESX server where all writes occur until the virtual machine
temporary_datastore is restored. All writes occur on this datastore until Storage vMotion is complete or until you
are finished with the virtual machine (such as for troubleshooting).
Note: This datastore must exist before you run nbrestorevm.

-S master_server Specifies a different master server, to restore a virtual machine from a backup that was
made by that master.
 Restore virtual machines with Instant Recovery 247
Instant Recovery options on the nbrestorevm command

Table 11-3 Options on the nbrestorevm command for Instant Recovery

(continued)

Option Description

-R rename_file_path The absolute path to a file that contains directives for restore to a different location.

The following are the entries that you can use in the file (each entry begins with change):

change esxhost to new_ESX_host

change resourcepool to new_resource_pool
change vmname to new_virtual_machine_name
change network to new_network

Note: The change line should be entered exactly as it appears in this list, except for the
variable at the end (such as new_virtual_machine_name). Replace that variable with the
new name.

For example:

change esxhost to esx01.prod4.com

change resourcepool to /Tech1/esx01.prod4.com/Res
change vmname to T1vm5

Each change line must end with a carriage return.

Note: If the file contains only one change line, (such as change vmname to T1vm5),
that line must end with a carriage return.

If you restore to a different ESX server, you may have to specify a different resource pool.
(The original resource pool cannot be used to create a VM on a different ESX server.) You
must specify a resource pool by including the following line in the rename file:

change resourcepool to path_to_new_resource_pool

-vmst Removes the virtual machine's tags.

-vmserver The name of the vCenter server that manages the target ESX host for the restore. To
vCenter_server restore to the same vCenter where the virtual machine originally resided, omit this option.

-vmproxy restore_host The host to perform the restore (a NetBackup client). The default is the host that performed
the backup.
 Restore virtual machines with Instant Recovery 248
Instant Recovery options on the nbrestorevm command

Table 11-3 Options on the nbrestorevm command for Instant Recovery

(continued)

Option Description

-disk_media_server Specifies which media server performs the Instant Recovery.

media_server
This option is useful if NetBackup storage is configured over several media servers, such
as for load balancing. Without the -disk_media_server option, the Instant Recovery job
may select any of the available media servers to do the restore. If only one of the media
servers is configured for Instant Recovery, specify that server with the -disk_media_server
option.

See “Notes on Instant Recovery for VMware” on page 242.

-vmpo Turns on the virtual machine after the restore.

-vmInstanceId Retains the Instance UUID of the original virtual machine (note that the Instance UUID is
a vCenter specific unique identifier of a virtual machine). The virtual machine is restored
with the same Instance UUID that it had when it was backed up.

If the restore of the virtual machine is to a standalone ESXi host, this option is disabled.

If a virtual machine with the same Instance UUID exists at the target restore location, a
message appears that the UUID is used already. In this case, the original instance UUID
is not restored and a new UUID is assigned for the virtual machine.

-vmsn Removes the virtual machine's network interface.

See “Restoring individual files with Instant Recovery for VMware while the current virtual
machine is running” on page 257.

-vmkeephv Retains the virtual machine's hardware version: the virtual machine is restored with the
same hardware version that it had when it was backed up.

Without this option, the virtual machine is restored with the default hardware version of the
target ESX server or vCenter. The default is typically the latest supported hardware version.

-vmid Retains the BIOS UUID of the original virtual machine.

-vmnewdiskuuid Generate new virtual machine disk UUIDs during an instant recovery. Use this option with
the –ir_activate option.

The VMs that activate with this option do not retain the new vmdk UUID during a subsequent
–ir_reactivate operation. In such a scenario, the VMDKs revert to their UUIDs at the
time of the backup.
 Restore virtual machines with Instant Recovery 249
Instant Recovery options on the nbrestorevm command

Table 11-3 Options on the nbrestorevm command for Instant Recovery

(continued)

Option Description

-ir_deactivate instant Removes the virtual machine from the ESX host. If no other virtual machine uses the
recovery ID NetBackup NFS datastore, this option removes that datastore and releases its resources
on the NetBackup media server.

This option is for removing the virtual machine when you are finished copying its files and
do not intend to keep it.
Note: If the virtual machine has been migrated to a production datastore by means of
vMotion, -ir_deactivate is the same as ir_done.

Note: If vMotion is still in progress for this virtual machine, you should cancel the vMotion
job before you use –ir_deactivate.

See “To restore a virtual machine with instant recovery” on page 251.

-ir_done instant recovery When vMotion migration of the virtual machine is finished, this option completes the virtual
ID machine instant recovery job. It also removes the NetBackup NFS datastore if no other
virtual machine uses it. When the datastore is removed, its resources are released on the
media server.

See “To restore a virtual machine with instant recovery” on page 251.

-ir_reactivate Reactivates a restored virtual machine: remounts the NetBackup NFS datastore, and from
instant_recovery_identifier the temporary datastore on the ESX host it registers the restored virtual machines on the
ESX host.
-ir_reactivate_all
instant recovery ID is the virtual machine's numeric identifier from the -ir_listvm output.
If more than one virtual machine needs reactivation: -ir_reactivate_all remounts the
NetBackup NFS datastores on the media server and reactivates the virtual machines. For
example: If VM1, VM5, and VM6 were mounted on the specified ESX server (-vmhost),
and their datastores were mounted on the specified media server (-media_server), those
three VMs are reactivated.

-vmhost vm_host When you reactivate virtual machines, this option specifies the ESX host on which the
virtual machines were mounted.

-media_server When you reactivate virtual machines, this option specifies the media server on which the
media_server_activate_vm NFS datastores that contain the backup images were mounted.

-force For the -ir_deactivate, –ir_reactivate, or –ir_reactivate_all option, -force suppresses

confirmation prompts.
 Restore virtual machines with Instant Recovery 250
Restoring a virtual machine with Instant Recovery for VMware

Table 11-3 Options on the nbrestorevm command for Instant Recovery

(continued)

Option Description

-s start_time -e end_time By default NetBackup selects the latest full backup plus Accelerator-enabled incremental
backups (if any).

This option limits the selectable backup images to those with the timestamps that fall within
the specified period. NetBackup selects the latest suitable backup image within the range.

Restoring a virtual machine with Instant Recovery

for VMware
You can use this procedure to do either of the following:
■ Copy files from a virtual machine backup.
■ Restore the full virtual machine.
In either case, you can restore the virtual machine to its original location or to an
alternate location.

Note: To avoid host name or IP address conflicts, shut down the current virtual
machine in your production environment before you start instant recovery.

To copy files while the current virtual machine is running, use a different procedure:
See “Restoring individual files with Instant Recovery for VMware while the current
virtual machine is running” on page 257.

Table 11-4 Basic steps for VMware instant recovery

Type of recovery Steps

Copy files or Basic steps are these:

troubleshoot an issue,
■ Restore the VM: Use nbrestorevm with the -ir_activate option.
then delete the
■ Copy files from the VM; or use the VM as a stand-in until the production host is back
restored virtual
online.
machine
■ Delete the VM and release the media server resources: Use nbrestorevm with the
-ir_deactivate option.

See the following procedure for command details.

 Restore virtual machines with Instant Recovery 251
Restoring a virtual machine with Instant Recovery for VMware

Table 11-4 Basic steps for VMware instant recovery (continued)

Type of recovery Steps

Restore and keep the Basics steps are these:

virtual machine
■ Restore the VM: Use nbrestorevm with the ir_activate option.
■ Transfer the virtual machine files to an ESX host: Use the Migrate option in vSphere
Client.

■ Release the media server resources: Use nbrestorevm with the ir_done option.

See the following procedure for command details.

See “Requirements for Instant Recovery for VMware” on page 241.

The detailed procedure follows.
To restore a virtual machine with instant recovery
1 On the master server, media server, or restore host, enter the nbrestorevm
command.
This command is in the following location:
UNIX, Linux: /usr/openv/netbackup/bin/
Windows: install_path\NetBackup\bin\
Enter the command as follows.
To restore the VM to its original location:
nbrestorevm -vmw -ir_activate -C virtual_machine -temp_location
temporary_datastore [-vmproxy VMware_access_host] -vmpo

To restore the VM to a different location:

nbrestorevm -vmw -ir_activate -C virtual_machine -temp_location
temporary_datastore [-vmserver vCenter_server] -R rename_file_path
[-vmproxy VMware_access_host] -vmpo

-C virtual_machine identifies the virtual machine by the name or ID that was

set in the policy's Primary VM identifier attribute for the backup. On the -C
option, specify the same type of identifier that was used in the policy: VM host
name, VM display name, VM BIOS UUID, VM DNS name, or VM instance
UUID.
The -R option provides the path to a file that contains directives for restore to
a different location.
See “Restoring a virtual machine to a different location with Instant Recovery
for VMware” on page 254.
 Restore virtual machines with Instant Recovery 252
Restoring a virtual machine with Instant Recovery for VMware

See “Instant Recovery options on the nbrestorevm command” on page 244.

The nbrestorevm command mounts the virtual machine's backup image as an
NFS datastore and makes the datastore accessible to the ESX host. It also
creates the VM on the ESX host. It then creates a snapshot of the virtual
machine.

Note: Storage lifecycle policies (SLPs) can use Auto Image Replication to
replicate a virtual machine backup image to another NetBackup domain. To
restore the virtual machine from the replicated image, you must include the
-vmproxy option on the command. Use the -vmproxy option to specify the
backup host (access host) that is in the domain where the virtual machine was
replicated. Without the -vmproxy option, nbrestorevm defaults to the backup
host in the original domain and the restore fails.

The following tasks appear in the vSphere Client interface. In this example,
dbl1vm5 is the virtual machine to be restored.

Note the following:

■ The nbrestorevm command creates a NetBackup job of type "VM Instant
Recovery."
See “Job types for Instant Recovery for VMware” on page 260.
■ If you cancel the instant recovery job or stop all NetBackup services, the
NetBackup NFS datastore is unmounted and its media server resources
are released.

Caution: The virtual machine is deleted from the ESX host.

2 In vSphere Client, turn on the virtual machine.

If you included the -vmpo option on the nbrestorevm command, the virtual
machine is already turned on.
 Restore virtual machines with Instant Recovery 253
Restoring a virtual machine with Instant Recovery for VMware

3 Browse and copy the virtual machine files as needed.

To copy files while the current virtual machine is running, use a different
procedure.
See “Restoring individual files with Instant Recovery for VMware while the
current virtual machine is running” on page 257.
4 If you do not want to keep the restored virtual machine, enter the following:
nbrestorevm –ir_listvm

In the output, find the VM Instant Recovery ID for the restored VM.
To remove the VM from the ESX host:
nbrestorevm –ir_deactivate instant recovery ID [-force]

where instant recovery ID is the virtual machine's numeric identifier from the
-ir_listvm output. -force is optional, to suppress confirmation prompts.
The VM is removed from the ESX host. If no other VM uses the NetBackup
NFS datastore, NetBackup removes that datastore and releases its resources
on the media server.
The following tasks appear in the vSphere Client interface. In this example,
dbl1vm5 is the virtual machine to be removed and datastore_V is the temporary
datastore that it used.

This step completes the VM Instant Recovery job. Skip the rest of this
procedure.
Step 5 uses Storage vMotion to move the virtual machine to a production
datastore. If vMotion is already in progress for this virtual machine, you should
cancel the vMotion job before you enter –ir_deactivate. Otherwise, vMotion
moves the virtual machine to a production datastore where –ir_deactivate
cannot remove it.
 Restore virtual machines with Instant Recovery 254
Restoring a virtual machine to a different location with Instant Recovery for VMware

5 To keep the restored virtual machine:

In vSphere Client, right-click on the restored virtual machine and select Migrate.
Select the migration type and the destination.
Note: For the destination, select a permanent (production) location for the
virtual machine. Do not select the temporary datastore that was used for the
instant restore.
Storage vMotion transfers the virtual machine data files from the NetBackup
NFS datastore to the datastore that you selected.
Note: You should migrate no more than one restored virtual machine at a time
per media server.
6 After the migration is complete, use vSphere Client to merge or consolidate
the virtual machine's redo log (or snapshot) files manually. See your VMware
documentation for details.
When the migration to the production datastore is complete, use the following
steps to unmount the NFS datastore and release its resources.
7 Enter the following:
nbrestorevm –ir_listvm

In the -ir_listvm output, find the VM Instant Recovery ID for the restored VM.
8 When the data migration is complete, enter the following:
nbrestorevm –ir_done instant recovery ID

where instant recovery ID is the virtual machine's numeric identifier from the
-ir_listvm output.
The -ir_done option completes the VM Instant Recovery job. It also removes
the NetBackup NFS datastore if no other VM uses it. When the datastore is
removed, its resources are released on the media server.
See “Restoring individual files with Instant Recovery for VMware while the current
virtual machine is running” on page 257.

Restoring a virtual machine to a different location

with Instant Recovery for VMware
This topic explains how to use the nbrestorevm command to restore a VM to a
different location.
The overall instant recovery procedure is available in another topic:
See “Restoring a virtual machine with Instant Recovery for VMware” on page 250.
 Restore virtual machines with Instant Recovery 255
Restoring a virtual machine to a different location with Instant Recovery for VMware

To restore a VM to a different location

1 Find the path to a resource pool at the restore destination. (If you already know
the full path, you can skip this step.)
Note: To restore to a different location, it is usually necessary to designate a
different resource pool. Here is an example of a resource pool path:
/TechOffice/host/F2/pl9.acme.com/Resources

To find the path, enter the following on the master server, media server, or
restore host:
UNIX, Linux:
/usr/openv/netbackup/bin/bpVMreq <restore_host> 11 0 <ESXi_server>
<VMserver_or_vCenter_server>

Windows:
<install_path>\NetBackup\bin\bpVMreq.exe <restore_host> 11 0
<ESXi_server> <VMserver_or_vCenter_server>

Note: The numeric value 11 0 is required and must be entered as shown.

For example:
bpVMreq battleship.acme.com 11 0 ESXi_pl9.acme.com
vC_p9vm3.acme.com

Where battleship.acme.com is the restore host, ESXi_pl9.acme.com is the

destination ESXi server, and vC_p9vm3.acme.com is the destination vCenter
server.
This command generates a path to an XML file in a temporary location. The
XML file lists all the available resource pools.
Here is an example of an XML file that bpVMreq creates:

<?xml version="1.0" encoding="UTF-8" standalone="no" ?>

<ResourcePoolList><ResourcePool Name="Resources"
Path="/TechOffice/host/F2/pl9.acme.com/Resources"
Type="ResourcePool"><ResourcePoolList/>
</ResourcePool></ResourcePoolList>

In this example, the path to the resource pool is

/TechOffice/host/F2/pl9.acme.com/Resources. Make a note of the path
for use in the next step.
 Restore virtual machines with Instant Recovery 256
Restoring a virtual machine to a different location with Instant Recovery for VMware

2 Create a text file with the following change entries.

Note: Each change entry helps to define the location for the restore. Each
change line must end with a carriage return.

change vmname to <new_virtual_machine_name> (The change vmname entry is optional)

change esxhost to <new_ESXi_host>

change resourcepool to <path_to_new_resouce_pool>

change networkname to <new_network> (The change networkname entry is

optional)

Enter each change line exactly as it appears in this list, except for the variable
at the end (such as new_virtual_machine_name). Replace the variable with
the new name. For example:
change esxhost to ESXi01.prod4.com

For the new_resouce_pool, use the path that was obtained in the first step of
this procedure. For example:
change resourcepool to /TechOffice/host/F2/pl9.acme.com/Resources

This text file is called the -R rename file, and is used with the nbrestorevm
command in the next step.
3 To restore the VM using the -R rename file, enter the nbrestorevm command
with the -R option as follows.
Note: The -R option specifies the path to the text file (rename file).
■ To restore to the same vCenter server but to a different ESXi host, enter
the following:
nbrestorevm –vmw –ir_activate –C <virtual_machine>
-temp_location <temporary_datastore> -R <rename_file_path>

■ To restore to a different vCenter server and a different ESXi host, enter the
following:
nbrestorevm –vmw –ir_activate –C <virtual_machine>
-temp_location <temporary_datastore> -R <rename_file_path>
-vmserver <vCenter_server>

To restore a VM after Auto Image Replication (AIR) to a disaster recovery (DR)

site: you must also include the -vmproxy option on the nbrestorevm command
to specify the restore host at the DR site.
 Restore virtual machines with Instant Recovery 257
Restoring individual files with Instant Recovery for VMware while the current virtual machine is running

Restoring individual files with Instant Recovery

for VMware while the current virtual machine is
running
You can use instant recovery to restore files individually from a virtual machine
backup. You can restore the virtual machine from its backup image and mount it
on a private network (such as a sandbox). This approach avoids the potential for
network conflicts with the virtual machine in your production environment. Another
virtual machine on a public network can be used as an intermediary, to copy the
files from the virtual machine on the private network.

Note: This procedure lets you restore files into a running VM. You do not need to
shut down the current virtual machine in your production environment before you
start this procedure.

Before you start this procedure, you need an intermediary virtual machine that has
a network connection to the public network or production network. In this procedure
you connect the intermediary to the private network where the restored virtual
machine is to be mounted.
At the end of the procedure, you can copy files from the restored virtual machine
to the intermediary virtual machine. Then the virtual machines on the public network
can access the files on the intermediary.
To restore individual files using instant recovery
1 Use vSphere Client to log on to the vCenter server.
You must use a logon that allows access to the files that you want to recover.
2 Create a vSphere standard switch.
This switch is for access to the ESX host from the sandbox or private network
where the VM is to be activated from its backup.

Note: The switch is for internal communication within the ESX host only.

For example, in vSphere Client 5:

■ Select the ESX host for communication between the restored virtual machine
and the intermediary virtual machine.
■ On the Configuration tab, in the Hardware pane, click Networking.
■ Click Add Networking.
 Restore virtual machines with Instant Recovery 258
Restoring individual files with Instant Recovery for VMware while the current virtual machine is running

■ Select Virtual Machine as the connection type.

■ Select Create a vSphere standard switch.
■ For Port Group Properties, Network Label, enter a name for the internal
switch (such as NB or NetBackup).
■ Click Finish.

3 On the intermediary virtual machine, add a network card (NIC) to be connected

to the vSphere standard switch.
Use this connection to retrieve files from the restored virtual machine that is
to be mounted on the private network.

Note: This intermediary virtual machine must already have a network connection
to the public network or production network.

For example, in vSphere Client 5:

■ Select the intermediary virtual machine.
■ On the Summary tab, click Edit Settings.
■ Click Add.
■ Select Ethernet Adapter.
■ For the Network label, select the private network that is created in step 2.
■ Click Finish.

4 If the intermediary's guest OS does not automatically assign an IP address for

the private network after step 3, note: You must manually configure the IP
address, default gateway, and subnet mask.
The intermediary should now be connected to both the public network and to
the private network where the virtual machine is to be restored.
 Restore virtual machines with Instant Recovery 259
Restoring individual files with Instant Recovery for VMware while the current virtual machine is running

5 Use the nbrestorevm command to restore the virtual machine.

nbrestorevm -vmw -ir_activate -C virtual_machine -temp_location
temporary_datastore -R rename_file_path -vmsn

-vmsn specifies that no network is enabled for the virtual machine when it is
activated from the backup image. Without the -vmsn option, network conflicts
with the production virtual machine may occur.
The file that is designated by -R rename_file_path specifies a different display
name or location for the restored virtual machine. You must change the virtual
machine name or location to avoid conflicts with the current virtual machine in
production. For example, to rename the virtual machine, the rename file can
consist of the following entry (ending with a carriage return):
change vmname to acme_vm5

Note: The words change vmname to are literals, followed by the actual name
to change to (such as acme_vm5).

See “Instant Recovery options on the nbrestorevm command” on page 244.

For other nbrestorevm options, see the man page or the NetBackup Commands
Reference Guide.
6 Add a network card (NIC) to the restored virtual machine and connect the NIC
to the vSphere standard switch from step 2.
7 Turn on the restored virtual machine.
8 If the guest OS does not automatically assign an IP address for the private
network, configure the IP address, default gateway, and subnet mask.
 Restore virtual machines with Instant Recovery 260
Job types for Instant Recovery for VMware

9 Set up file sharing (such as through FTP, NFS, or CIFS) between the restored
virtual machine and the intermediary virtual machine.
Then copy the files from the restored virtual machine to the intermediary virtual
machine. The current virtual machine in production can access the files.
10 If you do not want to keep the restored virtual machine, enter the following:
nbrestorevm –ir_listvm

In the -ir_listvm output, find the VM Instant Recovery ID for the restored virtual
machine.
To remove the restored virtual machine:
nbrestorevm –ir_deactivate instant recovery ID

where instant recovery ID is the virtual machine's numeric identifier from the
-ir_listvm output.

Job types for Instant Recovery for VMware

Instant Recovery jobs appear as the following job types in the NetBackup Activity
Monitor.

Table 11-5 VMware Instant Recovery job types in the Activity Monitor

Job type Description

VM Instant Recovery This job is the parent job for restoring a VM by means of Instant
Recovery.

To complete this job, you must enter one of the following:

nbrestorevm -ir_done instant recovery ID

nbrestorevm -ir_deactivate instant recovery ID

For details on these commands:

See “Restoring a virtual machine with Instant Recovery for

VMware” on page 250.

Activate Instant Recovery The parent VM Instant Recovery job starts an Activate Instant
Recovery job to create the VM on the ESX host.

Stop Instant Recovery This job runs when you use nbrestorevm -ir_done to remove
the NetBackup NFS datastore and release its resources on the
media server.

Deactivate Instant This job runs when you use nbrestorevm -ir_deactivate
Recovery to delete the restored VM from the ESX host.
 Restore virtual machines with Instant Recovery 261
Reactivating a restored virtual machine with Instant Recovery for VMware

Table 11-5 VMware Instant Recovery job types in the Activity Monitor
(continued)

Job type Description

Reactivate Instant This job runs when you use nbrestorevm with the ir_reconfigure
Recovery option to restart an interrupted virtual machine recovery.

Reactivating a restored virtual machine with

Instant Recovery for VMware
If an interruption occurs during an instant recovery (such as a restart of the host or
media server), the ESX connection to the media server may fail. In that case, it may
be possible to re-establish the connection and return the virtual machine to the
state it was in before the outage. Any transactions that occurred in the virtual
machine before the outage are retained.
 Restore virtual machines with Instant Recovery 262
Reactivating a restored virtual machine with Instant Recovery for VMware

To reactivate a restored virtual machine

1 If only one VM had been restored to the ESX host, enter the following:
nbrestorevm –ir_listvm

Find the VM Instant Recovery ID for the restored VM in the -ir_listvm output.
Then enter the following:
nbrestorevm –ir_reactivate Instant Recovery ID [-force]

where instant recovery ID is the virtual machine's numeric identifier from the
-ir_listvm output. -force is an optional parameter to suppress confirmation
prompts.
The ir_reactivate option remounts the NetBackup NFS datastore. From the
temporary datastore on the ESX host it registers the restored virtual machines
on the ESX host.
2 If more than one VM had been restored to the ESX host:
nbrestorevm –ir_reactivate_all -vmhost vm_host –media_server
media_server [-force]

Note: For multiple virtual machines, do not use the –ir_reactivate option. Use
–ir_reactivate_all.
The -vmhost option specifies the ESX host on which the virtual machines were
mounted. The -media_server option specifies the media server on which the
NFS datastores that contain the backup images were mounted. -force is an
optional parameter to suppress confirmation prompts.
The nbrestorevm -ir_reactivate_all command remounts the NetBackup NFS
datastores on the media server and reactivates the virtual machines.
3 When the virtual machine is reactivated, you can copy its files or migrate its
data to the ESX host.
See “To restore a virtual machine with instant recovery” on page 251.
4 If Storage vMotion was migrating the virtual machine files when the outage
occurred, restart the migration.
In vSphere Client, right-click on the restored virtual machine and select Migrate.
 Chapter 12
Use NetBackup for vCloud
Director
This chapter includes the following topics:

■ About NetBackup for vCloud Director

■ Notes on creating a NetBackup policy for vCloud

■ Notes on restoring virtual machines into vCloud Director

■ Restoring virtual machines into vCloud Director

■ Reducing the time required for VM discovery in a large vCloud environment

■ Virtual machine recovery dialog boxes for vCloud Director

About NetBackup for vCloud Director

NetBackup can back up VMware vCloud Director environments and restore virtual
machines into vCloud Director.
Table 12-1 describes the configuration requirements for backup of vCloud Director.

Table 12-1 Configuration for backup of vCloud Director virtual machines

Task Description (NetBackup Administration Console)

Enter NetBackup Media and Device Management > Credentials > Virtual Machine Servers.
credentials for the
See “Adding NetBackup credentials for VMware” on page 45.
vCloud Director server
and for its vCenter
servers.
 Use NetBackup for vCloud Director 264
Notes on creating a NetBackup policy for vCloud

Table 12-1 Configuration for backup of vCloud Director virtual machines

(continued)

Task Description (NetBackup Administration Console)

Configure the policy Select the following:

Clients tab
■ Select automatically through VMware Intelligent Policy query
■ Enable vCloud Director integration

With the Enable vCloud Director option, the policy selects only vCloud-managed virtual
machines for backup: it skips the virtual machines that are not in vCloud.

NetBackup collects information on the vCloud environment, such as its organizations, virtual
datacenters, and vApps. NetBackup also retrieves information about a vApp for later restore
of the vApp and its virtual machines.
Note: Enable vCloud Director integration makes several vCloud keywords available in
the policy Query Builder Field, for rule-based selection of virtual machines. If Enable vCloud
Director integration is not selected, NetBackup cannot use the keywords to locate virtual
machines in vCloud Director and the backup fails.

Note: The browsing icon (next to the Query Builder drop-down fields) may list non-vCloud
objects. If you select an object that is not in vCloud Director, it is excluded from the backup.

See “Configuring a VMware policy from the Policies utility” on page 68.

See “Notes on creating a NetBackup policy for vCloud” on page 264.

Note: The NetBackup for VMware restrictions also apply to vCloud Director objects.
VMware restrictions also may apply; see your VMware documention.

See “NetBackup for VMware: notes and restrictions” on page 27.

Notes on creating a NetBackup policy for vCloud

When you create a backup policy for vCloud virtual machines, note the following:
■ The configuration requirements are described in the following topic:
See “About NetBackup for vCloud Director” on page 263.
■ To back up all the existing vApp templates, use the vCDIsvAppTemplate
keyword in the Query Builder (vCDIsvAppTemplate Equal TRUE).
■ To back up specific vApp templates, use the vCDvApp keyword in the Query
Builder with appropriate operator and values to select the particular templates.
 Use NetBackup for vCloud Director 265
Notes on restoring virtual machines into vCloud Director

■ To allow a policy to back up virtual machines from multiple vCloud Director

organizations: On the VMware tab of the policy, click Advanced, and enable
Multiple organizations per policy.

Notes on restoring virtual machines into vCloud

Director
To restore a virtual machine into vCloud Director, note the following:
■ The Backup, Archive, and Restore interface allows the restore of one virtual
machine at a time.
You can use the nbrestorevm command to restore multiple virtual machines.
See “Using the nbrestorevm command to restore virtual machines into vCloud
Director” on page 341.
■ When you back up a virtual machine in vCloud Director, it is recommended the
VMware display name setting for Primary VM identifier on the VMware tab.
Use of the VM BIOS UUID setting is not recommended.
■ To restore a virtual machine into vCloud Director, the virtual machine must have
been in vCloud Director when it was backed up.
To restore a virtual machine into vCloud Director, the backup policy must have
been configured as described in the following topic:
See “About NetBackup for vCloud Director” on page 263.
■ When a vApp is restored, vCloud Director resets the vApp's expiration date. For
example: Assume the original vApp was created on the first day of the month
and was set to expire in 30 days. If the vApp is restored 15 days before its
expiration (on the 15th), vCloud resets the vApp to expire in 30 days from the
15th.
The VM administrator can reset the expiration date to its original date.
■ You cannot restore a virtual machine into an existing vApp template. VMware
sets this restriction.
■ After you restore a vApp template, the template cannot be changed and no
further virtual machines can be added to it (a VMware restriction). To restore a
vApp template that is to contain multiple VMs, you must restore all but one of
the VMs separately into a non-template vApp. Then restore the last virtual
machine by means of the Capture vApp as a template in catalog option.
See “Restoring a vApp template that has multiple virtual machines” on page 270.
■ You can restore the vCloud Director virtual machine into vSphere instead of
vCloud Director. On the Recovery Destination screen, select Alternate location
in vSphere.
 Use NetBackup for vCloud Director 266
Restoring virtual machines into vCloud Director

■ vCloud organization networks are not displayed on the Network Connections

screen for restore; only vSphere networks are displayed.
■ To restore into an existing vCloud Director vApp with the Capture vApp as a
template in catalog option, the vApp must be turned off.
■ vCloud backup images cannot be restored by means of the NetBackup vSphere
Web Client plug-in or the NetBackup vSphere Client (HTML5) plug-in. This type
of restore can be performed by means of the Backup, Archive, and Restore
interface.
■ vSphere 5 provides storage profiles for managing the datastores on which virtual
machines are created. When you restore into vCloud, the NetBackup restore
interface lets you select a target datastore but not a storage profile. If the target
datastore does not comply with the storage profile or if the datastore you select
has been disabled, note: NetBackup automatically migrates the virtual machine
to a compliant datastore if possible when you import to vCloud Director.

Note: At the end of a successful restore, NetBackup verifies that the virtual
machine is on compliant storage (on a datastore with the defined storage profile).

■ To ensure that any VM guest customizations are restored into vCloud Director,
you must set a NetBackup parameter. The parameter value specifies a wait
period in seconds so that the guest customizations can be restored successfully.
(The VMware API requires that the VMware Tools are installed and running,
but the state of the VMware Tools cannot be identified after the restore.
Therefore, we wait the specified amount of time so that the VMware Tools are
running in the initial restore environment.)
See “Ensuring that guest customizations can be restored in vCloud Director”
on page 320.

Restoring virtual machines into vCloud Director

This topic includes two procedures:
■ Restoring virtual machines into a vApp, or restoring a single virtual machine into
a vApp template.
■ Restoring a vApp template that has multiple virtual machines.
Note: You can also restore the virtual machine to an alternate location in vSphere.
See “Recovery Destination dialog boxRecovery Destination dialog box (vCloud
Director)” on page 276.
 Use NetBackup for vCloud Director 267
Restoring virtual machines into vCloud Director

Restoring a single virtual machine into a vApp or vApp template

1 Begin with the steps in the following procedure:
See “Restoring the full VMware virtual machine” on page 179.
When you start the Restore, the Recovery Destination screen appears.
2 Select the recovery destination:

Original location in You can restore a virtual machine into an existing vApp or into
vCloud Director a new vApp that replaces the existing one. You can also restore
a virtual machine that is in a vApp template.
Note: The restored vApp template can only contain one virtual
machine; no further virtual machines can be added.

Alternate location You can restore a virtual machine into an existing vApp or into
in vCloud Director a new vApp or vApp template.

To restore multiple virtual machines into a vApp template:

See “Restoring a vApp template that has multiple virtual

machines” on page 270.

3 On the Recovery Options screen, select the NetBackup recovery host and
transport mode for the restore.
See “Recovery Options dialog box” on page 182.
 Use NetBackup for vCloud Director 268
Restoring virtual machines into vCloud Director

4 On the Recovery vApp Options for vCloud Director screen, set the vApp
recovery options.
You can either restore the virtual machine into an existing vApp (if that vApp
is not a template), or you can create a new vApp.

For restore to original Select Restore into existing vApp, or Remove existing vApp
location and recreate the vApp.

Caution: If you remove the existing vApp, all virtual machines

in the vApp are deleted. You must select the Delete option to
acknowledge the deletion.

See “Recovery vApp Options for vCloud Director dialog box

(restore to original location)” on page 277.

To restore a virtual machine in a vApp template, select Remove

existing vApp Template and recreate the vApp Template.
The restored vApp template can only contain one virtual
machine; no further virtual machines can be added.

To restore multiple virtual machines into a vApp template:

See “Restoring a vApp template that has multiple virtual

machines” on page 270.

For restore to Select Restore into existing vApp or Create new vApp.
alternate location
You can enter details for the vApp to restore to.

See “Recovery vApp Options for vCloud Director dialog box

(restore to alternate location)” on page 279.
Note: To restore the virtual machine into a vApp template that
is to contain this virtual machine only, select Create new vApp
and Create vApp as a template in catalog.

To restore the virtual machine into a vApp template that is to

contain multiple virtual machines:

See “Restoring a vApp template that has multiple virtual

machines” on page 270.

5 For restore to an alternate location: On the Recovery Destination Options

for vCloud Director screen, select the ESX server, the virtual machine that
you want to restore, and a datastore.
See “Recovery Destination Options for vCloud Director dialog box (restore to
alternate location)” on page 281.
 Use NetBackup for vCloud Director 269
Restoring virtual machines into vCloud Director

6 On the Virtual Machine Options screen, select the appropriate options for
the virtual machine and its disk provisioning.
See “Virtual Machine Options dialog box” on page 184.
7 On the Network Connections screen, select the network for the restored
virtual machine.

Note: If the virtual machine was not connected to any networks during backup,
or if none of those networks are currently available, the dialog states: "No
networks available for selected virtual machine." If a network connection is
required, create it in vCloud Director after the restore completes.

Note: For restore to an alternate location (a different organization or vCloud

server), no networks are displayed in the dialog. If a network connection is
required, create it in vCloud Director after the restore completes.

See “Network Connections and Other Recovery Options dialog box (restore
to alternate location)” on page 195.
8 On the Perform Recovery screen, run a pre-recovery check.
To begin the restore click Start Recovery.
9 To restore additional virtual machines into the same vApp (not a vApp template),
repeat this procedure.
Note: On the Recovery vApp Options for vCloud Director screen, select
Restore into existing vApp. Make sure to select the vApp into which you
want to restore the additional virtual machines.
 Use NetBackup for vCloud Director 270
Restoring virtual machines into vCloud Director

Restoring a vApp template that has multiple virtual machines

1 Use the Backup, Archive, and Restore interface to restore all but one of the
virtual machines into a non-template vApp.
You can restore one virtual machine at a time. After you have restored the first
virtual machine, restore the second virtual machine with Restore into existing
vApp on the Recovery vApp Options for vCloud Director screen. Select
the same vApp into which you restored the first virtual machine. Step through
the restore screens to restore each virtual machine in this way, except for the
last virtual machine.

Note: Use the following steps to restore the last virtual machine and to copy
all the restored virtual machines into a vApp template.

2 On the Recovery Destination screen, select Alternate location in vCloud

Director.
3 On the Recovery vApp Options for vCloud Director screen, do the following:
■ Select Restore into existing vApp.
■ Select the vCloud server and the organization that includes the vApp into
which you have restored the other virtual machines.
■ If necessary, browse for the vApp into which you have restored the other
virtual machines.
■ Select Capture vApp as a template in catalog.
■ Select the catalog to contain the template vApp.
Note: The organization determines the available catalogs.
■ Enter a name for the vApp template.
■ Remove vApp after capture: Deletes the non-template vApp into which
you restored the other virtual machines at the beginning of this procedure.
Use this option to free up space on the datastore after the restore is
complete.
See “Recovery vApp Options for vCloud Director dialog box (restore to
alternate location)” on page 279.

4 On the Recovery Destination Options for vCloud Director screen, select

the last virtual machine that you want to restore into the template vApp.
See “Recovery Destination Options for vCloud Director dialog box (restore to
alternate location)” on page 281.
 Use NetBackup for vCloud Director 271
Reducing the time required for VM discovery in a large vCloud environment

5 On the Virtual Machine Options screen, select the appropriate options for
the virtual machine and its disk provisioning.
6 On the Network Connections screen, select the network for the restored
virtual machine.
See “Network Connections and Other Recovery Options dialog box (restore
to alternate location)” on page 195.
7 On the Perform Recovery screen, run a pre-recovery check.
To begin the restore click Start Recovery.
NetBackup copies the current virtual machine and the previously restored
virtual machines into a new vApp template. When the restore is complete, no
further virtual machines can be added to the template vApp.

Reducing the time required for VM discovery in a

large vCloud environment
NetBackup backup policies for vCloud Director use query rules to automatically
search and filter the vCloud environment. By default, the query rules search all the
vCloud Director (vCD) servers in your environment. If the environment contains
many vCloud servers with many vApps, VM discovery may take a long time. You
can speed up VM discovery by limiting the search to specific vCloud servers or
vApps.
The following is an example of a policy Query Builder rule that searches all vCloud
servers and all vApps:

vmware:/?filter=vCDvApp Contains "vapp1"

To limit the search to particular servers or vApps, insert an additional vCloud

expression in the Query Builder rule as explained in the following procedure.
To use the Administration Console to reduce the time required for VM
discovery in a vCloud environment
1 In the NetBackup Administration Console, open the vCloud Director policy.
2 On the Clients tab of the policy, make sure Select automatically through
VMware Intelligent Policy query and Enable vCloud Director integration
are selected.
3 In the Query Builder, click Advanced.
4 Under Query (Advanced Mode), create one or more rules to search for VMs
in specific vCloud Director servers or vApps.
 Use NetBackup for vCloud Director 272
Reducing the time required for VM discovery in a large vCloud environment

To create two or more rules, you must be in Advanced Mode (not Basic
Mode).
Each query rule must begin on its own line.
You can use the following types of rules:
■ To search for VMs in a particular vCloud server
vmware://<vCloud_server>?filter=<filter>

■ To search for VMs in a particular vApp or vApp template

vmware:/vApp/vapp-<vApp_id>?filter=<filter>
vmware:/vAppTemplate/vappTemplate-<vAppTemplate_id>?filter=<filter>

■ To search for VMs in a particular vApp or vApp template on a particular

vCloud server

vmware://<vCloud_server>/vApp/vapp-<vApp_id>?filter=<filter>

vmware://<vCloud_server>/vAppTemplate/vappTemplate-
<vAppTemplate_id>?filter=<filter>

The following detail from a Query Builder screen (Advanced Mode) shows
two rules for narrowing the search within a vCloud environment. The first rule
searches a single vApp on vCloud server vCD1.acme.com. The second rule
searches a different vApp on vCloud server vCD2.acme.com. For help in finding
the vApp identifiers, see the following topic:
Example Query Builder rules for searching specific vCloud servers or vApps
 Use NetBackup for vCloud Director 273
Reducing the time required for VM discovery in a large vCloud environment

5 For two or more search rules, you must enable multiple organizations for the
policy.
■ In the policy VMware tab, click Advanced.
■ In the VMware - Advanced Attributes dialog, enable Multiple
organizations per policy.
■ To use the nbdiscover command instead of the policy Query Builder, see
the following topic:
Examples of the nbdiscover command for searching specific vCloud servers
or vApps

Example Query Builder rules for searching specific vCloud

servers or vApps
In the NetBackup policy Query Builder, you can speed up discovery of vCloud VMs
by using the following types of query rules:
■ To search for VMs in a particular vCloud server (note the double forward slash):
vmware://<vCloud_server>?filter=<filter>
Example rule:
vmware://vCD1.acme.com?filter=vCDvApp Contains "vapp1"
NetBackup searches for VMs only in the vCD1.acme.com server.
 Use NetBackup for vCloud Director 274
Reducing the time required for VM discovery in a large vCloud environment

■ To search for VMs in a particular vApp or vApp template (note the single forward
slash):
vmware:/vApp/vapp-<vApp_id>?filter=<filter>
vmware:/vAppTemplate/vappTemplate-<vAppTemplate_id>?filter=<filter>
The vApp_id or vAppTemplate_id is the identifier on the end of the vCloud vApp
href. You can use a vCloud Director REST API query to find the identifier.
For example, the following is a REST API query for a vApp that is named
acmvappvm7:

https://acmvm5.acme.com/api/query?type=adminVApp&filter=
name==acmvappvm7

The following example is an excerpt from the API query Response:

href="https://acmvm5.acme.com/api/vApp/vapp-afaafb99
-228c-4838-ad07-5bf3aa649d42"

In this example, the vApp identifier for vApp acmvappvm7 is

afaafb99-228c-4838-ad07-5bf3aa649d42. You can use this identifier in a
NetBackup Query Builder rule as follows:

vmware:/vApp/vapp-afaafb99-228c-4838-ad07-5bf3aa649d42
?filter=Displayname Contains "prod"

Where Displayname Contains "prod" is an example filter for the rule.

■ To search for VMs in a particular vApp on a particular vCloud Director server:
vmware://<vCloud_server>/vApp/vapp-<vApp_id>?filter=<filter>
Example rule:

vmware://vCD1.acme.com/vApp/vapp-4c0d9722-80a4-4f19-b636-72ebf48e4e71
?filter=Displayname Contains "prod"

■ To search additional vCloud Director servers or vApps from the same backup
policy, include additional query rules in the Query Builder.

Note: To enter multiple rules in the Query Builder, you must be in Advanced
Mode.

Note: Start each rule on its own line.

Example of two rules in the Query Builder:

 Use NetBackup for vCloud Director 275
Reducing the time required for VM discovery in a large vCloud environment

vmware://vCD1.acme.com/vApp/vapp-4c0d9722-80a4-4f19-b636-72ebf48e4e71
?filter=Displayname Contains "prod"
vmware://vCD2.acme.com/vApp/vapp-5c0c9833-80a4-4f19-b636-72ebf48e4e63
?filter=Displayname Contains "prod"

Examples of the nbdiscover command for searching

specific vCloud servers or vApps
Use the following nbdiscover command format to search for VMs in specific vCloud
servers or vCloud vApps:
nbdiscover “<vCloud_query>” –job_info “snaparg=enable_vCloud=1”

Example 1. Search for VMs in vCloud server vCD1.acme.com only:

nbdiscover -noxmloutput "vmware://vCD1.acme.com?filter=DisplayName

Contains 'prod1'" –job_info “snaparg=enable_vCloud=1”

The -noxmloutput option displays one VM per line.

Example 2. Search for VMs in a vCloud vApp that has the following vApp identifier:
4c0d9722-80a4-4f19-b636-72ebf48e4e71

nbdiscover -noxmloutput "vmware:/vApp/vapp-4c0d9722-80a4-4f19-b636

-72ebf48e4e71?filter=DisplayName Contains 'prod1'" –job_info
“snaparg=enable_vCloud=1”

The following topic provides assistance in finding the vApp identifier:

See the section called “Example Query Builder rules for searching specific vCloud
servers or vApps” on page 273.
Example 3. Use two query rules to search for VMs prod1 and prod2 in a vCloud
vApp:

nbdiscover -noxmloutput "vmware:/vApp/vapp-4c0d9722-80a4-4f19-b636

-72ebf48e4e71?filter=DisplayName Contains 'prod1'" "vmware:/vApp/
vapp-4c0d9722-80a4-4f19-b636-72ebf48e4e71?filter=DisplayName Contains
'prod2'" –job_info “snaparg=enable_vCloud=1,multi_org=1”

Note the two rules, each enclosed with double quotes and separated by a space,
and the multi_org=1 option.
See “Notes on creating a VMware policy for vCloud Director” on page 330.
 Use NetBackup for vCloud Director 276
Virtual machine recovery dialog boxes for vCloud Director

Virtual machine recovery dialog boxes for vCloud

Director
The following topics describe the Backup, Archive, and Restore screens for restoring
a virtual machine into vCloud Director.
You can restore a virtual machine into an existing vApp or into a new vApp.
See “Recovery Destination dialog boxRecovery Destination dialog box (vCloud
Director)” on page 276.
See “Recovery vApp Options for vCloud Director dialog box (restore to original
location)” on page 277.
See “Virtual Machine Options dialog box for vCloud Director” on page 278.
See “Perform Recovery dialog box for vCloud Director” on page 278.
See “Recovery vApp Options for vCloud Director dialog box (restore to alternate
location)” on page 279.
See “Recovery Destination Options for vCloud Director dialog box (restore to
alternate location)” on page 281.

Recovery Destination dialog boxRecovery Destination dialog box

(vCloud Director)
You can select the type of destination for the virtual machine recovery: the original
location in vCloud, an alternate location in vCloud, or an alternate location in
vSphere.
■ Original settings of the virtual machine at backup
Shows the configuration of the virtual machine at the time of the backup.
When you restore the virtual machine to its original location, these configuration
settings are used and cannot be changed.
When you restore to an alternate location, you can change these settings after
you click Next.
■ vApp details
Shows the information about the virtual machine's vApp at the time of the backup.

vApp Name The name of the vApp that contained the virtual machine when
it was backed up.

vApp Status The status of the vApp (active or expired) at the time of the
backup. When the virtual machine is restored, its vApp becomes
active even if the current vApp in the restore location is expired.
 Use NetBackup for vCloud Director 277
Virtual machine recovery dialog boxes for vCloud Director

vCloud Catalog The name of the catalog that contains the vApp template. "..."
means that the vApp is not a template.

Other virtual A link to the names of the other virtual machines in the vApp, if
machines in vApp any.

■ Select Recovery Destination

You can restore to the original location in vCloud, to an alternate location in
vCloud, or to an alternate location in vSphere (not in vCloud).
See “Recovery Options dialog box (vCloud Director)” on page 277.

Recovery Options dialog box (vCloud Director)

Select the recovery host and transport mode. For a description of these options,
refer to the following topic:
See “Recovery Options dialog box” on page 182.
See “Recovery vApp Options for vCloud Director dialog box (restore to original
location)” on page 277.

Recovery vApp Options for vCloud Director dialog box (restore to

original location)
You can choose whether to restore the virtual machine into an existing vApp or to
remove the vApp and recreate it.

Restore into existing To restore the virtual machine into an existing vApp, select
vApp Overwrite the existing virtual machine.

If the existing virtual machine is already turned on, this option fails.
Note: If you restore a vApp template, this option is not available.

Remove existing For a vApp template, this option is labeled Remove existing vApp
vApp and recreate the Template and recreate the vApp Template
vApp
To remove the vApp (or vApp template) and recreate it, select
Delete all virtual machines to recreate this vApp.
Note: All virtual machines in the vApp are deleted.

Click on the numeric link to see the names of the virtual machines
that exist in the vApp.
 Use NetBackup for vCloud Director 278
Virtual machine recovery dialog boxes for vCloud Director

Virtual Machine Options dialog box for vCloud Director

This dialog is described in a previous topic:
See “Virtual Machine Options dialog box” on page 184.

Perform Recovery dialog box for vCloud Director

This topic applies to NetBackup for VMware. It describes how to restore a VMware
virtual machine.
Review the settings to be applied to the restored virtual machine.
■ Recovery settings
Lists your recovery selections, including the vCloud Director and vSphere
settings.
■ Run Pre-Recovery Check
Verifies the credentials and appropriate paths and connectivity, determines
whether the datastore or datastore cluster has available space, and reviews
other important requirements. You must run this check at least once. You can
proceed with the restore even if the check fails.
Table 12-2 describes the pre-recovery checks that are unique to vCloud Director.
The standard vSphere pre-recovery checks are described in another topic:
See “Perform Recovery dialog box” on page 187.
■ Start Recovery
Starts the recovery of the virtual machine

Table 12-2 Pre-recovery checks for vCloud Director

Type of check Description

vCloud server credential Validates the vCloud Director server credentials.

found

Verify login to vCloud Director Verifies that the NetBackup restore host can log on to the
server vCloud Director server.

Verify vCloud Organization Validates the selected vCloud Director organization to restore
to.

Verify vCloud ORGvDC Validates the selected organization virtual datacenter to

restore to in vCloud Director.

Verify template catalog If the option to capture the vApp as a template is selected,
validates the selected vCloud Director catalog to restore to.
 Use NetBackup for vCloud Director 279
Virtual machine recovery dialog boxes for vCloud Director

Table 12-2 Pre-recovery checks for vCloud Director (continued)

Type of check Description

Verify vApp powered off If the options to restore into an existing vApp and the capture
before capturing to template vApp as template are selected, verifies that the vApp is
powered off.

Recovery Options dialog box for vCloud Director (restore to alternate

location)
Select the recovery host and transport mode. For a description of these options,
refer to the following topic:
See “Recovery Options dialog box” on page 182.

Recovery vApp Options for vCloud Director dialog box (restore to

alternate location)
You can restore the virtual machine into an existing vApp (Restore into existing
vApp) or into a new vApp (Create a new vApp).
Table 12-3 describes the fields for specifying the vApp to restore to.

Note: The defaults are the server name, organization, and other details that were
recorded in the backup.

Table 12-3 Details for the vApp to be restored to

Field Description

vCloud Server Select the vCloud Director server for the restore.

The servers in the list are the ones for which NetBackup has
credentials of type VMware vCloud Director.

See “Adding NetBackup credentials for VMware” on page 45.

Organization Select the vCloud Director organization for the restore.

Org vDC Select the organization virtual datacenter.

 Use NetBackup for vCloud Director 280
Virtual machine recovery dialog boxes for vCloud Director

Table 12-3 Details for the vApp to be restored to (continued)

Field Description

vApp name Select or enter the name of the vApp for the restore.

If you restore into an existing vApp, click Browse to search for

a vApp.

See “Select a vApp dialog box” on page 280.

This option defaults to the original vApp that was recorded in the
backup.

Capture vApp as a This option is available when you restore to an existing vApp.
template in catalog
This option copies the vApp that is specified in the vApp Name
field into a new vApp template.

It also copies all of the vApp's virtual machines into the same
vApp template. Note that this operation may take a lot of time.

By default, the source vApp for the copy is retained after the copy;
you can have the source vApp removed after the copy. See the
Remove vApp after capture option.

Create vApp as a This option is available when you create a new vApp for the
template in catalog restore.

This option restores the virtual machine into a new vApp template.
The name of the new template is the same as the vApp that was
specified in the vApp Name field.

Catalog Select the catalog in which to place the vApp template. The
organization determines the available catalogs.

vApp template name When you restore into an existing vApp and capture the vApp as
a template:

Enter the name for the new vApp template.

Remove vApp after When you restore into an existing vApp and capture the vApp as
capture a template:

Select this option to remove the source vApp after the vApp copy
to the new vApp template completes. At the end of the copy, the
new template and its virtual machines are retained. The vApp
that was the source for the copy is deleted.

Select a vApp dialog box

Use this dialog to search the organization for a vApp.
 Use NetBackup for vCloud Director 281
Virtual machine recovery dialog boxes for vCloud Director

Specify vApp Enter the vApp name. As you type, the organization is expanded and
name the first name that contains the typed character(s) is highlighted.

Wildcards are not allowed.

Click Search to find further instances of the name.

vApps in You can drill into the organization and highlight a vApp.
Organization

Click OK to select the highlighted vApp.

Recovery Destination Options for vCloud Director dialog box (restore

to alternate location)
Specify details about the virtual machine to restore.
The vCloud options that you selected in the Recovery vApp Options for vCloud
Director dialog determine the vCenter server, resource pool, and datacenter.

ESX server Select the ESX server in which to restore the virtual machine.

Virtual machine name Enter the name for the restored virtual machine.

Datastore/ Datastore Specify manually: Select a datastore from the pull-down.

cluster
Or click Pick the datastore with the largest space available.

Virtual Machine Options (restore to alternate location)

This dialog is described in a previous topic:
See “Virtual Machine Options dialog box” on page 184.

Network connections (restore to alternate location)

The following topic describes the network information that appears on this dialog:
See “Network Connections and Other Recovery Options dialog box (restore to
alternate location)” on page 195.

Perform Recovery dialog box for vCloud Director (restore to alternate

location)
■ Recovery settings
Lists your recovery selections, including the vCloud Director and vSphere
settings.
 Use NetBackup for vCloud Director 282
Virtual machine recovery dialog boxes for vCloud Director

■ Run Pre-Recovery Check

Verifies the credentials and appropriate paths and connectivity, determines
whether the datastore or datastore cluster has available space, and reviews
other important requirements. You must run this check at least once. You can
proceed with the restore even if the check fails. More details are available on
the Pre-Recovery Check:
See “Perform Recovery dialog box for vCloud Director” on page 278.
See “Perform Recovery dialog box” on page 187.
■ Start Recovery
Starts the recovery of the virtual machine.
 Chapter 13
Best practices and more
information
This chapter includes the following topics:

■ NetBackup for VMware best practices

■ Best practices for VMware tag usage

■ Further assistance with NetBackup for VMware

NetBackup for VMware best practices

The following are best practices for NetBackup for VMware:
■ For a more efficient backup, the NetBackup media server and the VMware
backup host should be installed on the same host.
■ When creating virtual machines, use the same name for both host name and
display name. If the policy's Primary VM identifier option is changed, the existing
entries on the policy Clients tab still work.
■ VMware recommends that you run no more than four simultaneous backups of
virtual machines that reside on the same datastore.
■ Successful VMware snapshots depend on the following:
■ The amount of I/O that occurs on the virtual machine datastore. Backups
should be scheduled when relatively little I/O activity is expected. Reducing
the number of simultaneous backups can also help.
Limit access to the datastore per policy: Use the Limit jobs per policy
attribute in the NetBackup policy.
Limit access to the datastore globally (across all policies): Use the Host
Properties Resource Limit screen.
 Best practices and more information 284
NetBackup for VMware best practices

See “Setting global limits on the use of VMware resources” on page 52.
■ The design of the I/O substructure that is associated with each virtual machine
datastore. For correct I/O design and implementation, consult your VMware
documentation.

■ Make sure that the VMware backup host has enough memory to handle the
number of simultaneous backups that occur.
■ Include in a single NetBackup policy those virtual machines that use the same
datastore. This practice lets you control the amount of backup-related I/O that
occurs per datastore, to limit the backup effect on the target virtual machines.
■ NetBackup supports multiple backup hosts. When a single backup host is
saturated with a backup process, another backup host can be added to increase
backup throughput.
■ If a VM's disks are accessible to multiple ESX hosts, the disks can be accessed
for backup or restore through any of the ESX hosts. The ESX host may or may
not be the ESX host where the virtual machine is running or registered. All of
the following must be accessible to each other and should have DNS configured:
■ The vCenter server.
■ All ESX hosts under the vCenter that have access to the VM's vmdk files.
■ The backup host.

■ Upgrade to the latest version of VMware vSphere or Virtual Infrastructure.

See “NetBackup for VMware with deduplication” on page 284.

NetBackup for VMware with deduplication

For a VMware backup to a deduplication storage unit, select the Enable file
recovery from VM backup option on the VMware policy tab. This option provides
the best deduplication rates. Without the Enable file recovery from VM backup
option, the result is a lower rate of deduplication.
More information is available on the VMware options.
See “Backup options on the VMware tabVMware backup options” on page 73.
See “Configuring a VMware policy from the Policies utility” on page 68.
See “NetBackup for VMware best practices” on page 283.
 Best practices and more information 285
Best practices for VMware tag usage

Best practices for VMware tag usage

Use unique tag names whenever possible
The combination of Category and Tag uniquely identifies a tag association to a
virtual machine. Veritas, however, does not support selection of virtual machines
based on Category. As such, avoid the creation of tags with identical names in
different Categories.
When you create a VMware Intelligent Policy and select virtual machines based on
tags, identical tag names in different categories may have unintended consequences.
All virtual machines with the tag name are selected independent of the Category.
Be aware of this behavior as you create your VMware Intelligent Policies.

Use a master server or a media server as a discovery host

for VMware Intelligent Policies that use tags
Because tag queries require Java, the discovery host used for VMware Intelligent
policies must have Java installed. NetBackup master servers have Java installed
by default. Java is an optional component for NetBackup media servers and for
UNIX and Linux clients, and may or may not be installed in your environment.
Windows clients that are used as a discovery host require a separate installation
of the Remote Console.

Successful use of tags in a mixed vCenter Server

environment
NetBackup support for VMware tags begins at vCenter 6.0. In a mixed vCenter
Server environment (e.g. 5.x and 6.0), a VMware Intelligent Policy query that uses
the Tag keyword can return some virtual machines as Failed if the configuration
of the query requires tag evaluation on a 5.x vCenter. This behavior is observed
when you select Test Query during policy configuration. This behavior is also
observed when you run the policy. The parent (discovery) job exits with a non-zero
status and its details enumerate the virtual machines that Failed. The virtual
machines that the query has included are backed up normally.
You can use any of the options that are shown to create a query that uses tags but
does not report virtual machines on 5.x servers as Failed.
■ Use another field to limit tag evaluation to the supported vCenter Server versions.
Example: vCenterVersion GreaterEqual "6.0.0" AND Tag Equal
"Production"
Example: vCenter Equal "vcenter-123" AND tag Equal "Production"
■ Use the VMware Server List option under the Advanced Attributes on the
VMware tab to restrict the policy to a supported list of vCenter Servers.
See Table 5-7 on page 81.
 Best practices and more information 286
Further assistance with NetBackup for VMware

■ Enable the Treat Tags as unset if unable to evaluate option under the
Advanced Attributes on the VMware tab.
See Table 5-7 on page 81.

Place tag clause at the end of the VMware Intelligent Policy

query
Place the tag clause toward the end of the VMware Intelligent Policy query. This
configuration allows NetBackup to eliminate as many virtual machines as possible
before it evaluates the tag portion of the query. This organization optimizes the
performance of discovery and selection of virtual machines.
Consider the query: Powerstate Equal poweredOn AND Tag Equal "Production"
In the example, the first clause automatically eliminates all virtual machines that
are poweredOff. The query does not need to evaluate the tag clause of the query
for all of those virtual machines.

Tag backup and restore are best effort

Tag associations are part of the metadata of the virtual machine. NetBackup
considers virtual machine tag association protection a best effort backup. Any tag
collection errors are shown in the Activity Monitor for the virtual machine snapshot
job.

Further assistance with NetBackup for VMware

Table 13-1 Sources of information on NetBackup for VMware

Topic Source

Snapshot Client configuration NetBackup Snapshot Client Administrator's Guide.

List of supported combinations of NetBackupSnapshot Client Compatibility List

platforms and snapshot methods

Support information on NetBackup NetBackup Software Compatibility List

for VMware
 Chapter 14
Troubleshooting
This chapter includes the following topics:

■ Notes on troubleshooting NetBackup for VMware

■ Beware of unsupported backup host

■ How to determine the ESX network that NetBackup used for the backup or
restore

■ NetBackup logging for VMware

■ Configuring VxMS logging

■ Format of the VxMS core.log and provider.log file names

■ Configuring the VDDK logging level

■ Preventing browsing delays caused by DNS problems

■ Changing the browsing timeout for virtual machine discovery

■ Changing timeout and logging values for vSphere

■ Credentials for VMware server are not valid

■ Snapshot error encountered (status code 156)

■ The origin of the snapshot failure: NetBackup or VMware?

■ Conflict between NetBackup and VMware Storage vMotion with vSphere 5.0 or
later

■ Incremental backups of the VM may fail if the NTFS file system in the guest OS
is corrupted and the backup uses MSDP storage

■ The restore fails when you restore individual files to a virtual machine that has
NetBackup client software
 Troubleshooting 288
Notes on troubleshooting NetBackup for VMware

■ Backup or restore job hangs

■ VMware SCSI requirements for application quiesce on Windows

■ Mount point missing on a restored Windows virtual machine

■ Mount points not available when restoring files from a Linux virtual machine

■ Invalid client error when you restore files using NetBackup BAR interface installed
on the virtual machine

■ VMware virtual machine does not restart after restore

■ A restored VM may not start or its file system(s) may not be accessible

■ NetBackup job fails due to update tasks on the VMware server

■ The vSphere interface reports that virtual machine consolidation is needed

■ The Enable file recovery from VM backup option does not work if a volume disk
set contains a mixture of disk types

■ Linux VMs and persistent device naming

■ For a VMware virtual machine with Windows dynamic disks, a restore from
incremental backup fails with a Windows restore host and the hotadd transport
mode

■ Simultaneous hotadd backups (from the same VMware backup host) fail with
status 13

■ Troubleshooting VMware tag usage

■ Ensuring that guest customizations can be restored in vCloud Director

■ Troubleshooting vmdk restore to existing VM

■ Errors for backups of virtual machines on Virtual Volumes (VVols)

■ Issues with the CA certificate during installation of the NetBackup client on

VMware Cloud (VMC)

Notes on troubleshooting NetBackup for VMware

Note the following:
■ Make sure that the VMware Tools are installed and up to date on each virtual
machine. Otherwise, communication problems and other issues can result. For
example:
 Troubleshooting 289
Beware of unsupported backup host

■ When you select virtual machines on the policy Clients tab, NetBackup
cannot obtain the host name, IP address, or DNS name of the virtual machine.
Policy validation may fail.
■ The virtual machine cannot be quiesced in preparation for the snapshot.

■ You cannot restore individual VMware files onto the virtual machine itself, except
under certain conditions.
See “About restore of individual filesAbout restoring individual VMware files”
on page 223.
See “Configuring a VMware policy from the Policies utility” on page 68.

Beware of unsupported backup host

This topic applies to NetBackup for VMware.
Ensure that NetBackup supports your VMware backup host. For supported backup
host platforms, see the NetBackup Enterprise Server and Server OS Software
Compatibility List available from the following location:
NetBackup Master Compatibility List
For further VMware documentation on the backup host, see the information on
backup proxy in the following VMware article:
Best practices when using advanced transport for backup and restore
See “Notes on troubleshooting NetBackup for VMware” on page 288.

How to determine the ESX network that NetBackup

used for the backup or restore
If a virtual machine's disks are accessible to multiple ESX hosts, the disks can be
accessed through any of the ESX hosts. The ESX host that is used for the access
may or may not be the ESX host where the virtual machine is running or registered.
All of the following must be accessible to each other and should have DNS
configured:
■ The vCenter server.
■ All ESX hosts under the vCenter that have access to the virtual machine's vmdk
files.
■ The backup host.
 Troubleshooting 290
How to determine the ESX network that NetBackup used for the backup or restore

If all hosts are not accessible to each other, the backup or restore may not succeed.
In that case, you must determine which network NetBackup used for the backup
or restore.
Note: For an NBD transport mode backup through vCenter, NetBackup uses the
ESX network over which the ESX host was added or registered to the vCenter. For
an NBD transport mode backup directly from the ESX host, NetBackup uses the
ESX host’s DNS/IP network.
The VxMS provider logs contain information on the network that NetBackup used.
See “Configuring VxMS logging” on page 292.
Check the VxMS provider logs for messages similar to those in this example:

10:49:21.0926 : g_vixInterfaceLogger:libvix.cpp:1811 <INFO> : Opening file

[MYDATASTORE] TestVM/TestVM-000001.vmdk (vpxa-nfc://[MYDATASTORE]
TestVM/[email protected]:902)

10:49:22.0301 : g_vixInterfaceLogger:libvix.cpp:1811 <INFO> : DISKLIB-LINK :

Opened 'vpxa-nfc://[MYDATASTORE]
TestVM/[email protected]:902' (0x1e): custom, 41943040
sectors / 20 GB.

10:49:22.0301 : g_vixInterfaceLogger:libvix.cpp:1811 <INFO> : DISKLIB-LIB :

Opened "vpxa-nfc://[MYDATASTORE]
TestVM/[email protected]:902" (flags 0x1e, type custom).

10:49:22.0301 : vdOpen:VixInterface.cpp:480 <DEBUG> : Done with

VixDiskLib_Open(): 200346144
10:49:22.0301 : openLeafSnapshotDisks:VixGuest.cpp:475 <DEBUG> : vdOpen()
succeess
10:49:22.0301 : openLeafSnapshotDisks:VixGuest.cpp:476 <INFO> : Transport
mode in effect = nbd

VMware logs the messages starting with g_vixInterfaceLogger. Such messages

in the example indicate that TestVM-000001.vmdk is opened over the ESX host
network MyESX.xxx.xxx.com.
The following article contains related information:
■ Best practices when using advanced transport for backup and restore
http://kb.vmware.com/kb/1035096
 Troubleshooting 291
NetBackup logging for VMware

NetBackup logging for VMware

For log messages about VMware backup or VMware restore, see the following
NetBackup log directories.

Table 14-1 NetBackup logs that pertain to VMware backup and restore

Log directory Contains the Resides on

messages on

install_path\NetBackup\logs\bpbrm Backup and restore NetBackup master or media server

install_path\NetBackup\logs\bptm Backup and restore NetBackup media server

install_path\NetBackup\logs\bpfis Snapshot creation and VMware backup host

backup

install_path\NetBackup\logs\bpcd Snapshot creation and VMware backup host

backup

install_path\NetBackup\logs\bpbkar Backup VMware backup host

install_path\NetBackup\logs\bprd Restore NetBackup master server

install_path\NetBackup\logs\nbfsd Instant recovery NetBackup media server and VMware

backup host

install_path\NetBackup\logs\bpVMutil Policy configuration and on VMware backup or recovery host

restore

install_path\NetBackup\logs\bpVMreq Restore The client where the Backup, Archive,

and Restore interface is running.

install_path\NetBackup\logs\nbproxy Policy configuration VMware backup host

install_path\NetBackup\logs\ncfnbcs (originator Automatic virtual machine VMware backup host

ID 366) selection, and disabling and
re-enabling VMware
ncfnbcs uses unified logging. See the NetBackup
Storage vMotion during
Logging Reference Guide for information on how
backup or restore.
to use unified logs.

install_path\NetBackup\logs\ncfnbrestore Restore VMware recovery host

(originator ID 357)

ncfnbrestore uses unified logging. See the

NetBackup Logging Reference Guide for
information on how to use unified logs.
 Troubleshooting 292
Configuring VxMS logging

Table 14-1 NetBackup logs that pertain to VMware backup and restore
(continued)

Log directory Contains the Resides on

messages on

Windows: File mapping during backup VMware backup host

install_path\NetBackup\logs\vxms See “Configuring VxMS logging”

on page 292.
Linux:
Note: The use of VxMS logging can
/usr/openv/netbackup/logs/vxms
reduce the performance of the backup
host.

Note: Except for unified logging directories, these log directories must already exist
in order for logging to occur. If these directories do not exist, create them.

To create most of these log directories, run the following command on the NetBackup
servers and backup host:
Windows:
install_path\NetBackup\logs\mklogdir.bat

UNIX (on master or media servers):

/opt/openv/netbackup/logs/mklogdir

See “Configuring VxMS logging” on page 292.

More detail is available on snapshot logs and logging levels. See the NetBackup
Snapshot Client Administrator’s Guide.
A broader discussion of NetBackup logging is available. See the NetBackup Logging
Reference Guide.
See “Notes on troubleshooting NetBackup for VMware” on page 288.

Configuring VxMS logging

The following procedures describe how to configure VxMS logging for NetBackup.
Except as noted in this topic, you can also use the Logging Assistant (in the
NetBackup Administration Console) to configure VxMS logging. For details on the
Logging Assistant, see the NetBackup Administrator's Guide, Volume I.

Note: VxMS logging may require significant resources on the VMware backup host.
 Troubleshooting 293
Configuring VxMS logging

VxMS logging on a Linux backup host

To configure VxMS logging on a Linux backup host
1 Create the VxMS log directory:
/usr/openv/netbackup/logs/vxms

Note: For logging to occur, the VxMS directory must exist.

Note: If you have run the NetBackup mklogdir command, the VxMS log
directory already exists.
See “NetBackup logging for VMware” on page 291.

2 Add the following to the /usr/openv/netbackup/bp.conf file:

VXMS_VERBOSE=<numeric value of 0 or greater>

See Table 14-2 for the available logging levels.

3 To change the log location, enter the following in the bp.conf file:
vxmslogdir=path to new log location

Note: If the VxMS log location is changed, the Logging Assistant does not
collect the logs.
 Troubleshooting 294
Configuring VxMS logging

VxMS logging on a Windows backup host

To configure VxMS logging on a Windows backup host
1 Create the VxMS log directory:
install_path\NetBackup\logs\vxms

Note: For logging to occur, the VxMS folder must exist.

Note: If you have run the NetBackup mklogdir.bat command, the VxMS log
directory already exists.
See “NetBackup logging for VMware” on page 291.

2 In the Windows registry, create the DWORD registry entry VXMS_VERBOSE in

the following location:
HKEY_LOCAL_MACHINE > SOFTWARE > Veritas > NetBackup >
CurrentVersion > Config
3 To configure the logging level, set the numeric value of VXMS_VERBOSE to 0 or
greater. Larger numbers result in more verbose logs.
See Table 14-2 for the available logging levels.
4 To change the log location:
■ Open regedit and go to the following location:
HKEY_LOCAL_MACHINE > SOFTWARE > Veritas > NetBackup >
CurrentVersion
■ Create the registry entry vxmslogdir with a string value (REG_SZ). For the
string value, specify the full path to an existing folder.

Note: You can use NTFS compression on VxMS log folders to compress the
log size. The new logs are written in compressed form only.

Note: If the VxMS log location is changed, the Logging Assistant does not
collect the logs.

VxMS logging levels

Table 14-2 lists the VxMS logging levels.
 Troubleshooting 295
Format of the VxMS core.log and provider.log file names

Note: Logging levels higher than 5 cannot be set in the Logging Assistant.

Note: Logging levels higher than 5 should be used in very unusual cases only. At
that level, the log files and metadata dumps may place significant demands on disk
space and host performance.

Table 14-2 VxMS logging levels

Level Description

0 No logging.

1 Error logging.

2 Level 1 + warning messages.

3 Level 2 + informative messages.

4 Same as level 3.

5 Highly verbose (includes level 1) + auxiliary evidence files (.mmf, .dump, VDDK
logs, .xml, .rvpmem).

You can set the logging level for the VDDK messages.

6 VIX (VMware virtual machine metadata) dumps only.

7 VHD (Hyper-V virtual machine metadata) dumps only.

>7 Full verbose + level 5 + level 6 + level 7.

Format of the VxMS core.log and provider.log file

names
For the log files core.log and provider.log created by default during VxMS
logging, the NetBackup administrator's user name is inserted into the log file name.
Table 14-3 describes the format of the log file names.
 Troubleshooting 296
Configuring the VDDK logging level

Table 14-3 Format of VxMS core.log and provider.log file names

Platform VxMS log-file-name format

Windows VxMS-thread_id-user_name.mmddyy_tag.log

For example:

VxMS-7456-ALL_ADMINS.070214_core.log

VxMS-7456-ALL_ADMINS.070214_provider.log

UNIX, Linux VxMS-thread_id-user_name.log.mmddyy_tag

For example:

VxMS-27658-root.log.081314_core

VxMS-27658-root.log.081314_provider

See “Configuring VxMS logging” on page 292.

Configuring the VDDK logging level

The following NetBackup processes capture VDDK log messages:
■ bpbkar, bpbkarv, nbrestore
These processes write VDDK messages in the VxMS logs if the VxMS logging
level (VxMS_VERBOSE) is 5 or higher. By default, when VXMS_VERBOSE is
5 or higher, the VDDK messages are generated at the highest verbosity.
■ ncfnbcs, bpVMutil
These processes write VDDK messages in their own log directories if the
NetBackup global logging level is 5. By default, when the NetBackup global
logging level is 5, the VDDK messages in the ncfnbcs and bpVMutil logs are
generated at minimum verbosity.
To change the logging level (verbosity) of the VDDK messages
1 Check the following on the backup host:
■ Make sure the VxMS log directory exists and that the VXMS_VERBOSE
DWORD is set to 5 or higher.
See “Configuring VxMS logging” on page 292.
■ Make sure the bpVMutil log directory exists and that the NetBackup global
logging level is set to 5.
 Troubleshooting 297
Preventing browsing delays caused by DNS problems

Note: The ncfnbcs process uses unified logging: You do not have to
manually create a log directory for ncfnbcs. More information on unified
logging is available in the NetBackup Logging Reference Guide.

2 Enter the following on the backup host:

■ Windows:
Create the DWORD registry entry VDDK_VERBOSE in the following
location:
HKEY_LOCAL_MACHINE > SOFTWARE > Veritas > NetBackup >
CurrentVersion > Config
■ Linux:
Add the following to the /usr/openv/netbackup/bp.conf file:
VDDK_VERBOSE=numeric value

3 Set the numeric value of VDDK_VERBOSE as follows:

0 Panic (failure messages only).

1 Level 0 + error logging.

2 Level 1 + warning messages.

3 Level 2 + audit messages.

4 Level 3 + informational messages.

5 Highly verbose; level 4 + additional details.

6 Most verbose; level 5 + debug messages.

Preventing browsing delays caused by DNS

problems
This topic applies to NetBackup for VMware.
NetBackup may be unable to identify virtual machines when you use the Browse
for Virtual Machines dialog. Virtual machine host names may not be properly
configured in your Domain Name Server system (DNS), or the DNS system may
be slow. A timeout message may appear, or messages similar to the following may
appear in the NetBackup detailed status log:
 Troubleshooting 298
Preventing browsing delays caused by DNS problems

17:25:37.319 [12452.10360] get_vSphere_VMs: Processing vm 002-wcms

17:25:37.319 [12452.10360] get_vSphere_VMs: uuid
421a8b46-063d-f4bd-e674-9ad3707ee036
17:25:37.319 [12452.10360] get_vSphere_VMs: vmxdir [san-05] 002-wcms/
17:25:37.319 [12452.10360] get_vSphere_VMs: datastore san-05
17:25:37.319 [12452.10360] get_vSphere_VMs: IpAddress 172.15.6.133
17:25:41.866 [12452.10360] get_vSphere_VMs: retry_gethostbyaddr for
172.15.6.133 failed with The requested name is valid, but no data of
the requested type was found.

Note: NetBackup may be unable to determine the host names of the virtual machines
from their IP addresses (reverse lookup may fail).

To prevent browsing delays caused by DNS problems (Windows)

1 On the Windows desktop of the backup host, click Start > Run and enter
regedit.

2 To be on the safe side, make a backup of the current registry (File > Export).
3 Go to HKEY_LOCAL_MACHINE > SOFTWARE > Veritas > NetBackup >
CurrentVersion > Config and create a key that is called BACKUP.
4 Create a new DWORD under BACKUP, called disableIPResolution.
This registry key causes NetBackup to use the virtual machine's IP address
as the virtual machine's host name.
5 Use the NetBackup Browse for Virtual Machines screen to rediscover the
virtual machines. The host names should now be the IP addresses.
See “Browse for VMware virtual machines” on page 91.
 Troubleshooting 299
Changing the browsing timeout for virtual machine discovery

To prevent browsing delays caused by DNS problems (Linux)

1 On the Linux backup host, create (or open) the following file:
/usr/openv/netbackup/virtualization.conf

2 Add the following to the file:

[BACKUP]
"disableIPResolution"=dword:00000000

This entry causes NetBackup to use the virtual machine's IP address as the
virtual machine's host name.
Note: If the file already contains a [BACKUP] line, do not add another [BACKUP]
line. Any other lines that already exist under [BACKUP] should remain as they
are.
3 Use the NetBackup Browse for Virtual Machines screen to rediscover the
virtual machines. The host names should now be the IP addresses.
See “Browse for VMware virtual machines” on page 91.
See “Notes on troubleshooting NetBackup for VMware” on page 288.
The following applies if: the Primary Identifier in VMware Intelligent Policies is
selected as VM host name and Reverse name lookup is enabled in the
configuration setting.
In a large VMware environment, reverse name lookups can be very slow depending
on the number of virtual machines being discovered. You can change the
VNET_OPTIONS option to determine how many items NetBackup can cache. This
value is in the bp.conf file on UNIX and Linux, and the registry on Windows.
The third value 200 is the default number of entries to be cached. Each entry takes
about 1 kilobyte in memory. Available memory needs to be taken into account when
you change this value. The maximum number of allowed entries is 100000.
VNET_OPTIONS = 120 3600 200 40 3 1 30 10 1793 32 0 0

Use the nbgetconfig command to view the configuration settings. Use nbsetconfig
to change the settings.

Changing the browsing timeout for virtual

machine discovery
This topic applies to NetBackup for VMware.
When browsing for virtual machines in the Browse for Virtual Machines dialog,
the browsing timeout value may need to be increased.
 Troubleshooting 300
Changing timeout and logging values for vSphere

To change the browsing timeout value

1 On the host that runs the NetBackup Administration Console, open the following
file:
/usr/openv/java/nbj.conf

2 Change the value of the NBJAVA_CORBA_DEFAULT_TIMEOUT parameter.

By default, this parameter is set to 60 seconds:
NBJAVA_CORBA_DEFAULT_TIMEOUT=60

Increase the value to a higher number.

For more information on the nbj.conf file, see the NetBackup Administrator's
Guide, Volume I:

Http://www.veritas.com/docs/DOC5332
See “Browse for VMware virtual machines” on page 91.
See “Changing timeout and logging values for vSphere” on page 300.
See “Notes on troubleshooting NetBackup for VMware” on page 288.
See “Configuring a VMware policy from the Policies utility” on page 68.

Changing timeout and logging values for vSphere

This topic applies to NetBackup for VMware.
Table 14-4 lists the vSphere keys and their default values for various timeouts.
These values can be changed on the backup host (see the procedure in this topic).

Table 14-4 DWORD keys and defaults for vSphere timeouts

DWORD key name Default value (in seconds)

jobtimeout 900

poweroptimeout 900

snapshottimeout 900

registertimeout 180

browsetimeout 180

connecttimeout 300

The key and default for the vSphere API logging level are the following.
 Troubleshooting 301
Changing timeout and logging values for vSphere

Table 14-5 DWORD key and default for vSphere API log level

DWORD key name Default value

vmcloglevel 0 (no logging)

Changes to the vSphere API logging level affect the following logs on the backup
host:
■ For backups (snapshot creation): bpfis log
■ For restores: bpVMutil log
■ For virtual machine discovery: ncfnbcs log (originator ID 366)
The logs are in the following location on the backup host:
Windows: install_path\NetBackup\logs\
Linux: /usr/openv/netbackup/logs
To change vSphere timeouts and logging values on Windows
1 On the Windows desktop of the backup host, click Start > Run and enter
regedit.

2 To be on the safe side, make a backup of the current registry (File > Export).
3 Go to HKEY_LOCAL_MACHINE > SOFTWARE > Veritas > NetBackup >
CurrentVersion > CONFIG and create a key that is called BACKUP.
4 To change a timeout value, create a new DWORD under BACKUP, using the
appropriate registry name (such as jobtimeout or poweroptimeout).
Enter a value for the timeout.
5 To change the level of vSphere API logging, create a new DWORD called
vmcloglevel and enter the new logging value.

The allowed values are 0 through 6, where 0 is no logging and 6 is the highest
log level.
 Troubleshooting 302
Credentials for VMware server are not valid

To change vSphere timeouts and logging values on Linux

1 On the Linux backup host, create (or open) the following file:
/usr/openv/netbackup/virtualization.conf

2 To change a timeout value, enter a new dword line under [BACKUP], using the
appropriate name (such as jobtimeout or poweroptimeout). Include a value
for the timeout.
For example:

[BACKUP]
"jobtimeout"=dword:60

This example sets the job timeout to 60 seconds.

Note: If the file already contains a [BACKUP] line, do not add another [BACKUP]
line. Any other lines that already exist under [BACKUP] should remain as they
are.
3 To change the level of vSphere API logging, enter a dword line for vmcloglevel
with a logging value, under [BACKUP]. For example:

"vmcloglevel"=dword:6

The allowed values are 0 through 6, where 0 is no logging and 6 is the highest
log level.
See “Notes on troubleshooting NetBackup for VMware” on page 288.

Credentials for VMware server are not valid

This topic applies to NetBackup for VMware.
A number of issues can prevent NetBackup from gaining access to the ESX server
or vCenter server. When you add credentials for a VMware server (Add Virtual
Machine Server dialog), NetBackup validates the credentials when you click OK.
If the credentials do not pass validation, a pop-up message appears.
Problems can result for a variety of reasons, including the following:
■ An incorrect virtual machine server name. Make sure that the server name is
entered correctly.
More information is available about changing the server name that is stored in
NetBackup.
See “Changing the host name of a VMware server in NetBackup” on page 50.
■ An invalid user name or password. Make sure that a valid user name and
password were entered correctly.
 Troubleshooting 303
Snapshot error encountered (status code 156)

■ An incorrect port number. Make sure that the port number is correct on the
NetBackup Change Virtual Machine Server dialog. If the VMware server uses
the default port, no port specification is required. You can uncheck the Connect
using port number box.
■ You do not have enough privileges to perform backups or restores. (Note
however that lack of sufficient privileges may not cause the credential validation
to fail.)
For the minimum permissions needed to back up and restore with vStorage,
see the following:
http://www.veritas.com/docs/TECH130493
See “Adding NetBackup credentials for VMware” on page 45.
See “Notes on troubleshooting NetBackup for VMware” on page 288.

Snapshot error encountered (status code 156)

This topic applies to troubleshooting NetBackup for VMware.
The following table describes the VMware issues that relate to NetBackup status
code 156.

Table 14-6 Possible causes of status code 156

Causes of status Description and recommended action

code 156

NetBackup cannot obtain NetBackup may not be able to obtain the volume ID of a drive. In that case, none of the
the volume ID of a drive virtual machine drives are backed up. The backup fails with NetBackup status code 156.

The drive may be down.

A backup of the virtual You cannot run more than one backup per virtual machine at a time. If you start a second
machine is already active backup of the virtual machine while the first backup is active, the second job fails with a
status 156.

Recommended action: Wait until the first job completes, then run the second one.
 Troubleshooting 304
Snapshot error encountered (status code 156)

Table 14-6 Possible causes of status code 156 (continued)

Causes of status Description and recommended action

code 156

Cannot find virtual NetBackup cannot find the host name or VM display name of a virtual machine that is listed
machine name in the backup policy. The detailed status log may include the following error message:

Critical bpbrm (pid=<pid number>) from client <client name>:

FTL - snapshot creation failed, status 156.)

If the virtual machines do not have static IP addresses, you can configure NetBackup to
identify virtual machines by their VM display names or UUIDs. Examples of the environments
that do not use static IP addresses are clusters, and the networks that assign IP addresses
dynamically.

Note that NetBackup may have been configured to identify virtual machines by their VM
display names. In that case, make sure that the display names are unique and that they
do not contain special characters.

See “Primary VM identifier options (VMware)” on page 76.

The virtual machine is Through a vCenter server, NetBackup can back up the virtual machines that are turned
powered off off. You must provide credentials for NetBackup to access the vCenter server.

See “Adding NetBackup credentials for VMware” on page 45.

If NetBackup uses credentials for an ESX server instead of vCenter, it may not be able to
identify a turned off virtual machine. Note the following:

■ If the policy uses VM host name or VM DNS name as the Primary VM identifier,
NetBackup may not find the virtual machine. The backup fails.
■ If the policy uses VM display name or VM UUID as the Primary VM identifier, NetBackup
can identify the virtual machine. The backup succeeds.

The virtual machine has If a virtual machine with independent disks is in a suspended state, snapshot jobs fail.
one or more independent Messages similar to the following appear in the job details log:
disks and is in a
suspended state 01/12/2015 17:11:37 - Critical bpbrm (pid=10144) from client
<client name>: FTL - VMware error received: Cannot take a
memory snapshot, since the virtual machine is configured with
independent disks.

This issue results from a VMware limitation (SR#15583458301). More information is

available in the following VMware article:

http://kb.vmware.com/kb/1007532

As a workaround, change the state of the virtual machine to powered on or powered off,
and rerun the backup.
Note: Data on independent disks cannot be captured with a snapshot. The rest of the
virtual machine data is backed up.
 Troubleshooting 305
Snapshot error encountered (status code 156)

Table 14-6 Possible causes of status code 156 (continued)

Causes of status Description and recommended action

code 156

The virtual machine's disk The RDM is ignored (not backed up) and any independent disk is recreated but empty.
is in raw mode (RDM)
See “Configurations for backing up RDMs” on page 377.

The attempt to create a If the attempt to create a snapshot of the virtual machine exceeds the VMware timeout of
snapshot exceeded the 10 seconds, the snapshot fails with NetBackup status 156. This timeout may occur if the
VMware timeout virtual machine is configured with a large number of volumes. Note that the timeout may
be encountered even if the Virtual machine quiesce option was disabled.
Do one of the following:

■ Reduce the number of volumes within the virtual machine.

■ Install a NetBackup client on the virtual machine and select another backup method for
the policy (not the VMware snapshot method).

The virtual machine has The snapshot fails if the virtual machine has no vmdk file.
no vmdk file assigned
Virtual machines without vmdk files can occur in a vCenter Site Recovery Manager (SRM)
environment. If a replicated virtual machine has never been active, it is in passive mode
and may have no vmdk file(s).

You can enable the Ignore diskless VMs option on the VMware Advanced Attributes
tab of the policy. If this option is enabled: NetBackup does not back up a replicated (passive)
virtual machine in an SRM environment if that virtual machine has no vmdk files.

More information is available on the Ignore diskless VMs option.

See “VMware - Advanced Attributes dialog” on page 81.

 Troubleshooting 306
The origin of the snapshot failure: NetBackup or VMware?

Table 14-6 Possible causes of status code 156 (continued)

Causes of status Description and recommended action

code 156

The vmdk file has too Whenever a VMware snapshot occurs, a delta.vmdk file is created for each vmdk. If 32 or
many delta files more such delta files exist for a single vmdk file, a NetBackup backup of that VM may fail
(status 156). The NetBackup Activity Monitor job details contain messages similar to the
following:

02/06/2015 10:33:17 - Critical bpbrm (pid=15799) from

client fl5vm1_2012: FTL - vSphere_freeze: Unable to proceed with
snapshot creation, too many existing delta files(44).
02/06/2015 10:33:17 - Critical bpbrm (pid=15799) from client
fl5vm1_2012: FTL - VMware_freeze: VIXAPI freeze (VMware snapshot)
failed with 25: SYM_VMC_FAILED_TO_CREATE_SNAPSHOT
02/06/2015 10:33:17 - Critical bpbrm (pid=15799) from client
fl5vm1_2012: FTL - vfm_freeze: method: VMware_v2, type: FIM,
function: VMware_v2_freeze

To back up the VM, do the following:

1 Consolidate the VM's snapshots.

In the VMware interface, right-click on the VM and select Snapshot > Consolidate.
For more information, see your VMware documentation.

2 Verify that each of the VM’s vmdk files now has fewer than 32 delta files.

If the snapshot consolidation was not successful, see the following VMware article
for further assistance:
Committing snapshosts in vSphere

3 Rerun the NetBackup backup.

VMware snapshot If the NetBackup policy is enabled for virtual machine quiesce (the default), the VMware
quiesce operation failed snapshot operation in vSphere initiates a quiesce of the virtual machine. If snapshot quiesce
fails, the NetBackup job fails with status 156.

See “Notes on troubleshooting NetBackup for VMware” on page 288.

See “Configuring a VMware policy from the Policies utility” on page 68.

The origin of the snapshot failure: NetBackup or

VMware?
When a NetBackup snapshot job fails with status 156, the problem may originate
in your VMware environment rather than in NetBackup. You can begin to isolate
 Troubleshooting 307
Conflict between NetBackup and VMware Storage vMotion with vSphere 5.0 or later

the problem to one environment or the other by using vSphere Client to take a
snapshot of the VM. NetBackup support often uses this approach to investigate a
snapshot issue.
To identify the environment in which the snapshot error occurred
1 In the vSphere interface, right-click on the VM and click Snapshots > Take
Snapshot.
2 In the Take VM Snapshot for dialog, click Quiesce guest file system if the
NetBackup policy was enabled for virtual machine quiesce (the default).
In the NetBackup policy, the Virtual machine quiesce option is in the VMware
- Advanced Attributes dialog.

Note: In the Take VM Snapshot for dialog, make sure the Snapshot the
virtual machine's memory option is not selected. NetBackup does not use
that option.

3 Start the snapshot and check the Recent Tasks pane for snapshot status.
■ If the snapshot does not complete, the problem with the NetBackup snapshot
may be in the VMware environment. Consult your VMware documentation.
■ If the VMware snapshot is successful, the issue may be with NetBackup.
For relevant error messages, consult the NetBackup bpfis logs.
See “NetBackup logging for VMware” on page 291.
The following topic summarizes some common causes of 156 errors:
See “ Snapshot error encountered (status code 156)” on page 303.

Conflict between NetBackup and VMware Storage

vMotion with vSphere 5.0 or later
To avoid conflicts with Storage vMotion in vSphere 5.0 or later, NetBackup should
conduct backups through the vCenter server, not through the ESX host. A backup
directly through the ESX server may fail if Storage vMotion simultaneously migrates
the virtual machine’s files. In addition, the virtual machine’s snapshot files may be
stranded or other problems with the virtual machine may result. VMware has
acknowledged this issue.
If the backup fails, the NetBackup job details contain a message similar to the
following:
 Troubleshooting 308
Incremental backups of the VM may fail if the NTFS file system in the guest OS is corrupted and the backup
uses MSDP storage

Error opening the snapshot disks using given transport mode: Status 23.

To back up a virtual machine while its files are in the process of migration,
NetBackup must conduct the backup through the vCenter server.
To back up a virtual machine while its files are in the process of migration
1 In the NetBackup Administration Console, under Media and Device
Management > Credentials > Virtual machine servers, add NetBackup
credentials for the vCenter server.
Select a server type of VMware VirtualCenter Server.
2 Delete the credentials that are currently defined under the server type VMware
ESX server, and add those credentials as type VMware Restore ESX Server.
3 Rerun the backup.

Incremental backups of the VM may fail if the

NTFS file system in the guest OS is corrupted and
the backup uses MSDP storage
This error may occur in the following case:
■ The VMware or Hyper-V backup policy specifies Media Server Deduplication
Pool (MSDP) storage.
■ The backup schedule includes incrementals (differential or cumulative).
■ The NTFS file system in the VM’s guest OS is corrupted. For example, the file
system contains orphaned file record segments. The bptm debug log contains
messages similar to the following:

16:24:04.604 [5720.7068] <16> 6:bptm:5720:win2012-2.hrous.sen.com:

[ERROR] PDSTS: fbu_fill_bitmap: (1945) stream_offset(610848768)
+ length(8192) > (1946) stream_offset(610848768)
16:24:04.604 [5720.7068] <16> 6:bptm:5720:win2012-2.hrous.sen.com:
[ERROR] PDSTS: fbu_scan_buf: fbu_fill_bitmap() failed
(2060022:software error)

The incremental backups fail with status 84, media write error.
Use the chkdsk command to examine the file system in the guest OS and repair
it as needed. Then restart the VM and rerun the backup.
See “Notes on troubleshooting NetBackup for VMware” on page 288.
 Troubleshooting 309
The restore fails when you restore individual files to a virtual machine that has NetBackup client software

The restore fails when you restore individual files

to a virtual machine that has NetBackup client
software
This topic applies to NetBackup for VMware.
When you restore individual files to a virtual machine that has a NetBackup client,
make sure that a firewall does not interfere with the restore. If a firewall stops the
restore, turn off the firewall and retry the restore.
See “Restoring the full VMware virtual machine” on page 179.
See “Notes on troubleshooting NetBackup for VMware” on page 288.
See “Configuring a VMware policy from the Policies utility” on page 68.

Backup or restore job hangs

This topic applies to NetBackup for VMware.
NetBackup may have exceeded the maximum number of allowed VMware NFC
connections to the ESX server when it used the transport modes nbd or nbdssl.
Note that NetBackup uses one NFC connection per virtual disk on the ESX or ESXi
server.
If NetBackup is configured to access the ESX server directly (not through a vCenter
or VirtualCenter server), fewer connections are allowed. The following are the
maximum connections as set by VMware:

Table 14-7 VMware NFC connection limits for nbd or nbdssl transfers

ESX version Type of access to the ESX Maximum NFC connections

server allowed

ESX 4 Directly to ESX server 9

ESX 4 Through vCenter 27

ESXi 4 Directly to ESX server 11

ESXi 4 Through vCenter 23

ESXi 5 Directly to ESX server The maximum total for all NFC
connection buffers to an ESXi host
is 32 MB

ESXi 5 Through vCenter 52

 Troubleshooting 310
VMware SCSI requirements for application quiesce on Windows

Try a different transport type (such as SAN or hotadd). If a different transport type
is not available and NetBackup accesses the ESX servers directly, set up access
through a vCenter (or VirtualCenter) server. Use of a server increases the maximum
number of allowed connections. For example: With 27 connections, NetBackup can
access a virtual machine that has up to 27 disks, if the ESX 4 server is behind a
vCenter server.
Note that the connection limits are per-host (that is, per vCenter or ESX server).
For example, assume the following environment:
■ An ESX 4.0 server with three virtual machines.
■ Each virtual machine has ten virtual disks.
■ The virtual machines are behind a vCenter 4.0 server.
For a simultaneous backup of the three virtual machines, NetBackup requires 30
NFC connections. With a limit of 27 NFC connections per vCenter server, any of
the three backup jobs may hang.
These limits are described in the VMware Virtual Disk API Programming Guide:
VMware Virtual Disk API Programming Guide
See also the following section of the VMware vSphere 5 Documentation Center:
Virtual Disk Transport Methods
See “Restore notes and restrictions” on page 173.
See “Configuring a VMware policy from the Policies utility” on page 68.

VMware SCSI requirements for application quiesce

on Windows
For a snapshot that quiesces an application on a Windows VM, VMware imposes
the following disk requirements:
■ The VM's disks must be SCSI, not IDE.
■ The SCSI disks on the VM’s SCSI controller must not occupy more than half of
the total number of slots in the controller. Since the controller has a total of 15
slots, the number of disks in that controller must not exceed 7.
If these conditions are not met and the NetBackup VMware policy enables the
Virtual machine quiesce option, the backup may fail with status code 156.
For more information on this VMware requirement, refer to the following article
in the VMware vSphere 6.0 Documentation Center:
Windows Backup Implementations
 Troubleshooting 311
Mount point missing on a restored Windows virtual machine

Mount point missing on a restored Windows

virtual machine
This topic applies to NetBackup for VMware.
A Windows virtual machine may fail to write its mount point configuration to disk
(the mount point configuration remains in RAM). In that case, the mount point
information cannot be backed up. When the virtual machine is restored, the data
from the mounted volume is restored, but the mount point is absent from the restored
virtual machine.
Reassign the mount point on the restored virtual machine. To make sure the mount
point is correctly configured, restart the virtual machine.
See “Configuring a VMware policy from the Policies utility” on page 68.
See “Restore notes and restrictions” on page 173.

Mount points not available when restoring files

from a Linux virtual machine
This topic applies to NetBackup for VMware.
For Linux virtual machines, only the ext2, ext3, and ext4 file systems are supported
for individual file restore. If a partition is formatted with some other file system, the
backup succeeds but NetBackup cannot map the file system addresses of the files.
As a result, NetBackup cannot restore individual files from that partition. Only the
files that were on ext2, ext3, or ext4 partitions can be individually restored.
Note also: To restore individual files from their original mount points, the "/" (root)
partition must be formatted as ext2, ext3, or ext4. If the "/" (root) partition is formatted
with a different file system (such as ReiserFS), the mount points cannot be resolved.
In that case, you can restore ext2, ext3, or ext4 files from the /dev level (such as
/dev/sda1). You cannot restore the files from their original mount point level.
See “Restore notes and restrictions” on page 173.
See “Configuring a VMware policy from the Policies utility” on page 68.

Invalid client error when you restore files using

NetBackup BAR interface installed on the virtual
machine
This topic applies to NetBackup for VMware.
 Troubleshooting 312
VMware virtual machine does not restart after restore

If the virtual machine was backed up by display name or UUID, and the display
name is not the same as the host name, note: You cannot restore individual files
by means of the Backup, Archive, and Restore (BAR) interface if the interface is
installed on the virtual machine itself. The files can be restored if BAR is installed
on the master server or media server. In this case, BAR must not be installed on
the virtual machine that you want to restore to.
To restore files, the Destination client for restores field in the BAR interface must
have a valid host name or IP address.
See “Restoring individual filesRestoring individual VMware virtual machine files”
on page 224.

VMware virtual machine does not restart after

restore
This topic applies to NetBackup for VMware.
The virtual machine may have been configured as follows:
■ At the time of the backup, the virtual machine had a combination of SATA and
SCSI disks, or of SATA, SCSI, and IDE disks.
■ The guest OS resided on one of the SCSI disks.
The virtual machine when restored may attempt to boot from the SATA or the IDE
disk. The boot attempt fails with the message "Operating system not found."
VMware has identified this problem and will address it in a future release.
As a workaround, reconfigure the BIOS on the virtual machine to boot from the
correct SCSI disk.
See “Restoring the full VMware virtual machine” on page 179.
See “Restore notes and restrictions” on page 173.

A restored VM may not start or its file system(s)

may not be accessible
A restored VM may not boot up or its file system(s) may not be accessible in the
following case:
■ The VM's guest operating system is Windows 8,
■ The VM is restored from a block-level incremental backup image,
■ And the restore uses the hotadd transport mode.
 Troubleshooting 313
NetBackup job fails due to update tasks on the VMware server

As a result of a VMware issue in VDDK 5.5.x, the Windows NTFS Master File Table
on the restored VM may be corrupted. As a workaround, use a different transport
mode to restore the VM (not hotadd).
See “Restore notes and restrictions” on page 173.

NetBackup job fails due to update tasks on the

VMware server
Certain virtual machine update tasks on the VMware server may cause a NetBackup
job to fail. For example, a restore may fail when you use the nbrestorevm command
to restore multiple VMs at the same time. If the failed restore job creates the VM,
the VM is deleted.
On the backup or restore host, the bpVMutil log may contain a message similar
to the following:

Detail: <ManagedObjectNotFoundFault xmlns="urn:vim25"

xsi:type="ManagedObjectNotFound"><obj type="VirtualMachine">
vm-14355</obj>
</ManagedObjectNotFoundFault>

As a workaround, rerun the job for the operation that failed.

Note: If possible, avoid VMware maintenance activities during backup or restore

operations.

See “NetBackup logging for VMware” on page 291.

The vSphere interface reports that virtual machine

consolidation is needed
When NetBackup begins a virtual machine backup, it requests a VMware snapshot
of the virtual machine in vSphere. If the NetBackup policy is enabled for virtual
machine quiesce (the default), the VMware snapshot operation initiates a quiesce
of the virtual machine. If snapshot quiesce fails, the NetBackup job fails with status
156 and VMware snapshot delta files may be left behind in vSphere. Note: As a
result of leftover snapshot delta files, the vSphere status for the virtual machine
may warn that virtual machine consolidation is needed.
 Troubleshooting 314
The Enable file recovery from VM backup option does not work if a volume disk set contains a mixture of disk
types

VMware has acknowledged the problem of leftover delta files after a snapshot
quiesce failure. VMware has fixed this issue in certain ESXi versions. See the
following VMware article for more information on this issue:
Delta disk files (REDO logs) are left uncommitted after a failed quiesced snapshot
operation (2045116)
For a workaround for leftover delta files, see the following VMware article:
Committing snapshots when there are no snapshot entries in the Snapshot Manager
(1002310)

The Enable file recovery from VM backup option

does not work if a volume disk set contains a
mixture of disk types
If a VM is configured on a logical volume (LVM or LDM): Individual file recovery
does not work if the volume disk set contains a mixture of disk types:
■ Some of the logical volume disks are regular virtual disks (normal VMDKs).
■ Some of the disks in the same volume are independent disks or are physical
disks in raw device mapping mode (RDM).
The backup job succeeds but files cannot be individually restored from the file
systems that reside on the disk set (LVM or LDM). To be able to restore files
individually, reconfigure the VM's logical volumes to reside on regular virtual disks
(vmdk) only. Note that VMware does not make snapshots of independent disks or
RDM disks.

Linux VMs and persistent device naming

For Linux VMs without persistent device naming, multiple disk controllers (such as
IDE, SCSI, and SATA) may complicate the recovery of individual files. This issue
occurs because non-persistent device naming, such as /dev/sda and /dev/sdb,
may cause unexpected mount point changes after a restart. If the VM has a SCSI
disk and SATA disk, the Backup, Archive, and Restore interface may show incorrect
mount points for the VM's files. For example, the files originally under /vol_a might
appear under /vol_b when you browse to restore them. The restore is successful,
but the restored files may not be in their original directories.
As a workaround, search for the files on the restored VM and move them to the
proper locations.
 Troubleshooting 315
For a VMware virtual machine with Windows dynamic disks, a restore from incremental backup fails with a
Windows restore host and the hotadd transport mode

To prevent this issue on Linux VMs with multiple disk controllers, it is recommended
a persistent device-naming method for mounting the file systems. When persistent
naming is in place, device mounting is consistent and this issue does not occur
when you restore files from future backups.
For persistent device naming, you can mount devices by UUIDs. The following is
an example of the /etc/fstab file that contains the devices that are mounted by
UUIDs:

UUID=93a21fe4-4c55-4e5a-8124-1e2e1460fece /boot ext4 defaults 1 2

UUID=55a24fe3-4c55-4e6a-8124-1e2e1460fadf /vola ext3 defaults 0 0

To find the device UUIDs, you can use either of the following commands:
blkid

ls -l /dev/disk/by-uuid/

Note: NetBackup also supports the by-LABEL method for persistent device naming.

For a VMware virtual machine with Windows

dynamic disks, a restore from incremental backup
fails with a Windows restore host and the hotadd
transport mode
A restore of a Windows virtual machine by means of the hotadd transfer mode may
fail in the following case:
■ A backup is taken of a Windows virtual machine that has a dynamic disk group.
■ After the backup, another dynamic disk is added to the virtual machine's disk
group.
■ After the dynamic disk is added, an incremental backup is taken of the virtual
machine.
■ A Windows restore host is used with the hotadd transport mode to restore the
virtual machine from the incremental backup.
The restore fails when the Windows restore host tries to mount the dynamic disk
that was added after the first backup. Depending on the data that has already been
restored, Windows may detect the dynamic disk as Invalid or Foreign. Further
writes to an Invalid or Foreign disk are unsuccessful and the restore fails.
 Troubleshooting 316
For a VMware virtual machine with Windows dynamic disks, a restore from incremental backup fails with a
Windows restore host and the hotadd transport mode

The restore fails with status 1, "the requested operation was partially successful."
Messages similar to the following may appear in the VxMS provider logs:

14:10:18.0854 : vdWrite:../VixInterface.cpp:760 <ERROR> : Error

24488361628532739 in write with vdhandle 48870608 startsector
128 numsectors 1 14:10:18.0854 : write:VixFile.h:333 <ERROR>
: Returned error 3, offset 0x0000000000010000, length
0x0000000000000200 14:10:18.0854 : write:VixFile.h:334
<ERROR> : Returning: 11
14:10:18.0854 : vixMapWrite:../VixCoordinator.cpp:1259 <ERROR>
: Returning: 11
14:10:18.0854 : vix_map_write:../libvix.cpp:1826 <ERROR>
: Returning: 11

When they are enabled, VxMS logs are written in the following directory:
Windows:
install_path\NetBackup\logs\vxms

Linux:
/usr/openv/netbackup/logs/vxms

See “Configuring VxMS logging” on page 292.

Note: For successful restores from future incremental backups, run backups with
the Use Accelerator option in the policy.

Try any of the following workarounds to restore from the current incremental backup:
■ Use a Linux restore host (not Windows).
■ Use a different transport mode, such as NBD, NBDSSL, or SAN (not hotadd).
■ When the dynamic disk (the one that was added after the first backup) is mounted
for restore, manually set the disk to offline. When the disk is offline, NetBackup
can write data to it and successfully complete the restore.
See the remainder of this tech note for assistance with this workaround.
To determine when the dynamic disk is mounted for restore
◆ Use the Windows Disk Management utility (Control Panel > Administrative
Tools > Computer Management > Disk Management), or run diskpart in
administrator mode and enter the list disk option.
When Windows attempts to mount the disk, it labels the disk as Invalid or
Foreign.
 Troubleshooting 317
Simultaneous hotadd backups (from the same VMware backup host) fail with status 13

To use diskpart to take the dynamic disk offline

1 On the Windows restore host, run diskpart in administrator mode.
2 Enter list disk to list all disks and find the Invalid or Foreign disk.
3 Enter select disk disk ### to select the disk that is Invalid or Foreign.
4 Enter offline disk to offline the disk.
Example session:

DISKPART> list disk

Disk ### Status Size Free Dyn Gpt

-------- ------------- ------- ------- --- ---
Disk 0 Online 100 GB 1024 KB
Disk 1 Online 256 GB 56 GB *
Disk 2 Invalid 40 MB 40 MB * *
Disk 3 Offline 40 MB 40 MB

DISKPART> select disk 2

Disk 2 is now the selected disk.

DISKPART> offline disk

DiskPart successfully taken offline the selected disk.

Simultaneous hotadd backups (from the same

VMware backup host) fail with status 13
During simultaneous backups from the same VMware backup host, some of the
backups may fail with status 13, "file read failed." A hotadd backup of multiple disks
may take more time than the client-read timeout allows (the default is 300 seconds).
The delay may be caused by locking timeouts in the VMware VDDK.
In the NetBackup Activity Monitor, the detailed status log may include messages
similar to the following:

12/05/2014 06:43:53 - begin writing

12/05/2014 06:48:53 - Error bpbrm (pid=2605) socket read failed:
errno = 62 - Timer expired
12/05/2014 06:48:55 - Error bptm (pid=2654) media manager terminated
by parent process
 Troubleshooting 318
Troubleshooting VMware tag usage

The /NetBackup/logs/vxms log may include repeated instances of a VDDK

message similar to the following:

12/08/2014 05:11:35 : g_vixInterfaceLogger:libvix.cpp:1844 <DEBUG> :

[VFM_ESINFO] 2014-12-08T05:11:35.146-06:00 [7F1B1163F700 info Libs']
FILE: FileLockWaitForPossession timeout on '/var/log/vmware/hotAddLock.
dat.lck/M34709.lck' due to a local process '15882-26732358(bpbkarv)'

To prevent this issue, do one of the following:

■ Reduce the number of hotadd backups that run simultaneously.
■ Increase the client-read timeout on the media server as appropriate (15 minutes
or more):
In the NetBackup Administration Console, click NetBackup Management >
Host Properties > Double-click on media server > Timeouts > Client read
timeout.

Troubleshooting VMware tag usage

Tag associations are backed up and restored as part of the VMware backup process.
Backup and restore of tag associations is a best effort. Any tag collection errors
are shown in the Activity Monitor for the virtual machine snapshot job.
Because of the best effort on backup and restore of tag associations, you may
receive unexpected behavior. Please be aware of the following:
■ You can receive a NetBackup Status Code 0 even if the tag associations are
not captured during the backup. Because tag backup is best effort, this error is
not considered a failure that halts a backup. Any tag collection errors that occur
in the backup are shown in the Activity Monitor for the virtual machine snapshot
job.
■ You can receive a NetBackup Status Code 0 on a restore even if tag associations
are not restored. Any tag collection errors that occur in the backup are shown
in the Activity Monitor for the virtual machine snapshot job.
■ Restores receive a NetBackup Status Code 1 when:
■ The tag doesn't exist on the target vCenter Server. Be aware that NetBackup
restores tag associations to virtual machines by the tag identifier not tag
name.
■ The virtual machine was restored to a pre-6.0 vCenter Server.
■ Other VMware failures.
 Troubleshooting 319
Troubleshooting VMware tag usage

Query behavior with unsupported versions of VMware

Table 14-8 NetBackup query behavior for tag field keyword with unsupported
versions of VMware

Version of VMware Behavior Additional information

vCenter earlier than 5.1 Virtual machines are N/A

evaluated as if tags are not
set.

vCenter 5.1 and 5.5 Virtual machine selection is If tag information is required
marked as Failed. to determine virtual machine
selection, the virtual machine
is marked as Failed with
NetBackup Status Code
4266.

ESX server Virtual machine selection is If tag information is required

marked as Failed. to determine virtual machine
selection, the virtual machine
is marked as Failed with
NetBackup Status Code
4265.

List of possible Tag values not displayed in Query Builder

If the list of possible values does not return any results when the Tag Field is
selected in the Policy Query Builder, there are several causes of this problem.
■ Confirm that your version of vCenter Server is 6.0 or later.
NetBackup support for tags starts with VMware vCenter Server 6.0.
See “Notes and limitations for tag usage in VMware Intelligent Policy queries”
on page 41.
■ Confirm there are tags defined using the vSphere Web Client.
■ Confirm that all tags and categories have descriptions.
If one of the tags does not have a description, when you attempt to browse for
tags in the VMware Intelligent Policy, no tags are displayed.
Veritas has confirmed that this issue is resolved in vCenter Server 6.0 Update
1.
VMware Knowledge Base article: http://kb.vmware.com/kb/2124204
■ Confirm the system times of the discovery host and the vCenter Server are
synchronized.
VMware Knowledge Base article: http://kb.vmware.com/kb/2125193
 Troubleshooting 320
Ensuring that guest customizations can be restored in vCloud Director

Timeout issues in discovery job

In large VMware environments, you may experience timeout issues during the
discovery job or the test query. Review the NetBackup Activity Monitor for this
message:
09/21/2015 10:23:05 - Error nbpem (pid=13064) VMware vCloud Suite
SDK API Failed, msg = [This method requires authentication.], display
name = [display_name], server = [server_name]

This message indicates that the discovery job exceeded the bearer timeout. You
need to increase this timeout for the job to complete successfully.
To adjust the timeout on the vCenter
1 Open the VMware vSphere Web Client.
2 Select Administration > Single Sign-On > Configuration > Policies > Token
Policy.
3 Increase Maximum bearer token lifetime from the default 300 seconds.
Because each environment is unique, Veritasdoes not have any
recommendations on this value. Increase the value until the problem is
eliminated.

Ensuring that guest customizations can be

restored in vCloud Director
This topic applies to NetBackup for VMware.
NetBackup can back up VMware vCloud Director environments and restore virtual
machines into vCloud Director.
See “About NetBackup for vCloud Director” on page 263.
To ensure that any VM guest customizations are restored into vCloud Director, you
must set a NetBackup parameter, as follows:
■ On Windows, you must set a registry value.
See “To ensure that guest customizations can be restored in vCloud Director
on Windows” on page 321.
■ On UNIX and Linux, you must edit a NetBackup configuration file.
See “To ensure that guest customizations can be restored in vCloud Director
on Linux” on page 321.
The parameter value specifies a wait period in seconds so that the guest
customizations can be restored successfully. (The VMware API requires that the
VMware Tools are installed and running, but the state of the VMware Tools cannot
 Troubleshooting 321
Troubleshooting vmdk restore to existing VM

be identified after the restore. Therefore, we wait the specified amount of time so
that the VMware Tools are running in the initial restore environment.)
To ensure that guest customizations can be restored in vCloud Director on
Windows
1 On the Windows desktop of the backup host, click Start > Run and enter
regedit.

2 To be on the safe side, make a backup of the current registry (File > Export).
3 Go to HKEY_LOCAL_MACHINE > SOFTWARE > Veritas > NetBackup >
CurrentVersion > CONFIG and create a key that is called BACKUP.
4 Create a new DWORD under BACKUP and name it powerCycleInterval.
Enter a decimal value of 60 for the timeout.
5 Close the Windows Registry Editor.
To ensure that guest customizations can be restored in vCloud Director on
Linux
1 On the Linux backup host, create (or open) the following file:
/usr/openv/netbackup/virtualization.conf

2 Create a line in the file named [BACKUP] and then on a separate line create a
powerCycleInterval dword parameter with a value of 60, as follows:

For example:

[BACKUP]
"powerCycleInterval"=dword:60

Note: If the file already contains a [BACKUP] line, do not add another [BACKUP]
line. Any other lines that already exist under [BACKUP] should remain as they
are.
3 Save the file and then close it from the text editor.
See “Notes on restoring virtual machines into vCloud Director” on page 265.

Troubleshooting vmdk restore to existing VM

For virtual disk restores and in-place restores to an existing VM, be aware of the
following:
■ If the VMDKs cannot be attached to the target VM, the restored VMDKs are
retained on the temporary VM. The name of the temporary VM is available in
 Troubleshooting 322
Troubleshooting vmdk restore to existing VM

the job details in NetBackup. In the following job details example, the temporary
VM name is vCenter60vm1_rhel6.4_1465584674:

06/10/2016 13:51:17 - Info bpVMutil (pid=3400) Restoring [datastore1]

vCenter60vm1_rhel6.4/vCenter60vm1_rhel6.4_4.vmdk to [datastore1]
vCenter60vm1_rhel6.4/vCenter60vm1_rhel6.4_4-1465584677.vmdk
06/10/2016 13:51:38 - Info bpVMutil (pid=3400) Successfully created
virtual machine vCenter60vm1_rhel6.4_1465584674 with specified disks.
06/10/2016 13:51:41 - requesting resource @aaaab

You can access the data on the temporary VM.

If the temporary VM is retained after a restore failure, the restore job contains
a message similar to the following:

06/14/2016 15:29:06 - Info bpVMutil (pid=5225) attachDisksToExistingVM:

Unable to attach restored disks to target VM vCenter60vm2_rhel6.4
06/14/2016 15:29:06 - Info bpVMutil (pid=5225) attachDisksToExistingVM:
Temporary VM with restored virtual disks was left in place, it can be
used to access restored data

By default, NetBackup retains the temporary VM if the disks are not attached.
To change that behavior, set the DeleteRestoredVMOnError field to Yes in the
restore parameters file.
■ After a restore to an existing VM, the next backup of the VM backs up the
restored virtual disks. This backup may show a warning during collection of the
Changed Block Tracking (CBT) information.
■ For an In-place disk restore, raw devices (RDMs) and independent disks are
not deleted or replaced during restore. If the controller for these disks conflicts
with the disks being restored, the restore fails. The following example messages
are job details from a failed in-place restore:

May 07, 2020 10:26:21 AM - Warning bprd.sfr (pid=2425) Unable to

attach the restored disks to requested VM
May 07, 2020 10:26:21 AM - restored from image
InPlaceDiskRestoreDemo_1588837243; restore time: 0:00:50
May 07, 2020 10:26:21 AM - end Restore; elapsed time 0:00:50
May 07, 2020 10:26:21 AM - Info bpVMutil (pid=2673)
attachDisksToExistingVM: Controller scsi0-1 not available
to perform in-place disk restore. Aborting restore.
The requested operation was partially successful(1)
 Troubleshooting 323
Errors for backups of virtual machines on Virtual Volumes (VVols)

Errors for backups of virtual machines on Virtual

Volumes (VVols)
When troubleshooting backups of virtual machines on VVols, note the following:
■ Each NetBackup snapshot job creates a vSphere snapshot of the virtual machine.
■ To investigate snapshot failures, check the storage array’s VASA provider logs
as well as the vSphere error messages. (VASA is vSphere API for Storage
Awareness.)

Table 14-9 Errors for backups of virtual machines on VVols

Error Explanation and recommended action

message or
cause

vSphere Ensure that you have the required snapshot license from the array vendor. Licensing requirements
snapshot for vSphere snapshots vary from one type of VVol storage to another, depending on the array
creation fails for vendor.
a backup of a
Each NetBackup snapshot job creates a vSphere snapshot of the virtual machine.
VM on VVol.

There is insufficient space in VVol storage.

Space requirements vary from one array vendor to another. Consult the storage array documentation.
 Troubleshooting 324
Issues with the CA certificate during installation of the NetBackup client on VMware Cloud (VMC)

Issues with the CA certificate during installation

of the NetBackup client on VMware Cloud (VMC)
Failed to display CA certificate details

To fix this problem, do the following:

1 Click Cancel to cancel the installation.
2 Make sure the private DNS name was used to install the master server.
3 In the hosts file of the master server installed in AWS and of the backup host
installed in VMC, add the following:
■ The private IP and private DNS name of the NetBackup master server.
■ The IP and the DNS name of the backup host.
The hosts file location on Windows:
C:\Windows\System32\drivers\etc\hosts
The hosts file location on Linux:
/etc/hosts

4 Begin the NetBackup client installation again.

 Troubleshooting 325
Issues with the CA certificate during installation of the NetBackup client on VMware Cloud (VMC)

The target NetBackup server cannot be authenticated, or

the server name does not match any of the host names
that are listed in the server's certificate

To fix this problem, do the following:

1 Click Cancel to cancel the installation.
2 Follow the previous steps under "Failed to display CA certificate details."
3 If the error persists, see the following article:
Host validation fails when a NetBackup client tries to connect to the master
server
4 Begin the NetBackup client installation again.
 Appendix A
NetBackup commands to
back up and restore virtual
machines
This appendix includes the following topics:

■ Using NetBackup commands to create a VMware policy

■ Using the nbdiscover command to search the virtual environment

■ Using the nbrestorevm command to restore virtual machines into vSphere

■ Using the nbrestorevm command to restore virtual machines into vCloud Director

■ Using the nbrestorevm command to restore virtual machines with Instant

Recovery

■ Using the nbrestorevm command to restore virtual machine disks

■ The nbrestorevm -R rename file for restoring virtual machines

■ Logs for troubleshooting the nbrestorevm command

■ Query format and examples for searching virtual machine backups

■ Allowing other servers or clients to restore virtual machines

Using NetBackup commands to create a VMware

policy
This topic describes how to use NetBackup commands to create a policy to back
up virtual machines.
 NetBackup commands to back up and restore virtual machines 327
Using NetBackup commands to create a VMware policy

The NetBackup commands for policy creation are in the following directory:
Windows:
install_path\Veritas\NetBackup\bin\admincmd

UNIX or Linux:
usr/openv/netbackup/bin/admincmd

Run these commands on the NetBackup master server.

To use NetBackup commands to create a VMware policy
1 Create a blank policy.
bppolicynew policyName -M masterServer

For example:
bppolicynew pol1 -M nbserver.acme.com

2 Configure the VMware policy type and set the VMware snapshot options.
This step requires running the initialize form of the bpplinfo command in two
stages.

First bpplinfo policyName -set -pt VMware

stage:
Example for the first stage of bpplinfo:

bpplinfo pol1 -set -active -pt VMware -residence stu1

 NetBackup commands to back up and restore virtual machines 328
Using NetBackup commands to create a VMware policy

Second bpplinfo policyName -modify -use_virtual_machine 1

stage: -alt_client_name VMware_backup_host -snapshot_method
"VMware_v2" -snapshot_method_args
keyword=value,keyword=value,...

The -alt_client_name specifies the VMware backup host (NetBackup access

host). You can specify a particular backup host on this option, or specify
MEDIA_SERVER (a media server acts as the backup host).

The -use_virtual_machine, -alt_client_name, -snapshot_method

"VMware_v2", and-snapshot_method_args must be specified with all
parameters set. (Otherwise, the snapshot job fails.)
Note: -snapshot_method_args is a comma-delimited list of keywords
that define the VMware snapshot and policy. The keywords are described
in a separate topic:

See “VMware policy: bpplinfo options and keywords” on page 331.

Example for the second stage of bpplinfo (for a policy for vCloud Director):

bpplinfo pol1 -modify -use_virtual_machine 1 -alt_client_name

backuphost1.acme.com -application_discovery 1 -snapshot_method
"VMware_v2" -snapshot_method_args "file_system_optimization=1,
snapact=2,drive_selection=0,Virtual_machine_backup=2,
enable_vCloud=1,multi_org=0,disable_quiesce=0,nameuse=0,
exclude_swap=1,skipnodisk=0,post_events=1,ignore_irvm=0,
rLim=10,rTO=0,rHz=10,trantype=san:hotadd:nbd:nbdssl"

To back up virtual machines in vCloud Director, the bpplinfo command must

include the following:
■ -application_discovery 1
■ The snapshotMethodArgs list must include enable_vCloud=1.
In addition, NetBackup credentials must be specified for a vCloud Director
server.
3 Create a schedule for the policy.
bpplsched policyName -add scheduleLabel -st scheduleType

For example:
bpplsched pol1 -add Full -st FULL
 NetBackup commands to back up and restore virtual machines 329
Using NetBackup commands to create a VMware policy

4 Modify the schedule attributes.

bpplschedrep policyName scheduleLabel -(0..6) start duration

For example:
bpplschedrep pol1 Full -0 0 604800

5 Select the virtual machines to back up.

This step requires two commands: bpplclients and bpplinclude.

bpplclients policyName -add VM_to_back_up hardware operating_System

bpplinclude policyName -add ALL_LOCAL_DRIVES | filter_to_select_VMs

■ To back up a virtual machine through manual selection (for example):

bpplclients pol1 -add prvm1.acme.com Windows-x64 Windows2008

bpplinclude example -add ALL_LOCAL_DRIVES

In this example, prvm1.acme.com is the virtual machine to back up, including

all of its local drives (-add ALL_LOCAL_DRIVES). ALL_LOCAL_DRIVES is the
only file list option available for the VMware policy type.
■ To automatically select virtual machines through a query rule (for example):

bpplclients pol1 -add MEDIA_SERVER VMware VMware

bpplinclude pol1 -add "vmware:/?filter= Displayname Contains
'prvm'"

MEDIA_SERVER allows NetBackup to use a media server to discover and

select the virtual machines. As an alternative, you can use -add to specify
a particular discovery host.
In the bpplinclude example, the query rule "vmware:/?filter=
Displayname Contains 'prvm'" selects the virtual machines with the
display names that contain prvm. (Other characters can also appear in the
display names.)

Note: To back up virtual machines in vCloud Director, the policy must use a
query rule on the bpplinclude command to select virtual machines automatically.
For vCloud, do not specify a virtual machine on the bpplclients command.

bpplinclude has options for modifying the query rules in an existing policy.
See “bpplinclude options for modifying query rules in a VMware policy”
on page 333.
 NetBackup commands to back up and restore virtual machines 330
Using NetBackup commands to create a VMware policy

To test the bpplinclude query rule (vmware:/?filter) before you run the
backup
◆ Run the following:
Windows:
install_path\NetBackup\bin\nbdiscover -noxmloutput -policy
policyName -noreason

UNIX, Linux:
/usr/openv/netbackup/bin/nbdiscover -noxmloutput -policy
policyName -noreason

A + sign appears before the virtual machines that the query rule selects for the
backup: these virtual machines are included in the backup when the policy
runs.
The virtual machines to be excluded from the backup appear with a - sign.
The -noreason option omits explanations as to why the query excluded a virtual
machine or why the query failed. For explanations, do not include -noreason.
For more details on most of the command options, see the man page or the
NetBackup Commands Reference Guide.
See “Using the nbrestorevm command to restore virtual machines into vCloud
Director” on page 341.

Notes on creating a VMware policy for vCloud Director

Note the following when you create a policy for vCloud Director:
■ NetBackup credentials must be specified for a vCloud Director server.
■ The bpplinfo command must include the following option and value, to enable
the policy for automatic selection of virtual machines: -application_discovery 1
■ On the bpplinfo command, the snapshotMethodArgs list must include the
following attribute: enable_vCloud=1.
■ The policy must use a query rule, on the bpplinclude command, to select virtual
machines automatically.
For the complete steps--including these points--use the following procedure:
See “Using NetBackup commands to create a VMware policy” on page 326.
 NetBackup commands to back up and restore virtual machines 331
Using NetBackup commands to create a VMware policy

VMware policy: bpplinfo options and keywords

Table A-1 describes the options that are available on the bpplinfo command.
These options are set to default values if not specified.

Table A-1 bpplinfo options

Option Values Keyword label in policy GUI

-ignore_client_direct 0 enables client deduplication, 1 disables client The Attributes tab > Disable
deduplication client side deduplication

-application_discovery 0 disabled, 1 enabled Clients tab > Select

automatically through VMware
Intelligent Policy query

-discovery_lifetime seconds Clients tab > Reuse VM

selection query results for

-ASC_apps_attr exchange, mssql, sharepoint Enables the file-level recovery of

database data.
truncatelogs=0 disabled, 1 enabled (for Exchange or
SQL Server only) VMware tab > Application
Protection: Enable Exchange
To enable Exchange file recovery:
Recovery, Enable SQL Server
-ASC_apps_attr exchange: Recovery, or Enable SharePoint
Recovery
To enable Exchange, SQL, and SharePoint file recovery:

-ASC_apps_attr exchange:,mssql:,sharepoint:

To enable exchange recovery with truncate logs:

-ASC_apps_attr exchange:;truncatelogs=1

To enable exchange without truncate logs and mssql

with truncate logs:

-ASC_apps_attr exchange:,mssql:;truncatelogs=1

To enable exchange with truncate logs and mssql

without truncate logs (note the final colon):

-ASC_apps_attr exchange:;truncatelogs=1,mssql:

To enable both exchange and mssql with truncate logs:

-ASC_apps_attr
exchange:;truncatelogs=1,mssql:;truncatelogs=1

-snapshot_method_args keyword=value,keyword=value, ...

See Table A-2 on page 332.

Table A-2 describes the keywords that must be included in -snapshot_method_args.

 NetBackup commands to back up and restore virtual machines 332
Using NetBackup commands to create a VMware policy

Table A-2 Keywords and values for -snapshot_method_args

Keyword Value Name of keyword in policy GUI

disable_quiesce= 0 virtual machine quiesce is enabled, 1 VMware tab > Advanced > Virtual
virtual machine quiesce is disabled machine quiesce

drive_selection= 0 Include all disks, 1 Exclude boot disk, Exclude Disks tab
2 Exclude data disks

enable_vCloud= 0 disabled, 1 enabled Clients tab > Enable vCloud Director

integration (only available with automatic
selection of client)

exclude_swap= 0 disabled, 1 enabled VMware tab > Exclude swap and paging
files

file_system_optimization= 0 disabled, 1 enabled VMware tab > Exclude deleted blocks

ignore_irvm= 0 disabled, 1 enabled VMware tab > Advanced > Ignore Instant
Recovery VMs

multi_org= 0 disabled, 1 enabled VMware tab > Advanced > Multiple

organizations per policy

nameuse= 0 VM host name, 1 VM display name, 2 VMware tab > Primary VM identifier
VM BIOS UUID, 3 VM DNS Name, 4 VM
instance UUID

post_events= 0 No Events, 1 All Events, 2 Error Events VMware tab > Advanced > Post vCenter
events

rHz= 0 - 3600 VMware tab > Advanced > Snapshot

creation interval (seconds)

rLim= 0 - 100 VMware tab > Advanced > Snapshot retry

rTO= 0 - 1440 VMware tab > Advanced > Snapshot

timeout (minutes)

serverlist= VMware server names. Use a colon to VMware tab > Advanced > VMware server
separate the names. For example: list
serverlist=vc1.acme.com:vc2.acme.com

To specify no server list, enter

serverlist=

skipnodisk= 0 disabled, 1 enabled VMware tab > Advanced -> Ignore

diskless VMs
 NetBackup commands to back up and restore virtual machines 333
Using NetBackup commands to create a VMware policy

Table A-2 Keywords and values for -snapshot_method_args (continued)

Keyword Value Name of keyword in policy GUI

snapact= 0 Continue backup 1 Abort if any VMware tab > Existing snapshot handling
snapshot(s) exist 2 Remove NetBackup
snapshot(s) and continue backup 3 Abort
if NetBackup snapshot(s) exist

trantype= san, hotadd, nbd, or nbdssl VMware tab > Transport modes

If more than one, separate each with a

colon. Order of modes indicates priority.
Example:

trantype=nbd:hotadd

This example selects these two modes

and tries nbd first.

Virtual_machine_backup= 1 disabled, 2 enabled VMware tab > Enable file recovery from
VM backup

vmdk_ca= The name of the VMware custom attribute Exclude Disks tab > Perform custom
that specifies the disk(s) to exclude from attribute based selection
backups. For example,
vmdk_ca=NB_DISK_EXCLUDE_LIST.

Set the value for the custom attribute on

each virtual machine or managed host,
as appropriate.

The custom attribute must have comma

separated values of device controls for
the disks to be excluded. For example:
scsi0-0,ide0-0,sata0-0.

vmdk_list= Colon separated values of device Exclude Disks tab > Specific disk(s) to be
controllers for the disks to be excluded. excluded
For example:
scsi0-0:ide0-0:sata0-0.

See “bpplinclude options for modifying query rules in a VMware policy” on page 333.
See “Using NetBackup commands to create a VMware policy” on page 326.

bpplinclude options for modifying query rules in a VMware policy

The bpplinclude command has options for modifying the query rules in an existing
policy.
 NetBackup commands to back up and restore virtual machines 334
Using NetBackup commands to create a VMware policy

Table A-3 bpplinclude options for modifying query rules

Option Description

-addtoquery query_string ... Adds the specified query string to the end of the policy query rules, or creates a query
if none exists.

Quotes (") must be escaped (\).

Examples:

To add vm17 to the list of values in the query rules of policy1:

bpplinclude policy1 -addtoquery ,\"vm17\"

To create a query in a policy that does not have a query:

bpplinclude policy1 -addtoquery vmware:/?filter=Displayname

AnyOf \"grayfox7\",\"grayfox9\"
Note: Each quote (") is escaped with a backslash (\).

-addtoquery -f file_name Adds the entries to the query rules from the specified file, or creates a query if none
exists.

In the file, quotes (") do not need to be escaped.

Example:

To create a query in a policy that does not have a query:

bpplinclude policy1 -addtoquery -f qfile1

where the contents of qfile1 are:

vmware:/?filter=vCenter Contains "ROS" AND ESXserver Equal

"VM_test1" AND Powerstate Equal poweredOn
Note: The values "ROS" and "VM_test1" are not escaped.

Note: You can place entries on multiple lines in the file. All entries are added to the
end of the query (if a query already exists).
 NetBackup commands to back up and restore virtual machines 335
Using the nbdiscover command to search the virtual environment

Table A-3 bpplinclude options for modifying query rules (continued)

Option Description

-deletefromquery query_string Deletes the specified query string from the policy query rules.
...
Examples:

To delete vm27 from the list of values in the query rule of policy1:

bpplinclude policy1 -deletefromquery \"vm27\"

This example also deletes the comma preceding vm27 if such a comma exists in the
query rules.
Note: The -deletefromquery option deletes a comma if: the phrase in the query_string
does not begin or end with a comma and the character preceding the deleted string
is a comma.

To delete an entire query from the policy:

bpplinclude policy1 -deletefromquery vmware:/?filter=Displayname

AnyOf "grayfox7","grayfox9"

-deletefromquery -f file_name Deletes the file entries from the query rules.

Example:

To delete a query from a policy:

bpplinclude policy1 -deletefromquery -f qfile1

where the contents of qfile1 are:

vmware:/?filter=vCenter Contains "ROS" AND ESXserver Equal

"VM_test1" AND Powerstate Equal poweredOn
Note: The values "ROS" and "VM_test1" are not escaped.

Note: paths that contain wildcards must be enclosed in quotes.

See “Using NetBackup commands to create a VMware policy” on page 326.

Using the nbdiscover command to search the

virtual environment
To search for VMs, run nbdiscover on the master server.
UNIX and Linux:
/usr/openv/netbackup/bin/nbdiscover -noxmloutput "query"

Windows:
 NetBackup commands to back up and restore virtual machines 336
Using the nbrestorevm command to restore virtual machines into vSphere

install_path\NetBackup\bin\nbdiscover.exe -noxmloutput "query"

where -noxmloutput displays one VM per line. The query specifies either a
filter-based search rule or a browse operation.
See “Query format and examples for searching virtual machine backups” on page 350.
For example, to search for the backups of virtual machines that have "prod1" in
their display names:
nbdiscover -noxmloutput
"vmsearch:/;reqType=search?filter=VMDisplayName Contains 'prod1'"

For further descriptions of the command options, see the command's man page or
the NetBackup Commands Reference Guide.
See “Using the nbrestorevm command to restore virtual machines into vSphere”
on page 336.
See “Using the nbrestorevm command to restore virtual machines into vCloud
Director” on page 341.

Using the nbrestorevm command to restore virtual

machines into vSphere
To restore VMs, run the nbrestorevm command on the master server or recovery
host. The command is located in the following directory:
UNIX and Linux:
/usr/openv/netbackup/bin/nbrestorevm

Windows:
install_path\NetBackup\bin\nbrestorevm.exe

The following nbrestorevm options are used in these examples:

■ –vmw
Required for all VMware virtual machine restores.
■ –C virtual_machine_to_restore
The VM name must match the type of name that was selected in the Primary
VM identifier option of the backup policy. For example, if the VM was backed
up by its VM display name, specify the display name on the -C option.
■ -R absolute_path_to_rename_file
Restores to a different location. The required entries in the rename file depend
on the destination for the restore. See each example in this topic. Further notes
are available on the rename file.
 NetBackup commands to back up and restore virtual machines 337
Using the nbrestorevm command to restore virtual machines into vSphere

See “The nbrestorevm -R rename file for restoring virtual machines” on page 346.
■ -O
Overwrites the VMs and associated resources if they already exist.
■ –S master_server
Specifies the master server that made the backup (if different from the current
master).
■ -vmserver vCenter_server
Specifies a different vCenter server for the restore destination.
■ –vmproxy VMware_recovery_host
Specifies a different VMware recovery host (overrides the default backup host
that backed up the VMs).

Note: Storage lifecycle policies (SLPs) can use Auto Image Replication to
replicate a virtual machine backup image to another NetBackup domain. To
restore the virtual machine from the replicated image, you must include the
-vmproxy option. Use -vmproxy to specify the backup host (access host) that
is in the domain where the virtual machine was replicated.

■ -vmkeephv
Retains the hardware version upon recovery.
■ -vmid
Retains the BIOS UUID of the original virtual machine.
■ -vmInstanceId
Retains the Instance UUID of the original virtual machine (note that the Instance
UUID is a vCenter specific unique identifier of a virtual machine).
■ -vmsn
Removes the network from the restored VM.
■ -vmvmxd
Restores the .vmdk files to the vmx datastore (the configuration datastore).
■ -vmpo
Turns on the VM after the restore.
Following are examples of nbrestorevm, for various restore destinations and options.

A. Restore to the original location and overwrite the VM.

nbrestorevm –vmw –C VM_to_restore –O
 NetBackup commands to back up and restore virtual machines 338
Using the nbrestorevm command to restore virtual machines into vSphere

B. Restore to the original location, overwrite the VM, and

retain the BIOS UUID.
nbrestorevm –vmw –C VM_to_restore –O -vmid

C. Restore to the original location, overwrite the VM, and

retain the instance UUID.
nbrestorevm –vmw –C VM_to_restore –O -vmInstanceId

D. Restore to the original location, overwrite the VM, and

remove the network.
nbrestorevm –vmw –C VM_to_restore –O –vmsn

E: Restore to the original location from a VM backup that

a different master server made, and overwrite the VM.
nbrestorevm –vmw –C VM_to_restore –S master_server –O

F: Restore to the original vCenter and ESX, but to a

different datastore that also contains the VM configuration
files.
nbrestorevm –vmw –C VM_to_restore –R absolute_path_to_rename_file
-vmvmxd

For the -R rename file, create a text file with the following entry:
change datastore to new_vmx_datastore

where new_vmx_datastore specifies the path to the vmx datastore (configuration

datastore) for the restored VM. After the restore, this datastore contains the .vmdk
files of the restored VM.

Note: In the rename file, the change line must end with a carriage return.

G: Restore to a different vCenter server.

nbrestorevm –vmw –C VM_to_restore –R absolute_path_to_rename_file
–vmserver vCenter_server

For the -R rename file, create a text file with the following entries. Replace the
italicized values with the actual name for the restored VM and for the other vSphere
objects in this list:

change vmname to new_display_name

change esxhost to new_ESX_host
change datacenter to /new_datacenter
 NetBackup commands to back up and restore virtual machines 339
Using the nbrestorevm command to restore virtual machines into vSphere

change folder to /new_datacenter/new_folder

change resourcepool to new_datacenter/host/new_ESX_host/resources
/new_resource_pool
change datastore to new_vmx_datastore
change /original_vmdk_path to /new_datastore/

Note: In the rename file, each change line must end with a carriage return.

The last change entry in this list (change /original_vmdk_path to /new_datastore/)

designates the following: the original path to the .vmdk file, and the new datastore
for the .vmdk file of the restored VM. This change entry is required when you restore
to a different datastore (unless you use the nbrestorevm –vmvmxd option).
Depending on your VMware environment, this change entry may take the following
form:
change /original_datastore/original_VM_folder/original_vmdk.vmdk to
/new_datastore/

Note: If the change vmdk directive specifies a different path for some of the VM’s
vmdk files, the other vmdk files are restored to their original datastore.

See “The nbrestorevm -R rename file for restoring virtual machines” on page 346.

H: Restore to a different vCenter server that has multiple

networks.
nbrestorevm –vmw –C VM_to_restore –R absolute_path_to_rename_file
–vmserver vCenter_server

Create a rename file with the change entries listed for example G. Add the following
change entry:
change network to new_network, new_network(n)

Note: In the rename file, each change line must end with a carriage return.

I: Restore by means of a different recovery host, to a

different ESX server under the same vCenter Server. After
recovery, turn on the VM.
nbrestorevm –vmw –C VM_to_restore –R absolute_path_to_rename_file
–vmproxy VMware_recovery_host –vmpo

The –vmpo option turns on the restored VM.

 NetBackup commands to back up and restore virtual machines 340
Using the nbrestorevm command to restore virtual machines into vSphere

Include the following in the rename file:

change esxhost to new_ESX_host

To restore the .vmdk files to a different datastore, include the following entry in the
rename file:
change /original_vmdk_path to /new_datastore/

Note: In the rename file, each change line must end with a carriage return, even if
the file contains only one line.

Note: The rename file may require other change entries, depending on your vSphere
environment. See the available entries under example G.

Note: Configure the recovery host as a VMware Access Host on the master server.
Communication between the master server and the backup host or recovery host
(client) should be established before the restore.

J. Restore to a different location, to a standalone ESX

server
nbrestorevm –vmw –C VM_to_restore –R absolute_path_to_rename_file

Include the following in the rename file:

change esxhost to new_ESX_host

change datacenter to None
change folder to None
change resourcepool to /ha_datacenter//host/new_ESX_host/resources
change datastore to new_datastore
change /original_vmdk_path to /new_datastore/
change network to new_network

Note: For restore to a standalone ESX server, the resource pool entry must include
the literal string ha-datacenter in the pool path.

Note: The rename file may require other change entries, depending on your
vSphere environment. See the available entries under example G.
 NetBackup commands to back up and restore virtual machines 341
Using the nbrestorevm command to restore virtual machines into vCloud Director

K. Restore to a different location in a vSphere cluster,

retaining the VM hardware version
nbrestorevm –vmw –C VM_to_restore –R absolute_path_to_rename_file
–vmkeephv

Include the following in the rename file:

change resourcepool to
/new_datacenter/host/cluster/resources/new_resource_pool

Note: The rename file may require other change entries, depending on your vSphere
environment. See the available entries under example G.

For further descriptions of the nbrestorevm options, see the command's man page
or the NetBackup Commands Reference Guide.
The following topics contain additional information:
See “Logs for troubleshooting the nbrestorevm command” on page 350.
See “The nbrestorevm -R rename file for restoring virtual machines” on page 346.
See “Using the nbdiscover command to search the virtual environment” on page 335.
See “Using the nbrestorevm command to restore virtual machines into vCloud
Director” on page 341.

Using the nbrestorevm command to restore virtual

machines into vCloud Director
To search the virtual environment for VMs, use the nbdiscover command.
See “Using the nbdiscover command to search the virtual environment” on page 335.
To restore the virtual machines, run nbrestorevm on the master server or backup
host. The nbrestorevm command is located in the following directory:
UNIX and Linux:
/usr/openv/netbackup/bin/nbrestorevm

Windows:
install_path\NetBackup\bin\nbrestorevm.exe

The following command options are used in these examples:

■ –vmw
Required for all VMware virtual machine restores.
 NetBackup commands to back up and restore virtual machines 342
Using the nbrestorevm command to restore virtual machines into vCloud Director

■ -vcd
Required for all virtual machine restores into vCloud.
■ –C virtual_machine_to_restore
The VM name must match the type of name that was selected in the Primary
VM identifier option of the backup policy. For example, if the VM was backed
up by its VM display name, specify the display name on the -C option.
■ -R absolute_path_to_rename_file
Restores to a different location. The entries that are required in the rename file
depend on the destination for the restore. See each example in this topic. Further
notes are available on the rename file:
See “The nbrestorevm -R rename file for restoring virtual machines” on page 346.
■ -vcdovw
Overwrites an existing vApp.
■ -vcdrte
Restores to an existing vApp. (If this option is set but the vApp does not exist,
the restore fails.)
■ -vcdtemplate
Restores a vApp as a template.
■ -vcdlfree
Restores to the datastore that has the most free space.
■ -vcdremv
Removes a vApp after capture to a template.
■ -vmid
Retains the BIOS UUID of the original virtual machine.
■ -vmInstanceId
Retains the Instance UUID of the original virtual machine (note that the Instance
UUID is a vCenter specific unique identifier of a virtual machine).

A: Restore to the original vApp in vCloud and overwrite

the VM.
nbrestorevm -vmw -vcd -C VM_to_restore -vcdrte -O -vmInstanceId

The -O option overwrites the VM.

B: Restore to the original vApp in vCloud, overwrite the

VM, and retain the BIOS UUID.
nbrestorevm -vmw -vcd -C VM_to_restore -vcdrte -O -vmid
 NetBackup commands to back up and restore virtual machines 343
Using the nbrestorevm command to restore virtual machines into vCloud Director

The -O option overwrites the VM, and the -vmid option retains the original Instance
UUID.

C: Restore to the original vApp in vCloud, overwrite the

VM, and retain the instance UUID.
nbrestorevm -vmw -vcd -C VM_to_restore -vcdrte -O -vmInstanceId

The -O option overwrites the VM, and the -vmInstanceId option retains the original
Instance UUID.

D: Restore to the original vCloud location to an existing

vApp. Remove the existing vApp and create a new vApp.
nbrestorevm –vmw –vcd –C VM_to_restore –vcdovw

The –vcdovw option deletes the vApp and the VMs inside it, and creates a new
vApp.

Note: For restore to the original location, you must use either the –vcdovw option
or the –O option.

E: Restore to the original vCloud location in a vApp

template.
nbrestorevm –vmw –vcd –C VM_to_restore –vcdovw

Note: To restore a VM that resides in a vApp template to its original location, use
the –vcdovw option. If you use –O option, the restore fails with status 2820.

F: Restore to a different vCloud vApp but to the same

vCloud organization.
nbrestorevm –vmw –vcd –C VM_to_restore –R absolute_path_to_rename_file
–vcdrte

For the -R rename file, create a text file with the following entries. These entries
describe the restore location. Replace the italicized values with the actual values:

change vcdvapp to new_vcdvapp

change orgvdc to new_organization_VDC

Note: In the rename file, each change line must end with a carriage return.
 NetBackup commands to back up and restore virtual machines 344
Using the nbrestorevm command to restore virtual machines into vCloud Director

Note: If you want to associate the restored VM with a different organization VDC,
use the change orgvdc entry to specify the organization VDC.

G: Restore to a different vCloud organization.

nbrestorevm –vmw –vcd –C VM_to_restore –R absolute_path_to_rename_file
–vcdrte

Include the following in the rename file:

change organization to new_organization

change orgvdc to new_organization_VDC
change vcdvapp to new_vcdvapp
change resourcepool to /new_datacenter/host/new_ESX_host/resources
/new_resource_pool
change datastore to new_vmx_datastore

Note: To create the organization on a different vCenter, include the –vmserver

vCenter_server option on the command. You must also include other change
directives, such as for datacenter and ESX server.

See “The nbrestorevm -R rename file for restoring virtual machines” on page 346.

H: Restore to a different vCloud Director.

nbrestorevm –vmw –vcd –C VM_to_restore –R absolute_path_to_rename_file
–vcdrte –vmserver vCenter_server

Include the following in the rename file:

change vcdserver to new_vcdserver

change organization to new_organization
change orgvdc to new_organization_VDC
change vcdvapp to new_vcdvapp
change vcdvmname to new_vcdvmname
change esxhost to new_ESX_host
change datacenter to /new_datacenter
change resourcepool to /new_datacenter/host/new_ESX_host/resources
/new_resource_pool
change datastore to new_vmx_datastore
 NetBackup commands to back up and restore virtual machines 345
Using the nbrestorevm command to restore virtual machines into vCloud Director

I: Restore to a different vCloud vApp and capture it as a

template in a catalog.
nbrestorevm –vmw –vcd –C VM_to_restore –R absolute_path_to_rename_file
–vcdtemplate

Include the following in the rename file:

change vcdvapp to new_vcdvapp

change vcdcatalog to new_vcdcatalog

The vApp is restored to the original organization and organization VDC. To associate
the vApp with a different organization VDC, include this directive:
change orgvdc to new_organization_VDC

Note: After the template is created, no other VMs can be added to the template.

J: Restore to a different vCloud vApp, capture the vApp

as a template, and remove the vApp.
nbrestorevm –vmw –vcd –C VM_to_restore –R absolute_path_to_rename_file
–vcdtemplate –vcdremv

Include the following in the rename file:

change vcdvapp to new_vcdvapp

change vcdcatalog to new_vcdcatalog

The -vcdremv option removes the non-template vApp; the new vApp template is
retained.

Note: After the template is created, no other VMs can be added to the template.

K: Restore a vCloud vApp template that has multiple virtual

machines.
First, restore a virtual machine to a non-template vApp:
nbrestorevm -vmw -vcd -C VM_to_restore -vcdrte

Run this command for each VM but do not restore the last VM.
Next, restore the last VM, create a vApp template, and copy all the restored VMs
into that template:
nbrestorevm -vmw -vcd -C VM_to_restore -R absolute_path_to_rename_file
-vcdtemplate -vcdremv -vcdlfree
 NetBackup commands to back up and restore virtual machines 346
Using the nbrestorevm command to restore virtual machines with Instant Recovery

Include the following in the rename file:

change vcdvapp to vApp_template

The change directive provides the name for the new vApp template into which the
VMs are to be restored.
The -vcdremv option removes the non-template vApp into which you initially restored
the virtual machines. The -vcdlfree option selects the datastore that has the largest
available space.

Note: After the template is created, no other VMs can be added to the template.

See “Allowing other servers or clients to restore virtual machines” on page 353.

Using the nbrestorevm command to restore virtual

machines with Instant Recovery
Instant Recovery for VMware is described in the "Restore virtual machines with
Instant Recovery" chapter of this guide.
See “About Instant Recovery for VMware” on page 237.
See “Notes on Instant Recovery for VMware” on page 242.
See “Instant Recovery options on the nbrestorevm command” on page 244.
See “Restoring a virtual machine with Instant Recovery for VMware” on page 250.

Using the nbrestorevm command to restore virtual

machine disks
Virtual Machine Disk restore for VMware is described in other topics.
See “About VMware virtual machine disk restore” on page 196.
See “Restoring VMware virtual machine disks by using NetBackup commands”
on page 204.

The nbrestorevm -R rename file for restoring

virtual machines
The -R option on the nbrestorevm command is for restoring a VM to a different
location. It identifies a text file that is called the rename file. This file describes the
 NetBackup commands to back up and restore virtual machines 347
The nbrestorevm -R rename file for restoring virtual machines

restore location by means of change directives. Each change directive provides

information about a VMware attribute or object that defines the VM when it is
restored. For example: the change directives can set the display name for the
restored VM, the name of the datastore, and the name of the ESX server.
Note the following about the -R rename file:
■ The -R option must specify the absolute path to the rename file.
■ The -R option and rename file are not required when you restore the VM to its
original location with all its original settings.
■ Each change directive in the -R rename file must end with a carriage return. If
the file contains only one change directive, that line must end with a carriage
return.
■ The change directives to include in the rename file depend on the VMware
environment.
Unlike the Backup, Archive, and Restore interface, the nbrestorevm command
does not have its own validation checks. When you use the nbrestorevm -R
option, you must include all the change entries that the VMware environment
requires. Otherwise, the restore does not succeed or the VM does not function
correctly.
■ The change directives can be placed in any order in the file.
■ The change directive should be entered exactly as shown in Table A-4, except
for the variable at the end (such as new_display_name). Replace that variable
with the new name.
■ If a vSphere object does not apply to the restore destination, enter None as its
value. For example: For restore to a standalone ESX, enter change datacenter
to None.

■ The rename file overrides the nbrestorevm -vmsn and -vmvmxd options:
■ If the rename file includes a change directive for a .vmdk path and
nbrestorevm includes the –vmvmxd option, the –vmvmxd option is ignored.
The change directive determines the .vmdk path of the restored VM.
■ If the rename file includes a change directive to designate a network for the
restored VM and nbrestorevm includes the –vmsn option, –vmsn is ignored.

Table A-4 -R rename file change directives

-R rename file directives Description and notes

change vmname to new_display_name Sets the display name for the restored VM. If this entry is omitted,
the original display name is used.
 NetBackup commands to back up and restore virtual machines 348
The nbrestorevm -R rename file for restoring virtual machines

Table A-4 -R rename file change directives (continued)

-R rename file directives Description and notes

change esxhost to new_ESX_host Designates the ESX server to restore to. This entry is required
when you restore to a different ESX server.

change datacenter to /new_datacenter Designates the datacenter in which to restore the VM. If the
destination does not have a datacenter, enter None as the value:

change datacenter to None

change folder to Designates the folder in which to restore the VM. If the destination
/new_datacenter/new_folder does not have a folder, enter None as the value:

change folder to None

change resourcepool to For restore to a different ESX server, specify a different resource
/new_datacenter/host/new_ESX_host/resources pool. (The original resource pool cannot be used to create a VM
/new_resource_pool on a different ESX server.) You must specify a resource pool by
including the following line in the rename file:

For restore to a standalone ESX server, the change

resourcepool entry must include the literal string
ha-datacenter in the path:

change resourcepool to
/ha_datacenter//host/new_ESX_host/resources

change datastore to new_vmx_datastore Designates the vmx datastore (configuration datastore) of the
restored VM. The vmx datastore contains the configuration files
that describe the virtual machine.

This change entry is required when you use the –vmvmxd option
on the nbrestorevm command. The -vmvmxd option allows the
vmdk files to be restored to the same datastore where the vmx
file is specified.
Note: See the change /original_vmdk_path to /new_datastore/
directive in this table.
 NetBackup commands to back up and restore virtual machines 349
The nbrestorevm -R rename file for restoring virtual machines

Table A-4 -R rename file change directives (continued)

-R rename file directives Description and notes

change /original_vmdk_path to Designates the original path to the .vmdk file, and the new
/new_datastore/ datastore for the .vmdk file of the restored VM. As a rule, this
change entry is required when you restore to a different datastore.
Depending on your VMware environment, this change entry may
take the following form:

change /original_datastore/original_VM_folder/original_vmdk.vmdk
to /new_datastore/
Note: If you use the -vmvmxd option on nbrestorevm to restore
the .vmdk files to the vmx datastore, do not use this change
directive. By specifying the datastore for the vmdk files, this change
directive overrides the -vmvmxd option.

Note: If the change vmdk directive specifies a different path for

some of the VM’s vmdk files, the other vmdk files are restored to
their original datastore.

To restore more than one vmdk file to a specified datastore:

change /first_vmdk_path to /new_first_datastore/

change /second_vmdk_path to /new_second_datastore/
change /nth_vmdk_path to /new_nth_datastore

change network to new_network Designates the new network for the restored VM. To designate
multiple networks, enter the following:

change network to new_network, new_network(n)

change organization to Designates the vCloud Director organization to restore to.

new_organization

change orgvdc to Designates the virtual datacenter in the vCloud Director

new_organization_virtual_datacenter organization to restore to.

change vcdserver to new_vcdserver Designates the vCloud Director server to restore to.

change vcdvapp to new_vcdvapp Designates the vCloud Director vApp to restore to.

change vcdvapptemplate to Designates the vCloud Director vApp template to restore to.
new_vcdvapptemplate

change vcdvmname to new_vcdvmname Sets the display name for the restored VM in vCloud Director.

change vcdcatalog to new_vcdcatalog Designates the vCloud Director catalog to restore to.

See “Logs for troubleshooting the nbrestorevm command” on page 350.

 NetBackup commands to back up and restore virtual machines 350
Logs for troubleshooting the nbrestorevm command

Logs for troubleshooting the nbrestorevm

command
The following logs may contain helpful messages on nbrestorevm.

Table A-5 Logs with messages on nbrestorevm

Log directory Contains the messages Resides on

on

Windows The restore operation, such Master server

as for status code 2820.
install_path\NetBackup\logs\bprd

Linux, UNIX

/usr/openv/netbackup/logs/bprd

Windows The restore operation, such Recovery host

as for status code 2820.
install_path\NetBackup\logs\bpVMutil

Linux, UNIX

/usr/openv/netbackup/logs/bpVMutil

Windows Incorrect usage of the Host where

nbrestorevm command, such nbrestorevm is run
install_path\NetBackup\logs\nbrestorevm
as omission of a required
Linux, UNIX option.
/usr/openv/netbackup/logs/nbrestorevm

If the restore job reaches the point of creating the virtual machine, the job's detailed
status should contain a specific error message. vSphere Client shows an attempt
to create the virtual machine and displays an associated failure message.
See “Restoring a virtual machine with Instant Recovery for VMware” on page 250.

Query format and examples for searching virtual

machine backups
From the master server, the nbdiscover command can search virtual machine
backups based on a query rule. Compose the query manually and enclose it in
double quotes.
For example: Search for backups of the virtual machines that have "vm1" as their
display names:
 NetBackup commands to back up and restore virtual machines 351
Query format and examples for searching virtual machine backups

nbdiscover "vmsearch:/;reqType=search?filter=VMDisplayName Equal

'vm1'"

Table A-6 Query format with examples

Element Description

vmsearch:/ Specifies that the query searches virtual machine backups.

reqType The request type. The following request types are supported:

■ search
Specifies that the query is used for search. Requires a query rule
■ browse
Displays the hierarchy of the virtual machines that were backed up. Requires the
viewType.
■ listFields
Lists the supported field names and their attributes.
■ listOps
Lists the operations that a field supports. Requires the fieldName option.

viewType Specifies the type of hierarchical view (vcloud or vsphere). Used with the browse option.

Example - Show the hierarchy of vCloud:

nbdiscover "vmsearch:/;reqType=browse;viewType=vcloud"

Example - Show the hierarchy of vSphere:

nbdiscover "vmsearch:/;reqType=browse;viewType=vsphere"

fieldName Specifies the name of the field for which supported operators are to be listed. Used with the
listOps.

Example - List supported operators for VMDisplayName

nbdiscover "vmsearch:/;reqType=listOps;fieldName=VMDisplayName"

filter The query rule. nbdiscover searches virtual machine backups based on the rule.

Example:

nbdiscover -noxmloutput "vmsearch:/;reqType=search?filter=ScheduleType

Equal '0'"

See the following examples with sample output.

Example 1: Search for the virtual machines that have full backups.
nbdiscover -noxmloutput "vmsearch:/;reqType=search?filter=ScheduleType
Equal '0'"
 NetBackup commands to back up and restore virtual machines 352
Query format and examples for searching virtual machine backups

+ "demovm%20(8c879391-2917-4416-9373-bea7ec727717)"
+ "smallvm2%20(10a16222-8532-46ff-bd6a-9661240392f4)"

Example 2: Search backups of the virtual machines that have the display names
that contain "vm".
nbdiscover -noxmloutput
"vmsearch:/;reqType=search?filter=VMDisplayName Contains 'vm'"

+ "demovm%20(8c879791-2917-4428-9373-bea7ec727717)"
+ "small_vm%20(33e85579-7279-411f-b2f9-9fb570583911)"
+ "small_vm_percent_%25%20(33e85579-7279-411f-b2f9-9fb570583911)"

Example 3: Search for backups of the vCloud virtual machines that have "test" in
their vApp names.
nbdiscover -noxmloutput "vmsearch:/;reqType=search?filter= vCDvApp
Contains 'test'"

+ "demovm%20(8c879791-2917-4428-8213-bea7ec727717)"
+ "small_vm%20(61e83479-7246-411f-b1f9-9fb570546755)"
+ "small_vm_percent_%25%20(61e83479-7246-411f-b1f9-9fb570546755)"

Example 4: Search the virtual machine backups that occurred before 08/24/2012
11:05:56 AM. Specify the backup time in UTC.
nbdiscover -noxmloutput "vmsearch:/;reqType=search?filter=BackupTime
Less '1345871512'"

+ "demovm%20(8c846791-2917-4428-9373-bea7ec727437)"
+ "small_vm%20(22e85813-7246-433f-b2f9-9fb570585387)"
+ "small_vm_percent_%25%20(22e85579-7246-433f-b2f9-9fb570585387)"

Example 5: Search backups for virtual machines where the vCloud organization is
named "QA" and the vApp name contains "demo".
nbdiscover -noxmloutput "vmsearch:/;reqType=search?filter=vCDOrg
Equal 'QA' AND vCDvApp Contains 'demo'"

+ "demovm%20(8c879791-2917-4428-9373-bea7ec727717)"

Example 6: Show the hierarchy of a vCloud environment. The output is in XML.

nbdiscover "vmsearch:/;reqType=browse;viewType=vcloud"

<Start Iteration="vmsearch:/;reqType=browse;viewType=vcloud">
<VCDSERVER>
<NBU>
<NAME>hypervm1.acme.com</NAME>
 NetBackup commands to back up and restore virtual machines 353
Allowing other servers or clients to restore virtual machines

</NBU>
<VCDORG>
<NBU>
<NAME>Test Cloud</NAME>
</NBU>
<VCDORGVDC>
<NBU>
<NAME>Test Org</NAME>
</NBU>
<VCDVAPP>
<NBU>
<NAME>TestvApp</NAME>
</NBU>
</VCDVAPP>
</VCDORGVDC>
</VCDORG>
</VCDSERVER>
<StatusMsg NBUStatus="0" Severity="0"></StatusMsg>
</START>

This search found the following:

■ A vCloud Director server: hypervm1.acme.com
■ An organization: Test Cloud
■ A virtual datacenter: Test Org
■ A vApp: TestvApp
See “Using the nbrestorevm command to restore virtual machines into vCloud
Director” on page 341.

Allowing other servers or clients to restore virtual

machines
You can configure NetBackup to allow nbrestorevm to run from another host (such
as another server, client, or virtual machine). The enabled host can use nbrestorevm
to restore the virtual machine. The host must be configured as a peer of the
NetBackup master server that performed the backup.
You can configure the peer in either of the following ways:
■ In the NetBackup Administration Console, click Host Properties > Master
Servers > double click the master server > Servers. Add the peer on the
Additional servers tab.
 NetBackup commands to back up and restore virtual machines 354
Allowing other servers or clients to restore virtual machines

■ Add an empty peer_name file in the following location:

UNIX, Linux
/usr/openv/netbackup/db/altnames/peer_name
Windows
install_path\NetBackup\db\altnames\peer_name
The name of the file is the name of the server or client to be granted restore
privileges.
For instructions, see the topics on allowing clients to perform redirected restores,
in the NetBackup Administrator's Guide, Vol I.
 Appendix B
Configuring services for
NFS on Windows
This appendix includes the following topics:

■ About installing and configuring Network File System (NFS) for Granular
Recovery Technology (GRT)

■ About configuring services for NFS on Windows 2012 or 2016 (NetBackup for
VMware)

■ About configuring services for NFS on Windows 2008 and 2008 R2 (NetBackup
for VMware)

■ Disabling the Server for NFS (NetBackup for VMware)

■ Disabling the Client for NFS on the media server (NetBackup for VMware)

■ Configuring a UNIX media server and Windows backup or restore host for
Granular Recovery Technology (NetBackup for VMware)

■ Configuring a different network port for NBFSD (NetBackup for VMware)

About installing and configuring Network File

System (NFS) for Granular Recovery Technology
(GRT)
NetBackup Granular Recovery leverages Network File System, or NFS, to read
individual objects from a database backup image. Specifically, the NetBackup client
uses NFS to extract data from the backup image on the NetBackup media server.
The NetBackup client uses “Client for NFS” to mount and access a mapped drive
 Configuring services for NFS on Windows 356
About configuring services for NFS on Windows 2012 or 2016 (NetBackup for VMware)

that is connected to the NetBackup media server. The NetBackup media server
handles the I/O requests from the client through NBFSD.
NBFSD is the NetBackup File System (NBFS) service that runs on the media server.
NBFSD makes a NetBackup backup image appear as a file system folder to the
NetBackup client over a secure connection.

About configuring services for NFS on Windows

2012 or 2016 (NetBackup for VMware)
For instant recovery of virtual machines, the Services for Network File System (NFS)
must be installed on Windows media servers.

Table B-1 Configuring NFS in a Windows 2012 or 2016 environment

Action Description

Configure NFS on the On the media server do the following:

media server.
■ Stop and disable the ONC/RPC Portmapper service, if it exists.
■ Enable NFS.
See “Enabling Services for Network File System (NFS) on a Windows 2012 or 2016 media
server (NetBackup for VMware)” on page 356.
■ Stop the Server for NFS service.
See “Disabling the Server for NFS (NetBackup for VMware)” on page 368.
■ Configure the portmap service to start automatically at server restart.
Issue the following from the command prompt:
sc config portmap start= auto
This command should return the status [SC] ChangeServiceConfig SUCCESS.

Configure NFS on the On the restore host, do the following:

restore host.
■ Enable NFS.
See “Enabling Services for Network File System (NFS) on a Windows 2012 or 2016
restore host (NetBackup for VMware)” on page 360.
■ Stop the Server for NFS service.
See “Disabling the Server for NFS (NetBackup for VMware)” on page 368.

Enabling Services for Network File System (NFS) on a Windows

2012 or 2016 media server (NetBackup for VMware)
To perform VM instant recovery with a Windows 2012 or 2016 media server, you
must enable Services for Network File System. When this configuration is completed,
you can disable any unnecessary NFS services.
 Configuring services for NFS on Windows 357
About configuring services for NFS on Windows 2012 or 2016 (NetBackup for VMware)

See “About configuring services for NFS on Windows 2012 or 2016 (NetBackup
for VMware)” on page 356.
To enable Services for Network File System (NFS) on a Windows 2012 or 2016
media server
1 Open the Server Manager.
2 From the Manage menu, click Add Roles and Features.

3 In the Add Roles and Features Wizard, on the Before You Begin page, click
Next.
 Configuring services for NFS on Windows 358
About configuring services for NFS on Windows 2012 or 2016 (NetBackup for VMware)

4 On the Select installation type page, select Role-based or feature-based

installation.

5 Click Next.
6 On the Server Selection page, click Select a server from the server pool
and select the server. Click Next.
 Configuring services for NFS on Windows 359
About configuring services for NFS on Windows 2012 or 2016 (NetBackup for VMware)

7 On the Server Roles page, expand File and Storage Services and File and
iSCSI Services.
8 Click File Server and Server for NFS. When you are prompted, click Add
Features. Click Next.
 Configuring services for NFS on Windows 360
About configuring services for NFS on Windows 2012 or 2016 (NetBackup for VMware)

9 If the media server is also a restore host, on the Features page, click Client
for NFS. Click Next.

10 On the Confirmation page, click Install.

11 Disable any unnecessary services, as follows:
■ If you have a single host that functions as both the media server and the
restore host, you can disable the Server for NFS.
■ For a host that is only the NetBackup media server, you can disable the
Server for NFS and the Client for NFS.

12 Make sure that the portmap service is started and that its startup mode is set
to auto.

Enabling Services for Network File System (NFS) on a Windows

2012 or 2016 restore host (NetBackup for VMware)
To perform VM instant recovery with a Windows restore host, you must enable
Services for Network File System. When this configuration is complete, you can
disable any unnecessary NFS services.
 Configuring services for NFS on Windows 361
About configuring services for NFS on Windows 2012 or 2016 (NetBackup for VMware)

To enable Services for Network File System (NFS) on a Windows 2012 or 2016
restore host
1 Open the Server Manager.
2 From the Manage menu, click Add Roles and Features.

3 In the Add Roles and Features Wizard, on the Before You Begin page, click
Next.
 Configuring services for NFS on Windows 362
About configuring services for NFS on Windows 2012 or 2016 (NetBackup for VMware)

4 On the Select installation type page, select Role-based or feature-based

installation.

5 Click Next.
6 On the Server Selection page, click Select a server from the server pool
and select the server. Click Next.

7 On the Server Roles page, click Next.

 Configuring services for NFS on Windows 363
About configuring services for NFS on Windows 2008 and 2008 R2 (NetBackup for VMware)

8 On the Features page, click Client for NFS. Click Next.

9 On the Confirmation page, click Install.

About configuring services for NFS on Windows

2008 and 2008 R2 (NetBackup for VMware)
For instant recovery of virtual machines, the Services for Network File System (NFS)
must be installed on the Windows restore host and media servers.

Note: Before you configure NFS on the media server or restore host, look for the
ONC/RPC Portmapper service. If it exists, stop it and disable it. Otherwise, the
configuration of NFS Services for Windows fails.
 Configuring services for NFS on Windows 364
About configuring services for NFS on Windows 2008 and 2008 R2 (NetBackup for VMware)

Table B-2 Configuring NFS in a Windows 2008 or 2008 R2 environment

Action Description

Configure NFS on the On the media server do the following:

media server.
■ Stop and disable the ONC/RPC Portmapper service.
■ Use the Windows Server Manager to enable NFS.
See “Enabling Services for Network File System (NFS) on Windows 2008 or 2008 R2
(NetBackup for VMware)” on page 364.
■ Stop the Server for NFS service.
See “Disabling the Server for NFS (NetBackup for VMware)” on page 368.
■ Stop the Client for NFS service.
See “Disabling the Client for NFS on the media server (NetBackup for VMware)”
on page 369.
Note: If the media server is also a restore host, do not disable the Client for NFS.
■ Configure the portmap service to start automatically at server restart. Enter the following
from the command prompt:
sc config portmap start= auto
This command should return the status [SC] ChangeServiceConfig SUCCESS.

Configure NFS on the On the restore host, do the following:

restore host.
■ Enable NFS.
See “Enabling Services for Network File System (NFS) on Windows 2008 or 2008 R2
(NetBackup for VMware)” on page 364.
■ Stop the Server for NFS service.
See “Disabling the Server for NFS (NetBackup for VMware)” on page 368.

On the restore host, The hot fix is available at the following location:
install the hot fix for
http://support.microsoft.com/kb/955012
Client for NFS.
Note: Important Windows Vista hot fixes and Windows Server 2008 hot fixes are included
in the same packages. However, the hot fix Request page lists only Windows Vista. For the
hot fix package for either operating system, select the hot fix that is listed under Windows
Vista on the page. Always refer to the Applies To section in articles to determine the actual
operating system for each hot fix.

Enabling Services for Network File System (NFS) on Windows 2008

or 2008 R2 (NetBackup for VMware)
To perform VM instant recovery with a Windows restore host, you must enable
Services for Network File System. When this configuration is completed on the
media server and the restore host, you can disable any unnecessary NFS services.
 Configuring services for NFS on Windows 365
About configuring services for NFS on Windows 2008 and 2008 R2 (NetBackup for VMware)

See “About configuring services for NFS on Windows 2008 and 2008 R2 (NetBackup
for VMware)” on page 363.
To enable Services for Network File System (NFS) on Windows 2008 or 2008
R2
1 Open the Server Manager.
2 In the left pane, click Roles and, in the right pane, click Add Roles.

3 In the Add Roles Wizard, on the Before You Begin page, click Next.
 Configuring services for NFS on Windows 366
About configuring services for NFS on Windows 2008 and 2008 R2 (NetBackup for VMware)

4 On the Select Server Roles page, under Roles, check the File Services
check box.

5 Click Next.
6 On the Files Services page, click Next.
7 On the Select Role Services page, uncheck File Server.
 Configuring services for NFS on Windows 367
About configuring services for NFS on Windows 2008 and 2008 R2 (NetBackup for VMware)

8 Check Services for Network File System.

9 Click Next and complete the wizard.

10 For each host in your configuration, do one of the following.
■ If you have a single host that functions as both the media server and the
restore host, you can disable the Server for NFS.
■ For a host that is only the NetBackup media server, you can disable the
Server for NFS and the Client for NFS.

Note: Make sure that the portmap service is started and that its startup
mode is set to auto.

See “About configuring services for NFS on Windows 2008 and 2008 R2
(NetBackup for VMware)” on page 363.
 Configuring services for NFS on Windows 368
Disabling the Server for NFS (NetBackup for VMware)

■ For a host that is only a restore host, you can disable the Server for NFS.

Disabling the Server for NFS (NetBackup for

VMware)
To disable the Server for NFS
1 Open the Server Manager.
2 In the left pane, expand Configuration.
3 Click Services.

4 In the right pane, right-click on Server for NFS and click Stop.
5 In the right pane, right-click on Server for NFS and click Properties.
 Configuring services for NFS on Windows 369
Disabling the Client for NFS on the media server (NetBackup for VMware)

6 From the Startup type list in the Server for NFS Properties dialog box, click
Disabled.

7 Click OK.
8 Do this procedure for each media server and for the restore host.

Disabling the Client for NFS on the media server

(NetBackup for VMware)
After you enable Services for Network File System (NFS) on a host that is only a
NetBackup media server, disable the Client for NFS.
To disable the Client for NFS on the NetBackup media server
1 Open the Server Manager.
2 In the left pane, expand Configuration.
 Configuring services for NFS on Windows 370
Disabling the Client for NFS on the media server (NetBackup for VMware)

3 Click Services.

4 In the right pane, right-click on Client for NFS and click Stop.
5 In the right pane, right-click on Client for NFS and click Properties.
 Configuring services for NFS on Windows 371
Configuring a UNIX media server and Windows backup or restore host for Granular Recovery Technology
(NetBackup for VMware)

6 From the Startup type list in the Client for NFS Properties dialog box, click
Disabled.

7 Click OK.

Configuring a UNIX media server and Windows

backup or restore host for Granular Recovery
Technology (NetBackup for VMware)
For backups and restores that use Granular Recovery Technology (GRT), perform
the following configuration if you use a UNIX media server and Windows restore
host:
■ Confirm that your media server is installed on a platform that supports granular
recovery. See the NetBackup Software Compatibility List:
No other configuration is required for the UNIX media server.
■ Enable or install NFS on the restore host.
See “Enabling Services for Network File System (NFS) on a Windows 2012 or
2016 restore host (NetBackup for VMware)” on page 360.
 Configuring services for NFS on Windows 372
Configuring a different network port for NBFSD (NetBackup for VMware)

See “Enabling Services for Network File System (NFS) on Windows 2008 or
2008 R2 (NetBackup for VMware)” on page 364.
■ You can configure a different network port for NBFSD.

Configuring a different network port for NBFSD

(NetBackup for VMware)
NBFSD runs on port 7394. If another service uses the standard NBFSD port in your
organization, you can configure the service on another port. The following procedures
describe how to configure a NetBackup server to use a network port other than the
default.
To configure a different network port for NBFSD (Windows server)
1 Log on as administrator on the computer where NetBackup server is installed.
2 Open Regedit.
3 Open the following key:

HKEY_LOCAL_MACHINE\SOFTWARE\Veritas\NetBackup\CurrentVersion\Config

4 Create a new DWORD value named FSE_PORT.

5 Right-click on the new value and click Modify.
6 In the Value data box, provide a port number between 1 and 65535.
7 Click OK.
To configure a different network port for NBFSD (UNIX server)
1 Log on as root on the computer where NetBackup server is installed.
2 Open the bp.conf file.
3 Add the following entry, where XXXX is an integer and is a port number between
1 and 65535.
FSE_PORT = XXXX
See “Configuring a UNIX media server and Windows backup or restore host for
Granular Recovery Technology (NetBackup for VMware)” on page 371.
 Appendix C
The Reuse VM selection
query results option
This appendix includes the following topics:

■ About the Reuse VM selection query results option

■ The effect of virtual machine discovery on vCenter

About the Reuse VM selection query results

option
This topic applies to NetBackup for VMware.
The NetBackup Test Query screen lists the virtual machines that NetBackup
discovered in your virtual environment. Because the automatic selection feature is
dynamic, later changes in the environment may affect which virtual machines match
the query rules. For example: if virtual machines are added later, the current test
results may not be identical to the virtual machines that are selected when the
backup runs.
During the period you specify on Reuse VM selection query results for, NetBackup
reuses the current list of virtual machines as the backup list. It does not consult the
Query Builder or rediscover virtual machines.
The less often your virtual machine environment undergoes changes, the more
advantageous it may be to reuse the list of virtual machines for backups. In large
environments, discovery of virtual machines takes time and consumes resources
on the vCenter server.
Note the following about the Reuse VM selection query results option:
■ Determines how long the query results are reused (that is, how often the list of
discovered virtual machines is refreshed).
 The Reuse VM selection query results option 374
The effect of virtual machine discovery on vCenter

■ Controls how often NetBackup performs discovery on the vCenter server.

For the environments that contain many virtual machines, the discovery process
may increase the load on the vCenter server.
■ Has no effect on the NetBackup policy execution schedule.
■ Is invalidated if the query rules are changed or if the policy attributes that affect
discovery are changed. In that case, NetBackup rediscovers virtual machines
the next time the policy runs.
The following topic describes the policy attributes that affect discovery:
See “Effect of Primary VM identifier parameter on Selection column in Test
Query results” on page 142.
For example: assume that the Reuse VM selection query results for option is set
to 8 hours and your query selects turned-on virtual machines. If additional virtual
machines are turned on during the 8-hour period, they are not added to the policy's
backup list. The policy backs up the virtual machines that were last added to the
list (such as when the policy was created). After 8 hours, at the next execution of
the policy, the recently turned on virtual machines are discovered and added to the
backup list.
If the next backup occurs before the Reuse period expires, and a virtual machine
was renamed during the Reuse period, NetBackup backs up the renamed virtual
machine under its original name. Because the reuse period has not expired,
NetBackup does not rediscover virtual machines and therefore cannot identify the
virtual machine by its new name. (NetBackup identifies the virtual machine by its
instance UUID.) To have backed up the virtual machine with its new name, the
Reuse period should have been set to a shorter interval.
Note: The virtual machines that have been selected for backup are not displayed
on the policy Clients tab. To see which virtual machines NetBackup has selected
and backed up, refer to the following topics.
See “The effect of virtual machine discovery on vCenter” on page 374.
See “Using the Activity Monitor to monitor virtual machine backups” on page 153.
See “Using OpsCenter to report on virtual machine backups” on page 157.
See “Options for selecting VMware virtual machines” on page 109.
See “Test Query screen for VMware” on page 139.

The effect of virtual machine discovery on vCenter

In all but the largest environments, it may be advantageous to set the Reuse VM
selection query results for option so that discovery occurs more often.
 The Reuse VM selection query results option 375
The effect of virtual machine discovery on vCenter

If changes occur to virtual machine configuration (such as adding, deleting, or

moving vmdk files), it may be necessary to set Reuse VM selection query results
for to 0. With a setting of 0, NetBackup rediscovers the virtual machines and their
configuration each time the policy runs.

Note: If the vmdk files are reconfigured and the next backup runs without
rediscovery, NetBackup is not aware of the vmdk changes. It attempts to back up
the virtual machines in their previous configuration. The result may be an incorrect
backup.
 Appendix D
Backup of VMware raw
devices (RDM)
This appendix includes the following topics:

■ About VMware raw device mapping (RDM)

■ Configurations for backing up RDMs

■ About alternate client backup of RDMs

■ Requirements for alternate client backup of RDMs

■ Configuring alternate client backup of RDMs

About VMware raw device mapping (RDM)

This topic applies to NetBackup for VMware.
VMware raw device mapping mode (RDM) allows a virtual machine to directly
access physical disks. With raw device mapping, a VMware virtual machine can
use large storage devices such as disk arrays. Access to the data on an RDM disk
is faster than to a fully virtualized disk (vmdk file). An RDM disk can be locally
attached to the ESX server or configured on a Fibre Channel SAN.
NetBackup supports the disk arrays that are configured on a virtual machine as
RDMs.

Note: NetBackup cannot back up the RDM by means of a VMware backup host.

For notes and restrictions on NetBackup support for VMware RDM, see the following
Veritas tech note:
 Backup of VMware raw devices (RDM) 377
Configurations for backing up RDMs

Support for NetBackup in virtual environments

http://www.veritas.com/docs/000006177
See “Configurations for backing up RDMs” on page 377.
See “About alternate client backup of RDMs” on page 377.

Configurations for backing up RDMs

This topic applies to NetBackup for VMware and VMware raw device mapping mode
(RDM).
You can use either of the following NetBackup configurations to back up disk arrays
as RDMs:
■ Without Snapshot Client: Install a NetBackup client on the virtual machine. You
can configure NetBackup to back up the virtual machine and any RDMs as if
the client was installed on a physical host. Without Snapshot Client software on
the virtual machine, the features of Snapshot Client are not available. (This
configuration is not discussed in this NetBackup for VMware guide.)
■ With Snapshot Client: Install a NetBackup client and Snapshot Client software
on the virtual machine. Configure an alternate client backup.
See “About alternate client backup of RDMs” on page 377.

About alternate client backup of RDMs

This topic applies to NetBackup for VMware and VMware raw device mapping mode
(RDM).
Alternate client backup of an RDM consists of the following:
■ The RDM disk array contains the data to be backed up. Another host containing
NetBackup client software and Snapshot Client software must have access to
the disk array. This host is the alternate client. In this configuration, the virtual
machine is called the primary client.
■ A snapshot of the data is created on the disk array and is mounted on the
alternate client. The alternate client creates a backup image from the snapshot,
using original path names, and streams the image to the NetBackup media
server.
■ The alternate client handles the backup I/O processing; the backup has little or
no effect on the virtual machine. The media server reads the snapshot data from
the alternate client and writes the data to storage.
 Backup of VMware raw devices (RDM) 378
Requirements for alternate client backup of RDMs

■ The virtual machine and alternate client must be running the same operating
system, volume manager, and file system. For each of these I/O system
components, the alternate client must be at the same level as the primary client,
or higher level.
For complete requirements on alternate client backup, refer to the Policy
configuration chapter of the NetBackup Snapshot Client Administrator's Guide.
See “Requirements for alternate client backup of RDMs ” on page 378.

Requirements for alternate client backup of RDMs

This topic applies to NetBackup for VMware and VMware raw device mapping mode
(RDM).
To use NetBackup Snapshot Client to back up an RDM, note the following:
■ RDM devices must be configured in physical compatibility mode. You select this
mode when you create the RDM. Physical compatibility mode is not configured
in NetBackup.
For an introduction to RDM, refer to your VMware documentation. For example,
see the following VMware document:
ESX Server 3 Configuration Guide
■ NetBackup may require certain OS and array configuration, depending on the
guest OS and the array. For details, see the disk arrays chapter of the NetBackup
Snapshot Client Administrator's Guide.
■ NetBackup client software must be installed on the virtual machine.
■ The requirements for the NetBackup for VMware feature (a backup host and
the VMware snapshot method) do not apply to backups of RDM disk arrays. To
back up RDM disk arrays, you must configure a Snapshot Client alternate client
backup.
See “Configuring alternate client backup of RDMs” on page 378.

Configuring alternate client backup of RDMs

This topic applies to NetBackup for VMware and VMware raw device mapping mode
(RDM).
This procedure highlights key points in creating a NetBackup alternate client backup
of a disk array that is configured as an RDM. For broader details on disk array
configuration, refer to the NetBackup Snapshot Client Administrator's Guide.
 Backup of VMware raw devices (RDM) 379
Configuring alternate client backup of RDMs

To create an alternate client policy for a disk array that is configured as an

RDM
1 Select a policy type that is appropriate for the OS of the virtual machine and
for the type of data to back up.
Refer to the Policy configuration chapter of the NetBackup Snapshot Client
Administrator's Guide.
2 On the policy Attributes tab in the Administration Console, click Perform
snapshot backups and Perform off-host backup.
The following is a detail from the Attributes tab.

3 Select Alternate Client from the Use pull-down. Do not select VMware backup
host.
In the Machine field, enter the name of the host that is configured as an off-host
backup computer (the alternate client).
4 Click Options.
5 Select a snapshot method.
The VMware method does not apply to alternate client backup and is not
available in the list.
 Backup of VMware raw devices (RDM) 380
Configuring alternate client backup of RDMs

Select a snapshot method that is appropriate for the volume or array. For
example:
■ The HP_EVA_Snapclone method or other EVA method for an HP EVA
array.
■ The EMC_CLARiiON_Snapview_Clone or other CLARiiON method for an
EMC CLARiiON array.
■ FlashSnap.
For FlashSnap, the following must be installed: VxVM 3.2 or later for UNIX,
VxVM 4.0 or later for Linux and AIX, or VxVM 3.1 or later for Windows.
Also, volumes must be configured over the primary host’s disks. The VxVM
FlashSnap license must also be installed.
■ VSS (for Windows guest operating systems only).
The array may require additional OS and NetBackup configuration as described
in the disk arrays chapter of the NetBackup Snapshot Client Administrator's
Guide. The same guide contains more details on snapshot methods.
6 If required by an array snapshot method that you selected in the previous step,
specify the Snapshot Resources.
7 In the policy’s Clients list, select the virtual machine on which the array is
configured as an RDM.
8 In the policy’s Backup Selections tab, specify the disk that you want to back
up, or the files or volumes that reside on the disk.
See “Requirements for alternate client backup of RDMs ” on page 378.
See “About VMware raw device mapping (RDM)” on page 376.
See “NetBackup for VMware: notes and restrictions” on page 27.
 Index

Symbols B
156 (status code) 303 backup
fault tolerant VM 37
A host 15
how to configure 68
Accelerator 160
how to perform 151
and the NetBackup catalog 164
job hangs 309
logs for 165
monitoring with Activity Monitor 153
messages in the backup job details log 165
monitoring with OpsCenter 157
policy attribute 161
process overview 20
replacing image size with network-transferred
reduce size of 147
data in command output 169
backup host 74
reporting network-transferred data in command
adding to configuration 44
output 166
check if supported 289
Activity Monitor
backup media server 16, 18, 74
to monitor VM backups 153
backups
Advanced Mode
replacing Accelerator image size with
Query Builder 116
network-transferred data in command
ALL_LOCAL_DRIVES 303
output 169
alternate client backup
reporting network-transferred data for
configure for RDM 378
Accelerator 166
of RDM 377
Basic Mode
alternate location for restore 188
edit rule in Query Builder 115
AND vs OR 122
Query Builder 112
Join field in Query Builder 117
best practices 283
automatic selection of virtual machines 72
BIOS
about 102
reconfigure for booting 312
Advanced Mode 116
blacklist
Basic Mode 112
for dynamic multipathing on Linux 35
cache file for 373
BLIB 75, 99, 148
Clients tab 109
block-level incremental backup. See BLIB
examples 102, 118
blocks
notes on 105
exclude deleted 75
Query Builder fields 129
browse virtual machines 91
requirements 108
browsing timeout 299
task overview 108
testing 139
with multiple policies 121 C
cache file
for browsing virtual machines 93
catalog
and use of Accelerator 164
 Index 382

change VMware hostname in NetBackup 50 encryption of snapshot data transport 79, 184
Changed Block Tracking 148 equipment 289
Client Reports escape character
using OpsCenter 157 use of in Query Builder 138
Clients tab ESX server 16, 45
automatic selection of virtual machines 109 Exclude deleted blocks option 75
description of fields for queries 129 exclude disks from backup 88
code 156 error 303 exclude list
compound query rules 124 speed up discovery of virtual machines 93
configuration ext file systems, Linux 177
of auto selection of virtual machines 109, 112 extended attributes, restore of 177
overview for backup of VMs 24
configuration datastore 176 F
credentials
failed virtual machines
adding for VMware 45
in test query 141
troubleshooting 302
fault tolerant VMs 37
cross-platform restore 173
FIFO files, Linux 177
Cumulative Incremental Backup 71
file system
Linux, restore of 177
D firewall 309
datacenter
for restore 191 G
datacenter folders
grouping rules in Query Builder 116, 124
rules in Query Builder 126
guest customizations
datastore
ensuring their restore 320
in a datastore cluster 192
restore of 192
deduplication 284 H
delete host name 76, 94
vSphere Client snapshot 149 changing in NetBackup 50
Differential Incremental Backup 71 host properties
disable virtual machine quiesce 81 for VMware 44
Discovered Timestamp 94 hotadd transport mode 79, 184
disk array notes on 40
configured as RDM 378 hypervisor 21
display name 76, 94
DNS name 76 I
drives Ignore diskless VMs 81, 305
and reboot failure 312 in-place disk restore 197–198, 212, 216, 220–221,
dynamic multipathing 321–322
in a SAN environment 35 incremental backup
about 99
E requirements 71
edit rules in Query Builder 115 setting up 100
EMM database independent disk 22, 95, 305
remove VM host name from NetBackup 50 how to back up 28
Enable block-level incremental backup 75, 148 installing and configuring Network File System
encrypted drives 27 (NFS) 356
instant recovery 238
 Index 383

Intelligent Policy 102 Network File System (NFS), described 356

invalid client 312 NFC connections 309
IsSet operator examples 120 NFS share, Linux 177
NTFS file system
J for USB drives 27
job hangs 309
Join field in Query Builder 117 O
operators
K in Query Builder 137
OpsCenter
keywords
to monitor VM backups 157
in Query Builder 130
optimized storage 147
order of operations
L Query Builder 122
Last Update 93 rules in Query Builder 116
Linux overview
notes on 34 configuration steps 24
notes on restore 177 of backup process 20
quiesce 34 restore procedure 223
logging overwrite virtual machine 185
Accelerator 165
how to create folders for 292
VMware 291
P
parentheses
vSphere 300
for grouping rules in Query Builder 116
VxMS 292
for rules in Query Builder 124
LVM2 volume 34, 177
path length
maximum for Linux/UNIX files 34
M Perform Recovery dialog 196
MAC address 177 policy
manual backup 151 and cache file for auto selection of VMs 373
monitor backups configure for auto VM selection 112
with Activity monitor 153 for full virtual machine backup 68
with OpsCenter 157 multiple, for auto VM selection 121
mount point types 69
missing 311 wizard for creating 68
multiple policies Policy Configuration wizard 68
for auto selection of VMs 121 port number 49, 302
precedence 122
N prerequisites 26
nbdiscover command 139 Primary VM identifier
to test a query 106, 109, 111 setting 76
nbemmcmd 50 Primary VM identifier parameter
nbfsd port 372 and IsSet operator 120
NetBackup Accelerator. See Accelerator and multiple policies 121
Network Block Device (NBD) transport 79, 184 and testing query rules 141–142
Network Block Device (NBD) transport with SSL 79, and VM Name column in Test query 144
184 privileges 303
Network Connections dialog 195
 Index 384

Q resource pool
Query Builder for virtual machine restore 192
and Primary VM identifier parameter 142 resource pools
cache file for results 373 rules in Query Builder 125
configuration in Advanced Mode 116 restore
configuration in Basic Mode 112 about restoring individual virtual machines 204
description of fields 129 from backup 182, 188
edit rule in 115 guest customizations 320
enter rules manually 116 how to 179, 198
escape character 138 individual files 224
examples 118 into vCloud 276
IsSet operator examples 120 job hangs 309
keywords 130 maximum path length for Linux/UNIX files 34
operators 137 notes on 173
order of operations 122 procedure overview 223
quotation marks 138 virtual machine 179, 198
testing rules for 139, 141 vmdk 204
to select virtual machines for backup 102 restore vmdk
values 138 Changed Block Tracking warning during
wildcards 138 subsequent backup 322
with multiple policies 121 create restore parameters file 207
query builder determine backup ID from which to restore 205
to filter virtual machines for backup 72 restoring individual 213
quiesce validate restore parameters file 212
Linux file system 177 reuse file
Linux, notes on 34 for Query Builder results 373
quiesce virtual machine 81 Reuse VM selection query results for 373
quotation marks rules in Query Builder
use of in Query Builder 138 about 104
configuring 112
editing 115
R examples 104
RDM 22, 305 examples with parentheses 124
about 376 for datacenter folders 126
alternate client backup of 377–378 for resource pools 125
configurations supported 377 IsSet operator examples 120
reboot failure 312 manual entry of 116
Recovery Destination dialog 182 order of evaluation 116
recovery host 15, 183 order of operations 122
Recovery Options dialog 183, 189 testing 139
recovery wizard. See restore from backup. See testing failued 141
restore into vCloud run a backup 151
reduce size of backup 147
release level
using mixed levels 236 S
remove VMware host name from NetBackup EMM SAN transport mode 79, 183
database 50 SCSI drives
reports on backups and reboot failure 312
using OpsCenter 157 Select automatically through VMware Intelligent Policy
query 72, 102
 Index 385

Select ESX server dialog 193 V

Select Folder dialog 193 validate
server name credentials 45
changing in NetBackup 50 validate virtualization server's certificate 54
size of backups values
reducing 147 in Query Builder 138
snapshot vApp
vSphere Client browsing virtual machines within 93
how to delete 149 for virtual machine restore 192
socket files, Linux 177 vCenter server 16
SRM environment and replication 81, 305 vCloud
SSL encryption of snapshot data 79, 184 alternate location for restore 276
start a backup 151 NetBackup credentials for 48
status code 156 303 Virtual machine quiesce option 81
Storage Destination dialog 194 Virtual Machine recovery wizard
storage optimization 27, 147 restore to alternate location 188
how to delete snapshot for 149 vCloud restore to alternate location 276
Storage vMotion virtual machines
backup conflict 307 about restoring individual 204
auto selection from reuse file 373
T browsing 91
tag associations configure backup policy for 68
support 103 configure for auto selection of 112
use 103 how to back up 151
template VM 27 monitor backups of 153, 157
terminology 20 notes on query builder 105
Test Query screen 139 requirements for auto selection 108
and Primary VM identifier parameter 142, 144 Select automatically through VMware Intelligent
failed virtual machines 141 Policy query 72, 102
nbdiscover as alternative 106, 109, 111 testing auto selection of 139
thin provisioned disks 186 virtual network
timeout select for restore 195
browsing 299 VirtualCenter server 16
timeouts VM Name column in Test Query screen 144
vSphere 300 vmdk file 23
transport modes backup fails if none assigned 81, 305
and hung job 309 vmdk restore
for restore 183 about 196
hotadd 40 overwrite during restore 203, 222
troubleshooting process 204
tips 288 restore parameters file 214
troubleshooting 321–322
VMware
U backup environment 15
USB drives backup host 44
as the snapshot mount point 27 backup media server 18
UUID 76, 94 backups 151
configuration overview 24
credentials 45
 Index 386

VMware (continued)
ESX server 16, 45
Exclude Disk tab 86
host properties 44
introduction 13
logging 291
main features 13
policy configuration 68
required components 15
restores 179, 198
terminology 20
troubleshooting 291
VMware Intelligent Policy 102
VMware Intelligent Policy query 72
VMware policy 69
VMware policy tab 69
VMware Tools 23
VMX directory 176
vmx-07 71
vNetwork Distributed Switch 195
Volume Manager volumes 100
vSphere Client snapshot
how to delete 149
vStorage 23
incrementals 100
restore into vCloud 276
VxMS logging 292

W
whitelist
for dynamic multipathing on Linux 35
wildcards
use of in Query Builder 138
wizard
to create backup policy 68
to recover virtual machine 182, 276