Novell - 3005 - Advanced Novell Network Management Netware 6 - Instructor Guide v1.0
Novell - 3005 - Advanced Novell Network Management Netware 6 - Instructor Guide v1.0
Network Management
Novell Education www.novell.com
COURSE 3005
INSTRUCTOR GUIDE
100-004830-001
Version 1
Proprietary Statement Trademarks
Copyright © 2003 Novell, Inc. All rights reserved. Novell, Inc. has attempted to supply trademark information about company
names, products, and services mentioned in this manual. The following list of
No part of this publication may be reproduced, photocopied, stored on a retrieval
trademarks was derived from various sources.
system, or transmitted without the express prior consent of the publisher. This
manual, and any portion thereof, may not be copied without the express written
Novell, Inc. Trademarks
permission of Novell, Inc.
NetWare, the N-Design, and Novell are registered trademarks of Novell, Inc. in
Novell, Inc. the United States and other countries.
1800 South Novell Place
Provo, UT 84606-2399 CNA, CDE, CNA are service marks and CNE is a registered service mark of
Novell, Inc. in the United States and other countries.
Certified Novell Administrator, Certified Novell Engineer, eDirectory, IPX,
Disclaimer NCP, NetWare Core Protocol, NLM, Novell Certificate Server, Novell Client,
Novell, Inc. makes no representations or warranties with respect to the contents Novell Cluster Services, Novell Distributed Print Services, Novell iFolder, and
or use of this manual, and specifically disclaims any express or implied Novell Storage Services are trademarks of Novell, Inc.
warranties of merchantability or fitness for any particular purpose. ConsoleOne, GroupWise, ManageWise, NDPS, NDS, Novell Directory Services,
Further, Novell, Inc. reserves the right to revise this publication and to make and ZENworks are registered trademarks of Novell, Inc.
changes in its content at any time, without obligation to notify any person or
entity of such revisions or changes. Other Trademarks
Adaptec is a registered trademark of Adaptec, Inc. AMD is a trademark of
Further, Novell, Inc. makes no representations or warranties with respect to any
Advanced Micro Devices. Pentium is a registered trademark of Intel Corporation.
NetWare software, and specifically disclaims any express or implied warranties
Windows is a registered trademark of Microsoft Corporation. NetScanTools is a
of merchantability or fitness for any particular purpose.
trademark of Northwest Performance Software, Inc. Oracle is a registered
Further, Novell, Inc. reserves the right to make changes to any and all parts of trademark of Oracle Corporation. Btrieve is a registered trademark of Pervasive
NetWare software at any time, without obligation to notify any person or entity Software, Inc. Norton AntiVirus is a trademark of Symantec Corporation.
of such changes. VMware is a trademark of VMware, Inc.
This Novell Training Manual is published solely to instruct students in the use of
Novell networking software. Although third-party application software packages
are used in Novell training courses, this is for demonstration purposes only and
shall not constitute an endorsement of any of these software applications.
Further, Novell, Inc. does not represent itself as having any particular expertise
in these application software packages and any use by students of the same shall
be done at the students’ own risk.
Software Piracy
Throughout the world, unauthorized duplication of software is subject to both
criminal and civil penalties.
If you know of illegal copying of software, contact your local Software
Antipiracy Hotline.
For the Hotline number for your area, access Novell’s World Wide Web page at
http://www.novell.com and look for the piracy page under “Programs.”
Or, contact Novell’s anti-piracy headquarters in the U.S. at 800-PIRATES (747-
2837) or 801-861-7101.
Advanced Novell Network Management / Instructor Guide Contents
Contents
Course Setup
Set Up the Classroom . . . . . . . . . . . . . . . . . . . . . . . . . . . . Setup-1
Setup Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Setup-2
Setup Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Setup-5
Setup Time-Saving Procedure . . . . . . . . . . . . . . . . . . . . . . Setup-6
Setup Instructions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Setup-7
Setup for Section 12 . . . . . . . . . . . . . . . . . . . . . . . . . . . . Setup-50
Customize the Exercise to Meet Student Needs . . . . . . . Setup-50
Set Up the Network for the Exercise . . . . . . . . . . . . . . . . Setup-51
Introduction
Audience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Intro-1
Prerequisites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Intro-1
Certification. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Intro-2
Relationship to Other Courses in the Curriculum . . . . . . . Intro-2
Agenda. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Intro-2
Course Feedback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Intro-4
Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Intro-4
Digital Airlines Company Information . . . . . . . . . . . . . . . Intro-4
Digital Airlines eDirectory Tree . . . . . . . . . . . . . . . . . . . . Intro-7
Digital Airlines NetWare 6 Implementation Plan . . . . . . . Intro-8
Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
In-Place Upgrade . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
NetWare Accelerated Upgrade . . . . . . . . . . . . . . . . . . . . . . . . 1-3
Objective 1 Review How to Prepare for a Server
Migration to NetWare 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
Prepare the Workstation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-5
Prepare the Source (Original) Server . . . . . . . . . . . . . . . . . . . . 1-6
Prepare the Destination (New) Server . . . . . . . . . . . . . . . . . . . 1-7
Prepare Server Application Files . . . . . . . . . . . . . . . . . . . . . . . 1-9
Objective 2 Review How to Implement Novell Licensing . . . . . . . . . . . . 1-10
Server and User Licensing Models . . . . . . . . . . . . . . . . . . . . 1-11
How UAL Coexists with SCL . . . . . . . . . . . . . . . . . . . . . . . . 1-13
How the Licensing Models Differ . . . . . . . . . . . . . . . . . . . . . 1-15
License Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-16
Objective 3 Identify How to Perform a Migration . . . . . . . . . . . . . . . . . . . 1-18
Objective 4 Perform Post-Migration Tasks . . . . . . . . . . . . . . . . . . . . . . . . 1-39
Run the External Reference Check Process . . . . . . . . . . . . . . 1-39
Upgrade Existing NSS Volumes . . . . . . . . . . . . . . . . . . . . . . 1-40
Perform Other Post-Installation Tasks . . . . . . . . . . . . . . . . . . 1-41
Exercise 1-1 Upgrade a NetWare 4.11
Server to NetWare 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-42
Objective 5 Troubleshoot Post-Installation Issues . . . . . . . . . . . . . . . . . . . 1-65
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-68
Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
Objective 1 Upgrade Novell Network Management Tools . . . . . . . . . . . . . 2-3
Exercise 2-1 Upgrade Your Novell
|Network Management Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-5
Objective 2 Identify the Troubleshooting Features of Novell NetWork
Management Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-8
ConsoleOne Reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-9
Novell iMonitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-11
NetWare Remote Manager . . . . . . . . . . . . . . . . . . . . . . . . . . 2-12
Novell iManager . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-14
Objective 3 Identify the Purpose and Function of IP/IPX
Troubleshooting Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-17
NetWare IP/IPX Troubleshooting Tools . . . . . . . . . . . . . . . . 2-17
Client IP Troubleshooting Tools . . . . . . . . . . . . . . . . . . . . . . 2-25
TCP/IP Troubleshooting Example . . . . . . . . . . . . . . . . . . . . . 2-42
Exercise 2-2 Test Your Network. . . . . . . . . . . . . . . . . . . . . . . . 2-44
Protocol Analyzers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-47
TCP/IP Toolkits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-51
IP Addressing Calculators . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-54
Objective 4 Identify Additional Network Troubleshooting Resources . . . 2-57
Novell Web Site Resources . . . . . . . . . . . . . . . . . . . . . . . . . . 2-57
Shareware and Freeware . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-59
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-60
Exercise Answers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-61
Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
Objective 1 Identify Server Hardware and
Operating System Components . . . . . . . . . . . . . . . . . . . . . . . . 3-2
Identify Server Hardware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
Identify Operating System Components . . . . . . . . . . . . . . . . . 3-8
Exercise 3-1 Determine Hardware and Operating System
Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-18
Objective 2 Troubleshoot and Resolve NetWare Server Issues . . . . . . . . 3-21
Identify the Top Novell Technical Support
Server Issues and How to Resolve Them . . . . . . . . . . . . . . . 3-21
Identify Problems after Installation . . . . . . . . . . . . . . . . . . . . 3-22
Resolve Console Lock Ups . . . . . . . . . . . . . . . . . . . . . . . . . . 3-23
Resolve Hard Disk Errors and Access Problems . . . . . . . . . . 3-24
Resolve Application Monopolizing Server CPU . . . . . . . . . . 3-25
Resolve Server Memory Problems . . . . . . . . . . . . . . . . . . . . 3-26
Resolve Slow Server Response . . . . . . . . . . . . . . . . . . . . . . . 3-30
Identify Multiprocessing Problems . . . . . . . . . . . . . . . . . . . . 3-33
Find Tools for Managing Servers . . . . . . . . . . . . . . . . . . . . . 3-35
Exercise 3-2 Resolve Server Problems . . . . . . . . . . . . . . . . . . . 3-39
Objective 3 Troubleshoot and Resolve Critical Server Abends . . . . . . . . 3-47
What an Abend Is . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-48
What Types of Abends Occur . . . . . . . . . . . . . . . . . . . . . . . . 3-49
What an ABEND.LOG File Is . . . . . . . . . . . . . . . . . . . . . . . . 3-50
What a Core Dump Is . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-51
How to Respond to an Abend . . . . . . . . . . . . . . . . . . . . . . . . 3-52
How to Create and Submit a Core Dump for Analysis . . . . . 3-57
Exercise 3-3 Submit an ABEND.LOG File for Analysis . . . . . 3-64
Exercise 3-4 Create a Core Dump . . . . . . . . . . . . . . . . . . . . . . 3-66
Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1
Objective 1 Identify eDirectory Databases and Processes . . . . . . . . . . . . . 4-2
eDirectory 8.7 Databases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2
eDirectory Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-4
Post-Migration or Upgrade Issues That Affect
eDirectory Databases and Processes . . . . . . . . . . . . . . . . . . . . 4-6
Objective 2 Identify eDirectory Troubleshooting Steps . . . . . . . . . . . . . . . 4-8
Objective 3 Identify Partition and Replication Placement Design . . . . . . 4-15
Exercise 4-1 Adding Replicas with iManager . . . . . . . . . . . . . 4-17
Objective 4 Use iMonitor Reports to Obtain Server and eDirectory
Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-20
How to Review Report Options . . . . . . . . . . . . . . . . . . . . . . . 4-21
How to Run a Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-23
How to View Saved Reports . . . . . . . . . . . . . . . . . . . . . . . . . 4-25
Exercise 4-2 Verify eDirectory Status Using Reports . . . . . . . 4-26
Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1
The Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
Objective 1 Expand an NSS Storage Space . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
Increase the Number of Logical
Volumes in a Storage Pool . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
Overbook the Storage Pool . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-3
Increase the Size of a Storage Pool . . . . . . . . . . . . . . . . . . . . . 5-4
Objective 2 Configure NSS Volume Attributes . . . . . . . . . . . . . . . . . . . . . . 5-9
Objective 3 Mount a DOS Partition as an NSS Volume . . . . . . . . . . . . . . 5-11
How DOSFAT.NSS Works . . . . . . . . . . . . . . . . . . . . . . . . . . 5-12
How to Mount a DOS Partition Using DOSFAT.NSS . . . . . 5-13
How to Verify That the DOS Partition Is Mounted . . . . . . . . 5-14
Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1
Objective 1 Implement RAID 0 with NSS . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
What RAID Is . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
Hardware RAID versus Software RAID . . . . . . . . . . . . . . . . . 6-2
Hardware and Software RAID Levels . . . . . . . . . . . . . . . . . . . 6-3
How to Configure Software RAID 0 in NSS . . . . . . . . . . . . . . 6-6
How to Use NSSMU to Create RAID Arrays . . . . . . . . . . . . 6-10
Exercise 6-1 Configure a Software RAID Solution . . . . . . . . . 6-10
Objective 2 Configure Partition Mirroring and Duplexing in NSS . . . . . . 6-15
Configure Partition Mirroring and Duplexing in NSS . . . . . . 6-16
Troubleshoot Software RAID and Mirroring in NSS . . . . . . 6-19
Exercise 6-2 Mirror an NSS partition . . . . . . . . . . . . . . . . . . . . 6-20
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-22
Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1
Objective 1 Describe iFolder Configuration Files . . . . . . . . . . . . . . . . . . . . 7-1
How to Edit iFolder Configuration Files . . . . . . . . . . . . . . . . . 7-2
When to Edit iFolder Configuration Files . . . . . . . . . . . . . . . . 7-9
Objective 2 Perform iFolder Management Tasks . . . . . . . . . . . . . . . . . . . 7-14
Stop and Start the iFolder Server . . . . . . . . . . . . . . . . . . . . . . 7-14
Set iFolder Client and Server Policies . . . . . . . . . . . . . . . . . . 7-16
Change the Location of iFolder User Data . . . . . . . . . . . . . . 7-17
Add Contexts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-18
Add Additional Administrators . . . . . . . . . . . . . . . . . . . . . . . 7-19
Change the iFolder Server IP Address . . . . . . . . . . . . . . . . . . 7-20
Exercise 7-1 Perform Advanced iFolder
Management Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-22
Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1
Objective 1 Identify the Purpose and Advantages of
Implementing an NCS Solution . . . . . . . . . . . . . . . . . . . . . . . . 8-2
High Availability Terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-3
High Availability Definition . . . . . . . . . . . . . . . . . . . . . . . . . . 8-4
Computer System Outage Factors . . . . . . . . . . . . . . . . . . . . . . 8-7
Benefits and Features of an NCS High Availability Solution . 8-8
Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1
Objective 1 Verify NCS System Requirements . . . . . . . . . . . . . . . . . . . . . . 9-2
Hardware Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-2
Software Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-3
License Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-3
Shared Disk System Requirements . . . . . . . . . . . . . . . . . . . . . 9-4
Objective 2 Create a Cluster by Installing NCS . . . . . . . . . . . . . . . . . . . . . 9-5
Objective 3 Check Cluster Configuration Settings . . . . . . . . . . . . . . . . . . 9-13
Cluster ADMIN Object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-14
Cluster Object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-14
Master IP Address Resource Object . . . . . . . . . . . . . . . . . . . 9-23
Cluster Server Node Objects . . . . . . . . . . . . . . . . . . . . . . . . . 9-24
Exercise 9-1 Install and Check NCS on a 2-Node Cluster . . . . 9-25
Objective 4 Test and Monitor the Cluster . . . . . . . . . . . . . . . . . . . . . . . . . 9-34
Cluster State and Cluster Status Views . . . . . . . . . . . . . . . . . 9-34
Console Prompt Commands . . . . . . . . . . . . . . . . . . . . . . . . . 9-38
Exercise 9-2 Test the SBD Partition and Heartbeats . . . . . . . . 9-39
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-46
Exercise Answers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-49
Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-1
Objective 1 Configure NCS for High Availability File Access . . . . . . . . 10-2
Create a Shared Disk Partition . . . . . . . . . . . . . . . . . . . . . . . . 10-2
Create and Cluster-Enable an NSS Volume and
Pool on a Shared Storage Device . . . . . . . . . . . . . . . . . . . . . . 10-3
Cluster-Enable an Existing Pool or Volume on the
Shared Disk System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-6
Exercise 10-1 Create a Cluster-Enabled Volume
for High Availability File Access . . . . . . . . . . . . . . . . . . . . . . 10-10
Objective 2 Manage Resources in an NCS Cluster . . . . . . . . . . . . . . . . . 10-18
How to Migrate Resources . . . . . . . . . . . . . . . . . . . . . . . . . . 10-18
How to Troubleshoot Resource States . . . . . . . . . . . . . . . . . 10-20
Exercise 10-2 Test High Availability File
Access on the 2-Node Cluster. . . . . . . . . . . . . . . . . . . . . . . . . 10-23
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-29
Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-1
Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-1
Objective 1 Create a Disaster Recovery Plan . . . . . . . . . . . . . . . . . . . . . . 12-1
Planning for Hardware Failures . . . . . . . . . . . . . . . . . . . . . . . 12-2
Planning for Calamities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-3
Objective 2 Troubleshoot Network Problems . . . . . . . . . . . . . . . . . . . . . . 12-4
Exercise 12-1 Troubleshoot Network Problems . . . . . . . . . . . . 12-4
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-5
Index
Course Setup
x This course was tested for NetWare® 6 with Support Pack 2. If you use a later
support pack, you might need to adjust exercise steps.
Setup Requirements
Software ■ NetWare 6 OS CD
■ NetWare 6 Support Pack 2 CD
■ eDirectoryTM 8.7 CD
■ eDirectory Webapps CD
■ 3005.LDIF (on the Enhanced Learning
CD or CNI Net)
■ 3005LicenseFolders.EXE (NetWare 6
server and user licenses in
DA1\\DATA\SETUP on the DA1
VMwareTM server or CNI Net) or the
DAx license folders on the Enhanced
Learning CD (Setup)
Setup Diagram
Classroom
Hub
WS4 WS7
Windows 2000 Windows 2000
IPX Novell Client IPX Novell Client
BADBEEF DA4 DA7 BADBEEF
(empty box) (empty box)
2 hard disks 2 hard disks
SCSI adapter SCSI adapter
WS5 WS8
Windows 2000 Windows 2000
IPX Novell Client IPX Novell Client
BADBEEF DA5 DA8 BADBEEF
(empty box) (empty box)
2 hard disks 2 hard disks
SCSI adapter SCSI adapter
WS6 WS9
Windows 2000 Windows 2000
IPX Novell Client IPX Novell Client
BADBEEF DA6 DA9 BADBEEF
Arrange placement of each pair of servers (DA1 and DA2, DA4 and
DA5, DA6 and DA7, DA8 and DA9) so they are close enough for
the SCSI cables to reach from each server to the SCSI hard disk for
clustering.
Although these tasks are listed consecutively, you can save hours by
performing these tasks as follows:
1. Install Windows 2000 on the instructor workstation (WS1) and a
student workstation (WS4).
You can use a Microsoft scripted installation file (SIF) to run
the installation unattended while performing other setup tasks.
Setup Instructions
❑ IP address: 192.168.1.21
❑ Subnet mask: 255.255.255.0
x Any VMware servers you do not migrate during Setup and Section 1
must stay on and running throughout the course. The servers are
included in the eDirectory tree and are part of the replica ring.
Do the following:
1. Locate the 3005LicenseFolders.EXE file from one of the
following:
❑ CNI Net
❑ DA1 VMware server (DATA:\SETUP)
2. Double-click the file and extract the DAx folders to a temporary
directory.
(You can also copy the DAx folders from the SETUP\LICENSE
FILES directory on the Enhanced Learning CD).
3. Copy folders DA1 and DA2 (for the instructor servers) and DA4
to DA9 (for the student servers) to individual floppy disks
labeled with the same name.
Do the following:
1. At the instructor workstation, insert the NetWare 6 OS CD
(January 2002 SEL CD6A).
2. If it does not autorun, browse to the CD and launch
NWDEPLOY.EXE.
3. Double-click Network Preparation.
4. Double-click Step 2: View and Update NDS Versions.
5. Enter netware://Tree/DIGITALAIR-TREE.
6. Select Include Subordinate Containers; then select Next.
You should find that no servers need to be updated.
7. Select Exit.
8. Double-click Step 3: Prepare for NDS eDirectory 8.6.
9. Enter netware://Tree/DIGITALAIR-TREE.
10. Select Next.
Do the following:
1. Begin a pre-migration installation of NetWare 6:
a. Start the computer and insert the NetWare 6 OS CD.
After all VMware servers are on and running NetWare 4.11, prepare
DA1 by doing the following:
1. Using DSREPAIR, run the Unattended Full Repair and Report
Synchronization Status options:
x Most critical errors involve SMDR issues. Make sure Novell client
connections to both DA1 and TEMP1 are IPXTM.
Do the following:
1. Upgrade eDirectory:
a. Mount the eDirectory 8.7 CD (November 2002 SEL CD2A)
as a NetWare volume.
b. At the server console, load NWCONFIG.
c. In the Available Options menu, select Product Options.
d. Select Install a Product Not Listed.
e. To specify the path to the CD, press F3.
f. Specify the path to the directory where the installation
program can find the NDS8.IPS file by entering volume
name:NW.
Do the following:
1. Prepare the network for NetWare 6:
a. At the workstation, insert the NetWare 6 OS CD.
b. (Conditional) If it does not autorun, browse to the CD and
launch NWDEPLOY.EXE.
c. Double-click Network Preparation.
d. Double-click Step 2: View and Update NDS Versions.
e. Enter netware://Tree/DIGITALAIR-TREE.
f. Select Include Subordinate Containers; then select Next.
You should find that no servers need to be updated.
g. Select Exit.
h. Double-click Step 3: Prepare for NDS eDirectory 8.6.
i. Enter netware://Tree/DIGITALAIR-TREE.
j. Select Next.
k. Select DA1; then select Next.
l. When the update is complete, select Exit.
m. To close Deployment Manager select Cancel; then select
Yes.
2. Install NetWare 6 on server DA2:
a. Start the computer and insert the NetWare 6 OS CD.
b. Select your language.
c. Accept the license agreement.
d. Create a new boot partition and modify the size to equal 200
MB + the amount of RAM on the computer.
e. After the computer reboots and the boot partition is created,
accept the license agreement.
l. Select Yes to reboot the server after the files are copied.
m. Press Enter to continue.
n. When prompted, authenticate as admin.digitalair with the
password novell.
o. When the file copy is complete, wait for the server to restart.
Do the following
1. Upgrade eDirectory:
a. Mount the eDirectory 8.7 CD as a NetWare volume.
b. At the server console, load NWCONFIG.
c. In the Available Options menu, select Product Options.
d. Select Install a Product Not Listed.
e. To specify the path to the CD, press F3.
f. Specify the path to the directory where the installation
program can find the NDS8.IPS file by entering volume
name:NW.
For example: EDIR_8_7:NW.
g. Allow the files to copy.
h. On the Software License Agreement screen, press Esc to
continue.
i. Accept the license agreement.
j. On the License Agreement for JReport Runtime JInfonet
Software screen, press Esc.
k. Accept the Reporting license agreement.
l. Continue by pressing Esc.
m. Read the warning; then press Esc and allow the files to copy.
n. In the Administrator Name field, enter your full
distinguished name.
Do the following:
1. At the workstation use Internet Explorer to access iManager at
HTTPS://192.168.1.2:2200.
2. Under eDirectory iManager select DA2.
3. Use the following information to authenticate:
❑ User name: admin
❑ Password: novell
❑ Context: Digitalair
❑ Tree: Digitalair-tree
4. From the navigation frame on the left, expand License
Management; then select Install a License.
5. At the right of the Load license file field, select Browse.
6. From folder DA1 on your license diskette, select the user license
file; then select Open.
7. Continue by selecting Next.
8. Select an available user license; then select Next.
9. In the Location field, enter SLC.DIGITALAIR; then select
Install.
10. When the license is installed, select Done.
Do the following:
1. Configure DA2 as a DHCP server:
a. At the workstation use Internet Explorer to access iManager
at HTTPS://192.168.1.2:2200.
b. Select DA2 under eDirectory iManager.
g. Select Done.
h. Select Next 3 times to move through the remaining
configuration option windows.
i. From the Global DHCP Preferences screen scroll down to
the bottom and select Done.
j. When prompted that the operation is successful, select OK.
6. Start the DHCP service on DA2:
a. At the DA2 server console, enter DHCPSRVR.
b. Edit AUTOEXEC.NCF and place DHCPSRVR anywhere
after the MOUNT ALL command so the DHCP service
launches when the server is restarted.
7. Configure IP on the instructor workstation to obtain an IP
address automatically.
8. Verify that DHCP is functioning:
a. At the command prompt enter IPCONFIG /RELEASE;
then enter IPCONFIG /RENEW.
b. Verify the workstation is receiving an IP address assignment
from the DHCP server.
9. Configure DA2 as a DNS server:
a. In the left frame, expand DNS Management.
b. Select DNS Server Management.
c. In the drop-down list, select Create Server > OK.
d. In the Enter NCP Server Name field, enter
DA2.CORPORATE.SLC.DigitalAir.
e. In the Enter Host Name field, enter DA2.
f. In the Enter Domain Name field, enter DigitalAirlines.com;
then select Create.
g. When prompted that the operation is successful, select OK.
10. Create a DNS zone for the classroom network:
a. In the left frame, select Zone Management.
192.168.1.1 DA1.DigitalAirlines.com
192.168.1.2 DA2.DigitalAirlines.com
192.168.1.4 DA4.DigitalAirlines.com
192.168.1.5 DA5.DigitalAirlines.com
192.168.1.6 DA6.DigitalAirlines.com
192.168.1.7 DA7.DigitalAirlines.com
192.168.1.8 DA8.DigitalAirlines.com
192.168.1.9 DA9.DigitalAirlines.com
192.168.1.31 iFolder1.DigitalAirlines.com
192.168.1.32 iFolder2.DigitalAirlines.com
192.168.1.34 iFolder4.DigitalAirlines.com
192.168.1.35 iFolder5.DigitalAirlines.com
192.168.1.36 iFolder6.DigitalAirlines.com
192.168.1.37 iFolder7.DigitalAirlines.com
192.168.1.38 iFolder8.DigitalAirlines.com
192.168.1.39 iFolder9.DigitalAirlines.com
Import 3005LDIF
You build the classroom tree using the 3005.LDIF file included on
the Enhanced Learning CD.
1. If necessary, authenticate from your workstation as Admin.
2. Start ConsoleOne.
3. Import the 3005.LDIF file from the SETUP directory on the
Enhanced Learning CD:
a. In ConsoleOne, browse to the DigitalAir container.
b. Select Wizards > NDS Import/Export.
c. Select Import LDIF File > Next.
d. From the Select Source LDIF File screen, browse to and
select the 3005.LDIF file in the SETUP directory on the
Enhanced Learning CD.
(This file is also available from CNI Net or in
DATA:\SETUP on the DA1 server.)
e. Select Advanced.
f. Deselect Exit on Error; then select OK.
g. Select Next.
h. From the Select Destination LDAP Server screen, select
New.
i. In the Description field, enter DA Import.
j. In the Server DNS Name/IP Address field, enter 192.168.1.1.
k. In the Port field, enter 636.
l. In the Der File Containing Server Key Used for SSL
Communications field, browse to and select the
RootCert.der file from SYS:\PUBLIC on DA1.
m. In the User DN field, enter cn=admin,o=DigitalAir.
n. Select OK.
o. From the Select Destination LDAP Server screen, select DA
Import.
x If you receive an error stating that the client couldn’t connect to the
LDAP server, reboot the server and workstation; then run the
import again.
x You will see a few errors. These are instances where a container or
user object exists in the tree. You can disregard the errors. However,
if more than 18 errors occur, there was a problem with the import.
w. Select Close.
4. Refresh your tree view by pressing F5 and then verify that the
new containers and user objects were created.
In this course, you use DA2 to route between the classroom network
and the external network to provide Internet access. To configure
NAT, do the following:
1. Transfer control of the server LAN driver configuration to
INETCFG:
a. At the DA2 server console, enter INETCFG.
b. When prompted to transfer LAN configuration, select Yes.
c. When prompted to leave INETCFG and restart the server,
select Yes and wait while the server restarts.
2. At the server console prompt, enter CONFIG and make sure of
the following:
❑ Drivers for both network boards in DA2 are loaded
❑ A private IP address is assigned to the private board
❑ A public IP address is assigned to the public board
3. Configure dynamic NAT to enable public network access from
your private network:
a. At the server console, enter INETCFG.
b. Select Yes, Use the Fast Setup Method.
c. Press Esc; then select Go to INETCFG Main Menu.
d. Select Bindings.
e. Select the TCP/IP binding for the network board connected
to your organizational network segment.
f. Select Configure TCP/IP Bind Options.
g. Select Expert TCP/IP Bind Options.
h. Select Network Address Translation.
i. Change the status to Dynamic Only.
x If reinitializing fails, restart the server. You might also have to reset
the router after the server loads.
Do the following:
1. Install SCSI adapters in DA1 and DA2 and make sure there is a
SCSI cable for each.
2. Arrange placement of DA1 and DA2 so they are close enough for
the SCSI cables to reach from each server to the SCSI hard disk.
3. (Optional) Perform clustering setup steps from Sections 8, 9, and
10.
Student servers must have 2 hard disk drives with nothing installed
on them and must be prepared for the clustering exercises.
Do the following:
1. Run FDISK on each hard disk so there are no partitions
configured on them.
2. Install SCSI adapters in each server and make sure there is a
SCSI cable for each.
3. Make sure there is an external SCSI hard disk for every 2 servers.
4. Arrange the placement of servers so DA4 and DA5 are close
enough for the SCSI cables to reach from each server to the SCSI
hard disk; do the same for DA6 and DA7 and for DA8 and DA9.
This completes the steps you must do before teaching this class. The
following topic contains steps you perform after teaching
SECTION 11 and before teaching SECTION 12.
As instructor for the class, you are welcome to modify this setup to
include your own troubleshooting tasks or to customize the setup to
meet the needs of the students.
x If you do not migrate all NetWare 4.11 servers (DA4 - DA9), you must
modify this setup and the student network problems in Exercise 12-1 to
match the current state of your network.
For example, if you have not migrated DA8, you must implement step 1e on
another server (such as DA4) or eliminate the step.
Do the following:
1. Introduce communication issues into the LAN:
a. On DA4, edit AUTOEXEC.NCF and change the subnet
mask of the server to 255.255.255.252.
b. On DA5, edit AUTOEXEC.NCF and change the IP address
of the server to 10.0.0.4.
c. On DA6, unplug the LAN cable and edit AUTOEXEC.NCF
to change the subnet mask of the server to 255.255.255.252.
x You might want to wait until you have completed all setup tasks
before unplugging any LAN cables.
Introduction
Duration: 30 minutes In this course you learn advanced NetWare® network management
and troubleshooting skills for NetWare operating system
environments, directory services, and data storage and services.
You also learn how to set up an NCS SAN to test high availability
of resources.
Audience
This course is for students who have entry-level experience in
managing small LAN or WAN networks and meet the necessary
prerequisite knowledge.
Prerequisites
You must have an understanding of the following:
■ Foundations of Novell Networking, Novell Course 3001
■ Novell Network Management, Novell Course 3004
■ Terminology, hardware, and practices commonly used in
medium to large enterprise networks, such as routers, hubs,
switches, backbones, and subnets
Certification
This course helps you prepare for the following tests:
Agenda
This is a 5-day course.
Course Feedback
Your feedback is valuable to Novell Education. To provide feedback
on the course materials, use the web services tool at
http://www.novell.com/education/courses/feedback/index.html.
Scenario
This scenario is based on a WAN Digital Airlines, Inc. is a flight business that provides luxury charter
configuration. However, for and scheduled flight services for executives, government officials,
classroom exercise purposes, all athletic teams, and others needing private, flexible, secure, and
servers are connected over a LAN. catered air travel.
This results in students performing The following provides information about the Digital Airlines
some tasks in class (such as network, eDirectory tree structure, and tasks branch office network
creating a 2-node cluster) that administrators must perform to begin upgrading servers to
would not normally be done over a
NetWare 6:
WAN.
■ Digital Airlines Company Information
As you teach the course, make
sure students understand (when ■ Digital Airlines eDirectory Tree
appropriate) the limitations of ■ Digital Airlines NetWare 6 Implementation Plan
performing these tasks over a
WAN.
Digital Airlines, Inc. has been in business for 5 years. They have
■ 20 aircraft
■ 435 employees:
❑ 50 pilots
❑ 300 flight attendants
■ Marketing
■ Reservations
O=DigitalAir
Copying this document is strictly prohibited. To report suspected copying, please call 1-800-PIRATES.
OU=MARKETING OU=MARKETING OU=MARKETING
OU=MARKETING OU=MARKETING OU=MARKETING
OU=MARKETING
Version 1
Introduction
Advanced Novell Network Management / Instructor Guide Introduction
The network administrator for the Salt Lake City office administers
the following NetWare servers on a local LAN from a Windows
2000 workstation:
Network Hub
DA3
NW 4.11 SP9
IPX
DA1 0BADBEEF
DA2
NW 6 SP2
NW 6 SP2
DS 8.7 – Master Replica
IP and IPX
IP and IPX
192.168.1.2
192.168.1.1
2-Node Cluster
SCSI Storage
DA1 and DA3 currently hold DA1 IS.SLC.DIGITALAIR ■ Stores the ROOT master replica
replicas for servers across the for DigitalAir-Tree and currently
WAN. holds master replica for all
servers in the tree
If this design flaw is noted by the ■ Provides network backup for the
students, congratulate them and Salt Lake City LAN
note that this is a purposeful ■ Provides cluster-enabled iFolder
design flaw that will be corrected services for the Salt Lake City
by them in Section 4. office
Network Hub
DA4 IS.DEL.DIGITALAIR
DA5 IS.LGA.DIGITALAIR
DA6 IS.LON.DIGITALAIR
DA7 IS.SYD.DIGITALAIR
DA8 IS.TXL.DIGITALAIR
DA9 IS.TYO.DIGITALAIR
Duration: 5 hours In this section you learn how to migrate NetWare 4.x and
NetWare 5.x servers to NetWare 6.
Objectives
1. Review How to Prepare for a Server Migration to NetWare 6
2. Review How to Implement Novell Licensing
3. Identify How to Perform a Migration
4. Perform Post-Migration Tasks
5. Troubleshoot Post-Installation Issues
Introduction
In this section you perform a NetWare 6 migration.
In-Place Upgrade
Briefly review the in-place upgrade You can use the NetWare 6 installation program to perform an
material. Students performed an in-place upgrade of an existing NetWare 4 or NetWare 5 server to
in-place upgrade in Course 3004. NetWare 6.
After you access the NetWare 6 installation files, you follow the
instructions for installing a server found at
www.novell.com\documentation.
After you upgrade or install the first NetWare 6 server using the
NetWare 6 installation program, you can then use NetWare
Accelerated Upgrade to upgrade other servers in the tree.
x After the accelerated upgrade you must install Apache Web Server for the
following to operate: iFolder, NetWare Web Search, iManager, NetWare Web
Access, and NetStorage.
The process not only provides guidelines for how much hardware
you’ll need, but it also outlines the correct steps you must take to
prepare your file system and Directory for migration.
Define source server. Help The source server is the original NetWare server that contains the
students understand that the files, volumes, and eDirectory objects to be copied to the destination
source server is the original (old) server. Valid source servers can NetWare 4.11, 4.2, 5.0, 5.1, or 6
server. servers.
Do the following:
1. Verify that you have the Supervisor right to the source server’s
file system and Directory tree.
2. Make sure the destination (new) server can communicate with
the source server:
a. On the destination server, if it is using IPX, enter Display
Servers and make sure the source server is listed.
b. On the destination server, if it is using IP, enter Display SLP
Services and make sure the source server is listed.
b See TID 2961749 - Post SP1 NSS modules for NetWare 6. Despite what
the abstract implies, this patch can be applied before applying Support
Pack 1.
4. Update the source server with the latest NetWare support pack,
available at support.novell.com.
5. (Conditional) If you are migrating data from NetWare 4, make
sure the source server’s volumes have long name space support
added to all volumes to be copied.
To add long name space support to a NetWare 4.11 or NetWare
4.2 volume, enter the following at the server console:
LOAD LONG
ADD NAME SPACE LONG TO volume name
6. Load DSREPAIR and run the following options:
Unattended Full Repair
Time Synchronization
Report Synchronization Status
Point out that it is possible to finish Make sure these finish with no errors. However, in a mixed
a full unattended repair with errors NetWare 4.x and 5.x environment, it is possible to finish a full
due to schema mismatches. unattended repair with errors due to schema mismatches.
You perform these operations to ensure that the Directory is
healthy and stable before you migrate the server.
7. Make 2 full, verified backups of the eDirectory tree and the file
system.
8. Check reference material to verify that you are using the
supported hardware.
Define destination server. Help The destination server is the new computer that receives the data
students understand that the from the source server. This server must be installed into a
destination server is the new temporary tree.
server.
Point out that if students don’t 3. During installation of the destination server, create volumes on
create the same volumes on the the destination server that are the same size as, or larger than,
destination server that exist on the volumes on the source server.
source server, the volume’s files
won’t be migrated to the new
server.
x Volume names on the destination server must be the same as the volume
names on the source server.
Point out that the destination 5. Install the NetWare OS on the destination server using the
server must be installed using the Pre-Migration installation option.
Pre-Migration installation option.
6. Prevent time synchronization issues:
This option is available only in a. Configure the destination server as a secondary time source
NetWare 6, not NetWare 5.1. If by entering the following commands at the console prompt:
you were installing NetWare 5.1 as
the destination server, you would SET TIMESYNC TYPE = SECONDARY
not have a pre-migration option. SET CONFIGURED SOURCES = ON
SET TIMESYNC TIME SOURCES = SOURCE
SERVER NAME; or IP ADDRESS;
Server and user licenses are separate files. They can be installed
anywhere in the eDirectory tree.
(You can install licenses in the same container as the objects that
use them if all your users exist in a single container.)
Point out that the SCL model is not ■ The server connection license (SCL) model. Prior to the
available in NetWare 6. release of NetWare 6, Novell used the SCL model to regulate
licensed usage of NetWare and its services.
In the SCL model, users were granted access to network
services on a server basis. This meant that a single user might
use several connection licenses if the user concurrently
connected to multiple servers.
With the SCL model, you had to estimate the number of
connections a user might need to perform his or her job to
determine the number of connection licenses needed.
The number of nonuser objects requiring connections also had
to be accounted for in the license unit count.
Point out that a user license allows In the UAL model, each user is assigned a user license as they
a user unlimited access to initially log in to a NetWare 6 server. The user can then connect
NetWare 6 servers and their to any other NetWare 6 server in a single eDirectory tree
services. without requiring another license.
After the first assignment of a user license to a user object, that
license is reserved for that user as long as the user continues to
authenticate to the network.
x If you install user licenses in the same container as your users, all your
users must be in that container for the licenses to work.
If the user doesn’t log in again for 90 days or more, the license
is released and made available to the next user who needs a
license assignment.
x User licenses can be released from the originally assigned user through
iManager. This is helpful when license units have been assigned to users
who log in infrequently and you want to release the license for use by
others who require more regular network access.
In the UAL model, server licenses are still required for NetWare
6 servers.
During the installation of a NetWare 6 server, you install a
server license. Each server in the tree must have a unique server
license.
You can download additional server licenses from
www1.novell.com/eld/LRequest.jsp?ENCRYPTION=NW6.
Licenses downloaded from this site are demo server licenses
and are provided at no cost. This allows you to get a server
installed and running for your organization.
To purchase and download licenses that are not demo licenses,
you must establish a license agreement with Novell or an
authorized reseller.
For each NetWare 4.x or 5.x server THarvey logs in to, a NetWare
SCL unit is used.
The following shows the differences between the SCL model and
the UAL model:
Point out that a search for a Search for license A search starts at the A search starts at the
license starts looking up the tree user’s context and server’s context and
for the license. goes up the tree. goes up the tree.
If your users are
installed in different
containers, install the
licenses above the
users’ containers.
Point out that a UAL unit is not License released No. Yes.
released when a user logs out. when user logs out
License Types
Note: server licenses are free and The UAL model allows for the following 2 license types:
user licenses are paid for.
■ License agreement licenses. Large companies that require
many user licenses sign a license agreement with Novell. The
agreement stipulates the number of license units that can be
used before more licenses must be purchased.
Novell’s licensing agreements provide pricing breaks according
to the size of an organization.
After the original server’s file system and eDirectory database are
migrated, the original server is brought down and the NetWare 6
server reboots and assumes the name and identity of the original
server on your network.
Before you can migrate your data, you must first install a NetWare 6
server in a temporary eDirectory tree.
Define temporary tree. A temporary tree is a tree that contains one server with a basic
installation of NetWare and no additional products (other than SMS
or any other default products).
Do the following:
1. Install the migration wizard from the NetWare 6 OS CD.
Figure 1-2
Figure 1-3
Figure 1-4
Figure 1-5
9. Name the project and choose a place to save it; then select Next.
By default, the migration wizard saves all projects to
C:\PROGRAM FILES\NOVELL\NETWARE
MIGRATION WIZARD.
Figure 1-6
10. In the Select the Source NDS Tree window, select the Directory
tree that contains your source server; then select Next.
The following appears:
Figure 1-7
11. In the Select the Source Server window, select your source
server from the Directory tree.
12. In the Select the Destination NDS Tree window, select the
eDirectory tree that contains your destination server.
13. In the Select the Destination Server window, select your
destination server from the destination eDirectory tree.
Figure 1-8
14. Save your project and access the Project Window by selecting
Create.
The Project Window now appears:
Figure 1-9
Copy Volumes
Selecting Copy Volumes from the Project Window lets you copy
volumes from the source server to the destination server. You can
copy a volume as many times as you need to complete this step.
Before you can copy a volume from the source server to the
destination server, you must have a volume with the same volume
name created on the destination server.
x If you copy volumes in phases, at the final volume copy make sure you
select all volumes that you previously copied; then select Cancel.
Otherwise, the migration wizard restores trustee assignments only to the
last volumes that were copied.
Figure 1-10
2. Select each volume you want to copy and select Yes; then select
Next.
x If you decide not to copy any volumes, select No for all volumes; then
select Next and continue with “Edit Configuration Files” on 1-29.
Figure 1-11
If there are files in the source server’s SYS directories that you
want to use on the destination server, after the migration is
completed, copy the files from SYS:SYS.MIG into the
appropriate SYS directory on the destination server.
Figure 1-12
Figure 1-13
Figure 1-14
Figure 1-15
2. Select the configuration file you want to edit; then select Edit
File.
Figure 1-16
3. Copy and paste the commands from the configuration file on the
source server to the corresponding configuration file on the
destination server.
To change the IP address on your destination server to be the
same as the IP address of your source server, you must change
the IP address in 3 places: AUTOEXEC.NCF,
SYS:\ETC\HOSTNAME, and SYS:\ETC\HOSTS:
a. Copy the source server’s IP address and paste it into the
destination server’s AUTOEXEC.NCF file.
After you complete the volume copy and edit configuration files,
you can migrate eDirectory from the source to the destination
server.
During this phase, the source server is brought down and the
destination server is restarted. The destination server takes the
source server’s name and place in your network.
Figure 1-17
2. Back up all volumes that you are not planning to migrate and
complete the following tasks before continuing; then select Next:
❑ Make sure all volume data migrated successfully.
❑ Make sure all critical errors from the file copy are resolved.
❑ Make sure the time is synchronized on the servers in your
source tree and that eDirectory is synchronized.
For more information, see Step 2 in “Finish eDirectory
Migration” on 1-37.
3. When the Install License window appears, insert the license disk
and browse to or enter the path to the license file; then select
Next.
If you have an MLA, you can select the MLA instead of
inserting the license disk.
The following appears:
Figure 1-18
x If you select No to update the schema, by default the second Yes option
is enabled, which causes the migration wizard to compare the source
and destination schemas.
x The migration wizard does not check the health of the tree and will not
prevent you from continuing if your tree is unhealthy. If your tree is
unhealthy the migration might not complete successfully.
Figure 1-19
8. Copy the NICI files from the source server to the destination
server by following the on-screen instructions.
When the NICI configuration is complete, the following
appears:
Figure 1-20
9. Delete all user connections (except your own) to the source and
destination servers; then select Next.
10. In the Password Verification window, enter the passwords for the
source and destination trees; then select Next.
Figure 1-21
Figure 1-22
13. View the Error and Success logs; then select Done.
Use the Error log to see errors that occurred during migration.
If there were errors, use the Success log to determine how far
the migration progressed.
If migration failed, restore your servers to their original
configuration.
14. Check the destination server and verify that it has restarted and
taken on the name of the source server.
15. Reboot your workstation and log in to the former destination
server.
In this phase you upgrade the schema, fix volume objects, and
restore trustees to the destination server after the server is restarted.
Figure 1-23
3. Verify that you have completed the tasks by selecting Yes or No;
then select Next.
4. In the Password Verification window, enter the password for the
source tree; then select Next.
5. Resolve warnings or errors; then select Next.
The following appears.
Figure 1-24
Define external reference check The external reference check process is an eDirectory process that
process and external reference. verifies external references.
SET DSTRACE=ON
SET DSTRACE=+BLINK
SET DSTRACE=*B
If you switch to the Directory Services screen on the server, you can
watch the objects as they are processed.
b You can also run the external reference check process using iMonitor.
If the original server had NSS volumes, you need to upgrade them
to be compatible with the version of NSS shipped with NetWare 6.
x If you migrate volume SYS, SYS remains a traditional volume until you
upgrade it to NSS.
a 3 hours 30 minutes In this exercise you migrate a NetWare 4.11 server into NetWare 6.
Assign a VMware server to each Recall that the executive board of Digital Airlines has authorized an
pair of students. For example, DA4 upgrade of selected corporate NetWare servers to NetWare 6 to take
should be assigned the students advantage of technologies and services such as Novell Storage
with WS4, and so on. Services (NSS), iFolder, and Novell Cluster Services (NCS).
Because DIGITALAIR-TREE The network administrator for Digital Airlines has decided to have
recognizes DA4 - DA9, any all branch offices begin the transition to NetWare 6 by migrating
VMware servers not migrated your server to NetWare 6 on a new machine.
need to be running to avoid “can’t
communicate with DAx” error You must migrate your server to a new machine because the
messages and beeps.
hardware on your existing server will not support NetWare 6. Your
hardware has already been upgraded, so your job is to migrate your
Provide students with their Admin
context found in the Introduction. server and Directory data to the new box.
7. Enter enough space for your DOS partition using the following
formula:
200 + amount of RAM in machine
Ask your instructor for the amount of RAM in your machine.
8. Select Continue.
9. Verify that you want to create a boot partition by using the arrow
keys to select Continue.
10. Reboot your computer by pressing any key on the keyboard; then
allow your computer to reboot.
11. When prompted, reboot by selecting one of the following:
❑ To install from your IDE CD, press I.
❑ To install from your SCSI CD, press S.
❑ To install both, press B.
12. Configure basic installation parameters:
a. In the License Agreement for Jreport Runtime screen, press
F10 to accept the license.
b. In the Welcome screen, select Express; then press Enter to
switch to CUSTOM.
c. In the Welcome screen, select New Server; then press Enter
to switch to PRE-MIGRATION.
d. Press Tab.
e. Select Continue.
f. In the Pre-Migration Installation screen, select Continue.
g. In the Server Settings screen, select Continue.
h. In the Regional Settings screen, select Continue.
i. In the mouse and video selection screen, select Continue
and allow the files to copy.
13. Configure NetWare device drivers:
a. In the disk driver screen, select Continue.
f. Select OK.
g. Select Next.
17. Configure eDirectory using the following information:
DA4 IS.DEL.DIGITALAIR
DA5 IS.LGA.DIGITALAIR
DA6 IS.LON.DIGITALAIR
DA7 IS.SYD.DIGITALAIR
DA8 IS.TXL.DIGITALAIR
DA9 IS.TYO.DIGITALAIR
a. In the NDS Install screen, mark New NDS Tree; then select
Next.
b. In the Tree Name field, enter TEMPTREEx (where x = your
server number).
c. In the Context field, enter your context found in Table 1-3;
then press Tab.
d. Make sure your context appears in the Admin Context field.
e. In the Password field, enter novell.
f. In the Retype Password field, enter novell; then select Next
and allow eDirectory to install.
g. In the NDS Summary screen, select Next.
x If you receive an invalid license file dialog, select OK; then remove
the user license and continue with the exercise.
x Ignore time sync messages saying that single servers are incompatible
with Reference and Primary servers.
You can have students perform From the instructor’s workstation WS1, prepare your NetWare 4.11
these steps using RCONSOLE by server by doing the following:
loading RSPX and REMOTE on
each VMware 4.11 server. 1. Using DSREPAIR, do the following:
a. At the server prompt, enter LOAD DSREPAIR.
b. In Available Options, select Advanced options menu.
c. Select Repair local DS database.
d. Begin the repair by selecting F10; then select Yes.
e. When the repair is finished, view the current log file by
pressing Enter.
If you receive errors, have your instructor assist you.
f. Continue by pressing Esc twice; then press Enter.
g. Unlock the database file by pressing Esc; then return to the
Available options menu by pressing Esc.
h. In Available Options, select Report synchronization status.
If you receive errors, have your instructor assist you.
i. Exit DSREPAIR.
Username Admin
Password novell
Tree DigitalAir-Tree
Server DAx
Username Admin
Password novell
Tree TempTreex
Server TEMPx
11. In the Update NDS window, select the browse button at the right
of the text field.
12. In the NDS Tree Browser window, expand Novell Network.
16. Expand the containers until you get to your server’s container.
Remind students that no updates 19. When you receive a message saying that no servers were found
are necessary because that require an update, select OK.
NWDEPLOY was run when DA2
was installed. Normally you must perform the NDS preparation steps when
preparing NDS. If you are updating NDS in your work
environment, do the following; then continue with Step 20:
a. In the Select Servers to Update field, make sure Update NDS
is marked on your server’s row.
b. Select Next and allow files to copy.
c. Make sure Restart NDS is marked; then select Next.
d. After server DAx is updated, select Exit.
20. Select Exit.
If students are using Remote 28. Make sure DA1 appears in both the Available Servers and
Console and receive errors, cancel Selected Server fields.
the operation and abort the
installation. Normally you should see your server name appear in these
fields.
29. Select Next.
30. When you receive the message that the NDS tree is prepared for
eDirectory 8.6, select Exit.
Because your server does not contain a replica, you cannot
confirm this process.
However, if you were performing this in your work
environment, you would make sure you received the message
NDSEM Process: Complete on your server.
This message verifies that eDirectory on your server is prepared
for eDirectory 8.6.
31. Close Deployment Manager.
Do the following:
1. On WSx, select Start > Run.
2. Select Browse.
3. On the NetWare 6 OS CD, browse to
PRODUCTS\MIGRTWZD\MIGRTWZD.EXE.
4. Select Open.
5. Select OK and allow the files to extract to your workstation.
If you receive a critical error, resolve the error; then begin with
Step 1 of this part of the exercise. You can only perform Step 11
when you receive no critical errors.
If you receive an SMDR critical error, make sure you are
logged in with IPX on both of your connections; then begin
with Step 1 of this part of the exercise. If you still receive
critical errors, ask your instructor for assistance.
11. In the Ready to Copy Files window, copy the file system to the
destination volume tree by selecting Migrate; then allow the file
trustees to back up and the volume files to migrate.
12. Notice on your server the message that login was disabled.
13. In the Copy Volumes Status window, verify that the file copy
was completed with no critical errors.
14. In the Copy Volumes Status window, review the Error log by
selecting View Error Log.
15. In the Copy Volumes Status window, review the Success log by
selecting View Success Log.
16. Scroll to the end of the Success log and verify that volume DATA
was migrated.
Remember that open files cannot be migrated. This is normal.
17. In the Copy Volumes Status window, select Done.
x Do not replace the Server ID number with the IPX internal net number.
To clear up any schema sync 10. In the Ready to Migrate NDS window, begin the eDirectory
issues at the DA1 server console, migration by selecting Migrate.
enter the following:
Notice that your source server (DAx) shuts down during
Set DSTRACE=+SCHEMA migration, because it has been moved to your destination server.
Set DSTRACE=+SYNC 11. In the Migrate NDS Results window, view the Error and Success
Set DSTRACE=*SCHEMA
logs; then select Done.
Set DSTRACE=*SS
Set DSTRACE=*H 12. Close the current project by selecting Close.
13. Check the destination server and verify that it has restarted and
taken on the name of the source server.
14. Close the migration wizard.
Keep the VMWare servers that are Finish the eDirectory migration by doing the following:
not migrated running for the rest of
the course. 1. On your workstation, open the migration wizard by selecting
Start > Programs > Novell > NetWare Migration Wizard >
Turn off only the migrated NetWare Migration Wizard.
VMWare servers when the 2. In the About Novell NetWare Migration Wizard window, select
migration is complete.
OK.
3. Make sure Open Last Project is selected; then select OK.
4. In the Getting Started Migrating window, select Close.
5. In the Project Window, select Finish NDS Migration.
6. In the Continue NDS Migration window, mark Yes; then select
Next.
7. In the Password field, enter novell; then select Next.
8. In the Ready to Continue Migrate NDS window, finish the
eDirectory migration by selecting Continue.
9. In the Continue Migrate NDS Results window, select View
Error Log.
10. Select View Success Log.
11. Scroll to the bottom of the success log and verify that the
migration completed; then close the log.
12. In the Continue Migrate NDS Results window, select Done.
Alternative Installation: During After upgrading your server’s OS, install the latest support pack for
classroom setup, copy the Support NetWare 6.
Pack CD to DA1\DATA\NW6SP2.
Do the following:
Have students install the support
pack by entering 1. Mount the NetWare 6 SP2 CD as a NetWare volume on DAx.
DA1\DATA:\NW6SP2 instead of 2. At the server console, enter NWCONFIG.
completing Steps 1 and 6.
3. In Configuration Options, select Product Options.
4. In Other Installation Actions, select Install a product not listed.
5. To specify the directory path, press F3.
IDE CD Drives 6. In Specify a directory path, change A:\ to NW6SP2: (include the
colon).
If you are using older IDE CD
drives in the classroom, instruct 7. Press Enter.
students to select No for step 14. 8. In the Novell Terms and Conditions screen, press Esc to
continue.
This prevents newer drivers from
causing problems with older 9. Accept the license agreement by selecting Yes.
equipment.
10. In the License Agreement for JReport Runtime JInfonet
software, press Esc to continue.
11. Accept the license agreement for JReport Runtime by selecting
Yes.
12. Install the NetWare Support Pack version 6.0.2 by pressing
Enter.
Alternative Installation: During During the upgrade, you installed eDirectory 8.6. Now you need to
classroom setup, copy the upgrade eDirectory 8.6 to eDirectory 8.7.
eDirectory 8.7 CD to
DA1\DATA\EDIR_8_7. Do the following:
Have students install eDirectory by 1. Mount the eDirectory 8.7 CD as a NetWare volume.
entering 2. At the server console, load NWCONFIG.
DA1\DATA:\EDIR_8_7\NW
instead of completing Steps 1 3. From the Available Options menu, select Product Options.
and 7.
4. Select Install a Product Not Listed.
5. (Conditional) If you receive the Close the Previously Specified
Paths screen, press Esc.
6. Specify the path to the CD by pressing F3.
7. Specify the path to the NW directory where the installation
program can find the NDS8.IPS file by entering volume
name:NW.
For example, EDIR_8_7:NW.
8. Allow the files to copy.
14. Read the warning; then press Esc and allow the files to copy.
16. In the Password field, enter your password and allow the files to
copy and your server to reboot.
17. Authenticate to the Directory and allow the files to copy.
24. In the Summary screen, select Finish and allow the files to copy
26. When the Installation Complete screen appears, remove the CD;
then select Yes and allow the server to reboot.
27. At the server console, enter the following:
SET DSTRACE=ON
SET DSTRACE=+BLINK
SET DSTRACE=*B
This ensures that all your trustees are restored.
28. Switch to the Directory Services screen and verify that the
external reference check process has succeeded.
29. From your server console, enter RESET SERVER.
Point out that students update the Now you upgrade the Novell Client to enable IP and IPX.
client to show them the tasks
involved in performing a thorough Do the following:
migration.
1. On your workstation, insert the Novell Client CD into your CD
drive.
2. In the Client Installation window, select your language.
3. Select Novell Client 4.83 for Windows NT/2000/XP.
4. In the Novell Client Installation window, mark Custom
Installation; then select Next.
5. In the Components to Install window, select Next.
6. In the Protocol Preference window, make sure IP and IPX is
marked; then select Next.
7. In the Login Authenticator window, make sure NDS is selected;
then select Next.
8. Complete the installation by selecting Finish; then allow the files
to copy.
9. On the Installation Complete window, select Reboot.
At the end of this exercise, perform 15. Verify that your license was successfully installed; then select
a full unattended repair on DA1 to Done.
resolve network problems caused
by the migrations. 16. Close the iManager window.
(End of Exercise)
x For issues that can happen before and during installation and migration, see
Known Issues at the NetWare 6 documentation web site at
www.novell.com/documentation/lg/nw6p/index.html.
Summary
The following is a summary of the objectives in this section:
Duration: 3 hours 30 minutes In this section you learn about the tools available for
troubleshooting a Novell network and how to troubleshoot problems
in a mixed IP/IPX LAN environment.
Objectives
1. Upgrade Novell Network Management Tools
2. Identify the Troubleshooting Features of Novell NetWork
Management Tools
3. Identify the Purpose and Function of IP/IPX Troubleshooting
Tools
4. Identify Additional Network Troubleshooting Resources
Introduction
As you consider troubleshooting problems on the network, you
realize how complex your network is. The following identifies
common network communication problems:
■ Workstations can’t communicate with the server
■ Connections are dropped periodically
■ The web browser cannot access a web site
■ Slow network response time
With the knowledge you gained from previous courses, you should
have a firm understanding of the components that make up a
network, and be able to determine where in the network
communication process a specific problem might occur.
x You must use iManager navigational buttons. The browser’s back and
forward buttons do not work.
Alternative Installation: For Step Update iManager 1.2 to iManager 1.5 by doing the following:
2, you can also enter JAVA -CP
{volume_name}: installs 1. Mount the Web Apps CD as a NetWare volume.
\nwMonitorInstall.jar install. 2. At the server prompt, enter EDIRWEBAPPS:WEBAPP.NCF.
3. Select your language; then select OK.
If students have problems
4. In the Novell eDirectory Web Applications screen, select Next.
accessing iManager, do the
following: 5. Accept the license agreement by marking I accept the terms of
the License Agreement; then select Next.
1. Open the HOSTNAME file and
verify that all instances of TEMPx 6. Deselect Novell eGuide - eDirectory White Pages; then select
have been changed to DAx. Next.
2. Open SYS:\APACHE\CONF.
3. Open ADMINSERV.CONF;
then search and replace any
b For a demonstration of Novell eGuide, see www.novell.com.
4. Reset the server and access 9. In the iManager screen, select Next.
iManager again. 10. In the Detection Summary screen, accept the default settings by
selecting Next.
11. In the Pre-Installation Summary screen, select Install and allow
files to install.
12. After iManager is installed, select Done.
13. After the web applications are installed, launch the web browser
by selecting Done.
14. In your browser, select the Getting Started link.
(End of Exercise)
x You can access all Novell web management tools (except iMonitor) from
NetWare Web Manager (https://your server IP address:2200). This includes
management tools for services such as iFolder.
However, if you change the port numbers for services (such as iFolder) you
must change the port numbers in SERVERS.ORG and SERVERS.XML (in
SYS:\WEBAPPS\WEBADMIN) for the management tool link to work
properly in the NetWare Web Manager interface.
ConsoleOne Reports
Introduce ConsoleOne reports as With the release of ConsoleOne 1.3 for Windows, you can generate
a troubleshooting tool. predefined reports to help you troubleshoot network problems. The
other platforms do not support the Reports functionality.
After you extend the schema to support reporting you must install
the desired reports. After the reports are installed, they appear in the
tree, as shown in the following:
Figure 2-1
You must extend the schema and The following predefined report categories are available:
install reporting before you can
demonstrate the reports option in ■ eDirectory General Object Reports. The reports for general
ConsoleOne. objects include NetWare file servers, print servers, and printers.
These reports provide information and status for each object.
■ eDirectory User Security Reports. This report catalog
contains report forms that let you generate reports on
eDirectory login and rights security for users in your tree. The
following reports are available:
❑ Disabled User Accounts
❑ Users Locked by Intruder Detection
❑ Security Equivalence
❑ Template Security Settings
❑ Trustee Security Settings
❑ Trustee Assignments
x You must extend the schema to access the menu items for most ConsoleOne
snap-ins.
Novell iMonitor
Figure 2-2
Using iMonitor, you can monitor your servers from any location
where a web browser is available.
Tell students that they use iMonitor iMonitor lets you look at the eDirectory environment in depth on a
for most administration tasks and partition, replica, or server. You can also examine what processes
troubleshooting in this section. are taking place, when they are happening, what their results are,
and how long they take.
Show each of the following The following is a list of troubleshooting tools in iMonitor:
screens as you introduce the
troubleshooting tools in iMonitor. ■ Reports
■ Trace
■ Repair
Tell students that NetWare NetWare Remote Manager lets you use a web browser to securely
Remote Manager is designed for access NetWare servers from any workstation and perform specific
server administration, but it can server management tasks.
link to iMonitor for eDirectory
administration as well. You can access NetWare Remote Manager from a web browser by
entering https://your server IP address:8009.
Figure 2-3
Using Novell Remote Manager, you can monitor the health of your
servers, their processes, and CPU usage. You can also perform
common server management tasks such as
■ Mounting and dismounting volumes
■ Managing server connections
■ Configuring SET parameters
■ Viewing the server configuration
■ Accessing files on volumes and DOS partitions
■ Shutting down, restarting, and resetting your server.
Using the Console Screens feature, you can view and run all the
console screens just as though you were using the keyboard at the
server console.
Novell iManager
Open iManager and review the Novell iManager 1.5 is a web-based application for managing,
administration options. maintaining, and monitoring eDirectory using wired and wireless
devices.
iManager version 1.5, which ships
with eDirectory 8.7 on the Web Traditionally, eDirectory has been managed through utilities such as
Applications CD, includes a NetWare Administrator and ConsoleOne. These tools are platform
number of new tools. specific and allow management by browsing all objects in a tree.
Figure 2-4
Most of these tools ship with the operating systems being discussed
or are available for download.
CONFIG
Entering CONFIG at the server console is one of the first things you
should do when troubleshooting your server. This utility returns the
following information:
■ The file server name
■ The internal network number of the file server
■ The loaded LAN drivers
■ The hardware settings on all network boards
■ The node (station) addresses of the network boards
■ The communication protocol bound to the network board
■ The network number of the cabling scheme for a network board
■ The frame type assigned to the board (more than one frame type
can be assigned to Ethernet and Token-Ring boards)
■ The board name assigned
NSLOOKUP
Demonstrate NSLOOKUP and Use NSLOOKUP at the server console to identify your DNS
point out the DNS server name configuration, to diagnose DNS setup problems, or to identify DNS
and IP address. problems in an application.
To demonstrate NSLOOKUP, For example, to view the SET commands on your server, enter
enter HELP to show the available SET ALL.
commands. Enter 2 or 3
commands and discuss the 4. When you finish using NSLOOKUP, exit the NSLOOKUP
information shown. utility by entering EXIT.
You can configure the RESOLV.CFG file with the correct DNS configuration
information and then exit and reload NSLOOKUP.
In the HOSTS file, a single host can have from one to ten aliases.
The alias is another name for the same system. Typically, this is a
shorter name. For example, the host Sales could have the following
address and aliases:
139.0.9.5 sales sa
TCPCON
TCPCON lets you view the configuration and statistics for the
SNMP target only, and is an excellent troubleshooting utility for
viewing errors that occur within the TCP/IP stack.
Display Local Traps Shows the local SNMP trap log entries, if
any exist. Requires SNMPLOG.NLM on the
server. Cannot be used to view remote trap
log entries.
PING
PING is typically the first tool to use when connectivity issues exist.
If you can't access a server, try pinging it.
When you initiate a PING, the ICMP echo (ICMP type 8) and
ICMP echo reply (ICMP type 0) packets are used to verify
communication between the devices.
From the server console, you can ping another server by entering
one of the following:
■ PING IP_address
■ PING host name
Figure 2-5
The first line shows the PING test results to a device named
CORPFS1. This test is successful and is supplying good responsive
roundtrip times.
The second line shows that a test to a device with the IP address
10.0.0.2 is failing. The source has sent out 7 packets, but it has not
received anything back. The completion rate is 0%.
The PING utility provides the high and low roundtrip times, as well
as the last roundtrip time, average roundtrip time, and the current
trend of roundtrip times.
DEBUG
The NetWare server supports many debug screens that can help you
identify and resolve TCP/IP-based problems.
x Many of these debug screens are I/O intensive and can affect server
performance. Use these screens with caution.
IPCONFIG
Figure 2-6
The syntax is
IPCONFIG [/? | /all | /release [adapter] | /renew
[adapter]]
PING
Similar to the PING utility on the server, you can use this command
from the Windows command line.
This is one of the first tools you should use for any problem that
appears to be caused by a lack of connectivity between network
devices.
-w timeout Defines the time (in milliseconds) to wait for each reply
The station will ping its own IP stack, as shown in the following:
Figure 2-7
If it does not see that its own stack is active, it cannot communicate
on the network.
When the workstation cannot ping its own stack, do the following:
1. Access My Network Places > Properties > Local Area
Connection > IP Protocol.
2. Identify the following:
❑ Verify the IP address information is acquired using one of
the following methods:
❑ The IP address is automatically assigned.
❑ A dedicated IP address has been entered and the
information is correct.
❑ Verify the DNS settings are established using one of the
following methods:
❑ The DNS server address is automatically assigned.
❑ A dedicated DNS server address is assigned and the
information is correct.
Figure 2-8
ROUTE
The next time Michelle sends a packet to Drake, she will find a
network entry for 204.10.11.0 in her routing tables that indicates she
should forward such traffic to router 204.10.10.4.
Ping 204.10.10.3
Redirect Router
d
ar
rw
Fo
Drake
Router
204.10.11.5
204.10.11.0
Figure 2-10
Figure 2-11
Figure 2-12
-w timeout Sets the wait time (in milliseconds) for each reply.
NSLOOKUP
Figure 2-14
ARP
-g Same as -a.
NETSTAT
Figure 2-15
“I waited, and waited, and waited. I found that I could not connect
to IANA to validate the list. This was a perfect time to test some
troubleshooting tools.”
Here are the steps the author used to isolate the problem:
1. Ping www.iana.org. Results: Successful.
2. TRACERT to www.iana.org. Results: Successful.
3. Use NSLOOKUP to obtain the IP address for www.iana.org.
Results: 128.9.160.83.
4. Portscan 128.9.160.83. Result: HTTP port 80 is not active on the
host.
Figure 2-16 shows the result of the port scan that day. As you can
see, the host is running.
The HTTP port (used for the web server) is not functional. The Port
Scanner indicates that the HTTP daemon is not loaded.
Figure 2-16
The following shows what the port scan looks like on a normal day
when the web server is running:
Figure 2-17
❑ IP address
❑ Subnet mask
4. Where would you typically record this information and why?
11. What command should you use to identify the path packets might
take to go from your workstation to novell.com?
13. From your server, use the IPTRACE command to identify the
path from your server to the instructor server.
(End of Exercise)
Protocol Analyzers
Installing and demonstrating a Protocol analyzers (also called network analyzers or sniffers)
protocol analyzer (such as capture packets on the cabling system and show conversations and
LANalyzer) is much more effective individual packets in a readable format.
than discussing it.
Protocol analysis let you listen in on network communications to
Some students might not be determine the health of that network.
familiar with the term protocol
analyzer, but they might be familiar To understand protocol analyzers, you need to know the following:
with sniffer.
■ Analysis Session Procedure
■ Use of Protocol Analysis
Packets Packets
Trace Buffer
Decodes
Decodes
Display Display Display
Filter Filter Filter
Plug
You can use the display filter to view the packets you are interested
in. All the captured packets are still in the trace buffer; you have just
chosen to view a portion of them.
Captured packets are sent through a capture filter. You can set the
capture filter to grab all packets being sent to and from the server's
hardware or software address. You can then set a display filter to
filter out everything but broadcast traffic.
The resulting filtered buffer would show only broadcast traffic from
the server. This is one example of how to use protocol analysis to
identify a possible problem.
Troubleshooting
Optimization
When you test an application or plan for growth, you can determine
how much bandwidth a single user requires to run a particular
application on the network.
TCP/IP Toolkits
IP Addressing Calculators
Several tools can help you calculate subnet addressing for TCP/IP.
The IP Subnet Calculator from WildPackets is a freeware calculator
that offers a range of functions.
Figure 2-19
Access www.novell.com and show The Novell web site at http://www.novell.com has links to
the links described. Show students troubleshooting tips and tools. The following are a few of the
your favorite links and have them available resources:
explore the links available.
■ Solutions. You can access white papers that describe Novell
business solutions that enhance and improve your existing
systems, allowing you to meet new business requirements more
quickly, and make sure every dollar you spend on technology
translates directly into lower costs and higher revenues.
■ Products. You can access products listed from A–Z. From this
page, you can link to any product for information, FAQs,
highlights, what’s new, and Novell AppNotes.
■ Training. You can access information about Novell Education
products, services, certifications, and events. You can also
access Novell AppNotes.
■ Support. You can access online support resources, including
the following:
❑ Novell documentation. You can access all product
documentation online. The product documentation includes
troubleshooting sections and error code information.
Figure 2-20
b To access tools, demos, and evaluation copies of commercial tools that work
with or enhance the functionality of many Novell products, access the Novell
Cool Solutions site at http://www.novell.com/coolsolutions/tools.
Summary
The following is a summary of the objectives in this section.
Exercise Answers
11. What command should you use to identify the path packets
might take to go from your workstation to novell.com?
TRACERT
13. From your server, use the IPTRACE command to identify the
path from your server to the instructor server.
Results will vary.
Duration: 3 hours
Setup: The files needed for this In this section you learn how to troubleshoot and resolve problems
section should have been that occur on the server.
migrated to volume DATA from the
NetWare 4.11 server.
Introduction
Now that you have upgraded servers and learned about Novell and
IP management and troubleshooting tools, you are ready to learn
about troubleshooting problems that occur on servers, resolve
critical server abends, and resolve server communication issues.
Bus Types
Similarly, a server must have a wide enough bus to meet high usage
peaks.
Mass Storage
You discuss RAID later in the Servers typically support either software or hardware RAID. This is
course. discussed in “Troubleshoot Software RAID and Mirroring in NSS”
on 6-19.
Processor Capacity
Students might want to know how NetWare 6 supports up to 32 processors, with a minimum
many processors are commonly requirement of a Pentium II or AMDTM K7 processor.
used. Before class, you might want
to review the product information The Xeon, Pentium 4, and Pentium III processors are commonly
and processor capacity currently used in servers.
offered by server vendors.
During installation, NetWare detects multiple processors by reading
the multiprocessor (MP) configuration table in BIOS and then
determining which of the available NetWare Platform Support
Modules (PSMs) matches the MP hardware platform.
You can then load the PSM or run NetWare on Processor 0 only.
The installation program will modify the STARTUP.NCF file to
load the PSM whenever the server is started.
Compaq and other vendors also provide a PSM for their system
requirements.
Memory Capacity
Scalability
Failure Recovery
■ Multitasking
■ Multiprocessing
■ Processor Load Balancing
■ Pre-Emption
The server has a load template that defines what NLMs will be
loaded at each stage of the boot process.
■ LOADSTAGE 2
■ LOADSTAGE 3
■ LOADSTAGE 4
■ AUTOEXEC.NCF
■ LOADSTAGE 5
Demonstrate the LIST STAGE You can view the modules that are loaded in each stage by entering
command. LIST STAGE at the server console prompt. This can be helpful
when troubleshooting problems that occur when the server starts.
Demonstrate module colors by After the server boots, NLMs can be viewed with the MODULES or
typing M at the console of DA1 and M command. You will notice they are color coded, which is useful
discuss the blue, red, and white when troubleshooting.
color codes and load locations.
The following provides a description of each color:
Cyan (light blue) Loaded NLM was bound into SERVER.EXE and
loaded from there.
For example, PVER500.NLM, is bound into
SERVER.EXE.
During the NetWare 6 boot process, the module will load and
indicate the success of the load.
Demonstrate the Logger Screen. The Logger Screen can be used to review the load status of
modules.
Students might want to know what Pressing F1 on the Logger Screen shows the navigation keys. F2
PUB EXISTS indicates when they saves the output as a text file to C:\NWSERVER\LOGGER.TXT.
view this screen. It is short for
public symbols.
Kernel
This message is informational
(yellow) and indicates that the
The kernel is the core of a network OS. It provides fundamental
public symbols required for that
operating OS, such as handling interrupts and the I/O system.
NLM have already been loaded.
In NetWare 4, there were 2 kernels, one for uniprocessing and one
for multiprocessing. Starting with NetWare 5, the 2 kernels were
integrated into one.
Threads
The kernel maintains a data structure called the run queue, which
contains threads that are ready to be executed by the processor, as in
the following:
Processor 0
Server 20
Multithreading
Excecution Time
For example, thread 1 executes code, then it yields and waits while
thread 2 executes code. Thread 2 then yields to thread 1 to process
another string of code. This process continues until both threads
completely process.
Processor 1
Excecution Time
Multitasking
Multiprocessing
Analogy: The scheduler performs Aside from the processors, the component that drives the
actions similar to a police officer functionality of multiple processor servers is called the scheduler.
directing traffic. The scheduler determines how to distribute threads.
Memory
2 3
1
Processor 0 6
Software
Kernel
4 5 Either
Processor 1
Scheduler
Threads
Pre-Emption
Do the following:
1. From NetWare Remote Manager, select Disk / LAN Adaptors
from the Manage Hardware section.
2. Select Storage Statistics.
3. Review the Current and Peak statistics and record the values in
the following:
IO Request Rate:
Do the following:
1. From NetWare Remote Manager, under Manage Hardware
select Disk / LAN Adaptors.
2. Select Network Statistics and view the statistics.
3. Select Packets per Second Graph to monitor the current traffic.
4. Close the graph.
12. Check the Disk Controller upgrade the drivers to the latest
versions.
Do the following:
1. From NetWare Remote Manager, under Manage Hardware,
select Processors.
2. View the information for Processor 0.
If your server has multiple processors, they will also be listed.
(End of Exercise)
The disk driver has ■ Make sure both the HAM and CDM driver are
not been loaded. loaded.
■ Check the logger screen to see if the modules
loaded correctly or try loading the modules
from the server console prompt.
■ Verify that the Disk Adaptors list in NetWare
Remote Manager matches actual hardware.
Click the Disk / LAN Adaptors link in the
navigation frame; then click the Info icon for
each disk controller in your server.
■ Try loading the driver from the server console
prompt on the server.
A hard disk is not ■ Check the cables between the hard disks and
installed or cabled the controller boards. Make sure that Pin 1 of
correctly. each cable is attached to Pin 1 of each
connector.
A general rule is that the red stripe on the
cable is closest to the power connector on the
drive.
The communication ■ Check the power cables and make sure they
channel between are seated correctly in the power sockets on
the controller the hard disks.
interface board, the ■ Verify that no interrupt conflicts exist.
disk coprocessor ■ Repair or replace the faulty hardware.
board, and the hard ■ Verify that the driver is current.
disk is not
functioning.
The Network cabling ■ Check the cable with a cable tester and
is faulty. replace faulty cabling.
The server hard disk ■ Check the Hot Fix status of all hard disks that
is slow or faulty. use the traditional file system.
To view the status, load MONITOR.NLM, and
from Available Options, select Storage
Devices, Hot Fixed Partition.
A non-zero count for Used Hot Fix Blocks
indicates problems on the drive.
The server is low on ■ Free memory and then add more if required.
memory.
The volume has too ■ Purge deleted files by doing the following:
many deleted files From NetWare Remote Manager, select
that have not been Volumes, the Volume Information icon next the
purged. volume you want to delete files on, Purge
Deleted Files.
■ You can also set the Purge attribute on files
you want to be purged.
In this section you learn how to show processor status and start and
stop individual processors in a multiprocessing server. Only
secondary processors can be handled this way.
You can start or stop secondary processors at any time while the
server is running by using the console commands START
PROCESSORS and STOP PROCESSORS or by clicking the Start
Processor or Stop Process link on the Processor Information page in
NetWare Remote Manager.
Stop Processors
Start Processors
You will find hundreds of tips, articles, Q&A, and free tools.
AppNotes
b Alexander SPK v4.1 for NetWare runs on NetWare 3.x through 6. For
more information, see http://www.alexander.com/.
a 30 Minutes
Verify that PROBLEMS.NLM was After Digital Airlines upgraded to NetWare 6, the help desk
migrated in volume DATA received some support calls. As the network administrator, you also
(EXERCISE\SECTION3) to each noticed a few problems. The issues include performance,
student’s NetWare 6 server. communication, and file access.
In this part of the exercise, you were able to locate an NLM using
too much memory and unload it remotely through NetWare Remote
Manager.
3. Leave the Trace Portal window open (do not press Enter).
4. Locate the source of the problem:
a. Select Health Monitor.
b. Select CPU Utilization and view the graph.
5. Notice that the CPU Utilization status is Bad and the utilization
is staying at about 98-100%.
6. Select Diagnose Server > Profile / Debug and note that
PROBLEMS.NLM is causing the problem.
7. Unload the NLM:
a. Select PROBLEMS.NLM from the list.
b. Select UNLOAD > OK.
8. Select Health Monitor.
9. CPU Utilization status should now indicate Good.
In this part of the exercise, you were able to locate and resolve a
server CPU hog using NetWare Remote Manager.
It is now the day after the upgrade and you decide to check the
console error log of your upgraded server:
1. From NetWare Remote Manager, select Health Monitor and
notice that the list of items monitored do not report a problem.
2. Select Reports / Log Files.
3. Select System Error Log File.
While scrolling through the log file, you notice the following
error near the end of the file:
“Free detected corrupt trailing redzone for node
0xCCB72140, node size 24”
From an NTS Knowledgebase query you determine that the
cause is a portion of memory has been improperly overwritten
by an errant program.
The specific module mentioned in the article is not running on
the server, so you decide to troubleshoot the problem further.
You suspect that it might be an application named
PROBLEMS.NLM that was installed prior to the upgrade.
4. Select List Modules.
5. Locate PROBLEMS.NLM in the list.
6. Select the Alloc Memory number for that NLM.
7. From the PROBLEMS.NLM Allocation Summary, select
Display Memory Allocation Information by Size.
8. View the Corruption Count column for any non-zero counts and
select that item.
A corruption count indicates that the header or footer of the
allocated memory has been corrupted.
9. Select the Address link to further view the data.
10. Switch to the PROBLEMS.NLM screen on the server.
You now know that there is a corruption problem, you located the
application causing the problem, and you know the size of memory
corrupted and what data written to it.
You can further resolve the problem when you contact the vendor
and provide them with this information.
The Digital Air help desk reports that the file SYS:TESTDB.DAT
on your server cannot be updated because it is being held open by
someone. Find out who has the file open and alert the person to
close the file.
6. Select Connections.
7. Select the user name for one of the connections that have the file
locked.
8. From the Connection Information screen, you can see that the
user has the file open.
9. In the Send Message box, enter a message asking the user to
close the file.
10. There was no response from the user, so you decide to clear the
connection so that the file lock is removed:
a. Select Connections.
b. Next to the user that you sent the message to select Clear
Connection.
11. Select Volumes.
14. From the Global Lock Information section, notice the Use Count
is now 3.
In this part of the exercise, you were able to find out who had a file
open, communicate with that user, and free the file for others to use
by clearing the connection of the user that had the file locked.
14. Scroll down to the server process threads named Server 00,
Server 01, and so forth, and notice that the Thread State is
Waiting for work for all server processes.
(End of Exercise)
What an Abend Is
After an abend occurs and the server abruptly halts, disk I/O
requests and other processes might be in an incomplete state.
A server that has abended will have the number of times the server
has abended added to the server name on the console. For example,
DA1 <1>:
■ Abended NLM
■ Running process
■ Stack limit and pointer
■ Stack trace
■ Modules list
Define the term core dump. If an abend continues to occur, it might be necessary for Novell
Technical Support to analyze a capture of the server’s memory (a
core dump).
The term core dump comes from the mainframe environment where
RAM was referred to as core memory. The process of saving
(dumping) a snapshot of a server’s memory to disk is referred to as
creating a core dump.
The core dump contains the following information that can be used
to analyze the problem:
■ Processes. All processes on the server at the time of the abend
are included in the core dump. The state of these processes can
be running, waiting to run, or not in use. A history of what the
process has done (call stack) is also preserved in the core dump.
■ Loaded modules. This includes module information, code, and
data of all NLMs.
■ Allocated memory. This includes the memory allocated by
processes included in the core dump.
■ Cache memory. Memory that is available for allocation by
modules or processes can also be included in the core dump.
You can respond to the abend manually or have the server respond
automatically.
When you respond manually, the server determines the nature of the
abend and shows the appropriate response option on the screen,
along with additional options for bringing down the server or
executing a core dump. You must execute an option to respond to
the abend.
x Sometimes an abend (or a faulty NLM) can cause the server console to stop
functioning. In this case, the abend message is not shown and you cannot
enter commands at the server console prompt.
Demonstrate how to set the The default parameters on the server are set to respond to an abend
parameters in NetWare Remote automatically. The server automatically recovers from most abends
Manager. and continues functioning normally.
Users can save their files before the server is restarted and file
system corruption can be avoided because volumes can be properly
dismounted.
Auto Restart After 0. Do not try to recover from the abend. The
Abend = 1 (values 0 server is left in a halted state. This option is
– 3) discussed more in “Respond to an Abend
Manually” on 3-54.
1 (Default). For software abends, NMIs, and
Machine Check Exceptions: attempt to recover
from the problem, bring down the server in the
configured amount of time, and then restart the
operating system.
For other exception abends, suspend the
faulting process and leave the server up.
2. For all software and hardware abends,
attempt to recover from the problem, bring down
the server in the configured amount of time, and
then restart the operating system.
3. For all software and hardware abends,
immediately restart the server.
Auto Restart After This setting indicates how many minutes the
abend Delay Time = server will wait after an abend occurs before
2 (Range: 2 to 60 going down and restarting itself.
minutes) In most cases, the server can recover but is in a
critical state that requires a restart.
The purpose is to prevent data loss by giving
users time to save files and log out before the
server is restarted.
x Several options have the same first letter (such as R, S, or X). In a given
abend situation, the option list includes only one option for any given first
letter.
Turn off and back on If the console has been secured, you must turn
to restart. the server off and then back on to restart the
server.
Demonstrate how to view the You can use NetWare Remote Manager to view the memory
memory configuration from configuration. It will show total memory, cache memory, and how
NetWare Remote Manager (from much memory is used by the file system, swap files, NLMs, and
Manage Server, select View virtual memory.
Memory Config).
There are several options when creating a core dump that can affect
the size of the file. You can include all memory or exclude cache
memory. You can also use compression, which reduces the disk
space needed.
The size of the full image file is approximately equal to the total
RAM installed in the server.
x DIAG500.NLM 2.1.0 or later also excludes NSS cache when the exclude
cache option is selected. This results in a smaller core dump image.
The default method is to save the core dump locally to the server’s
DOS partition as C:\COREDUMP.IMG.
Using an add-on utility called DBNET, you can also save the core
dump to a remote server or workstation. DBNET is included in the
NetWare 4, 5, and 6 support packs.
Core dumps can be analyzed by Novell engineers and are often the
key to finding software bugs.
The core dump must still point to the cache memory pages that
have been excluded. These are referred to as phantom pointers.
The page size is based on the memory chip, such as 4K.
b Users can’t access the server while the NetWare Internal Debugger is
active.
After you start the core dump, you are asked to specify the DOS
drive letter and file path that the memory image file will be written
to.
The drive can be any writable DOS device that contains enough
storage space. The device must be set up not only before the server
abends but before the server is booted.
After the server is running, the image file can also be copied to a
workstation or network drive using NetWare Remote Manager.
For servers not running NetWare Remote Manager, this can be done
using IMGCOPY.NLM or any other third-party NLM that provides
this functionality.
This method is useful when the DOS partition is not large enough to
hold the core dump. This method also greatly decreases the time
that the server must remain in a downed state for diagnostics.
DIAG500.NLM NetWare 5x
Integrated into SERVER.EXE
Before you send the file for analysis, make sure that your core dump
■ Contains useful information:
❑ For abends, verify Auto Restart After Abend = 0. This helps
ensure the core dump contains information right after the
abend occurs.
❑ For performance problems, verify the core dump is taken
during the time the problem is occurring.
■ Can be opened with the NetWare Virtual Debugger.
A senior Support Engineer will analyze the memory image file and
recommend a solution.
a 10 Minutes As the network administrator, you noticed one of the servers had
experienced an abend. As a first step, you decided to submit the
ABEND.LOG file to Novell’s Abend Analysis System to see if it is
a known issue.
Internet access is required for this An NLM will be loaded on the server to simulate the abend.
exercise.
Do the following:
1. On your server, use NetWare Remote Manager to load
PROBLEMS.NLM:
a. From the Manager Server options select Console Screens.
b. Select Console Screens.
c. Load PROBLEMS.NLM.
2. From the menu select Page Fault.
❑ Abend message
❑ Registers
❑ Abended NLM
❑ Running process
❑ Stack limit and pointer
❑ Stack trace
❑ Modules list
(End of Exercise)
a 30 Minutes As the Digital Airlines network administrator, you have been unable
to resolve an abend on your server after installing a new application.
When the core dump is started, You have opened an incident with Novell Technical Support to help
you might want to suggest the resolve the problem. The Support Engineer helping you has
students take a break for a few requested that you send a core dump image for analysis as soon as
minutes. the server abends.
You might want to take a core Complete the following:
dump prior to class to estimate the
time required based on the ■ Part I: Save a Core Dump Image to a Local Drive
classroom hardware being used.
■ Part II: Save a Core Dump Image to a Remote Drive
■ Part III: Validate a Core Dump Image
Do the following:
1. From NetWare Remote Manager, change Auto Restart After
Abend = 0 to keep the server from automatically restarting.
This lets you obtain a core dump of the memory after the abend
occurred. Do the following:
a. Select Set Parameters.
b. Select Error Handling.
c. Select the value for Auto Restart After Abend.
d. Change the value to 0.
e. Select OK.
2. Load PROBLEMS.NLM on your server console.
3. From the menu select Page Fault.
4. Review the abend message.
5. Copy the diagnostic image to disk (core dump) by selecting Y.
6. As the Coredump Type select 2 - Full W/o Cache (All Server
Memory Except File Cache).
7. For Compress Coredump select 1 -Yes.
8. Accept the default path (C:\COREDUMP.IMG) by selecting
Enter.
9. Allow the core dump process to complete.
This process might take several minutes to complete. After the
core dump completes, you are prompted to create another core
dump.
10. Do not press Y.
Do the following:
1. From DAx\\DATA\EXERCISE\SECTION3, copy
DBNET6.NLM and IMGHOST.NLM to SYS:SYSTEM.
2. From DAx\\DATA\EXERCISE\SECTION3, copy
IMGHOST.EXE to C:\COREDUMP.
(These files are also available in EXERCISES\SECTION 3 on
your Enhanced Learning CD.)
3. On your workstation launch IMGHOST.EXE.
4. On your server console load DBNET6.NLM.
8. Press Enter and notice the server console prompt now indicates
<1> abend has occurred.
9. Review the abend message and notice that the offending process
is suspended.
10. Enter the NetWare Internal Debugger by selecting Right-Shift +
Alt and then Left-Shift + Esc.
11. At the # prompt, enter .C to start the core dump.
12. As the Coredump Type select 2 - Full W/o Cache (All Server
Memory Except File Cache).
13. For Compress Coredump select 1 -Yes.
17. Record the filename of the core dump image for later reference:
(End of Exercise)
■ Check the server to make sure the CPU use is not remaining at
or close to 100%.
■ Verify that DSREPAIR.NLM is not loaded and locking the
eDirectory database.
With Client 32 and Windows 95/98, the login window appears only
if a file server is located. If you have NWEnableLogging=True, then
NIOS.LOG will show A Server Could Not Be Found.
Monitor Servers
You also need much of this information if you call Novell Technical
Support. You might want to keep documentation in a notebook or
online in a database.
Demonstrate accessing the Server Using NetWare Remote Manager, you can keep a Server Personal
Personal Log Book. Log Book at SYS:SYSTEM/NRMUSERS.LOG.
Select Reports / Log Files > Server Personal Log Book to enter and
track changes made to the server or to log information you want to
keep to track server performance or history.
a 20 Minutes In this exercise, the instructor will introduce a problem and you will
need to troubleshoot the following as a group:
Complete this exercise as a group. ■ Part I: The Server Is Showing Errors
Set up the problem at the ■ Part II: The Workstation Cannot Login
beginning of each part. ■ Part III: The Admin Account Has Been Deleted
Present the problem for discussion ■ Part IV: Workstation Connections Are Dropping Periodically
and let the students come to a
■ Part V: You Cannot Map Drives
conclusion. Guide them as needed
and then discuss a solution.
Setup: Simulate a -625 error by Your server is showing an error message Unable to Communicate
removing the network cable from with Server .DA2.DIGITALAIR.
DA2 and show the console screen
on the overhead. When you run a DSREPAIR you are getting -625 errors for server
DA2.
See “Resolve Server-to-Server 2. List the items you can check to locate the source of the problem.
Communication Problems” on
3-71.
Setup: Simulate a login problem by Users are reporting that they cannot log in to server DA2. They can
using the PROBLEMS.NLM to attach to it, but drive mappings fail.
dismount volume SYS on DA2.
1. Add the drive mapping MAP K:=\\DA2\SYS\ to your
administrator login script.
2. Log in to your admin account.
You have checked the router and cabling and everything seems to be
functioning properly.
Guide the students using standard 1. What is the most likely source of the problem?
troubleshooting methods. If
needed, provide them with hints
that will lead them to the suspect,
an outdated network board driver.
Solution: Update the LAN driver to 2. What can you do to correct the problem?
the latest version by downloading
it from the manufacturer’s web
site.
(End of Exercise)
Summary
The following is a summary of the objectives in this section:
Objective Summary
Objective Summary
Objective Summary
Exercise Answers
3. Review the Current and Peak statistics and record the values in
the following:
Answers will vary.
Part II: Review the Network Board Driver Statistics and Version.
17. Record the filename of the core dump image for later reference:
Answers will vary. The format of the filename will be a
hexadecimal value with an IMG extension (for example,
1021213AA.IMG).
2. List the items you can check to locate the source of the
problem.
See “Resolve Server-to-Server Communication Problems” on
3-71.
Duration: 3 hours In this section you learn how to monitor and troubleshoot
eDirectory.
Objectives
1. Identify eDirectory Databases and Processes
2. Identify eDirectory Troubleshooting Steps
3. Identify Partition and Replication Placement Design
4. Use iMonitor Reports to Obtain Server and eDirectory
Information
5. Perform Health Checks
Introduction
When you make changes to eDirectory, the changes are replicated
throughout your eDirectory tree. The size of your eDirectory tree,
the number of servers, and the number and replicas you have,
determine the time it takes to distribute these changes.
You should perform regular status reports and health checks on your
system to ensure it is trouble free.
The files that comprise the eDirectory database are shown in the
following:
NDS.DB
NDS.01
(.02, .03, etc., as needed)
eDirectory Processes
When you make changes to eDirectory, you should make sure that
the following processes are complete before you make additional
changes:
■ Time synchronization. Time synchronization is very important
to eDirectory. All servers in a tree must be synchronized to the
same time source. If they are not, collisions will occur when
objects are being synchronized in replicas.
Synchronizing time across the network lets you maintain
consistent time stamps.
The most common time stamp problem is with synthetic time.
Synthetic time occurs when an eDirectory object has a
modification time stamp ahead of current network time.
If the period between current time and the synthetic time is
small this problem will correct itself. However, if the period is
large you might need to resolve the problem manually.
Ask students if they use similar If eDirectory does not correct itself, then you should use most, if not
troubleshooting steps to resolve all, of the following eDirectory troubleshooting steps:
the problems they experience.
■ Step 1: Identify the Scope of the Problem
■ Step 2: Determine the Cause of the Problem
■ Step 3: List Possible Solutions to the Problem
■ Step 4: Assess Possible Solutions
■ Step 5: Implement a Solution
■ Step 6: Verify That the Problem Is Resolved
■ Step 7: Document the Resolution to the Problem
■ Step 8: Avoid Repeating the Problem
Before you can resolve a problem, you need to know the extent of
the problem.
Demonstrate accessing these log The following shows how Novell Remote Manager provides easy
files using Novell Remote access to the logs:
Manager > Diagnose Server.
Figure 4-2
Not all eDirectory issues are obvious, and an eDirectory error code
might only be a symptom of the real issue.
After you determine what process is having the problem and why,
you can formulate a solution to the problem.
x Removing and reinstalling eDirectory should be used as a last resort, after all
other possible solutions have been evaluated.
Ask coworkers and others what actions they would take. If possible,
test the solutions in a lab environment before implementing them.
The tools used to implement a solution are often the same tools used
to diagnose the problem.
After the actions you took to resolve an eDirectory issue have been
processed, verify that the problem is resolved.
Demonstrate adding information to Novell Remote Manager has a user friendly interface to the
the server log as a documentation NRMUSERS.LOG file, which is the server log.
procedure.
NRMUSERS.LOG allows you to
■ Prevent the same problem in the future
■ Find a resolution to the same problem quickly in the future
■ Provide insight into other problems your network might have
Although steps 7 & 8 are similar, The final step in the process is to avoid repeating the problem.
the purpose of step 8 is to
reinforce the benefits of You might find that an eDirectory problem occurs by an inadvertent
documenting the problem. or inappropriate action or that no proper procedure has been defined
to perform the action properly.
Whatever the cause of the problem, you want to make sure it does
not get repeated.
You should
■ Document the problem and the solution.
■ Establish procedures and policies to ensure that people who
administer or use the eDirectory tree will do so in a consistent
and established manner.
■ Take precautions, such as restricting access to servers.
As you review these log files, you question if partitions and replica
placement meet the needs of the merged organizations.
Figure 4-3
a 15 minutes In this exercise, you create a replica and change the replica type to
better suit your needs.
DA4 IS.DEL.DIGITALAIR
DA5 IS.LGA.DIGITALAIR
DA6 IS.LON.DIGITALAIR
DA7 IS.SYD.DIGITALAIR
DA8 IS.TXL.DIGITALAIR
DA9 IS.TYO.DIGITALAIR
Figure 4-4
Figure 4-5
10. Verify your server name; then use the browse button to locate
and select your location container (such as DEL.DIGITALAIR).
11. Specify that this will be a Read/Write replica; then select OK.
Figure 4-6
The replica view dialog appears, with the Master replica type
now available.
15. Select Master; then select OK.
(End of Exercise)
Tell students they will use some of In the Agent Summary screen, the status of your server is shown by
these reports to verify their system the icon at the left of the server address. The server status is
in an exercise. represented by the color of the signal light: green means the server
is functioning properly, yellow indicates a problem, and red
signifies that communication cannot happen.
You can also run reports to determine the status of the server and
eDirectory.
Demonstrate running reports. Tell When you select the Report icon you will notice that there are 2
students that they will run a report Report options (Reports and Report Config), as shown in the
in the next exercise. following:
Figure 4-7
Figure 4-8
Move your cursor over an icon and The list of preconfigured reports appears.
point out the various icons on the
reports screen. Use one of the following icons to access report options:
❑ Red arrow. Use to run the report.
❑ Check boxes. Use to configure report options.
❑ Question mark. Use to access report help.
Figure 4-9
This report lists the obituaries that match the specified criteria.
4. Select Run Report.
Figure 4-10
a 15 minutes As the system administrator for your Digital Airlines office, you
have just upgraded a server from NetWare 4.11 (with NDS 6.11) to
NetWare 6 (with eDirectory 8.7).
Discuss the questions in this You need to confirm with Mark Bassil, the vice president of IT, that
exercise as a group to facilitate the upgrade is complete.
students comparing results and
helping each other determine To verify that the schema has been properly updated and distributed
causes and solutions. to all partitions and replicas, run the Server Information report.
Do the following:
1. Access iMonitor using the IP address for your server
https://192.168.1.x:8009 (where x = your server number).
2. Enter your userid and password: then select Login.
3. Select the NDS iMonitor link.
4. Select the Report icon.
If this is the first report, no reports are shown.
5. Select Report Config.
6. From the list of runable reports, select the Configure Report
icon for Server Information.
7. Accept the default report options by selecting Run Report.
13. From the list of runable reports, select the Configure Report
icon for Server Information.
14. Select the Try: IPX option; then run the report a second time.
15. What are the current errors for any IPX servers in your network
(such as DA3)? Why have they changed?
(End of Exercise)
You can use iMonitor to perform the same tasks you might have
performed using the following tools: DSTrace, DSRepair,
DSBrowse, and NDS Manager.
Figure 4-11
Figure 4-12
Figure 4-13
Figure 4-14
3. From the list of runable reports, select the option to run the Agent
Health report.
The report results appear similar to the following:
Figure 4-15
4. When problems exist, you can select the Health Check > Agent
link to see the reason for the marginal status, as seen in the
following:
Figure 4-16
The 603 error lets you know that an attribute is not defined. The
625 error tells you that the communication process for the
schema is not happening, most likely due to a LAN problem.
These errors are not critical at this time, but if left unattended,
they could cause problems. To identify the full scope of the
problem, check the partition and replication error you saw.
5. Use the Back button to return to the report status.
6. Select the Partition/Replication link.
If there are no errors, the Result area will show a green
indicator. Because there are errors in this example, the
following appears:
Figure 4-17
This opens the details for each replica. You can see that
replication is not happening, as shown in the following:
Figure 4-18
Figure 4-19
Figure 4-20
Demonstrate running Trace. In most cases, the default trace line prefixes will meet your
needs.
2. Specify the DS Trace Options; then scroll to the bottom of the
screen and select Trace On.
These settings are stored for this trace run.
After you specify the options, you must tell the agent what
process to perform these options on.
3. Select Agent Configuration using the navigation button at the
left of the Trace Configuration button.
After you select Trace On and select Agent Configuration, a
new icon appears on the navigation bar. This is the Trace
button, which you use to access all Trace options.
4. Select Agent Triggers; then specify the background process you
want to trace.
5. Select Submit to begin the trace.
6. Select the Trace icon.
Figure 4-21
Figure 4-22
When you find problems in your eDirectory tree, you can initiate
the Repair process to correct them.
As you learned in Course 3004, you can run Repair on the server
where you are running iMonitor.
Figure 4-23
x If Repair is loaded at the server, the Repair option in iMonitor cannot run.
Figure 4-24
As you review the log file, notice that the databases are checked,
errors are identified, and are then corrected.
a 20 minutes In this exercise you manually perform a health check, and then run
an Agent Health Report.
The results of the manual health check are identical to the results
you receive when you run the Agent Health Report.
Discuss the questions in Part I as You perform the following Health Check procedures:
a group to facilitate student
understanding of the results. ■ Part I: Check Schema Synchronization
■ Part II: Check Agent Status
■ Part III: Perform Health Check on Subsequent Servers
■ Part IV: Run an Agent Health Report Check
12. From the list of trace links on the left, select Trace Live.
13. List and describe the 2 health check categories and their function:
❑
14. From the health check you just performed, are you satisfied that
eDirectory on your server functioning properly?
x You might see a 628 error for DA3 because your server can’t synchronize to
that server.
You can use one of the following to ensure that all servers in your
tree are functioning properly:
■ Perform DSTrace and DSRepair at the server console.
■ Change the URL in iMonitor to pull the information from a
specific server.
■ Use the links in Replica synchronization to view each server.
Do the following:
1. On the left under the Links heading, select Agent Health.
On the right, a Health Check list appears.
2. Select an option (such as Partitions/Replica) to run a report.
3. What additional information is shown?
4. When would you use the Agent Health Report check feature
instead of performing your own health check with the options in
Parts I - III?
(End of Exercise)
Figure 4-25
Figure 4-26
Figure 4-27
Figure 4-28
From the Trace Live screen, you can select the Red text to be
linked to the Error section in the Help file, as seen in the
following:
Figure 4-29
Figure 4-30
This shows the error letting you know that the attribute
STATUS has been changed and does not match the schema.
3. From your browser, access http://www.novell.com; then select
Support > Knowledgebase.
4. Perform a search using 608 Illegal Attribute as your search
criteria.
A number of TID references are listed, relating to the problem
and possible solutions.
5. Using the Error Codes help screen and Knowledgebase, what
could have caused the problem?
1.
2.
3.
1.
2.
3.
In this scenario, it was determined that you should repair the user’s
record. The following shows the repair dialog:
Figure 4-31
The Repair Single Object option is selected and the user’s record
appears. This happens because the user’s record was being viewed
prior to selecting Repair.
Figure 4-32
Figure 4-33
4. How did the problem occur? (You might not know how the
problem was introduced to your system, but if you do, record it
so it can be prevented.)
6. What additional actions did you take to prevent the problem from
occurring again?
The problem you just experienced and resolved can occur again in
the future for a number of reasons.
With the problem and solution documented, you can save yourself
and your peers a great deal of time if this problem occurs again.
Consider how you would respond to the following:
1. In addition to documenting the problem, what can you do to
avoid duplication of problems?
Summary
The following is a summary of the objectives in this section:
Objective Summary
Objective Summary
Objective Summary
Objective Summary
Exercise Answers
15. What are the current errors for any IPX servers in your network
(such as DA3)? Why have they changed?
The remote server is up because the NetWare 4.11 server uses
IPX. By using the IPX protocol, iMonitor can communicate
with this server.
13. List and describe the 2 health check categories and their
function:
❑ Agent. To view information about time synchronization and
the state of eDirectory on the server.
❑ Partition. To view health information, such as replica
synchronization and replica ring detail, for each partition
on the server.
14. From the health check you just performed, are you satisfied that
eDirectory on your server functioning properly?
Yes
4. When would you use the Agent Health Report check feature
instead of performing your own health check with the options in
Parts I - III?
Answers will vary
Possible Cause
Solutions
4. How did the problem occur? (You might not know how the
problem was introduced to your system, but if you do, record it
so it can be prevented.)
❑ This error occurred when an attribute was changed on the
users record illegally.
5. What steps did you follow to resolve the problem?
The following troubleshooting steps were used:
❑ Step 1: Identify the Scope of the Problem
❑ Step 2: Determine the Cause of the Problem
❑ Step 3: List Possible Solutions to the Problem
❑ Step 4: Assess Possible Solutions
❑ Step 5: Implement a Solution
❑ Step 6: Verify That the Problem Is Resolved
❑ Step 7: Document the Resolution to the Problem
❑ Step 8: Avoid Repeating the Problem
Duration: 1 hour 30 minutes In this section you learn about advanced Novell Storage Services
(NSS) management tasks.
Objectives
1. Expand an NSS Storage Space
2. Configure NSS Volume Attributes
3. Mount a DOS Partition as an NSS Volume
4. Use VCU to Create an NSS Volume from a Traditional Netware
Volume
5. Resolve Common NSS Errors
6. Restore a Deleted Logical Volume
7. Describe Storage Area Networks and Network Attached Storage
Introduction
As data storage needs on the network grow, satisfying those needs
becomes more complex. Novell Storage Services (NSS) is a very
scalable and flexible file system that can meet these needs, but it can
require advanced NSS management skills.
In this section you learn about these skills and 2 other storage
architectures and technologies: Storage Area Networks (SANs) and
Network Attached Storage (NAS).
The Scenario
At Digital Airlines, you are the network administrator for one of the
branch offices and have installed NetWare 6 on your servers.
You might want to demonstrate To increase the storage space on a server, you can increase the
how to create a volume as a number of logical volumes in an NSS storage pool.
review of Course 3004.
In Course 3004, you learned that a storage pool is a specified
amount of space from various storage devices used to contain
multiple NSS logical volumes.
Figure 5-1
POOL
Dynamic limit allows these to expand as necessary.
You might want to demonstrate Another method for increasing the amount of storage space on a
how to overbook a pool as a server is to overbook the storage pool.
review of Course 3004.
In Course 3004 you learned that although the size of an individual
logical volume cannot exceed the size of a storage pool, the sum of
multiple logical volumes in the pool can exceed the pool size.
You can limit some volumes to a certain size and allow others to
grow as necessary in the pool.
Using NSS, the sum of Volume A and Volume B can actually exceed
the size of the storage pool.
Figure 5-2
x While you can increase the size of a pool, you cannot reduce the size.
ConsoleOne
Demonstrate how to use ConsoleOne lets you complete NSS management tasks from the
ConsoleOne to increase the size Media tab in a Server Object Properties dialog:
of a storage pool.
Figure 5-3
Remote Manager
Demonstrate how to use Remote Remote Manager lets you complete NSS tasks from anywhere you
Manager to increase the size of a have a web browser and Internet access.
storage pool.
Figure 5-4
NSSMU
Demonstrate how to use NSSMU NSS Management Utility (NSSMU) lets you complete the same
to increase the size of a storage tasks at the server console that you can complete using ConsoleOne
pool. or Remote Manager at a workstation.
Figure 5-5
6. Press Enter.
7. Note the amount of space in the pool.
Figure 5-6
x Do not enable features you do not need. Many of these features will
affect the performance of NSS.
After loading the appropriate hardware drivers, you can mount the
removable disk as an NSS volume and create a core dump on the
disk. Because it is a DOS-formatted disk, the file can then be read
from the disk on any Windows computer.
Demonstrate how to mount a DOS To mount existing DOS partitions on your NetWare 6 server as NSS
partition. volumes, do the following:
1. At the server console prompt, enter SET AUTO RESTART
AFTER ABEND = 0.
x If you do not turn off automatic abend recovery, you risk corrupting the
DOS drive's FAT tables.
Figure 5-7
To verify that the DOS partition was mounted properly, look for the
following:
■ After loading DOSFAT.NSS, you should see a message stating
** DOSFAT_C mounted successfully on the server console.
■ At the console prompt, enter VOLUMES.
If the module loaded correctly, the volume appears as
DOSFAT_x, where x is the drive letter, such as DOSFAT_C.
If NSS cannot determine the drive letter, the volume appears as
DOSFAT_0 or DOSFAT_1.
The VCU utility creates a new NSS volume and then copies data
(keeping the same file structure) from the source traditional volume
to the new NSS logical volume. VCU can copy volumes with long
name space applied.
Because VCU creates a new volume, you must have adequate space
on your server.
x You must have enough space for both the traditional volume and the new
logical volume.
Use VCU only when server demands are low (such as after working
hours), after you disconnect all other users and disable login, and
after you create a backup of the volume.
x After you copy traditional volume data to a logical volume in NetWare 6, you
cannot access the new logical volume using prior versions of NetWare.
Create a small traditional volume; To copy a volume, load VCU.NLM, specify the volume to copy, and
then demonstrate how to copy an then specify the NSS pool where you want to create the new NSS
NSS volume from the traditional volume. Use the following syntax:
volume.
VCU volume pool
Do the following:
1. Create a new pool called NEWPOOL:
a. From ConsoleOne, browse to your server.
b. Right-click your server; then select Properties.
c. Select Media > NSS Pools.
d. Select New.
e. In the Name field, enter NEWPOOL; then select Next.
f. Mark the unpartitioned space.
g. In the Used field, enter a size (in MB) equal to or greater than
the size of volume DATA; press Enter or Tab; then select
Next.
h. Select Finish.
i. If prompted with a warning concerning the Hot Fix size,
select Yes.
2. Verify that NEWPOOL was created:
a. At the server console, enter NSS /SPACE.
b. Verify that NEWPOOL appears in the list.
Do the following:
1. Access the Media Properties page for your server object:
a. From ConsoleOne, browse to your server.
b. Right-click your server, then select Properties.
c. Select Media > NSS Pools.
2. Note the size of the NEWPOOL pool:
a. Select the NEWPOOL pool.
b. Record the total amount of space in the pool:
Do the following:
1. Mount your DOS partition as an NSS volume:
a. At the server console enter SET AUTO RESTART AFTER
ABEND = 0.
b. Enter LOAD DOSFAT.NSS.
2. At the server console, enter VOLUMES.
3. Verify that DOSFAT_C mounted successfully by mapping a
drive to the volume and viewing the files.
(End of Exercise)
For example, NSS cache parameters are best left at their default
value. If you change them without consulting with Novell, you
could lose large amounts of data.
Also, SYS should always be the only volume in its pool. Create
other pools for other volumes.
The following are errors you might experience while working with
NSS:
NSS does not NSS can only use At the server console, enter
recognize a what the Media SCAN FOR NEW DEVICES;
device. Manager then enter LIST DEVICES.
recognizes. If the device is not listed or
appears as an unbound
object, most likely the device
is malfunctioning or the
appropriate driver is not
loaded.
Make sure the correct board
driver (*.HAM) and device
driver (*.CDM) are loaded.
NSS does not let ■ Your storage ■ Make sure you have enough
you create a devices might free space to create another
storage pool or a not have storage pool or logical
enough free volume.
logical volume.
space to ■ Before you create a logical
create more volume, create a storage
storage pools pool.
or logical
volumes. ■ Create an NSS partition for
your storage pools and
■ All logical logical volumes.
volumes might
not be part of
the same
storage pool.
■ NSS might not
own the free
space you
want to use for
a storage pool.
b You can access a list of all NSS error codes and possible solutions at
http://www.novell.com/documentation/lg/nwec/index.html?treetitl.html
Demonstrate how to use VERIFY. Whenever you need to check the integrity of an NSS file system of a
pool, you can use VERIFY.
Ask coworkers and others what actions they would take. If possible,
test the solutions in a lab environment before implementing them.
x Only use REBUILD as a last resort to recover the file system because it could
cause loss of data.
Before using REBUILD, try restoring the pool from a tape backup
first. If this does not work, contact Novell for help in using
REBUILD.
The REBUILD and VERIFY utilities both generate a log file at the
root of the DOS drive.
These solutions can be expensive, but they are usually less costly
than losing the data. The best way to avoid this situation is to
consistently back up the data.
After you implement a solution, test the system to make sure the
issue is resolved. Continue to monitor the situation to ensure the
problem does not recur.
a 10 minutes In this exercise, you encounter an NSS error, research the solution,
and fix the problem.
You thought you had enough storage space on the system, but one
day you received the following error: 20103 zERR OUT OF
SPACE.
Do the following:
1. From a web browser, access http://www.novell.com
/documentation/lg/nwec/index.html?treetitl.html.
2. Under Novell Storage Services (NSS) Error Codes, select List
of Codes.
3. Select 20103 zERR OUT OF SPACE.
4. Record the action recommended in the list:
Do the following:
1. Access the Media Properties page for your NSS logical volumes:
a. From ConsoleOne, browse to your server.
b. Right-click your server, then select Disk Management >
NSS Logical Volumes.
2. Access the NSS Attributes page for volume DATA and
determine whether the Salvage Files property is enabled:
a. Select DATA.
b. Select Properties.
c. Verify that Salvage Files is selected.
3. Delete unnecessary salvaged files:
a. From your workstation, map a drive to volume DATA.
b. In Windows Explorer, right-click DATA; then select Purge
Files.
c. Select Purge Subdirectories; then select Yes.
d. Check volume DATA again.
Notice that your data is still there.
(End of Exercise)
You must retrieve the volume before the delay time elapses;
otherwise, the volume is purged from the system, and you can no
longer restore it.
x If you delete a storage pool, you delete all volumes in that pool. Volumes
deleted in this manner cannot be restored.
The default setting for the purge delay time is 2 days. After this time
expires, NSS purges the volume. You can change the purge delay
time to extend or reduce the time for the automatic purging cycle.
x The purge delay change command is not persistent. The parameter is lost if
the server is restarted. To make the change permanent, add the command to
the server’s AUTOEXEC.NCF file.
The acronyms SAN and NAS are sometimes confused, but you must
know the capabilities of each and the differences between the 2 for
later sections in this course where you will implement a SAN for
clustering.
Figure 5-8
TCP/IP
LAN
Because the SAN is not part of the LAN, it is not slowed by the
normal network traffic on the LAN.
Backup and transfer of data within the SAN is very fast and efficient
because of the high-speed, large-block transfers that Fibre Channel
allows.
Figure 5-9
Application Servers
Workstations
File Server File Server
RAID Tape
Files are saved or retrieved directly from the NAS appliance. The
complexities of the file system are hidden from the user. Because of
this, a NAS appliance is often a great place for hosting home
directories for users.
Keep in mind that while a NAS appliance stores and retrieves files
more quickly than a general-purpose server, network traffic does
impact the performance.
Summary
The following is a summary of the objectives in this section:
Duration: 1 hour In this section you learn to configure and troubleshoot a RAID
solution with NSS.
Objectives
1. Implement RAID 0 with NSS
2. Configure Partition Mirroring and Duplexing in NSS
3. Troubleshoot Software RAID and Mirroring in NSS
Introduction
You are the network administrator at a Digital Airlines branch
office. With your servers upgraded to Netware 6, you want to
quickly and effectively resolve some server performance and file
storage issues.
What RAID Is
Hard disks are mechanical devices and slowly wear out with use.
Every hard disk has an associated Mean Time Before Failure
(MTBF).
To guard against data loss from a disk failure in your server, you can
implement RAID.
Hard drive
Hard drive
NSS also fully supports RAID arrays created using hardware RAID
adapters.
The RAID stripe size is the amount of data the file system places on
a disk before moving to the next disk. The stripe size ranges
between 4 KB and 256 KB in increments of 2 KB.
The size of the stripe units depends on the application for which the
array is used. For example, if the system will store large files, such
as graphics or digital video, the stripes are generally small, around
512 bytes.
The small size of the stripes ensures that a single file spans as many
disks as possible. This ensures that the files can be manipulated
quickly because modifying the file will require reading and writing
to all disks in the array at the same time.
x Remember that RAID 0 improves and enhances performance but does not
provide fault tolerance.
Emphasize that each segment in You can use RAID 0 for both logical and traditional volumes.
the RAID 0 configuration should
come from a different device or
performance suffers. Create a Software RAID 0 Array
Figure 6-2
Figure 6-3
Figure 6-4
12. Select Increase Size; then select the device you want to add to
the RAID configuration.
13. Select Finish.
x The restriping process takes time to complete, depending on the size of the
disk devices involved. As a result, file system performance is impacted
during the restriping process.
You can also use NSS Management Utility (NSSMU) to create NSS
RAID arrays. This is done from the server console, as shown in the
following:
Figure 6-5
Before starting the exercise, make During this exercise, you use 2 hard drives installed on your
sure both IDE controllers are NetWare 6 server. Before you start, do the following:
recognized by the computer BIOS
(if you are using 2 controllers) and ■ If there are 2 IDE controllers for the hard drives, check the
that a second driver is loaded for computer BIOS settings to make sure the computer recognizes
the second hard drive. both hard drive IDE controllers.
■ Using NWCONFIG, make sure there is a driver for each hard
drive. You might need to load a driver for the second hard
drive.
Figure 6-6
Figure 6-7
7. Make sure the stripe size is 64 (KB) and the RAID type is RAID
0; then select Finish.
The new RAID device is listed with settings shown to the right.
8. Select Increase Size.
A screen similar to the following appears:
Figure 6-8
11. To verify that you have 2 raid device segments, select Show
Segments.
12. On the right under RAID information, find the Device ID and
record it to use later in the exercise:
x The restriping process takes some time to complete, depending on the size of
the disk devices involved. As a result, file system performance is impacted
during the restriping process.
Do the following:
1. From the Server Properties window, select Media > NSS
Logical Volumes.
2. Select New.
3. For the volume name enter VOLUME1 for the volume; then
select Next.
4. Select POOL1.
You can also select unpartitioned space at this point. If you do,
NSS creates an NSS partition and a storage pool for your
volume.
5. Select Allow volume quota to grow to the pool size; then select
Next.
6. Select Finish.
7. From the DAx server console enter VOLUMES and verify that
VOLUME1 was mounted.
(End of Exercise)
Mirroring your partitions lets you protect critical data by storing the
same data on 2 separate disks using the same disk controller. If one
disk goes down, the system uses the other disk.
■ The file system adjusts the hot fix size to the allowable ranges
to make the data area identical to the other partitions in the
mirror group.
■ Mirrored partitions must both be marked for sharable for
clustering.
The partitions you add to a group cannot be part of an existing
group—they must be individual mirrored objects.
Figure 6-9
4. Select New.
Figure 6-10
The physical size (combined data and hot fix size) of the
partition must be at least 100 KB, but no more than 120 MB
larger than the data size of the existing partitions in the mirror
group.
9. To mirror the partition, select Mirror; then select one of the
following options:
❑ Create New Mirror. This option allows the partition to be
part of a mirror group. You do not actually create the group
until you add another mirrored partition to the partition you
are creating.
❑ Existing Mirror Group. This option shows a list of
existing mirrored groups that are compatible in data area
size. This option lets you add this new partition to one of
the mirrored groups in the list. If you select this option, also
select the ID of the mirrored partition.
10. (Optional) Enter a label for the partition.
a 10 minutes In this exercise, you learn to provide fault tolerance for sensitive
data by mirroring an NSS partition. Do the following:
You should have 2 devices per 1. From your workstation start ConsoleOne.
machine. If that is not possible,
demonstrate the procedure for the 2. Right-click DAx; then select Properties.
students. 3. Select Media > Partitions > New.
4. Select a device to create a partition on (either the hard drive or
RAID).
5. Select the type of partition as NSS.
6. For the volume size enter 300 MB.
7. Make sure Hot Fix is selected with a setting of 100 KB.
Remember that this is required for mirroring to work. If the
existing partitions don't have a hot fix area, they can't be
mirrored.
8. Make sure Create New Mirror Group is selected.
9. In the Label field enter MIRROR1.
This is important because it will help you identify the mirrored
or duplexed partition later.
10. Select OK.
11. On the left, locate and select the partition you just created (look
for the MIRROR1 label on the right).
This partition is probably located near the bottom of the list.
12. Record the Mirror ID number:
14. For the size, select MB from the drop-down list; then enter 300.
18. From server DAx, switch to the console prompt and notice that
the partitions were not synchronized; when remirroring occurred,
they were synchronized.
19. Return to the Properties Window and select Media > NSS
Logical Volumes.
20. Select New.
22. Select the mirrored space you just created using the mirror ID
number as a guide.
23. Select Allow volume quota to grow to the pool size; then select
Next.
24. In the Name field enter POOL2; then select OK.
26. From your server console prompt list the MIRROR volume by
entering VOLUMES.
(End of Exercise)
Summary
The following is a summary of the objectives in this section:
Objective Summary
Duration: 4 hours In this section you learn advanced iFolder administration tasks and
troubleshooting steps.
Objectives
1. Describe iFolder Configuration Files
2. Perform iFolder Management Tasks
3. Maintain and Troubleshoot the iFolder Client
4. Maintain and Troubleshoot the iFolder Server
SYS:Apache\iFolder\Server\HTTPD.CONF
SYS:Apache\iFolder\Server\HTTPD_ADDITIONS_NW.CONF
iFolder configuration files are text files. To edit them you use a text
editor, make changes, and save the file in the correct location.
Before you edit the contents of these files it is useful to know how
they are structured.
HTTPD.CONF
This is the default Apache configuration file. This is a long text file
that does not need to be discussed in its entirety to understand how
to edit iFolder configuration settings.
If you look at this file on your DAx server, it looks like this:
Listen 192.168.1.x:80
■ In the main server section, the entry is
ServerName IP address
If you look at this file on your DAx server, it looks like this:
ServerName 192.168.1.x
■ The virtual hosts section contains the following:
<IfModule mod_tls.c>
SecureListen IP address:443 “SSL
CertificateIP”
</IfModule>
If you look at this file on your DAx server, this entry looks like
this:
<IfModule mod_tls.c>
SecureListen 192.168.1.x:443 “SSL
CertificateIP”
</IfModule>
x The term virtual host refers to the practice of maintaining more than one
server on one machine.
HTTPD_ADDITIONS_NW.CONF
#
# Edit the xxx.xxx.xxx.xxx to your IP address
# =================================
<VirtualHost 192.168.1.1:443>
DocumentRoot "SYS:\apache\iFolder\DocumentRoot"
<Directory "SYS:\apache\iFolder\DocumentRoot">
Options Indexes FollowSymLinks MultiViews
AllowOverride None
Order allow,deny
Allow from all
</Directory>
<location /iFolderServer>
SetHandler ifolderserver-form-handler
</location>
#
# iFolder Server LDAP Settings
#
# Edit the LdapHost and LdapSecondaryHost
# if the SSL LDAP port of 636 is used, you must
# enter the path to the LdapRootCert.
# ==============================================
LdapHost da1.digitalair.com
LdapPort 636
LdapLoginDnContext "O=DigitalAir"
# -or -
# LdapLoginDnContext ",ou=xxxx,o=xxxx"
LdapRootCert "SYS:\apache\iFolder\server\
RootCert.der"
# Suggestion: Enter IP address of the iFolder
# server so that you can use NWAdmin to add
# "shared" accounts
# LdapSecondaryHost %LdapSecondaryHost%
# LdapSecondaryPort %LdapSecondaryPort%
# LdapSecondaryLoginDnContext
"%LdapSecondaryLoginDnContext%"
# -or -
# LdapSecondaryLoginDnContext ",ou=xxxx,o=xxxx"
# LdapSecondaryRootCert "%LdapSecondaryRootCert%"
#
# iFolder Volume \ directory for user files
#
# Edit the iFolderServerRoot
# =================================
iFolderServerRoot SYS:\iFolder
#
# iFolder Admin Settings for Server Management
# Console
#
# Edit the iFolderAdmin lines
# ==================================
iFolderAdminName admin
#
# iFolder Server Secure Port
#
# Edit the ServerSecurePort
# ==================================
ServerSecurePort 443
</VirtualHost>
The file is divided into 2 sections: the first section gives the
configuration of the nonsecure virtual host; the second section gives
the configuration of the secure virtual host.
These sections create 2 virtual hosts that are run by the same
Apache server. The sample code shown above configures a
nonsecure virtual host to use port 80 and a secure virtual host to use
port 443, both on the same server.
Aside from the different port numbers, the configurations for each
virtual host are identical. For iFolder to function, each virtual host
must have the same login contexts, server name, admin name, and
server root location.
In most cases you edit this file by using the iFolder server
management console, which writes changes to this file. In some
cases you must make the changes manually to the file.
There are 2 ways to add more user contexts in the iFolder server
management console. You can individually specify each
container or you can use the Search Subcontainer option.
You add login contexts from the LDAP configuration page in
the iFolder server management console.
When you individually specify each container, you enter the
contexts in the DN field, as shown in the following:
Figure 7-1
The syntax you use for entering contexts must follow LDAP
conventions, which are different from eDirectory conventions.
eDirectory requires that contexts be written with period (.)
separators, like this:
OU=IS.OU=SLC.O=DIGITALAIR
LDAP requires that contexts be written with comma (,)
separators, like this:
OU=IS,OU=SLC,O=DIGITALAIR
If you enter more than one context in the DN field, separate
each context with a semicolon (;) and no spaces, as shown in
the following example:
O=DIGITALAIR;OU=SLC,O=DIGITALAIR;OU=IS,
OU=SLC,O=DIGITALAIR
If you only have 1 or 2 contexts to configure for iFolder users,
manually specifying them is the easiest way to accomplish this.
Figure 7-2
■ Admin names indicate the names of users that can log in to the
iFolder server management console.
You add authorized administrators from the Admin Sessions
page in the iFolder server management console, as shown in the
following:
Figure 7-3
This entry exists in both the nonsecure and secure virtual host
sections of HTTPD_ADDITIONS_NW.CONF. You must make
the change to both entries.
You must also manually move the SYS:\IFOLDER directory to
DATA:\IFOLDER.
Before you stop iFolder you should stop the iFolder server from
synchronizing. This task is performed from the iFolder server
management console.
Figure 7-4
After you stop synchronization, you stop the iFolder server at the
server console by entering the following command, which
immediately terminates the iFolder service:
STOPIFOLDER
Doing the stop synchronization first insures that all iFolder data files
are closed when the server goes down. It also makes bringing the
server back up much faster.
The server must then perform a data integrity check of all iFolder
data on the server. On an iFolder server that hosts a lot of data, this
check can take 45 minutes or longer to perform.
STARTIFOLDER
You can also view this screen to view the log of iFolder events that
have taken place during the current iFolder session.
The iFolder server management console lets you set several client
and server policies. These policies affect the way users interact with
iFolder and the way servers perform.
Figure 7-5
Figure 7-6
The default location for iFolder user data is volume SYS, but it is
rarely advisable to use volume SYS to store user data. iFolder users
can potentially fill up the volume and bring down the server.
Add Contexts
You add user login contexts through the iFolder server management
console by doing the following:
1. Access the iFolder server management console through
https://iFolder server IP address or https://DNS
name/iFolderServer/Admin.
This URL is case sensitive.
2. Log in as an authorized admin; then select LDAP.
3. Make changes to the DN field; then select Update.
#LdapLoginDnContext “O=DigitalAir”
LdapLoginDnContext “O=DigitalAir;OU=SLC,O=DigitalAir”
You can avoid this by making a backup copy of the original that you
can use to replace the overgrown file. You can also open the file and
delete the extra remarked out lines, but you should have a backup
copy in case you make a mistake.
x Admin user names and passwords cannot contain special characters. All
characters must belong to the UTF 8 character set.
This might happen if port conflicts occur between iFolder and other
services that must use the primary IP address. iFolder can use a
secondary IP address so it is logical to change the IP address that
iFolder is using.
SYS:\APACHE\IFOLDER\DOCUMENTROOT\
IFOLDERCLIENT.EXE must be modified to deliver the new IP
address when new users download it.
Do the following:
1. Change the necessary entries in HTTPD.CONF.
2. Change the necessary entries in
HTTPD_ADDITIONS_NW.CONF.
Do the following:
1. Mount the NetWare 6 CD on the server.
2. From the graphical console, select Novell > Install.
3. Select Add.
4. Browse to the NetWare 6 volume and highlight PRODUCT.NI;
then select OK.
5. Select OK again to run the product installation program.
6. From the Components screen, select Clear All.
12. From the Summary window, make sure iFolder is in the list of
products to be installed; then select Finish.
13. Complete the installation by selecting Close.
Do the following:
1. Mount the Support Pack CD as a NetWare volume on DAx.
2. At the server console, enter NWCONFIG.
3. In Configuration Options, select Product Options.
4. In Other Installation Actions, select Install a Product Not
Listed.
5. Continue by pressing Esc.
6. To specify the directory path, press F3.
7. In Specify a Directory Path, change the default setting to
NW6SP2: (include the colon).
8. Press Enter.
9. In the Novell Terms and Conditions screen, press Esc to
continue.
10. Accept the license agreement by selecting Yes.
Do the following:
1. At the workstation, use a supported browser to access
https://DAx.DIGITALAIRLINES.COM:52443/
iFolderServer/Admin.
The iFolderServer/Admin part of this URL is case sensitive.
2. From the Security Alert window, select Yes.
3. Authenticate as admin.
Do the following:
1. From the iFolder server management console, at the bottom of
the General Info page, select Stop Synchronization.
2. Close the browser.
3. To stop iFolder, at the DAx server console prompt enter
STOPIFOLDER.
4. Verify that the iFolder server is no longer running:
a. Press Ctrl + Esc.
b. Note that the Apache for NetWare screen is not listed.
c. To return to the server console, enter 1.
5. Verify that the iFolder web site is not functioning:
a. At the workstation, launch Internet Explorer.
b. For the URL enter
http://DAx.DIGITALAIRLINES.COM:52080.
You might receive a message that the page cannot be
displayed or nothing at all appears.
6. To start iFolder again, at the DA1 server console enter
STARTIFOLDER.
Note that Apache for NetWare screen states that iFolder
initialized successfully.
Do the following:
1. Download the iFolder client:
a. From the iFolder web page, select Download.
b. Select Open.
c. Select Run This Program from its Current Location; then
select OK.
d. From the Security Warning window, select Yes.
e. From the Welcome screen, select Next.
f. Choose your language, then select Next.
g. Accept the license agreement by closing the browser
window and selecting Yes.
h. From the Choose Destination Location window, select Next.
i. When prompted, select Finish.
j. Close the ReadMe file.
k. Restart the workstation by selecting Finish.
2. Create an account for Admin:
a. When the workstation restarts, authenticate to eDirectory.
b. When the iFolder window appears, select Continue.
c. In the Login window, enter admin for the username and
novell as the password; then select Login.
Do the following:
1. From the workstation, map a drive to the root of your server’s
volume SYS.
2. Use Notepad to browse to and open SYS:\APACHE\
IFOLDER\SERVER\HTTPD_ADDITIONS_NW.CONF.
3. Change the data location from SYS to DATA:
a. In the nonsecure virtual host configuration portion of this
file, find iFolderServerRoot SYS:\iFolder.
b. Change this entry to iFolderServerRoot DATA:\iFolder.
c. Repeat this change for the same entry in the secure virtual
host configuration portion of this file.
d. Save the file and exit Notepad.
4. Create data files in Admin’s iFolder home directory:
a. Using the shortcut on the desktop, open Admin’s iFolder
home directory.
b. Inside the folder, right-click; then select New > Bitmap
Image.
c. Repeat Steps a and b to create any other files.
Do the following:
1. View the current iFolder user account contexts:
a. At the workstation, use Internet Explorer to access the
iFolder server management console at
https://DAx.DIGITALAIRLINES.COM:52443
/iFolderServer/Admin.
This URL is case sensitive.
b. Log in with the username admin and the password novell.
The username is case sensitive.
c. From the left column, select LDAP.
2. Attempt to log in as a user from xxx.Digitalair:
a. Right-click the iFolder trayapp icon; then select Logout.
Do the following:
1. Prove that you cannot create an account for a user in one of your
subcontainers by attempting to log in as a user from the
CUSTRSVC container (see Table 7-2 for your username).
DA4 JKURMI
DA5 SDEES
DA6 KCHILDS
DA7 BFULLER
DA8 HWAARLE
DA9 KCHUNG
Do the following:
1. Add the user from Table 7-1 as an authorized iFolder
administrator:
a. From the left column in the iFolder server management
console, select Admin Sessions.
b. In the Authorized Admins field, place your cursor after
admin.
c. Without entering any spaces, enter ;username.
For example, for username KSINGH you would enter
;KSINGH. The semicolon is required.
x Remember that the case you use here is the case you must use when
logging in to the iFolder server management console.
d. Select Update.
2. Verify that the user you added can access the iFolder server
management console:
a. From the left column, select Login/Logout; then from the
right select Logout.
b. From the left column, select Login/Logout.
c. Enter username and use novell as the password.
d. Select Admin Sessions.
e. Note that the Current Admin is username.
f. Log out of the iFolder server management console.
Do the following:
1. At the workstation, log out of iFolder.
2. Add a secondary IP address that the iFolder server can use:
a. At the DAx server console, enter ADD SECONDARY
IPADDRESS 192.168.1.x (see Table 7-3 for the IP address).
DA4 192.168.1.34
DA5 192.168.1.35
DA6 192.168.1.36
DA7 192.168.1.37
DA8 192.168.1.38
DA9 192.168.1.39
a. Open SYS:\APACHE\IFOLDER\SERVER\
HTTPD_ADDITIONS_NW.CONF.
b. Find the following lines
<VirtualHost DAx.DIGITALAIRLINES.COM:52080>
<VirtualHost DAx.DIGITALAIRLINES.COM:52443>
and change them to
<VirtualHost iFolderx.DIGITALAIRLINES.COM:80>
<VirtualHost iFolderx.DIGITALAIRLINES.COM
:443>
c. Find both instances of
ServerName DAx.DIGITALAIRLINES.COM
and change them to
ServerName iFolderx.DIGITALAIRLINES.COM
d. Find both instances of
ServerSecurePort 52443
and change them to
ServerSecurePort 443
e. Save the file.
5. Change the DNS name that is configured in
IFOLDER_NAV.HTML:
a. Use Notepad to open SYS:\APACHE\IFOLDER\
DOCUMENTROOT\HTML\IFOLDER_NAV.HTML.
b. Find the line that contains
“https://DAx.digitalairlines.com:52443/applet/
java.htm”
and change it to
“https://iFolderx.digitalairlines.com:443/applet
/java.htm”
c. Save the file.
If the iFolder service starts with the 6. Stop and start the iFolder service.
error that it cannot resolve the
iFolder server’s hostname, make 7. Log in to iFolder as any of the users you have already logged in
sure RESOLVE.CFG is not empty. as:
a. At the workstation, right-click the iFolder Trayapp icon;
If RESOLVE.CFG is empty have then select Login.
students copy the contents from
another server where the file is not b. In the Novell iFolder Login window change the server name
empty. to iFolderx.DIGITALAIRLINES.COM (remove the port
number); then select Login.
iFolder is now running on the new IP address and is using
the standard ports: 80 and 443.
Do the following:
1. Use FIXUP.NLM to assign a new DNS name to
iFolderClient.exe:
a. At the DAx server console enter FIXUP
iFolderx.DIGITALAIRLINES.COM SYS:\APACHE
\IFOLDER\DOCUMENTROOT\IFOLDERCLIENT.
EXE.
b. View the Logger screen to verify that FIXUP loaded and
reported no errors.
2. Remove iFolder from the workstation so you can verify that
iFolderClient is using the new IP address:
a. Log out of iFolder.
b. Delete all iFolder home directories and their shortcuts.
c. Select Start > Settings > Control Panel > Add/Remove
Programs.
d. Select Novell iFolder 1.03.
e. Select Change/Remove; then select Yes.
f. After uninstall completes, restart the workstation.
Do the following:
1. Access the iFolder server management console at
http://iFolderx.DIGITALAIRLINES.COM/iFolderServer
/Admin.
2. Log in as admin.
3. Select Client Policies.
4. For Encryption, select the following:
❑ On
❑ Enforced
❑ Hidden
5. For Save Password, select the following:
❑ On
❑ Enforced
6. For Save Pass Phrase, select the following:
❑ On
❑ Enforced
7. Select Update Policy; then select Refresh.
DA4 RBHAT
DA5 KFULLMER
DA6 ASANDERS
DA7 DCROCKETT
DA8 IBLUNCK
DA9 HNAGAI
x If you experience difficulty logging in, make sure you have configured
subcontainer search properly. See the steps to Part VIII: Enable
Subcontainer Search.
11. While still logged in as your user from FLIGHTOPS, check the
current client quota:
a. Right-click the iFolder icon in the System tray.
b. Select Account Information.
c. Select the Account Information tab.
d. In the Server Information box, note that this user has 200 MB
total space on the server.
e. Close the Account Information window and log out of
iFolder.
12. Return to the iFolder server management console; then select
Server Policies.
13. Change the Initial Client Quota to 300 MB.
DA4 MJAIN
DA5 DDECKER
DA6 RHAYMOND
DA7 JASTIN
DA8 KBAHR
DA9 MYAMADA
d. In the Server Information box, note that this user has 300 MB
total space on the server.
e. Close the Account Information window; then log out of
iFolder.
17. Check the client quota for a user that already has an account:
a. Log in again as the user from FLIGHTOPS and check that
user’s client quota.
Note that it still says 200 MB. This has not changed
because the setting you changed only affects the initial
client quota.
b. Close the Account Information window and log out of
iFolder.
18. Change an individual user’s client quota:
a. Return to the iFolder server management console.
b. Select User Accounts.
c. Select the user from FLIGHTOPS.
d. Change the Disk Quota to 300 MB.
e. Select Change; then select Refresh.
f. Log in again as the user from FLIGHTOPS and check the
user’s client quota.
Note that it now says 300 MB.
g. Close the Account Information window and log out of
iFolder.
(End of Exercise)
When the user is logged in, the iFolder client watches for new files
and for changes made to existing files in the iFolder home directory
and transmits those changes to the server.
If the original file is lost from the user’s computer, there is always
an up-to-date backup of the file on the iFolder server.
The iFolder client goes through the following steps as it keeps data
synchronized with the iFolder server:
1. When a user enters a username and password on a workstation,
the iFolder client sends them to the iFolder server encrypted with
RSA Encryption.
2. The iFolder server takes the user name and password and
performs an LDAP bind to the LDAP servers.
Figure 7-7
4. If the client discovers through the comparison that there are new
files or changes to files on the server or client, the new data on
the server is downloaded, and any new data on the client is
uploaded.
5. When the iFolder server receives the new data it increments the
sync index.
The sync index indicates the current state of file system. As
long as the sync index on the client matches the sync index on
the server, there has been no change. The sync index is only 4
bytes and is a very small operation on the server.
6. When the same user connects to the iFolder server using another
computer (such as at home or on a laptop), iFolder compares
indexes. If it sees that the sync index on the server is different
from the sync index on the client, it downloads the changes.
Now the files in both computer’s iFolder home directories are
the same.
7. While logged in to iFolder, the user creates a document or
modifies an existing file. The iFolder client watches for file
system changes and is notified of the new or modified file.
The new file or changes to the existing file are then uploaded to
the iFolder server.
8. When the iFolder server receives the data it increments the sync
index.
The server is ready for the next client login.
New Files Are Not Synchronizing from the Client to the Server
Users might discover that new files they thought were being
synchronized from the client to the server are actually not on the
server, while at the same time changes to existing files are being
synchronized.
Users discover this condition when they try to access a file while
using a different computer than the one they created the file on. The
cause of this problem is corrupted file maps and dirmaps.
Users might discover that files that have been synchronized to the
server from one computer are not being synchronized when they log
in from another computer.
The most likely cause is that the second computer does not have
sufficient disk space for the synchronization process to complete.
As files are downloaded from the server they are placed in the user’s
working home directory before being copied to their iFolder home
directory.
The purpose of the conflict bin is to save files that have been deleted
or overwritten by the synchronization process. If a user expects to
find a file in the conflict bin but it is not there, you should look for
the following possible causes:
■ The conflict bin is too small. The default setting for the
conflict bin is 25 MB. If the conflict bin is full or the deleted
files are more than 25 MB, they will not be placed in the
conflict bin.
There is no way to solve this problem after it happens, but it can
be prevented by allocating more space to the conflict bin. This
is a client policy setting that you can change at any time.
Users can also increase the size of their conflict bin by doing
the following:
1. Right-click the iFolder icon.
2. Select View Conflict Bin.
3. Select File > Properties.
Figure 7-8
When many users are accessing iFolder through the browser client,
you might need to increase the number of threads per child.
The default setting for threads per child is 150. The maximum
setting is 2048. Threads are used to maintain user connections to the
iFolder server. The iFolder client does not use persistent
connections.
If most of your users are using the client, you do not need to have 1
thread per user. 50 threads can service 1,200 concurrent iFolder
client connections.
If the server certificates that provide security for the admin to log in
become corrupted, the server management console is not available.
Users can still log in using the client because the client uses port 80.
But browser-based access is not available because the secure port is
used.
A port conflict can arise because iFolder and iPrint both use port
443 by default as their secure port. The port resolver should prevent
this problem.
When users forget their pass phrase they can no longer get into their
iFolder account. Access to their accounts can be restored by deleting
all user data from the server and then allowing the user to log in
again and resynchronize with the server.
Do the following:
1. Make sure the user has a complete local copy of their data.
2. Make sure the user also has a connection to the server with
adequate bandwidth for completing the resynchronization.
3. Find out which folder the user’s data is saved in on the iFolder
server:
a. Find the user’s account in iFolder server management
console on the User Account page.
b. Place your mouse on the user’s name.
c. Note that in the status line at the bottom of the browser
window there is a long hexidecimal number.
Example:
User=9E226380764BAA07696D656DF6F7B191.htm
Figure 7-9
Figure 7-10
Figure 7-11
d. Open the user’s directory and verify that this is the correct
directory.
Within each user’s directory is a CONTROL.DAT file.
Open the CONTROL.DAT file with Notepad and you will
see the user’s name in the last characters of the first line.
This is the name of the user that this directory belongs to.
5. Delete the user’s directory.
Delete the entire directory that belongs to the user that forgot
the pass phrase. For example, for the user associated with
9E226380764BAA07696D656DF6F7B191, you would delete
the 9E226380764BAA07696D656DF6F7B191 folder.
6. Have the user login to iFolder.
iFolder will recognize that this user doesn’t have a directory on
the server and it will create one for the user.
The user enters a new pass phrase and the iFolder client
initiates the synchronization process that places all the user’s
data back on the server. This time the data is encrypted with the
new pass phrase as the key.
(iFolder 2.0 has an added feature that allows the administrator
to restore a user’s forgotten pass phrase.)
Using port 389 is a legitimate option when the LDAP server and the
iFolder server are running on the same physical computer. In that
scenario no communications take place over the wire so no
encryption is required.
Figure 7-12
4. Select OK.
Summary
The following is a summary of the objectives in this section:
Duration: 2 hours In this section, you learn the purpose and advantages of clustering
data and applications and how to design and set up a 2-node NCS
cluster.
Objectives
1. Identify the Purpose and Advantages of Implementing an NCS
Solution
2. Design and Set Up an NCS Cluster Configuration
Introduction
Consider introducing this section Your employees, customers, and partners need access to data,
by demonstrating the migration of applications, web sites, and other services 24 hours a day, 7 days a
a cluster-enabled volume while week, 365 days a year.
playing the video used in Exercise
10-2. Keeping that data online, along with critical applications that
depend on that data, requires an intelligent approach to system
As part of the demo, have students design that includes clustering services.
run the video (located on the
DA1/DA2 cluster-enabled SCSI Novell Cluster Services (NCS) 1.6 is a multinode clustering system
hard drive) from their workstations for NetWare 6, and is enabled for eDirectory. NCS ensures
while migrating the
availability and manageability of volumes, applications, server
cluster-enabled volume
licenses, and services.
Scenario
Enlist the assistance of students Because of the increase in company data storage and service
who have experience with NCS availability requirements, the company CIO has given you the task
1.5 or 1.6 clustering installations to of researching the use of clustering as a possible solution.
share their experience.
You know that NetWare 6 provides a 2-node NCS clustering license,
but you’re not sure what advantages NCS provides.
You want to set up a 2-node cluster in your lab for testing and
demonstration purposes but funds are not available to purchase a
commercial clustering hardware solution.
These terms are used later in this You should understand the following about high availability:
objective when introducing the
purpose and benefits of an NCS ■ Resource. Any service or data that can be migrated from one
high availability solution. server to another in a cluster
For example, you cannot migrate a physical printer from one
They are introduced here to server to another, but you can migrate a service such as iPrint
provide students with a basic
that provides access to the printer.
vocabulary to help you determine
what students know about high ■ Service. A resource that is available to a customer or employee
availability solutions, and to from a server
prevent student interruptions in
this objective. A server is a host for services. Network administrators care
about servers. Users care about services; they do not care about
These terms give students the servers until a service goes down.
ability to explain the benefits of
high availability to management,
Examples of services include printing, file access, web services,
and the ability to discuss high and email.
availability with clustering ■ Availability. The percentage of total system time that a service
consultants. is accessible for normal use
■ Uptime. The duration of time a service is functioning
■ Outage. The loss of a computer service
■ Downtime. The duration of an outage (planned or unplanned)
When determining high availability, downtime duration can be
a critical factor. For example, a system that sustains 10 outages
of 10 seconds duration each has a higher availability than a
system that has one 10 minute outage.
■ Reliability. The amount of time before a system is expected to
fail
■ Mean time between failures (MTBF). The average time
(usually listed in hours) that a device or system works without
failure
You can calculate the MTBF by dividing the total number of
operating hours by the total number of failures.
Emphasize that high availability High availability generally means one or more of the following to
should always be viewed from the your company management, employees, partners, and customers:
standpoint of the customer or
employee. ■ 24x7x365. This represents the ability to access resources 24
hours a day, 7 days a week, 365 days a year.
The term 24x7 is often used to represent 24x7x365, with 365
days a year assumed.
■ 24x7x365 at 100%. This represents 100% availability of
resources all the time, and is what most organizations and
management personnel define as high availability.
When computing this in terms of hourly outage costs and the yearly
loss at five 9s availability, see the following (as compiled by Stratus
Technologies):
Ask students how often their Although NCS provides all software features necessary to manage
servers fail or how often they bring and configure a high availability clustering solution, other factors
down their servers and for what contribute significantly to making services highly available.
reasons (such as maintenance or
software upgrades). These factors often cause computer system outages and include the
following:
Make sure students understand ■ Physical. Physical faults or hardware failures
that NCS and NetWare are
normally the most reliable ■ Design. Design errors in both the hardware and software you
components in a high availability want to cluster-enable
clustering solution. ■ Operations. Errors caused by operations personnel or users
This is especially true when ■ Environmental. Power or cooling system failures, failures of
configuring a 2-node SCSI cluster external network connections, natural disasters, and so on
with a SAN.
■ Reconfiguration. Scheduled maintenance, upgrades, or
The SCSI hard drive and adaptor configuration changes
cards are not designed specifically
for clustering, and must be In addition, there are single points of failure, such as one power
carefully configured before source or one hub. The more single points of failure, the greater the
clustering software (such as NCS) risk for maintaining high availability.
can properly use the SAN.
Although many factors can cause an interruption, NCS solves the
problem of unavailable services due to an abended server.
NCS 1.6 includes the following to help you ensure high availability:
After reviewing the features of ■ Multinode all-active cluster (up to 32 nodes). NCS lets you
NCS, you might want to discuss configure up to 32 NetWare servers (nodes) into a
with students the benefits of high-availability cluster, where resources can be dynamically
implementing NCS clustering in switched or moved to any server in the cluster.
their own network environment.
Services can be assigned across the cluster to different servers.
Any NetWare server in the cluster can restart resources from a
failed server in the cluster.
■ Multiprocessor and multithreading enabled. NCS 1.6 is more
efficient than ever because NetWare 6 is not just multiprocessor
enabled, it’s also multithreaded.
Each processor can be maximized to execute commands faster
and more efficiently, providing faster network throughput that
delivers 24 hours a day, every day of the year.
■ Consolidation of applications and operations. NCS lets you
tailor a cluster to the specific applications and hardware
infrastructure that fit your organization.
You can also reduce unplanned and planned outages for
software and hardware maintenance and upgrades.
In addition, you can lower costs by consolidating applications
and operations onto a cluster. Customers find they can reduce
the number of servers used to provide services by 50% or more.
■ Flexible resource management. You can configure resources to
automatically switch or be moved when a server fails, or you
can move them manually to troubleshoot hardware or balance
the workload.
■ Shared storage support. NCS provides support for shared SCSI
devices or Fibre Channel SANs.
After being implemented, SANs and NASs require 75% less personnel
to maintain than traditional direct attached storage.
While discussing NCS clustering, You should understand the following terms when discussing a
students might become confused clustering solution:
as you begin to use the words
node and server interchangeably. ■ Cluster. A group of servers linked together in a dedicated
network to minimize the loss of service by reducing or
A node in a cluster is a physical managing failures and minimizing downtime
server that is cluster-enabled.
■ Node. A server in a cluster
However, clustering consultants ■ Cluster resource. A server resource, application, or network
and administrators often use both service with a dynamic location managed by clustering
words to mean the same thing software
when referring to a cluster-enabled
server. In NCS, a cluster resource can only be assigned to one node at a
time.
■ Shared storage device. A device (such as external hard drives,
disk arrays, and Fibre Channel disks) in a cluster used to store
shared cluster resources
x If you use SCSI cards and a SCSI hard drive to configure a 2-node
cluster, make sure each card and the hard drive are assigned a
different SCSI ID number.
Also, make sure you purchase enclosures and cables with the
correct pin-outs.
Some students might be confused Typical cluster configurations include a shared disk system
about the spelling and use of the connected to all servers in the cluster. If a server fails, another
term fibre. server is assigned the resources. This gives users continuous access
to resources such as data, applications, and services.
Fibre Channel refers to a protocol
that uses fiber cable or copper wire You can use 2 basic shared disk system configurations:
to transmit data from the nodes to
the shared storage device. ■ Fibre Channel Cluster Configuration
■ SCSI Hard Drive Cluster Configuration
Figure 8-2 (slide) Network Hub NIC = Network Interface Card (Board)
SCSI = SCSI Adapter Card
Node 1 Node 2
NetWare 6 NetWare 6
However, external SCSI hard drives and SCSI adaptor cards are not
designed specifically for clustering.
x If the SCSI adaptor cards and drivers are multi-initiator enabled, they can
share a SCSI hard drive, but be careful about configuration issues (see
Troubleshooting a 2-Node NCS SCSI SAN in this section for details).
You might want to use the Cluster You need to know the following NCS system terms:
State view (ConsoleOne) or the
Cluster Status view (NetWare ■ Master Node
Remote Manager) to identify and ■ Cluster-Enabled Volumes and Pools
reinforce some of these terms.
■ Shared Storage Device
■ Cluster Resource
■ Heartbeats, Tics, Poison Pills, and the Split Brain Detector
(SBD)
■ Fan-Out Failover
Master Node
Make sure students understand The shared storage device in an NCS cluster is where customers and
that if employees or customers employees access files on the SAN.
need direct read/write access to
data and files, you should store the For example, if employees need access to an EMAIL volume, you
files in a cluster-enabled volume can mount EMAIL as a volume on the shared storage device and
on the shared storage device. instruct employees to map a drive to that copy of EMAIL.
However, if the service or By placing EMAIL on the shared storage device, you ensure that
application maintains the public services associated with the volume are always available, no
employee or customer data (such matter which node fails or which node you take offline.
as synchronized iFolder data), you
should create a cluster resource Other examples of placing files on the shared disk include web sites,
that accesses the volume for the print drivers, and iFolder user data files.
service or application.
By moving all customer and employee file access to the shared
storage device, you can reduce the number of servers needed in your
network and reserve files on the local hard drive of a node (such as
volume SYS) for network administrators.
Cluster Resource
Make sure students understand A cluster resource is an object in eDirectory that represents an
that resources in a cluster are no application or other type of service (such as DHCP or the master IP
longer dedicated to a particular address) that you can migrate or failover from one node to another
server. They become server in an NCS cluster.
independent in a cluster.
The cluster resource object includes scripts for unloading the
Although a resource is initially service from one node and loading it on another node.
assigned to a server, that resource
can be migrated at any time to In most cases, make sure the service is installed on all nodes in the
another server. cluster that will host the service.
NCS uses heartbeats on the LAN, tics on the SAN, and a split brain
detector (SBD) on the shared storage device, and poison pills to
keep all services highly available on the cluster when a node fails:
■ A heartbeat is a small IP packet sent periodically over the LAN
(not the SAN) by the master node and the slave nodes in the
NCS cluster.
The master node sends out a multicast heartbeat to all slave
nodes. Each nonmaster node (slave node) sends out a unicast
heartbeat to the master node.
The nodes monitor the heartbeat of other nodes in the cluster at
a tolerance rate of 8 seconds (default setting).
The tolerance rate is the amount of time a node waits for a
heartbeat from another node before taking action that results in
casting off (abending) the failed node.
■ A tic (Transport Independent Checking) is a type of heartbeat
sent over the SAN by a node. The tic writes an epoch number to
the node’s sector in an SBD partition on the shared storage
device.
The failed node can only join the cluster again when you reboot the
server and the node starts running the cluster protocol.
x For details on heartbeats, split brains, and poison pills, see TID 10053882.
NCS determines when a slave node fails and casts off the failed
node through the following process:
1. The master node monitors the heartbeats of all other nodes in the
cluster to determine if they are still “alive.” The master node also
reads the epoch numbers for all nodes in the cluster.
2. If a heartbeat is not received from a slave node within 8 seconds
(the default tolerance rate), the master node and remaining slave
nodes create a new cluster membership view without the failed
node.
3. Each node in the new membership (including the master node)
updates its epoch number by 1 in the SBD partition.
This causes a split brain, because the epoch number for the
isolated node is one less than for the rest of the nodes.
4. NCS uses this information in the SBD to vote between the 2
cluster memberships.
The cluster membership that has the most nodes wins. If there
are equal nodes in both views, the side with the membership
that contains the previous master node wins.
In the special case of a 2-node cluster, if one of the nodes can
still communicate over the LAN and the other node can’t, the
node with the good connectivity wins.
5. The nodes in the surviving cluster membership write a special
token to the sector in the SBD partition for the losing node.
In this case, the token is written to the sector for the failed slave
node.
6. The losing node reads the special token, and then abends by
taking a poison pill. The poison pill causes a self-inflicted abend
that stops all processes on the node.
Abending ensures that nodes on the losing side cannot corrupt
the new, healthy cluster.
7. The new cluster (minus the failed node) migrates the resources
(volumes and services) assigned to the failed node to other nodes
in the cluster and services continue uninterrupted for customers
and employees.
This only happens in a 2-node cluster where both master and slave have a
vote and NCS can’t determine which node is communicating. In a 3-node
cluster, the slaves have 2 votes and will cast off the master node. For details
on these counters, see TID 10057437.
SBD Partition
Each node sends a heartbeat over the LAN and writes an epoch
number in a tic to the SBD partition. Node 1 is the master node and
the current epoch number for all nodes is 2.
Suppose the network board for node 3 fails. The following occurs to
maintain the health of the cluster:
1. The master node (node 1) listens for a heartbeat over the LAN
from nodes 2 and 3.
2. After 8 seconds (the default), the master node notifies node 2
over the LAN that there is a new cluster membership view that
does not include node 3.
Node 3 continues to maintain the old cluster membership view
and the old epoch number (2).
3. The master node and node 2 write a new epoch number (3) to the
SBD partition.
4. NCS uses the information in the SBD partition to vote between
the 2 cluster membership views.
5. Because there are 2 nodes in the new cluster membership view
with a new epoch number, node 3 loses.
6. The master node and node 2 write a special token to the sector in
the SBD sector for node 3.
7. Node 3 reads the special token, takes a poison pill, and abends.
8. NCS migrates the resources on node 3 to the master node and
node 2.
SBD Partition
Fan-Out Failover
Emphasize that the clustering You will probably want to distribute, or fan out the volumes and
administrator and consultant are resources to several nodes based on factors such as load balancing
responsible for making sure that and the availability of installed applications.
resources are configured to fan out
correctly across the cluster when a For example, suppose you configure a 3-node NCS cluster as
node fails. follows:
NCS lets you configure this ■ DHCP server and iFolder resources assigned to node 2
resource distribution, but does not
■ 2 web site resources assigned to node 2
detect the network traffic, data, or
services load of each node nor ■ 2 email resources assigned to node 3.
does it adjust the fanning out of
resources. The following illustrates how this setup might look:
The following shows how the email resources migrate when node 3
fails (based on the network administrator configuration):
Node 3
Node 1 Node 2
When the problems with node 3 are resolved and the network
administrator starts running node 3, Email A and Email B remain
running on nodes 1 and 2 because the resources are not configured
to automatically failback to node 3.
These rules focus on the To avoid serious problems with migrating and failing over services,
importance of cluster-enabling all you must follow several rules when managing an NCS SCSI SAN.
nodes in a cluster. If a node is not
clustered, serious data and When you install NCS with shared storage, each shared storage
volume loss occurs. device is assigned a global unique ID (GUID), and a flag is set on
the device to help Netware 6 distinguish between local server
storage devices and shared storage devices.
When working with shared storage, you must observe the following
rules or risk data corruption or volume loss:
■ Don’t attach a noncluster server to the shared storage device
unless you isolate the storage so the noncluster server has
access only to its own volumes.
All servers attached to the shared storage device (whether in the
cluster or not) have access to all volumes on the shared storage
device unless you specifically prevent such access.
NCS manages access to shared volumes for all cluster nodes but
cannot protect shared volumes from being corrupted by
noncluster servers.
x Novell does not maintain a list of cards and drivers that are
multi-initiator enabled.
x For external SCSI devices such as a hard drive, the SCSI ID usually is
set with a switch on the back of the device.
For a 2-node NCS SCSI cluster, the hard drive must not be
terminated. The 2 adaptor cards are at the ends of the SCSI bus
and should be terminated.
However, if an adaptor card fails, termination also fails. You can
avoid this situation by using an external terminator at each end
of the SCSI bus.
(Most newer SCSI devices use autotermination.)
■ Have you done a low-level format on the SCSI hard drive?
Every SCSI hard disk must be physically low-level formatted,
partitioned, and logically formatted before it can be used to
store data.
Most SCSI drives are pre-formatted at the factory. However, if
you connect a used SCSI hard drive to the NCS nodes for
clustering, you must perform a low-level format before you can
use the drive.
Because a low-level format destroys all data on the drive, make
sure you back up the data before performing a low-level format.
In addition, you might want to use the SCSI utility to configure the
following BIOS settings of your SCSI adaptor cards:
Make sure the BIOS settings for both SCSI adaptor cards are
identical. You can then adjust the settings to resolve problems with
the SAN.
b For tips on configuring SCSI adaptor cards for a NCS 2-node SCSI SAN, see
http://developer.novell.com/research/sections/netsupport/abend/2001/april/s
pv.htm and http://developer.novell.com/research/sections/netmanage/tips/
2001/November/t011101.pdf.
a 40 minutes Now that you know more about clustering with NCS, you are ready
to set up a 2-node NCS cluster in your lab using an external SCSI
hard drive for a SAN.
In this exercise, you design the 2-node SCSI cluster and set up the
SAN for the installation of NCS.
Note that the cluster name (DACluster) is the same for all clusters,
but the context (such as IS.DEL.DIGITALAIR) and IP address are
different for each cluster. First server and second server are
references used in the exercise.
Emphasize that designing a Using Figure 8-2 as reference and the information in Table 8-4,
cluster for a production draw a 2-node cluster configuration that uses the following:
environment is critical to the
success of the cluster. ■ A LAN that connects 2 NetWare 6 servers with a hub
■ A SCSI SAN with a shared SCSI hard drive
■ Two Windows 2000 workstations connected to the LAN
through the hub
Use cluster names, node names, and IP addresses from the table.
Now set up the SAN. In addition to connecting the SCSI hard drive
to your 2 servers (such as DA4 and DA5), you must also remove all
partitions from the SCSI hard drive in preparation for installing the
SBD partition.
Do the following:
1. Bring down NetWare 6 on your servers; then turn off both
servers.
You can bring down NetWare 6 on a server by pressing Ctrl +
Esc, entering 1 for System Console, and then entering DOWN
at the console prompt.
2. Plug in the SCSI hard drive to a power source; then make sure the
drive is off.
3. Connect both servers to the SCSI hard drive enclosure using the
SCSI cables.
4. Turn on the SCSI hard drive.
5. Turn on your servers.
6. As each server boots, look for the name of the SCSI adaptor card
(such as Adaptec 2940) and the keystrokes for accessing the
adaptor card BIOS (such as Ctrl+A).
7. Access the BIOS configuration utility for the SCSI adaptor card
on each server.
8. Select the option that shows the SCSI ID numbers for the SCSI
components associated with the server (the hard drive and the
adaptor card).
Each adaptor card and the SCSI hard drive should have a
different SCSI ID number.
For example, your SCSI hard drive might be assigned a SCSI
ID number of 0 (the same on each server) while the SCSI
adaptor card should be assigned a different SCSI ID (such as 6
or 7).
9. (Conditional) If 2 of the components use the same SCSI ID
number, change the number for one of the components; then
reboot the server.
Part III: Check the Drivers for the SCSI Card and Hard Drive
Now that the SCSI hard drive is connected to both servers and
running, and you have checked the SCSI ID numbers, you can
configure the SAN for NCS. Do the following:
1. After NetWare 6 loads (and the GUI interface appears) on both
servers, make sure a driver for the SCSI adaptor card and a driver
for the SCSI hard drive are installed on your servers by doing the
following on each server:
a. Start the NetWare Configuration utility by entering
NWCONFIG at the console prompt.
You should see SCSIHD and the driver for your SCSI card
listed with a Currently Loaded status.
f. Exit NWCONFIG by pressing Esc until an Exit the Install
message appears; then select Yes.
2. At the console prompt for each server, enter SCAN FOR NEW
DEVICES; then enter LIST DEVICES.
After a few moments, you see the SCSI hard drive listed with a
device ID number (such as 0x0005 or 0x000C) and a name
(such as FUJITSU MAE3091LP).
The device ID number might be different on each server, but the
hard drive name should be the same.
x If you see Unbound Device instead of the SCSI hard drive name, the
driver for the SCSI hard drive has not been properly installed.
3. Record the name of the SCSI hard drive and the device ID
number for each server:
DAx:
DAx:
When you connect a SCSI hard drive for clustering that contains
partitions and data, you must initialize the drive. This prepares the
drive for installing the SBD clustering partition and for creating
pools and volumes.
(Make sure you back up valuable data from the hard drive before
initializing it.)
x Initializing the hard drive removes the partitions and rewrites the MBR. This
works most of the time in class for this exercise.
Do the following:
1. Start ConsoleOne from your second server; then log in as admin
to your second server.
For example, if you are clustering DA4 and DA5, your second
server is DA5.
2. Right-click the second server object (in IS.xxx.DIGITALAIR
where xxx = your location container); then select Properties.
For example if you are clustering DA4 and DA5, right-click
DA5 in IS.LGA.DIGITALAIR.
3. List the devices connected to the server by selecting Media >
Devices.
4. Select the device ID number of your SCSI hard drive.
The name of the SCSI hard drive appears in the Description
field.
(End of Exercise)
Summary
The following is a summary of the objectives in this section:
Exercise Answers
The following is the suggested design for your 2-node SCSI cluster:
LAN
Windows 2000 Windows 2000
Network Hub
DACluster
192.168.1.83
SAN
NetWare 6 NetWare 6
192.168.1.81 192.168.1.82
Duration: 2 hours 30 minutes In this section you learn how to install, monitor, and test a 2-node
NCS cluster for use in a non-production lab environment.
Objectives
1. Verify NCS System Requirements
2. Create a Cluster by Installing NCS
3. Check Cluster Configuration Settings
4. Test and Monitor the Cluster
Introduction
After designing a 2-node cluster with a SCSI shared disk drive and
setting up and configuring the SAN, you can install and test NCS on
the 2-node cluster.
Hardware Requirements
You might want to use these The following lists minimum hardware requirements for installing
requirements as a checklist and NCS 1.6 on a 2-node NCS cluster:
have students confirm that their
hardware and software meet the ■ Two NetWare 6 servers
requirements. ■ 256 MB of memory on all servers in the cluster (512 MB is
recommended for failing multiple applications to the same
For example, as you discuss
server)
memory requirements for the
servers, have students check each ■ At least one local disk device (not shared) for SYS on each
server to make sure the server
requirement is met.
If you are configuring a 2-node SCSI cluster, you must have a SCSI
adaptor card installed in each server. In addition, the external SCSI
hard drive and each SCSI adaptor card must have a unique SCSI ID.
Software Requirements
License Requirements
NCS requires a Cluster Server License for each server that is part of
the cluster. The Cluster Server License allows a server to join a
cluster. Cluster Server License objects are created in the same
eDirectory context as the Cluster object.
x If the disks in the shared disk system are not configured to use mirroring or
RAID 5, a single disk error can cause a volume failure. NCS does not protect
against such faults.
After running the installation the first time to create a cluster, run
the installation again to add servers to your cluster or to upgrade
NCS software on a cluster.
Figure 9-1
Figure 9-2
Figure 9-3
Figure 9-4
6. Enter the name for the cluster object you are creating and specify
the eDirectory tree and context where you want it created; then
select Next.
The NCS Cluster Node Modification screen appears:
Figure 9-5
7. Add the servers you want in the cluster to the NetWare Servers
in Cluster list by doing one of the following:
❑ Enter the name of the server in the NetWare Servers box;
then select Add to Cluster.
❑ Select the browse button, find and select a server, and select
Add. Repeat this for each server you want in the cluster.
When you finish, select OK.
The installation program detects each server and then adds the
server name and IP address to the list.
You can remove a server you added to the list by selecting the
server and selecting Remove.
If the server you are adding has more than one IP address, you
are prompted to select the IP address you want NCS to use.
8. When you finish adding servers to the list, select Next to
continue.
The Cluster IP Address Selection screen appears:
Figure 9-6
Figure 9-7
11. Specify whether your cluster has a shared disk system; if so,
select the drive where you want the special cluster partition
created.
NCS requires a special cluster partition on the shared disk
system. You are also given the option of mirroring the partition
for greater fault tolerance.
x To create the SBD partition you must have at least 10 MB of free space
that is not part of an NSS partition on one of the shared disk drives. If
no free space is available, the shared disk drives can't be used by NCS.
Figure 9-8
13. Choose whether you want the servers you are adding to your
cluster to start NCS software after the installation.
If you choose to not start NCS software on each server that you
upgrade or add to your cluster, you must manually start the
server after the installation or you must reboot cluster servers to
automatically start the server.
You can manually start NCS by entering LDNCS at the server
console on each cluster server.
14. Continue by selecting Next.
Figure 9-9
Figure 9-10
Cluster Object
Some students might be confused The cluster object contains several objects necessary for configuring
about services and servers in a and running the cluster. The cluster objects include the following:
cluster. This is especially true of
the role of the master node. ■ Master IP address resource. When you install a cluster, you
assign an IP address to the cluster. The IP address and the
Help students understand that the scripts for loading and unloading the IP address are stored in
master node is determined by this object.
where the Master IP Address
resource is running, and not by a
The master IP address resource object is new to NCS 1.6. The
specific node configured as the address is always assigned to the master node and allows the
master node in the cluster. cluster to advertise on the LAN as though it were a virtual
server.
■ Cluster server nodes. Configuration settings for each server in
the cluster are stored in a node object. These settings include
the server IP address.
When you first start the cluster, NCS waits for a specific amount of
time and for a specific number of nodes to join the cluster before
starting. The specific number of nodes is called a quorum.
Figure 9-11
Cluster Protocol
You can use the Cluster Protocol tab pages to view or edit the
transmit frequency and tolerance settings for all nodes in the cluster,
including the master node.
The master node is generally the first node brought online in the
cluster. However, if that node fails, any of the other nodes in the
cluster can become the master.
Figure 9-12
For example, if you set this value to 8 and the master node
does not receive an “I’m alive” signal from a node in the
cluster within 8 seconds, that node is cast off from the
cluster.
You might want to increase the tolerance value if there is
significant traffic on the LAN and you want to ensure that
the master node waits long enough before initiating the
cast-off process.
You might want to decrease the tolerance value if you feel
the master node is not responding quickly enough to meet
the design specifications of the cluster.
However, unless there is some compelling reason to change
this setting, keep the tolerance value at 8 seconds. This
setting is optimal for most installations of NCS cluster.
❑ Master Watchdog. You can set Master Watchdog to specify
the amount of time between transmits for the master node
in the cluster.
For example, if you set this value to 1, the master node in
the cluster transmits an “I’m alive” signal to all other nodes
in the cluster every second.
❑ Slave Watchdog. You can set Slave Watchdog to specify the
amount of time the master node has to signal that it is alive.
For example, if you set this value to 8 and the nonmaster
nodes in the cluster do not receive an “I’m alive” signal
from the master within 8 seconds, the master node is cast
off from the cluster and one of the other nodes becomes the
master node.
❑ Max Retransmits. You can set this option to the number of
times the master node waits for a heartbeat from another
node before casting if off from the cluster.
4. Save the configuration settings by selecting OK.
5. Restart the cluster.
x You should not make any changes to the configuration settings unless you
check with Novell Technical Support or a qualified NCS clustering
consultant.
When you install NCS, you assign an IP address to the cluster. The
cluster IP address normally does not need to be changed, but you
can change it if needed.
Resource Priority
Some students might have You can use the Resource Priority configuration settings to control
questions about the colors the order in which multiple resources start on a given node when the
associated with the resource cluster is brought up or during a failover or failback.
priority list.
For example, if a node fails and 2 resources fail over to another
The colors have no specific node, the resource priority determines which resource loads first.
significance in relation to the order
or placement of resources in the This is useful for ensuring that the most critical resources load first
list. and are available to users before less critical resources.
Figure 9-13
You can automatically send email messages for cluster events such
as cluster and resource state changes or nodes joining or leaving the
cluster with cluster email notification.
You can enable or disable email notification for the cluster and
specify up to 8 administrator email addresses for cluster
notification.
Figure 9-14
You can view information such as loading and unloading scripts for
the cluster; start, failover, and failback node settings; and nodes
associated with the cluster by showing the
Master_IP_Address_Resource object properties.
Make sure students understand To view the settings in ConsoleOne, do the following:
that the Master_IP_Address
Properties dialog is for viewing 1. Select the cluster object.
properties only. 2. On the right side of the ConsoleOne display screen, right-click
the Master_IP_Address_Resource object.
NCS updates configuration
information in the dialog. 3. Select Properties.
4. In the Properties dialog, select the Scripts tab.
The Properties dialog appears:
Figure 9-15
You can view or edit the cluster server node number or IP address of
the selected node. You can also view the context for the NetWare
Server object.
Figure 9-16
a 40 minutes With the SCSI SAN connected to the NetWare 6 servers, you are
ready to set up the clustering software.
In this exercise you install NCS for the 2-node SCSI clustering
solution you designed and set up in section 8.
x You can remove a server you added to the list by selecting the
server and selecting Remove.
For example, if you are using DA6 and DA7 in your cluster, make sure
DA6 holds a replica of DA7 (and vice versa); then start the installation
again.
15. (Optional) View the Readme file by selecting View from the
Installation Complete dialog.
16. End the installation by selecting Close from the Installation
Complete dialog.
17. Exit NetWare Deployment Manager by selecting Cancel; then
select Yes.
18. Remove the NetWare 6 OS CD from the workstation.
Review the property values When you finish installing a cluster, check the basic cluster and
students record for the cluster and node properties to make sure the correct settings are included in the
the nodes to make sure students cluster object.
understand the impact of those
values on the operation of the You might also want to record these settings for future reference.
cluster.
For example, to check and record the settings for the DACluster
object, do the following:
1. On both workstations, make sure a shortcut exists for
ConsoleOne.
2. Start ConsoleOne using the ConsoleOne shortcut on your first
server; then open the IS container where your first server is
located.
Quorum Timeout
Membership
Protocol Heartbeat
Tolerance
Master WatchDog
Slave WatchDog
Max Retransmits
Management IP Address
Port
Notice that the node number is 0 even though your first server is
listed as node 1 in the Cluster Membership Monitor. When
identifying nodes internally, NCS starts counting nodes from 0.
8. When you finish, select Cancel.
Your first and second servers appear in the assigned list. If your
first server (currently the master node) fails, the master IP
address resource is migrated to your second server and the
cluster IP address continues to be broadcast on the network.
15. Close the Properties dialog by selecting Cancel.
(End of Exercise)
View and discuss the Cluster State ConsoleOne and NetWare Remote Manager provide a special status
view in ConsoleOne and the screen for cluster objects that you can use to perform tasks such as
Cluster Status view in NetWare view the cluster state, migrate resources, check an events log, and
Remote Manager. print or save an HTML report (in ConsoleOne only) on the cluster
state.
Figure 9-17
The cluster object name and the epoch number appear at the top of
the view. The epoch number indicates the number of times the
cluster state has changed. The cluster state changes every time a
node joins or leaves the cluster.
An icon for each cluster server appears in the middle; the resources
running in the cluster appear in a list at the bottom. Gauges indicate
the percentage of nodes and resources available in the cluster.
Event Log
Figure 9-18
Every time the cluster state changes, a new event is added to the
event log.
You can perform the following while viewing the event log:
■ Sort events in the log by selecting the column headings of the
table
■ Reverse the sort order by pressing the Shift key while selecting
a column heading
■ Save the event log to a file
The event log is stored on the SBD cluster partition on the shared
storage device, which ensures that the event log is always available.
Figure 9-19
You can save this report to an HTML file for printing or viewing
with a web browser.
You can use the following to view the status of the cluster and SBD
partition, and to load and unload clustering:
a 30 minutes After installing NCS, you decide to run some tests to check the
health and reliability of the cluster.
In this exercise you test the SBD partition and heartbeat monitoring
on the 2-node SCSI clustering system you have configured by
observing a node abend.
The following lists information you might need when testing the
partition:
x If your second server is cast off (abends) instead of your first server,
NCS has had a problem reading the statistics on the network board
drivers and can’t determine which node is still communicating over
the LAN.
Because there are only 2 nodes with 1 vote each, NCS breaks the
tie by keeping the master node (your first server) alive and casting
off the slave node (your second server).
11. Compare the abend message in the log entry to the abend
message you recorded for your first server earlier in the exercise
(step 3e).
The messages are the same.
12. Exit the Abend log by pressing Esc twice and selecting Yes.
(End of Exercise)
Summary
The following is a summary of the objectives in this section:
1. Verify NCS System Before installing NCS, your system must meet
Requirements the following requirements:
■ Hardware Requirements
Minimum hardware for installing NCS 1.6 on a
2-node NCS cluster include 2 NetWare 6
servers, 256 MB of memory each server (512
MB recommended), and 1 local disk device
(not shared) for SYS on each server.
■ Software Requirements
Software includes all nodes in the same
eDirectory tree and running NetWare 6, all
servers configure for IP and on the same IP
subnet, and an IP address for the cluster and
each resource or volume.
■ License Requirements
NCS requires a cluster server license for each
server that is part of the cluster. The license
allows a server to join a cluster.
■ Shared Disk System Requirements:
■ At least 15 MB of free disk space available
on the shared disk system for creating the
SBD partition
■ The shared disk system set up and
functioning according to the
manufacturer's instructions
■ The disks in the disk system configured in
a mirroring or RAID 5 configuration to add
fault tolerance to the system
2. Create a Cluster You must run the NCS installation when you do
by Installing NCS the following:
■ Create a cluster
■ Add nodes to an existing cluster
■ Upgrade NCS software in an existing cluster
The installation does the following:
■ Creates a cluster object in eDirectory
■ Installs NCS software on servers you specify
for your cluster
After running the installation the first time to
create a cluster, run the installation again to add
servers to your cluster or to upgrade NCS
software on a cluster.
4. Test and Monitor To test and monitor the cluster state, you need
the Cluster to learn about the following:
■ Cluster State and Cluster Status Views
ConsoleOne and NetWare Remote Manager
provide a special status screen for cluster
objects that you can use to perform tasks such
as view the cluster state, migrate resources,
check an events log, and print or save an
HTML report (in ConsoleOne only) on the
cluster state.
■ Console Prompt Commands
To view a list of commands, enter HELP
CLUSTER or HELP SBD at a server console
prompt, or enter HELP and a command (such
as HELP CLUSTER VIEW) to see a
description and example.
Exercise Answers
Quorum Timeout 60
Membership 2
Protocol Heartbeat 1
Tolerance 8
Master WatchDog 1
Slave WatchDog 8
Max Retransmits 30
Duration: 1 hour 30 minutes In this section you learn how to cluster-enable and test high
availability access of data using NCS on a NetWare network.
Objectives
1. Configure NCS for High Availability File Access
2. Manage Resources in an NCS Cluster
Introduction
After creating an NCS cluster, you need to create and configure
cluster resources to make them available to customers and
employees.
Although the Sharable for When you add a device to the SAN, NetWare 6 detects that the
Clustering option lets students device is shared storage and identifies it as sharable for
configure local disk devices as clustering.
shared, they should avoid doing
this to keep the SAN isolated from If NetWare does not detect a device as shared storage on the
local drives on server nodes. SAN, you need to select this option.
Device names are not changeable and might be labeled
The only time students need to
select the Sharable for Clustering
something like 0x2 or 0x1.
option is if NetWare does not 5. On the Media tab, select Partitions; then select New.
detect a device as shared storage
on the SAN. 6. Select the device where you want to create the partition (the same
device you selected in step 3).
7. Specify the size of the partition and make sure of the following:
❑ NSS is selected as the partition type (the default)
❑ Hot Fix and Mirror are selected
❑ Create New Mirror Group is selected
8. Create the partition by selecting OK.
eDirectory objects for the volume, pool, and virtual server are
created in the same container as the NetWare server. In
addition, a cluster resource object for the NSS pool is created in
the cluster object.
For example, if you create a cluster-enabled USERS pool with a
MEDIA volume for DACLUSTER, the following objects are
created:
The first 2 guidelines are targeted The following are guidelines for cluster-enabling a pool or volume
at students who are already in NCS 1.6:
familiar with previous versions of
NCS and need to know what’s new ■ Cluster-enabled volumes no longer appear as cluster resources.
in NCS 1.6. The load and unload scripts in cluster resource objects apply to
pools (not volumes).
■ Each cluster-enabled NSS pool requires its own IP address for
the virtual server. This means that each cluster-enabled volume
does not have an associated load and unload script or an
assigned IP address.
■ The first volume you cluster-enable in the pool cluster-enables
the pool where the volume resides.
After a pool is cluster-enabled, you must cluster-enable the
other volumes in the pool if you want them to be mounted on
another node during a failover.
■ When a node fails, any cluster-enabled pools being accessed by
that node are migrated to other nodes in the cluster.
All volumes in the pool are migrated with the pool, but only
volumes that have been cluster-enabled are mounted. Any
volumes in the pool that are not cluster-enabled must be
mounted manually.
For this reason, volumes that aren’t cluster-enabled should be in
separate pools that are not cluster-enabled.
■ If you want each cluster-enabled volume to be its own cluster
resource, each volume must have its own pool.
■ If a server application does not require NetWare client access to
volumes, cluster-enabling those pools and volumes might not
be necessary.
■ Pools should be deactivated and volumes should be dismounted
before being cluster-enabled.
a 20 minutes When you initialized the SCSI hard drive, you removed all existing
partitions. Installing NCS created an SBD partition on the hard
drive to monitor clustering.
Although it is easier to create and At this point, you could use the rest of the hard drive as a partition
cluster-enable an NSS pool and for an NSS pool. However, all data and services on the hard drive
volume using NetWare Remote will migrate or failover to the same node if they are stored in the
Manager, this exercise focuses on same partition and pool.
using ConsoleOne to perform
these tasks. You decide to create at least 2 partitions on the SCSI hard drive to
test migration and failover of part of the data on the hard drive from
NetWare Remote Manager is one node to the other.
more task-oriented and less
focused on Directory objects. By In this exercise, you create a partition that uses half of the available
using ConsoleOne, students can
storage space on the SCSI hard drive, and then create a
observe and explore more
MULTIMEDIA pool and VIDEO volume in that partition.
conveniently the eDirectory
objects that support NCS
clustering.
Part III: Verify That the Cluster Objects Are Created for the
Volume
To verify that the cluster-enabled objects have been created for the
MULTIMEDIA pool and volume VIDEO, do the following:
1. From one or both workstations, make sure you are logged in as
Admin to one of the servers; then start ConsoleOne.
2. Find the following in IS.xxx.DigitalAir for your first server:
❑ DACluster_VIDEO (cluster-enabled volume)
❑ DACLUSTER_MULTIMEDIA_SERVER (virtual
server)
❑ DACLUSTER_MULTIMEDIA_POOL (cluster-enabled
pool)
3. Right-click DACluster_VIDEO; then select Properties.
4. Select General > Identification.
5. Verify the name of the host server on the Identification tab page.
You see the DACLUSTER_MULTIMEDIA_SERVER virtual
server listed.
6. Close the Properties dialog by selecting Cancel.
7. Right-click DACLUSTER_MULTIMEDIA_POOL; then
select Properties.
8. Select Other; then expand the Host Server attribute.
9. Verify the name of the host server.
The pool object is associated with the same virtual server.
In the Network address field you see the IP address you entered
when cluster-enabling the volume (such as 192.168.1.24).
With the pool and volume associated with a virtual server,
VIDEO is always available in the cluster because the virtual
server IP address continues to be broadcast as the resource
migrates from node to node.
13. Close the Properties dialog by selecting Cancel.
(End of Exercise)
Migrating resources lets you balance the load and evenly distribute
applications among the servers in your cluster.
This topic introduces the When running or testing an NCS cluster, you can view valuable
troubleshooting table. You do not information about cluster resource states from the Cluster State
need to cover every state, view in ConsoleOne or from the Cluster Status view in NetWare
description, and possible action Remote Manager.
listed.
The first solution to most error messages listed (such as Comatose
or NDS Sync) is to take the resource offline and then bring it online
again.
If this does not work, take the resource offline, check the
configuration settings (especially the load and unload scripts), and
then bring the resource online again.
Exercise 10-2 Test High Availability File Access on the 2-Node Cluster
During classroom setup, you In this exercise you copy a video file to volume VIDEO and then
should have created a run the video from a Windows 2000 NetWare 6 workstation while
C:\MARKETING VIDEO folder on migrating the MULTIMEDIA_SERVER resource and failing a
each student workstation with a node.
video file that runs 1–2 minutes.
Specifically, you do the following:
This folder is also available in
EXERCISES\SECTION 10 on the ■ Part I: Verify That Volume VIDEO Is Cluster-Enabled
Enhanced Learning CD.
■ Part II: Test the Effects of Migrating Volume VIDEO while
Playing a Video
Before starting the exercise,
instruct students on how to start ■ Part III: Test the Effects of Failing a Cluster Node while
the video for Parts II and III. Playing a Video
For example, if you are using WS4 and WS5, you can use WS4 as
your first workstation and WS5 as your second workstation.
Try closing the video player, opening it again, and then start the
video file. Because the migration is complete, you should be able
to access and play the video again.
Part III: Test the Effects of Failing a Cluster Node while Playing
a Video
(End of Exercise)
Summary
The following is a summary of the objectives in this section:
1. Configure NCS for When configuring NCS for making data and
High Availability files highly available to users, you
File Access
■ Create a Shared Disk Partition. Before
creating disk partitions on shared storage
devices in your SAN, you must install NCS. You
should carefully plan how you want to
configure your shared storage prior to
installing NCS.
■ Create and Cluster-Enable an NSS Volume
and Pool on a Shared Storage Device.
Although storage pools must be created prior
to creating volumes, you can create and
cluster-enable an NSS volume and pool by
using the Create a New Logical Volume option
on the Media tab of the server properties
dialog.
■ Cluster-Enable an Existing Pool or Volume on
the Shared Disk System. To cluster-enable an
existing NSS volume and pool, you need to
know the following:
■ What Happens to eDirectory Objects
during Cluster-Enabling
■ How to Cluster-Enable an Existing Volume
and Pool
■ Guidelines for Cluster-Enabling the Pool
and Volume
Duration: 2 hours In this section, you learn how to cluster-enable and test service
solutions such as DHCP and iFolder on your 2-node NCS cluster.
Objectives
If you are running out of time to 1. Identify Cluster-Aware and Cluster-Naive Applications
complete this section, briefly
introduce each objective but focus 2. Identify How to Cluster-Enable an Application
on cluster-enabling an application 3. Identify How to Assign Nodes to a Resource
and accessing load and unload
scripts. 4. Identify How to Set Start, Failover, and Failback Modes
5. Identify How to View and Edit Load and Unload Scripts
Remember that the overall
objective for this module is to 6. Identify How to Find NCS Configuration and Troubleshooting
prepare students for a successful Information
experience implementing a 2-node
cluster on their own.
Repeat this step for all servers you want assigned to the
resource.
You can also use the Left-arrow button to unassign servers from
the resource.
4. Click the Up- and Down-arrow buttons to change the failover
order of the servers assigned to the resource or volume.
You can set Start, Failover, and Failback modes using ConsoleOne
or NetWare Remote Manager.
The default for both Start and Failover modes is AUTO; the
default for Failback mode is DISABLE.
4. Make sure the resource runs only on the master node in the
cluster by selecting the Master Only box.
If the master node in the cluster fails, the resource fails over to
whichever node becomes the master.
You can use any commands in a load script that would be used in an
NCF file run from the server console. If you don’t know which
commands to add to your load script, consult the documentation for
the application or resource.
Load and unload scripts are created for disk pools when you
cluster-enable them. Because of this, it might not be necessary to
configure or change the scripts for a pool.
Many TIDs reference NCS for NetWare 5.x, but the content is valid
for NetWare 6. The following TIDs address common clustering
problems and questions:
TID 10053882 is very useful for 10053882 Novell Cluster 14 Jan 2002 A detailed explanation
understanding the algorithm NCS Services: the of heartbeats, node
uses to cast off a cluster node. It Gory Details failure, false node
also includes several failure, split brains,
troubleshooting tips. false split brains, and
poison pills. A support
guide is also included.
For students to perform this You also do not need to use a secondary IP address for the DHCP
exercise, they must install and resource because DHCP uses the IP address of the server to
configure DHCP on both broadcast its service over the network.
cluster-enabled servers.
Because the DHCP service is already running on DA2 in your
You might want to use this classroom LAN, the instructor might decide to demonstrate this
exercise as a demonstration for exercise by cluster-enabling DHCP on DA1 and DA2.
cluster-enabling a resource.
If you have time in class (or in a self-study environment), try
installing DHCP on your own servers and then use the steps that
follow to cluster-enable the service.
The following lists information you might need during the exercise;
x To cluster-enable DHCP in this exercise, you must install the service on both
servers before beginning Part I.
(End of Exercise)
By placing the iFolder user data on the shared storage device, you
can migrate iFolder from one node to another when you need to
perform maintenance on an iFolder server. You can also ensure high
availability if the server hosting iFolder fails.
Your first server and your second server already have iFolder
installed with the IP addresses listed in Table 11-5.
x For this exercise, you can select which workstation you want to use as
your first (and second) workstation.
You can activate the pool by using ConsoleOne. View the properties
for your first server; then select Media > NSS Pools > IFOLDER
> Activate. Refresh the NetWare Remote Manager screen to view
the pool and volume.
e. When you finish editing, select File > Save; then exit the
Editor.
9. Copy HTTPD.CONF and HTTPD_ADDITIONS_NW.CONF
from your first server to SYS:\APACHE\IFOLDER\SERVER
on your second server:
a. On your second workstation desktop, right-click My
Network Places; then select Open.
b. Double-click Novell Connections > DigitalAir-Tree >
DIGITALAIR > xxx > IS> your first server_SYS > Apache
> iFolder > Server.
For example, if your cluster includes nodes DA4 and DA5,
you would double-click Novell Connections >
DigitalAir-Tree > DIGITALAIR > DEL > IS> DA4_SYS
> Apache > iFolder > Server.
c. Select and copy HTTPD.CONF and
HTTPD_ADDITIONS_NW.CONF.
d. From the Address drop-down list select your location
container for your second server; then double-click IS.
For example, if your cluster includes nodes DA4 and DA5,
you would select LGA > IS.
e. Double-click your second server_SYS > Apache > iFolder
> Server.
f. Paste HTTPD.CONF and
HTTPD_ADDITIONS_NW.CONF into the SERVER
folder, replacing the existing files.
10. Restart your first server and your second server by entering
RESTART SERVER at the console prompt.
11. After NetWare 6 loads on both servers, check the Cluster Status
view in NetWare Remote Manager.
All resources in the Cluster Resource list are running, except
for the IFOLDER resource.
Do the following:
1. From NetWare Remote Manager, bring the IFOLDER resource
online by selecting IFOLDER in the Cluster Resource list; then
select Online and begin the refresh.
NCS loads and runs the IFOLDER resource on your first
server.
2. From your first server console prompt, enter VOLUMES.
USERDATA is mounted on your first server.
3. From NetWare Remote Manager, migrate the IFOLDER
resource to your second server.
4. After the IFOLDER resource is running on your second server,
at your second server console prompt enter VOLUMES.
USERDATA is now mounted on your second server.
5. On your second server, check the load messages for Apache
server by pressing Ctrl + Esc and selecting Apache for
Netware.
If Apache server loaded successfully, you see an “iFolder server
initialization complete” message.
7. Try out your test plan and share your results with the rest of the
students.
(End of Exercise)
Summary
The following is a summary of the objectives in this section:
4. Identify How to Set You can configure the start, failover, and
Start, Failover, and failback of cluster resources to happen
Failback Modes
manually or automatically.
You can set Start, Failover, and Failback modes
using ConsoleOne or NetWare Remote
Manager.
Exercise Answers
Objectives
Prior to beginning this section, 1. Create a Disaster Recovery Plan
introduce problems into the
network by completing the steps in 2. Troubleshoot Network Problems
“Setup for Section 12” on
Setup-50.
Introduction
In this section, you develop a disaster recovery plan and then
troubleshoot server and communication problems on the network.
Don’t wait until a disaster strikes to determine how you will restore
service. Doing so could cost you your job. Creating a disaster plan
involves the following:
■ Planning for Hardware Failures
■ Planning for Calamities
If a server hard disk fails and you don’t have a recent backup to
restore from, it could cost your job.
■ To further protect data, consider implementing a RAID solution
that incorporates redundant disks. If a particular disk in a RAID
array fails, the redundant disk can take over, ensuring the data
remains intact.
■ To make sure NetWare services remain available, consider
implementing a clustering solution. If a server in the cluster
goes down, other servers in the cluster can take over, ensuring
services remain available to end users.
In this exercise, help students In this exercise, you must troubleshoot and repair every problem on
isolate each problem and then the network.
guide them to the proper solution.
The symptoms being reported to the Digital Airlines help desk
However, avoid giving students include:
too much information. Make them
draw upon their experience in ■ Users in all locations can’t log in to DA4, DA5, DA6, DA7,
class to arrive at a solution. DA8, or DA9.
■ Three users can’t log in to DA3.
Verify that students fix every bug
you introduce into the system. ■ Local administrators report various time error messages on
several servers.
■ iFolder users on DA7 can’t log in with their iFolder clients.
■ iFolder users on DA8 and DA9 report that all of their files are
missing.
■ Administrators in Sydney report that workstations aren’t
receiving DHCP addresses.
(End of Exercise)
Summary
Objective Summary
Remember that this is a standard and not all layers in the OSI model
are used in every communication.
The following shows the OSI layers and the data packet reference
names:
UNIX
Application Messages & Packets
Presentation H1 Packets
Session H2 Packets
Transport H3 Datagrams, Segments & Packets
Network H4 Datagrams & Packets
Data Link H5 Frames & Packets
Physical Bits & Packets
Figure B-2
Computer A Computer B
IP Routing
Every packet being processed by a TCP/IP host has a source and
destination IP address. The router examines the destination address
on each packet, compares it with entries in its local routing table,
and decides what action to take.
There are three code paths that the packet can take:
x This occurs when you are using a software router on NetWare, Linux, or
Windows, etc.
The route table can maintain 4 types of routes. They are listed as
follows in the order that they are searched for a match:
1. Host (a route to a single, specific destination IP address)
2. Subnet (a route to a subnet)
3. Network (a route to an entire network)
4. Default (used when there is no other match)
Figure C-1
For example, the decode in the previous figure breaks down the
DHCP options to list the magic cookie, message type, client
identifier, hostname, and parameter request list.
If a DHCP decode was not available with LZFW, you would need to
look up the field lengths and values in the DHCP specifications
(RFC 2131) and manually decode the packets.
■ Capture filters. Capture filters define the traffic that should be
copied from the network into the trace buffer.
For example, you could set a capture filter to capture all TCP/IP
traffic to and from a workstation, or you could set a capture
filter to capture all IPX/SPX traffic.
Figure C-2
Figure C-3
Some alarms are more critical than others. The following list
identifies the more critical alarms:
❑ Utilization Percentage. The percentage of the bandwidth
(available roadway) that is used up.
For example, on Ethernet networks, performance degrades
significantly when use is above 40%. If utilization is high,
you should also consider the collision/fragment error count.
Standalone Analyzers
Standalone analyzers are the most popular type of analyzer. You can
use a portable analyzer in various locations.
Router
Hub Hub
To view traffic on the other side of the router, LZFW would need to
be moved to a hub port on the other side of the router.
Router
Hub Hub
Analyzer Analyzer
Agent Agent
Because an analyzer agent has been placed on each server, you can
capture data on both sides of the filtering device (router). In this
case, you can see all traffic on each side of the router, accounting for
the entire network.
x Be sure you examine the traffic crossing the network. Make sure you are not
causing problems (such as adding extra load onto the network).
Hubbed Network
Bridged Network
A B
Port 1: A and B
Port 2: C and D
Analyzer
1
C D
Bridge 2
If a packet from client A is destined for client B, the bridge will not
forward the packet because both devices are on the same port (port
1) of the bridge.
Switched Network
If you plug your analyzer into a switch port and classify the traffic,
you'll see only broadcast and multicast packets and any packets
specifically addressed to your analyzer's board address.
In the following figure, the client and server share a virtual circuit
between them. Their traffic flows from port 1 to port 3 and vice
versa. Their traffic is not sent to any other ports.
Port 1 Port 3
Virtual
Circuit
Server Client
If you plug your analyzer into the switch, you won't be able to
observe the communications because the switch is isolating local
conversations.
Switch
Port 1 Port 3
Virtual
Circuit
Server Client
x You must either connect the switch to the hub's crossover port or use a
crossover cable to connect the 2.
The problem with this solution is that you must move the
hub/analyzer combination around from device to device if you
want to look at multiple devices on the network.
■ Analyzer agents. Analyzer agents are used by distributed
analyzers. These agents are typically software programs loaded
on switches to enable them to capture traffic from all ports and
send the data to a management console.
The agents allow you to manage switched traffic from a central
location. Unfortunately, this type of feature might make the
switch too expensive.
■ Port spanning or mirroring. Port spanning or mirroring
enables you to configure the switch to send a copy of any port's
traffic down another port, specifically the port your analyzer is
connected to.
Port 1 Port 3
Port 7
Routed Network
Router
Hub Hub
WAN Links
How you place an analyzer on the WAN link depends on the WAN
link and the analyzer solution.
Solution 1: Connect the WAN analyzer to the router and CSU using
a Y-cable:
Analyzer
T1 or E1
POD Analyzer
Index
A C
adaptor Setup-30, 3-20, 3-33, 8-7, 8-14–8-15, cable Setup-46, Setup-51, 3-24, 3-31–3-33,
8-28–8-32, 8-34–8-36, 9-2, 11-23 3-76, 3-83, 4-11, 5-21, 8-13, 8-34,
address 2-22, 2-30, 7-21, 7-49, 8-16, 10-15, 9-42–9-43, 10-28, A-3–A-4, C-12,
10-17, 11-33, C-3 C-15
administration 2-5, 2-16 cache 3-58–3-59, 3-67, 3-69
administrator Intro-1 casting off 8-17, 8-25, 9-43
agent 4-30 class Setup-49, Setup-51, Intro-4, 2-54–2-55,
3-1, 3-6, 3-66, 4-4–4-5, 4-7, 4-49,
process status 4-33 4-56, 4-68–4-69, 8-38, 11-13, 12-4
alias 2-19 client 2-25
Apache 1-4, 2-7, 7-1–7-9, 7-15–7-17, cluster Intro-3–Intro-4, Intro-8–Intro-9,
7-20–7-21, 7-25, 7-27–7-28, Intro-11, 1-17, 1-42, 7-22, 8-1–8-3,
7-33–7-35, 7-48, 7-54, 11-2, 8-5, 8-7–8-30, 8-32–8-33,
11-22–11-23, 11-25–11-28, 11-30 8-39–9-49, 10-1–10-12,
asynchronous 6-5 10-14–11-16, 11-18–11-25,
11-27–11-33, 12-3
B cluster-enable 8-7, 9-1, 9-3, 9-26, 10-1,
10-3–10-4, 10-6–10-10, 10-12,
10-15, 10-29, 11-1–11-3, 11-5, 11-9,
background 4-28, 4-34, 4-38–4-40, 4-64, 11-13–11-15, 11-20, 11-22,
4-68, 7-40, 11-11 11-30–11-31
backup 5-24 clustering Setup-5, Setup-46, Setup-52, 1-17,
bandwidth 2-51, 3-3, 5-30, 7-42, 7-47, 7-50, 3-7, 5-11, 5-29, 6-17, 8-1–8-3, 8-5,
A-5, C-4 8-7–8-10, 8-15, 8-21, 8-24, 8-30,
binary 3-64 8-32, 8-34, 8-38, 9-3, 9-5, 9-7,
bindery 1-36 9-11–9-12, 9-14, 9-16, 9-19,
9-25–9-29, 9-31, 9-34, 9-38–9-39,
block 3-29 9-41, 10-2–10-3, 10-10, 10-14,
bootable Setup-16, 5-13 10-19, 10-25, 11-2, 11-4, 11-6, 11-8,
BorderManager 11-2 11-10–11-12, 11-15, 11-17, 12-3
bound Setup-18–Setup-19, 1-6, 1-9–1-10, cluster-naive 11-2–11-3, 11-30
1-49, 2-18, 2-26, 3-9–3-11, 3-22, compatibility Setup-51, 1-66, 3-1
3-71, 7-49, 10-18
D
deactivate 5-11, 5-24
debug 1-9, 2-12, 2-14, 2-17, 2-24–2-25, 2-60,
3-26, 3-42, 3-46, 3-54
decompression 1-8 F
device 8-10, 8-12, 8-27, 9-2, 9-46, C-5, C-8
failback 11-7
DHCP Setup-9, Setup-31,
Setup-36–Setup-39, Setup-49, failover 11-6–11-7
Setup-52, Intro-8–Intro-9, 2-5, 2-30, FAT 5-12–5-13, 5-34
2-53, 8-12, 8-17, 8-24, 9-15, FCS 8-11
11-1–11-2, 11-13–11-20, 11-30,
12-4, C-2 file
diagnose 2-18, 3-24–3-25, 3-28, 3-30, 3-40, system Setup-21, 1-4–1-8, 1-18, 1-27,
3-42, 3-62, 4-11, 4-13, 4-22 1-46, 1-55, 3-29, 3-32, 3-48,
3-53, 3-58, 5-1, 5-3, 5-10,
directory Setup-12–Setup-14, Setup-17, 5-23–5-24, 5-32, 5-35, 6-3,
Setup-23, Setup-26–Setup-28, 6-6–6-7, 6-9, 6-14, 6-17, 7-43,
Setup-33–Setup-34, Setup-42, 10-6, 12-2
Setup-52–Intro-1, 1-4–1-8, 1-21,
1-23, 1-25–1-26, 1-33, 1-40, 1-42, flush 1-41, 5-10, 10-6
1-45, 1-51, 1-57, 1-59–1-63, 1-65,
2-52, 3-10, 3-29, 3-33, 3-64,
3-68–3-69, 3-71, 4-5, 4-28, 4-43, G
5-10, 5-17, 7-5–7-8, 7-14, 7-18,
7-23, 7-27–7-28, 7-40–7-42, generate 2-9–2-11, 2-60, 3-23, 3-28, 3-71,
7-44–7-45, 7-51–7-52, 9-27, 10-10 4-12, 4-21–4-22, 4-42, 5-25, 8-26
dismounting 2-13 global Setup-38–Setup-39, 3-44–3-45, 7-3,
8-27, A-1
DNS Setup-31, Setup-36–Setup-42, Setup-47,
Intro-8–Intro-9, 2-5, 2-17–2-19, unique ID 8-27
2-26, 2-29, 2-36, 7-18–7-21, GUI Setup-17, 1-59, 3-29, 8-36
7-33–7-36 GUID 8-27
Domain
Name Service 1-46 H
driver 5-21
HAM 1-65, 3-24, 5-21
E hardware Setup-1–Setup-3, Setup-46, Intro-1,
1-1–1-2, 1-4, 1-7, 1-18, 1-25, 1-42,
encrypted 7-40, 7-42, 7-53, B-3 2-2, 2-18, 2-49, 2-62, 3-2–3-3,
3-5–3-6, 3-8–3-9, 3-14, 3-18–3-20,
entry 2-31 3-24, 3-33, 3-47, 3-49–3-50, 3-53,
epoch 8-17–8-20, 8-22, 8-25, 8-41, 9-35, 3-56, 3-66, 3-74–3-75, 3-79,
9-38–9-39, 9-42, 9-44 4-15–4-16, 5-12, 6-2–6-3, 6-5–6-6,
export Setup-42 6-22, 8-2, 8-7–8-8, 8-11–8-12,
8-14–8-15, 8-24, 8-30, 8-36, 8-40,
external Setup-2, Setup-44, Setup-46, 9-2–9-3, 9-46, 11-6, 11-23–11-24,
1-39–1-40, 1-59, 1-63, 1-69, 3-27, 12-2–12-3, A-2, A-5, B-6, C-8–C-10
4-23, 4-29, 4-34, 4-64, 4-66, 5-21,
8-7, 8-10, 8-14–8-15, 8-29–8-30, header 3-43
8-32, 9-2, 9-9, 9-14, A-3