Welcome to PowerPath Configuration and Administration.

Copyright © 1996, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012, 2013 EMC Corporation. All Rights
Reserved. EMC believes the information in this publication is accurate as of its publication date. The information is subject to change
without notice.
THE INFORMATION IN THIS PUBLICATION IS PROVIDED “AS IS.” EMC CORPORATION MAKES NO REPRESENTATIONS OR WARRANTIES OF
ANY KIND WITH RESPECT TO THE INFORMATION IN THIS PUBLICATION, AND SPECIFICALLY DISCLAIMS IMPLIED WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

Use, copying, and distribution of any EMC software described in this publication requires an applicable software license.
EMC2, EMC, Data Domain, RSA, EMC Centera, EMC ControlCenter, EMC LifeLine, EMC OnCourse, EMC Proven, EMC Snap, EMC SourceOne,
EMC Storage Administrator, Acartus, Access Logix, AdvantEdge, AlphaStor, ApplicationXtender, ArchiveXtender, Atmos, Authentica,
Authentic Problems, Automated Resource Manager, AutoStart, AutoSwap, AVALONidm, Avamar, Captiva, Catalog Solution, C-Clip, Celerra,
Celerra Replicator, Centera, CenterStage, CentraStar, ClaimPack, ClaimsEditor, CLARiiON, ClientPak, Codebook Correlation Technology,
Common Information Model, Configuration Intelligence, Configuresoft, Connectrix, CopyCross, CopyPoint, Dantz, DatabaseXtender, Direct
Matrix Architecture, DiskXtender, DiskXtender 2000, Document Sciences, Documentum, elnput, E-Lab, EmailXaminer, EmailXtender,
Enginuity, eRoom, Event Explorer, FarPoint, FirstPass, FLARE, FormWare, Geosynchrony, Global File Virtualization, Graphic Visualization,
Greenplum, HighRoad, HomeBase, InfoMover, Infoscape, Infra, InputAccel, InputAccel Express, Invista, Ionix, ISIS, Max Retriever,
MediaStor, MirrorView, Navisphere, NetWorker, nLayers, OnAlert, OpenScale, PixTools, Powerlink, PowerPath, PowerSnap, QuickScan,
Rainfinity, RepliCare, RepliStor, ResourcePak, Retrospect, RSA, the RSA logo, SafeLine, SAN Advisor, SAN Copy, SAN Manager, Smarts,
SnapImage, SnapSure, SnapView, SRDF, StorageScope, SupportMate, SymmAPI, SymmEnabler, Symmetrix, Symmetrix DMX, Symmetrix
VMAX, TimeFinder, UltraFlex, UltraPoint, UltraScale, Unisphere, VMAX, Vblock, Viewlets, Virtual Matrix, Virtual Matrix Architecture, Virtual
Provisioning, VisualSAN, VisualSRM, Voyence, VPLEX, VSAM-Assist, WebXtender, xPression, xPresso, YottaYotta, the EMC logo, and where
information lives, are registered trademarks or trademarks of EMC Corporation in the United States and other countries.

All other trademarks used herein are the property of their respective owners.

© Copyright 2013 EMC Corporation. All rights reserved. Published in the USA.

Revision Date: April, 2013

Revision Number: MR-1WP-PPADMN.5.8_v2.0

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 1
 This course focuses on the details for implementing and managing a SAN environment using
PowerPath. It reviews the commands used to display and manipulate the PowerPath
configuration as well as troubleshooting of problems within a SAN.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 2
 The objectives for this module are shown here. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 3
 The objectives for this lesson are shown here. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 4
 PowerPath has traditionally been viewed only as path management software supporting load
balancing and failover. With the addition of migration and encryption functionality,
PowerPath has expanded its scope of functionality as well as its supported platforms.
PowerPath should now be viewed as a family of technology having two distinct functions.
Two Products support the core function of Path Management and Optimization, while the
other two products perform a Data Protection function:

PowerPath Multipathing, sometimes referred to simply as PowerPath, is the same trusted

product it has always been. The product automates data path management and failover and
recovery, as well as optimizes load balancing to ensure application availability and
performance. PowerPath Multipathing runs on Windows, Solaris, HP-UX, and Linux.
PowerPath/VE automates and optimizes data path pools in virtual environments such as
VMware to ensure business process availability and performance and the ability to scale-out
mission-critical applications.
PowerPath Migration Enabler leverages the same underlying technology as PowerPath
Multipathing to enable data mobility technologies to perform data migrations to and from
any network or supported array type. This includes EMC and qualified non-EMC storage. The
product allows completion of these tasks without application downtime.
PowerPath Encryption, sometimes also called VNX Host Encryption, uses software agents on
hosts and centralized RSA® Data Protection Manager appliances for key management to
secure data from the host to disk on storage.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 5
 The PowerPath filter driver groups all equivalent paths into a volume path set, creates a
volume path set for each multi-ported LUN in the array, and then populates the path set with
all the available paths to the device.
Once PowerPath creates a volume path set, it can use any path in the set to service an I/O
request. If a path fails, PowerPath can redirect an I/O request from that path to any other
viable path in the set. This redirection is transparent to the application. Occasionally, a short
delay occurs due to the low level disk driver trying to recover from an error or waiting for a
request to time out.
To distinguish between paths to the same volume and paths to a different volume,
PowerPath uses a unique volume - or LUN - serial number.
PowerPath performs load-balancing on a path set with more than one active path, based on
the path set’s load-balancing policy. By balancing the I/O load, PowerPath takes full
advantage of all active paths to the LUN. Applications run faster when the I/O is not
constrained to a single path as is the case without PowerPath installed on the host.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 6
 PowerPath cannot redirect I/O to an alternate path until the HBA driver recognizes the path
failure and returns a failure message to the PowerPath driver. Therefore, it is necessary to set
the timeout value in the HBA driver and HBA according to the EMC Support Matrix and
PowerPath documentation. Setting these values incorrectly can increase the length of time
PowerPath requires before failing over an I/O request.
Load-balancing occurs only in one direction; from the host to the storage system. There is no
load-balancing from the storage system to the host.
PowerPath maintains a table of available paths in its host memory space and refers to this
table when determining which path to use.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 7
 Each of the OS platforms that PowerPath supports requires a specific syntax for the device
name. The table lists the valid device naming conventions by operating system.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 8
 This illustration shows an example of multi-pathing on a Symmetrix DMX. In an active/active
storage array, if multiple interfaces exist to a LUN, they all provide equal access to the logical
device. Active-active means all interfaces to a device are active simultaneously.
In a configuration that includes an active-active array, PowerPath can spread the work load
across both paths. In addition, PowerPath can fail over across both paths to the LUN.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 9
 In this example, a LUN is assigned to port 0 and port 1 on storage processor, or SPA. In this
active/passive system, SPA is designated as the primary or active route to the device and all
I/O is directed down the paths through SPA to the device. PowerPath load balances I/O
across these active paths as shown by the green arrows.
The LUN can also be accessed through SPB but only after the device has been reassigned to
Storage Processor B. This path is referred to as a passive path. PowerPath does not send I/O
down the passive paths. These paths are highlighted with red rectangles.
With all active paths to the LUN unavailable, the active paths to an interface card logical
device must be moved to another storage processor. This reassignment is initiated by the
other functioning interface. When PowerPath is installed on the host, the reassignment is
initiated by PowerPath, which instructs the storage system to make the reassignment.
On a CLARiiON array, these reassignments are known as trespassing. Trespassing can take
several seconds to complete; however, I/Os do not fail during this process. After devices are
trespassed, PowerPath detects the changes and seamlessly routes data via the new route.
After a trespass, logical devices can be trespassed back to their original paths.
In order to take advantage of all array interfaces, the devices should be assigned equally
among interface cards. This ensures that the full capacity of the array interfaces are being
used.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 10
 This lesson explains how PowerPath monitors the paths from the HBA to a LUN and the state
and mode in which these paths can be placed. The objectives for this lesson are shown here.
Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 11
 In this illustration, the path from one HBA to the LUNs is highlighted in red. This path is able
to simultaneously access LUN’s A and B because any HBA port to any storage array port
connection can be made through switching in the SAN network.
This being the case, PowerPath monitors the state of the HBA as part of any logical path
through the SAN it can take to any of these LUNs. From the PowerPath driver perspective, an
HBA path to a LUN can be in one of the following states: optimal, degraded, or failed.
In this example, the red-boxed HBA has a valid path to each of the two devices, A and B.
The PowerPath command powermt display reports the state of all logical paths that
originate from an HBA.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 12
 The HBA marked with a red X has been disabled using the powermt disable hba command.
The command disables all paths configured through the specified HBA. These logical paths
are shown in red. When an HBA is disabled, all I/O on the HBA port is stopped, and all
devices configured through this port are closed. These devices are reported as dead in the
powermt display dev command output.
In this example, all logical paths through the HBA to devices A and B are closed (dead) and
are no longer accessible through the disabled HBA. There remains at least one viable path
set through another HBA to devices A and B. As shown by the yellow arrow, PowerPath
now sends I/O to the devices through paths originating through the other HBA port.
The PowerPath command, powermt display hba_mode, reports the mode for a path
that originates from an HBA.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 13
 In this illustration, the port on the storage array marked with a red X has been disabled using
powermt set port disable=true dev=A.
When you disable a storage array port, all logical paths all to devices configured through that
port are marked dead, and I/O is no longer sent to the port. To disable a port, enter the
name of a native-named device configured through the port. To determine which devices
are configured through a particular port, use the powermt display and powermt display dev
command output.
Note: A disabled port state persists across host reboots; however, the complete suppression
of I/O from the host to the disabled array port is not guaranteed during early boot (of the
host).
In this example, the storage port was disabled and device B became unavailable. The logical
path to device A was also marked as dead because it was accessible through the same array
port. Since PowerPath manages all HBA and storage ports and logical paths, devices A and B
can still be accessed through another array port which also has logical paths available
through the SAN to these devices.
The PowerPath command powermt display port_mode reports the mode of all ports
on connected arrays.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 14
 In this illustration, the port on the storage array marked with a red X has been disabled using
powermt set port disable=true dev=A.
When you disable a storage array port, all logical paths all to devices configured through that
port are marked dead, and I/O is no longer sent to the port. To disable a port, enter the
name of a native-named device configured through the port. To determine which devices are
configured through a particular port, use the powermt display and powermt display dev
command output.
Note: A disabled port state persists across host reboots; however, the complete suppression
of I/O from the host to the disabled array port is not guaranteed during early boot (of the
host).
In this example, the storage port was disabled and device B became unavailable. The logical
path to device A was also marked as dead because it was accessible through the same array
port. Since PowerPath manages all HBA and storage ports and logical paths, devices A and B
can still be accessed through another array port which also has logical paths available
through the SAN to these devices.
The PowerPath command powermt display port_mode reports the mode of all ports
on connected arrays.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 15
 PowerPath also manages the state of each path to each LUN independently. From the
PowerPath driver's perspective, a path is alive or dead.
In the example, the logical path to LUN A is in yellow and includes the HBA, cable to the
switch, the switch, the cable from the switch to the array, and device A. The path to LUN B is
not highlighted because PowerPath manages that logical path independently from the path
to LUN A.
The PowerPath powermt display dev command reports the state of the logical paths
to a LUN.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 16
 Here are the possible modes in which a logical path to a LUN can be placed.
When a path is active, load-balancing is performed for a device with more than one active
path, based on the load-balancing and failover policy set for the device. As shown in the
illustration, for active/passive and ALUA-compliant storage systems, I/O is load-balanced
among the ports of the storage processor which owns the LUN.
When a path is in standby mode, the path is held in reserve. Being set to standby does not
mean a path will not be used. It only means the weight of the path is heavily adjusted to
preclude its use in normal operations. A standby path still can be selected if active paths fail
or it is determined that there is extreme degradation in all active paths.
A path that is reported as unlic indicates that unlicensed PowerPath is running for a
Symmetrix, VNX, or CLARiiON storage system. In this scenario, all paths are marked
unlicensed except one path to each SP. Unlicensed paths cannot become candidates for path
failover. Policies are discussed later in this module.
Use the PowerPath powermt set mode command to modify the mode of the path to the
LUN.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 17
 Here are the possible modes in which a logical path to a LUN can be placed.
When a path is active, load-balancing is performed for a device with more than one active
path, based on the load-balancing and failover policy set for the device. As shown in the
illustration, for active/passive and ALUA-compliant storage systems, I/O is load-balanced
among the ports of the storage processor which owns the LUN.
When a path is in standby mode, the path is held in reserve. Being set to standby does not
mean a path will not be used. It only means the weight of the path is heavily adjusted to
preclude its use in normal operations. A standby path still can be selected if active paths fail
or it is determined that there is extreme degradation in all active paths.
A path that is reported as unlic indicates that unlicensed PowerPath is running for a
Symmetrix, VNX, or CLARiiON storage system. In this scenario, all paths are marked
unlicensed except one path to each SP. Unlicensed paths cannot become candidates for path
failover. Policies are discussed later in this module.
Use the PowerPath powermt set mode command to modify the mode of the path to the
LUN.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 18
 These are the key points covered in this module. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 19
 The objectives for this module are shown here. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 20
 The objectives for this lesson are shown here. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 21
 A command line interface (CLI) is provided to manage the PowerPath environment. The base
management commands are listed here. In practice, a command is followed by specific
parameters – also known as arguments - to perform the desired operation. The command
used most often in administering PowerPath is the powermt command. For hosts running a
supported Windows Server OS, administrators have the choice of using the PowerPath CLI or
the PowerPath Administrator GUI.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 22
 Here is a partial list of the powermt commands and their functions.
This list can be displayed by issuing the powermt help command at the OS command line.
Note that powermt config is not available on Windows. This is because for Windows,
PowerPath automatically configures devices at installation and afterward when necessary.
The PowerPath Family CLI and System Messages Reference Guide provides complete
information for all commands, including command line syntax and arguments within each
supported operating system.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 23
 Here is a a partial list of the powermt commands and their functions.
This list can be displayed by issuing the powermt help command at the OS command line.
Note that powermt config is not available on Windows. This is because for Windows,
PowerPath automatically configures devices at installation and afterward when necessary.
The PowerPath Family CLI and System Messages Reference Guide provides complete
information for all commands, including command line syntax and arguments within each
supported operating system.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 24
 This slide lists the access level that is required to run the PowerPath management utilities.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 25
 There are two methods of starting the PowerPath Administrator Console for Windows. In
the first method, you start the console from the Start menu as shown in the example on the
left. The second method is shown on the right. Right-click the PowerPath Monitor Taskbar
icon shown in the ‘systray’. You may also start PowerPath Administrator by double-clicking
the PowerPath Monitor icon.
The PowerPath Monitor continuously monitors the status of the PowerPath configuration
and can be set to alert with visual and audible cues when the status of the PowerPath
configuration changes. Consult the PowerPath Installation and Administration Guide for
Windows for details on how the PowerPath Monitor icon reports configuration status.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 26
 The PowerPath Administrator graphical user interface for Windows Server hosts consists of
two applications, the PowerPath Administrator and PowerPath Monitor.
The PowerPath Administrator runs as a snap-in program within Microsoft Management
Console and is a graphical user interface that is used to configure, monitor, and manage
PowerPath devices.
The illustration shows a sample of the PowerPath Administrator Console. The PowerPath
Administrator has two panes. The scope pane displays PowerPath objects in a hierarchical
list that can be expanded or collapsed. The result pane provides a view of configuration
statistics for PowerPath objects selected in the scope pane. PowerPath Monitor (available on
the taskbar of the Windows desktop) continuously monitors the status of the PowerPath
configuration. PowerPath Monitor can be set to alert when there are operational status
changes.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 27
 PowerPath Administrator icons represent the PowerPath objects in the configuration. The
type of icon reflects the state or status of the particular object selected.
The table lists the PowerPath Administrator icons that are documented in the EMC
PowerPath for Windows Installation and Administration Guide

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 28
 Here is an example of the structure of a typical PowerPath command and argument fields
when using the CLI interface.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 29
 The objectives for this lesson are shown here. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 30
 Use the powermt display command to display information about HBAs and/or devices
configured for and managed by PowerPath. It is available on all storage systems and
operating system platforms that PowerPath supports.
Under the Host Bus Adapters, ### is the number PowerPath assigned to the HBA. The value
is preserved across boots but is not preserved after configuration changes.
The Host Bus Adapters Hardware Path is the host specific hardware address where the HBA
is installed. The values in the HW Path column are platform specific.
This example shows the results of executing the command on a Solaris host. There are two
HBAs in the host. Each HBA has a path to twenty-four LUN’s.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 31
 This example shows the same command output as shown on the previous slide.
The first three lines of output lists the storage_system_type logical device count. This is the
total number of unique logical devices sorted by array type that are configured by PowerPath
and accessible by the host. In the example, eight LUN’s are Symmetrix devices, six are
CLARiiON, and four are HP Storage Works xp.
The I/O paths Summary is the status of the paths originating from this HBA. An optimal
status means all paths are alive (usable). A status of degraded means one or more, but not
all, paths from this HBA are dead (not usable). A status of failed means all paths are dead
and no I/O is passing through this HBA.
The I/O Paths Total is the total number of paths that originate from this HBA. The maximum
number of logical devices supported by one HBA is platform specific. Due to zoning, the total
number of paths may exceed the number of logical devices in a complex SAN topology as
they do here.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 32
 The Stats Dead displays the total number of paths originating from this HBA that are dead.
These paths are dead (not usable) because of a path failure or a change to the configuration.
This value is modified when a path fails or is restored.
The Stats IO/sec displays the average number of I/Os sent across each HBA each second. This
field is blank for powermt display, unless it is used with the every parameter.
Subsequent powermt display iterations show values in the I/O sec column. This
parameter is discussed later in this course.
The Stats QIO’s is the total number of outstanding I/Os on the HBA now.
The Stats Errors is the total number of times any logical I/O paths on this bus transitioned
from alive to dead. This is always equal to or less than the total number of HBA I/O path
errors. It is cleared at system boot time or when powermt restore executes.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 33
 The class=<Array Type> parameter can be added at the end of any powermt display
command to limit the results to only show devices/paths from the array you specify. If the
value is not specified, the class variable defaults to all. This default can be modified by
altering the PP_DEFAULT_STORAGE_SYS environment variable within the OS.
This table lists the codes for each array class.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 34
 Shown is the powermt display dev=1a output for a LUN that resides in a Symmetrix
array. 1a is a shortened version of the full PowerPath pseudo name emcpower1a. The
information that is shown is similar to that for a CLARiiON array but is Symmetrix specific.
The output shows that there is only one path through each HBA to LUN 21.
The load-balancing policy is Symmetrix Optimized which is the default policy for Symmetrix
arrays. These are discussed later in this course. The Storage System ID is the serial number
of the Symmetrix frame.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 35
 The next step is to use the output of a powermt display paths command to
determine the number of paths from the HBA to the array. The example shows the output
for the powermt display paths command on a Windows host.
The command output points out that there is one path from each HBA to the Symmetrix and
two paths from each HBA to the CLARiiON. The output also provides the array interface.
In this example, each HBA has eight logical devices mapped to a fibre adapter (FA) port in the
Symmetrix, and 12 mapped to the CLARiiON, with 6 going to each Service Processor.
Note that the SP ports, such as SP A0, are listed more than once. This means that each HBA
is zoned to two SP ports, SPA0 and SPB0. This is an example of complex zoning. The total
number of disks listed in the powermt display class=clar output is twelve which
equals the total seen by any one HBA.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 36
 The next step is to look at the output of a powermt display ports. The output is shown on
the slide.
The output shows the same output of the powermt display paths except that it is a sum of
the number of devices that have a path through the HBA to the array port. Otherwise, the
port number and storage system ID are already known.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 37
 The objectives for this lesson are shown here. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 38
 Autoprobe identifies failed paths by periodically probing live paths; especially paths that have
not been recently used. Testing helps prevent application I/O from being sent on dead paths
that PowerPath otherwise would not detect as dead. As a result, I/O timeout and retry delays
are reduced. Periodic testing of live paths is a lowpriority task. It does not test all paths
within a specific time, but rather tests all paths within a reasonable time.
This testing scheme minimizes the impact that the testing has on application I/O. PowerPath
does path testing on the paths that have not been tested for at least one hour and on idle
paths that have not been used for I/O within the last minute.
If one probe fails down a path, it kicks off errorchecking routines discussed earlier in the
PowerPath Path Testing after a Failure slide.
Typically, all live idle paths are tested at least hourly, although this is not guaranteed. On an
active system, failed paths are recognized immediately when an application I/O fails. In this
case, path testing is triggered promptly because of the I/O failure.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 39
 If they pass the test, PowerPath autorestore periodically tests dead paths and automatically
restores them to service. Like periodic testing of live paths, periodic autorestore is a low
priority task. It is not designed to restore a path immediately after it is repaired, but rather to
restore the path within a reasonable time after it is repaired. Dead paths are tested when the
path test process is run, provided the paths have not been tested for at least one hour. This
frequency limits the number of I/Os that fail, so the impact on normal operations is
negligible.
Using a combination of autoprobe and autorestore features means that if a path fails, it is
marked failed but the application continues to run. Autorestore then tests the path’s viability
and automatically restores the connection.
The time it takes for all paths to be restored varies greatly. In lightly loaded or small
configurations, paths typically are restored within an hour after they are repaired (on
average, much sooner). In heavily loaded or large configurations, it may take several hours
for all paths to be restored after they are repaired because periodic autorestore is
preempted by higher priority tasks.
The fastest way to restore paths is to restore paths manually using the PowerPath CLI
command, powermt restore.
Use powermt set periodic_autorestore=on/off to enable or disable PowerPath
autorestore. Autorestore is enabled when PowerPath is installed.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 40
 PowerPath periodically runs the path test process. This process sequentially visits each path
and tests it if it meets the following requirements:
Live paths are tested periodically if they have not been tested for at least one hour and are
idle.
Dead paths are tested periodically if they were marked for testing at least one hour ago
and are idle.
Any paths are marked for testing as a result of the conditions listed here.
Tests are spaced out, so at least one path on every HBA and every port is tested at least
hourly. The more paths that need testing, the longer it takes to complete the path test
process. As a result, it is hard to predict exactly when a path will be tested.
When a path state change is detected, it is propagated quickly to all related paths. After all
paths are visited and those marked for testing have completed their tests, the process sleeps
for 10 seconds, and then restarts. This 10 second period is a compromise between using
non-application system resources (CPU and I/O cycles) and keeping the state of paths current
so that the maximum number of paths are always available for use.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 41
 The objectives for this lesson are shown here. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 42
 PowerPath has an automated load-balancing policy configuration mechanism. When the
operating system boots and PowerPath is initialized, device-specific information on the
managed LUNs is read by PowerPath. The optimized load-balancing policy is automatically
set on a per-LUN basis. If multiple arrays are accessible on the SAN and zoned to the
PowerPath host, each LUN that is masked to the sever will be assigned the policy that
provides the best load-balancing algorithm. There are multiple load-balancing policies
intended to support various application requirements. Optimized policies are the default
policies for EMC Symmetrix, and VNX devices. Use of other policies for EMC arrays should be
used only under the direction of EMC Customer Support.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 43
 EMC has developed multiple load-balancing policies intended to support the most common
application requirements in production and test environments. Here is a list of the default
load-balancing policies available when the PowerPath installation is licensed:
CLARiiON Optimized - EMC proprietary algorithm that is designed for VNX and CLARiiON
models. This policy is recommended for all VNX and CLARiiON implementations.
Asynchronous Logical Unit Access (ALUA) and non-ALUA modes are also supported. When
used with ALUA-compliant arrays, this policy recognizes the difference between optimized
and non-optimized paths as specified in the ALUA design. The Symmetrix Optimized and
CLARiiON Optimized policies are the default policies for Symmetrix, and VNX and CLARiiON
arrays.
Symmetrix Optimized - EMC proprietary algorithm that is designed for past and present
Symmetrix models. This policy is recommended for all Symmetrix implementations.
Adaptive - A load-balancing and failover policy for PowerPath devices in which I/O requests
are assigned to paths based on an algorithm that takes into account path load and logical
device priority.
Note that Symmetrix Optimized and CLARiiON Optimized policies should be used by most
EMC customers for Symmetrix, VNX, and CLARiiON arrays, and Adaptive used with VPLEX
arrays. Use of other policies with EMC arrays should be considered only under the direction
of EMC Customer Support. For more information on other PowerPath policies, consult the
EMC PowerPath Product Guide and Release Notes for the PowerPath version installed.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 44
 Here is listed the load-balancing policies that can be assigned for devices belonging to all
supported arrays when PowerPath is licensed.
Least Blocks and Least I/O policy are based on observed I/O load to the device. These
policies are commonly used in large database environments. This policy looks at the number
of blocks going through the path or the size of the data overall going through, regardless of
the number of I/O’s.
Round Robin is useful when all active paths to a LUN will be used at the same time. I/O
requests are assigned to each available path in rotation.
StreamIO uses the same path to the volume that was most recently used for I/O. If the next
I/O exceeds the volume’s threshold value, a new path is selected based on the adaptive
policy algorithm. This policy applies to all arrays that PowerPath supports.
All of these polices include path failover to any logical path that is part of a volume path set
to the device.
For more information on each policy, consult the PowerPath Product Guide and Release
Notes for the version of PowerPath installed.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 45
 Here are special load-balancing and failover policies that can be used on licensed and
unlicensed PowerPath installations. All of these policies are used in special circumstances
because they disable load balance and/or failover capability.
With the No Redirect policy in effect, PowerPath does not load balance or support path
failover. This policy is not available on platforms attached to a VNX or CLARiiON array. It is the
default policy for Symmetrix systems on platforms with a valid PowerPath Base license. It is
also the default policy for all third-party storage systems on platforms without a valid
PowerPath license. No Redirect policy is not recommended for use in production
environments. It can be used as a diagnostic tool.
When the Request policy is used, PowerPath does not load balance but does support path
failover to an alternate path. Request is the default policy for CLARiiON storage systems on
platforms with a valid PowerPath Base license. It is also available on platforms with a valid
PowerPath license that are attached to Symmetrix, CLARiiON, Invista, or supported 3rd party
arrays.
The Basic Failover policy protects against VNX and CLARiiON SP failures, Symmetrix FA port
failures, and backend failures. It allows non-disruptive upgrades to work when running
PowerPath without a license key. It does not protect against HBA failures when load-
balancing is not in effect. I/O routing on failure is limited to one HBA and one port on each
storage system interface. For more information on each policy, consult the PowerPath
Product Guide and Release Notes for the version of PowerPath installed.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 46
 Use the powermt display dev CLI command or the disk view in the PowerPath
Administrator GUI to determine the current load-balancing policy for a device. To change the
load-balancing policy, use the powermt set policy CLI command.
In the example, the powermt set policy CLI command is used to change the load-
balancing policy for device dev=11 from Symmetrix optimize to round robin. A powermt
display dev=11 is run to verify that the load-balancing policy has changed.
Load-balancing policies can be set on a device by device basis or for all devices. One device
or a subset of devices can be using one load-balancing policy and another subset can be
using another. The powermt set policy command also accepts the class parameter to
modify only the devices belonging to an array class.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 47
 Shown here are deprecated powermt CLI command options.
The ability to set the basic failover and noredirect policies has been deprecated in this
release. The powermt set policy=bf or set policy=nr commands work, but
information is removed from the command usage and help files to discourage use. The bf or
nr policy cannot be set from the GUI.
The ability to set the I/O priority for a device has also been deprecated in this release. Users
should not use powermt set priority as this command is no longer supported.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 48
 This example illustrates one of the several methods to change load-balancing policies using
the PowerPath Administrator GUI.
Right-click a disk in the Disks tree and choose Change Policy to in the drop-down menu. A
drop-down menu displays the policies that can be set on the disk Choose a policy from the
list to change the current policy.
In the Disk Properties window on the right, the result of changing the load-balancing policy
is shown. The Disk Properties window is accessed by right-clicking on the disk and choosing
Properties from the drop-down menu.
You may also change the load-balancing policy by choosing the Policy tab in the Disk
Properties window. The window has a box that lists the available polices. Choose a policy
and click Apply to change the policy.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 49
 The objectives for this lesson are shown here. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 50
 Write throttling is enabled to limit the number of queued writes to the common I/O queue in
the HBA driver; instead, the writes are queued in PowerPath. As a result, read requests do
not get delayed behind a large number of write requests. Write throttling is disabled by
default.
powermt set write_throttle_queue sets the write throttling queue depths for a storage
system port connected to a specified device. The queue depth setting limits the number of
writes to all devices enabled for write throttling that can be outstanding (from PowerPath's
perspective) on the storage system port. The queues are allocated within PowerPath, one per
storage system port.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 51
 PowerPath for Windows now has the same capability as other PowerPath host platforms to
set or change the threshold for switching paths when the “si” (streamio) policy is in effect.
Also, two new PowerPath event messages have been added to indicate if a bus has died or
come back to life.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 52
 PowerPath allows you to form a channel group of dedicated paths to a logical device, to
increase application performance. (Note, however, that reserving paths for one application
makes those paths unavailable for other applications, potentially decreasing their
performance.)
Channel groups keep a second set of paths in reserve in case the first set fails. As a form of
manual load balancing, channel groups reserve bandwidth more precisely than automatic
means. Channel groups require at least two paths; they work best in environments with
more than two paths and at least two separately managed applications on the same host
that use different logical devices. You create a channel group by using the powermt set
mode command to label a group of paths to a logical device as active or standby. An
application accessing one or more logical devices designates one group of paths as active
and another group as standby. A second application accessing different logical devices
designates the first group of paths as standby and the second group as active. Each
application has its own dedicated group of active paths, while the overall configuration
provides channel failover protection.
If a path in an application’s active group fails, the application’s I/O is redirected
automatically to another active path in the group. If all paths in the active group fail, the
application’s I/O is redirected automatically to the standby paths.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 53
 PowerPath allows you to form a channel group of dedicated paths to a logical device, to
increase application performance. (Note, however, that reserving paths for one application
makes those paths unavailable for other applications, potentially decreasing their
performance.)
Channel groups keep a second set of paths in reserve in case the first set fails. As a form of
manual load balancing, channel groups reserve bandwidth more precisely than automatic
means. Channel groups require at least two paths; they work best in environments with
more than two paths and at least two separately managed applications on the same host
that use different logical devices.
You create a channel group by using the powermt set mode command to label a group of
paths to a logical device as active or standby. An application accessing one or more logical
devices designates one group of paths as active and another group as standby. A second
application accessing different logical devices designates the first group of paths as standby
and the second group as active. Each application has its own dedicated group of active
paths, while the overall configuration provides channel failover protection.
If a path in an application’s active group fails, the application’s I/O is redirected automatically
to another active path in the group. If all paths in the active group fail, the application’s I/O is
redirected automatically to the standby paths.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 54
 Use the powermt set mode=standby hba=xx dev=xx CLI command to set an
adapter into standby mode. This example shows how to set adapter 2304 into standby
mode. A powermt display dev=11 is used to verify that the adapter is in standby
mode. Use powermt set mode=active hba=xx dev=xx to set the mode back to
active.
By setting an adapter into standby mode, the adapter is not used to satisfy I/O’s unless all
active paths through the available adapters are unavailable. PowerPath automatically uses
the standby path when it is needed to satisfy I/O requests but only when all other resources
are unavailable.
Note that the powermt display output shows no paths in standby mode. The reason is
because you set paths to standby mode at the device level.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 55
 This example illustrates one of several methods to set the path mode to standby using the
PowerPath Administrator GUI. The first step is to click a disk in the scope pane. Choose
Change Mode to and Standby from the drop-down menu. The Disk Properties window
shows the path to the disk is now in standby mode.
You may also set the path to standby mode using the Disk Properties window by choosing
the Mode tab and selecting the radial button next to Standby.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 56
 The objectives for this lesson are shown here. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 57
 PowerPath may be used to disable host HBA and array ports.
Disable HBA Port stops I/O to a port on a HBA and closes all devices configured through the
port. Disable a HBA when you want to replace a HBA. A disabled HBA is not persistent across
host reboots.
Disable array Port enables or disables a storage system port, and closes all devices
configured through the port. When you disable a port, all paths configured through the port
are marked dead, and I/O is not sent to the port. It is useful when I/O needs to be isolated
away from the port in order to perform maintenance on the port hardware. For example,
disabling the appropriate array ports to isolate paths to LUNs accessed through a disk array
controller which is to be replaced.
The disabled port state persists across host reboots; however, the complete suppression of
I/O from the host to the disabled array port is not guaranteed during early boot of the host.
When you disable a port with powermt set port disable in an R1/R2 boot failover
environment, the disabled port does not persist through a failover or failback. The command
may be used with AIX and Solaris platforms that are connected to storage by Fibre Channel,
HP-UX platforms connected by Fibre Channel and iSCSI, and Linux platforms connected by
Fibre Channel and iSCSI, but only where iSCSI is implemented at the host using iSCSI
HBAs.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 58
 The example shows the steps to follow when disabling a HBA. First powermt display
hba_mode is executed to determine the current status of the HBAs and the devices
accessible from the HBA. Next a powermt disable hba=1280 is executed. Finally, a
powermt display hba_mode is re-run to view the results of disabling the HBA.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 59
 The example shows the steps to follow when enabling a HBA. First, a powermt display
is executed to verify the HBA is dead. Notice here that the error count is 0. This is because
the dead paths are due to a user configuration change and are not unexpected. This is a clue
when diagnosing dead paths whether someone has inadvertently disabled the HBA or if it is
truly a dead path.
To enable a HBA, use the powermt enable hba=1280. The result of enabling the HBA
is shown on the bottom of the slide.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 60
 The example shows the steps to follow when disabling an array port. First, the powermt
display command is executed to determine the current status of the paths from the HBAs to
the array port and the devices accessible from the port. Next, the powermt disable array
port, powermt set port_disable=true dev=c2t28d5so is executed. Note that
the device name of one of the devices accessed through the port is specified. This device
name is only used to identify the array port that it uses. The command will disable the port
for all devices, not just this one.
We then see the result of a powermt display port_mode, which shows us all ports accessible
by this host and their modes. In this example, we can see that port FA5bA has been set to
Disabled.
Another thing to keep in mind with this command is that it only disables the array port for all
devices on this host. Any other server accessing this same array port is not affected.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 61
 This example shows the steps to follow when enabling an array port. First, the user must do a
powermt display ports to show that all 3 paths through FA 5bA are dead. Notice here that
there are no errors counted. This is because the paths are dead due to user changes, not
unexpected changes. Use the powermt set port_disable=false dev=c2t28d5s0
to enable the port. You can use this device or any other device that uses this array port. Use
the powermt display port_mode to verify the array port is enabled.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 62
 Hosts running the SUN Solaris 10 operating system may display an error message when using
PowerPath to enable an array port that is already in the disabled state. In this case, you can
use the Solaris ‘cfgadm’ and ‘devfsadm’ commands to help resolve the issue. The ‘cfgadm’
command is used to display as well as configure system devices, and the ‘devfsadm’
command is used to dynamically reconfigure system device tables without having to reboot
the host. Using these commands, along with the PowerPath command shown, will allow you
to change array port status from disabled to enabled. Use them on SPARC or Intel hosts
running Solaris 10 when this error is encountered.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 63
 The objectives for this lesson are shown here. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 64
 Use powermt save to save the PowerPath settings and configuration data in an
ASCIIformatted file. The command forces PowerPath to save all logical device serial numbers,
policy, priority, write throttle setting, write throttle queue depth, and pseudo device names.
Powermt save also records the mode for each configured path and the periodic autorestore
setting for each storage system class. For performance tuning purposes, you can use
powermt save to save different configurations under different filenames.
By default, PowerPath saves the configuration file to the powermt.custom file. (On Windows,
the file is saved in %SystemRoot%\system32\drivers\etc. On Unix systems, the file is saved in
/etc.).
To save the configuration to a userspecific configuration file, use powermt save
file=<filename>. If a filename is not specified, the information is saved in the default file
powermt.custom.
Use powermt load to load a saved configuration file. Executing powermt load without
specifying a file name loads the default powermt.custom file from the default location.
The saved configuration is also loaded when the system is booting. The exception is AIX,
which always loads default settings at boot time. If the default filename is not used, the
configuration does not load automatically at boot time. In this case, you must load the
configuration manually with powermt load.
This example shows how to execute a powermt save and powermt load using both
the CLI and the Administrator GUI.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 65
 Use powermt config to configure all detected logical devices as PowerPath devices and
add these devices to the PowerPath configuration. The command also creates pseudo
devices on operating system platforms that support pseudo devices such as Solaris and AIX.
The powermt config command also configures all detected paths to PowerPath devices and
adds these paths to the PowerPath runtime configuration. Powermt config also configures all
detected third-party storage system logical devices as PowerPath devices if their storage
system classes are set to ‘managed’.
The command adds paths to logical devices based on the storage system frame serial
number and the logical device serial number. Together, these values uniquely identify a
logical device. By default, powermt config adds devices under PowerPath’s control with the
Symmetrix optimization, CLARiiON optimization, or Adaptive load-balancing and failover
policy, a priority of 0, write throttling set to off, and a write throttle queue depth of 256.
Powermt config does not remove previously configured paths when they become dead
paths. Use powermt remove to remove these paths.
The powermt config command is not available with PowerPath for Windows platforms, as
Windows does this automatically.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 66
 This slide lists the steps needed to add a new device to the PowerPath configuration. The
procedure applies to managed storage system classes only. It does not apply to any storage
system class that has been excluded from PowerPath control using the powermt unmanage
command. Also note that the procedure is an overview and therefore does not include the
fine points. Consult the PowerPath Product Guide for detailed information on each step.
Also, the procedure to use a new logical device is platform-specific and may require service
interruption.
First, use the PowerPath CLI or Administrator GUI to view the existing environment. Verify
that all devices and paths are correctly configured to the host. Next, make the devices visible
to the host. This step could entail a number of steps as shown on the slide. Verify the devices
are correctly seen by the host before continuing with the next step.
Then use powermt config to reconfigure the PowerPath environment and add the new
devices. When the reconfiguration is complete, use the PowerPath CLI or GUI to verify the
devices have been added correctly. The next step is to adjust device-specific values such as
load-balancing policy and device priority.
Always save the PowerPath configuration using the powermt save command. This
ensures that PowerPath always loads the correct configuration.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 67
 Shown here is an example of using powermt config to add a device to the PowerPath
configuration. The powermt display output provides a view of the configuration prior to
adding the new device. Next, a new device is added using an array management tool. The
next step is to execute a powermt config to add the device to the PowerPath
configuration. The subsequent powermt display output now includes the new
device(s).
Before using powermt config to add a device, it must be visible to the operating system.
The example screenshots were shot on a Solaris host. When making devices available to
Solaris, verify that the sd.conf file includes an entry for the target and LUN. Next, run the
operating system commands to configure the device. In the example, drvconfig and disks
commands were used.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 68
 In this example, the before and after view of adding a device using the PowerPath
Administrator GUI is shown. On Windows, there is no need to use powermt config because
PowerPath automatically configures new devices. Prior to addition to the PowerPath
configuration, the device must be seen in Windows device manager.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 69
 These are the general steps that should be followed when removing a path from the
PowerPath configuration.
Use the PowerPath CLI or Administrator GUI to view the existing environment. Next, use the
powermt display command sets to identify the paths that you wish to remove. In other
words, determine the HBA, zone, etc, that you want to remove. The output of this step is the
pathname for the path that is to be removed.
Use the powermt set mode command and specify the pathname found in the previous
step to set the path mode to standby. This is optional, but a good best practice as it ensures
that the paths are not in use when you unplug cables or HBAs. The next step is to remove the
components of the path whether it is a zone, or cable, etc. Next, run the powermt
restore command to verify that the path you removed in the previous step is dead.
The next step is to run the powermt check command to remove the dead paths from the
configuration file. When devices and adapters are added or removed, the configuration file
may not reflect these recent changes. Running the powermt check command synchronizes
the configuration file with the running configuration. If a path is marked dead or the serial
numbers encoded in the path configuration information do not match the serial numbers on
the logical device, powermt check prompts the user to remove the path.
Next, use the operating system specific commands to remove the paths and all devices down
the path. This step includes removing the operating system device file and the PowerPath
device that corresponds to the device(s) being removed. Always remove the PowerPath
device file first. Save the configuration changes and then reboot the system for these to take
effect. When the reboot is finished, check the new PowerPath configuration.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 70
 Shown here is an example of the powermt check command discussed in the previous
slide. Notice that there are 27 total and one dead device through HBA 2304 on the first
powermt display. Then we see only 26 devices after running the check command.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 71
 These are the key points covered in this module. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 72
 The objectives for this module are shown here. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 73
 The objectives for this lesson are shown here. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 74
 This slide lists the steps a PowerPath Administrator can follow to view current path I/O
performance and use the information to tune the PowerPath environment.
Start by viewing performance at HBA level by using the powermt display every=xx
command. The value after the every parameter specifies how many seconds between the
time the command is run. Any integer between 1 and 86400 is accepted as an argument for
the every parameter.
Use the information in the report to determine if the I/O is unbalanced across HBAs.
When unbalanced I/O is found, the administrator should use the powermt command with
the dev option to drill down and investigate the devices that are part of the path from the
HBA. First, look for dead path(s). If dead paths are present, the alive paths are handling more
of a share of the I/O. Check the load-balancing policy and the device priority. Try adjusting
the load-balancing and device priority values and check the results using powermt
display. Load-balancing and device priority settings can usually be changed during
operation without interrupting the application.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 75
 The every=n parameter can be added at the end of any powermt display command to
reprint the results of the display command every n seconds. When using this feature, keep in
mind that interface delay time must be taken into account. For example, if you are connected
through a hyper terminal and the screen is printing slowly, you are better off setting n to a
higher value to avoid output that is constantly scrolling.
Here we see output from the command after five seconds.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 76
 This example shows the output of the powermt display every=n command.
Note that the first time the report runs the I/O per sec field reports “”.
In the example, the number of I/O’s per second and the QIO’s column is shown in the red
box. The IO/Sec column displays the average number of I/Os sent across this bus each
second.
The Stats QIOs column reports the total number of outstanding I/Os on this HBA now.
In the example, there are no dead paths to the devices. So, you can discount a performance
issue because of dead paths to devices.
There are a small number of QIO’s, four on each path, listed in the report. These outstanding
I/O’s may point to a problem where the paths are at maximum utilization and tuning may be
needed. In order to tune the paths, the powermt display dev command is used to
view load-balancing and device priority settings for the disks attached to the system.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 77
 This illustration shows the I/O per second HBA performance view using the Administrator
GUI.
The IOs/sec and QIOs columns are boxed in red.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 78
 In this example, a portion of the powermt display dev=all every=1 output is
shown. There is a sample output for a Symmetrix and CLARiiON array.
When you specify the device option of the powermt display command, the IO/Sec column is
removed. However, the QIOs field is available to use as a gauge on how the paths to the
device are utilized.
In the Symmetrix portion of the output, there is a queued I/O on one path to the device
suggesting that there is a high amount of traffic being sent down that path. This device may
be a candidate for tuning by adjusting the load-balancing policy for the device. You may also
see what effect of changing the device priority setting has on device performance.
The CLARiiON device has zero QIO’s on all paths. There is probably no need to tune these
devices.
This command can be used on a single device as well to limit the amount of data shown on
each repeat of the every parameter. Simply replace dev=all with dev=n where n is the
pseudo device you want to look at.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 79
 When an underperforming device is found, PowerPath can be used to adjust load-balancing
and device priorities to improve performance. Information on how to modify load-balancing
and device priority settings using the PowerPath CLI and Administrator GUI is covered later in
this course.
It is important to understand the application I/O profile before changing load-balancing or
device priority values. For example, some applications perform sequential I/O while others
perform random I/O.
When adjusting device priorities, the key is to know in which devices the application data
resides. It is equally important to determine which devices contain the data the application
needs efficient access to. An example is a database log device. The idea is not to set a higher
device priority on any of the other devices that share the path to the database log device.
Doing so improves performance to the wrong devices at the expense of the database log
device. A better solution is to raise the priority of the database log device.
Also listed are other methods of improving I/O performance to a disk. The list is not
complete but provides a good starting point.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 80
 Shown here are a few of guidelines to keep in mind when modifying load-balancing policies.
In most cases, the default load-balancing policies for the array class delivers the best
performance. They are optimized to use with devices for the array types they support. These
policies are Symmetrix optimize (so), CLARiiON optimize (co), and adaptive (ad). Others, such as
round robin, are great for balancing I/O across all paths. Least blocks and least I/O policies are
best for applications that perform sequential I/O or block I/O’s. The basic failover and request
polices should not be used on a fully licensed PowerPath environment. The reason is that these
policies enable only failover functionality.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 81
 The objectives for this lesson are shown here. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 82
 Starting with version 5.0, PowerPath monitors the time it takes I/O requests to complete on
all paths in your environment. When you enable or disable path latency monitoring, this
setting applies globally to all paths.
As shown here, the powermt display options command reports whether path
latency monitoring is enabled, and the path latency threshold, if one is set.
You may monitor the latency on all paths to your environment by first enabling the path
latency monitor. The next step is to set a path latency threshold to set a time interval in
seconds within which I/Os should complete. Each customer environment is different and
therefore different thresholds may be used.
Use powermt display latency to display the latency information including the last and
longest time required to deliver an I/O. The diagnostic information reported by path
monitoring allows you to determine normal I/O times for your environment, which can help
to determine an appropriate path latency threshold.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 83
 PowerPath 5.0 includes a new I/O latency monitoring capability on UNIX and Linux platforms
that PowerPath supports. Path latency enables an administrator to set time threshold in
seconds when a host I/O should complete. To use the path monitoring facility, it must be
turned on using a PowerPath CLI command. Administrators can use another PowerPath CLI
command to monitor the actual current and maximum latency and any incident where I/O
latency exceeds the threshold logged to the system log.
Path latency monitoring is a good tool to use when observing I/O performance on SAN’s.
Some SAN configurations that may benefit from I/O latency scrutiny are listed on this slide.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 84
 Use the powermt display options command to check whether path latency
monitoring is turned on or off. This example illustrates the command usage and
representative output of a system that has been turned off. The default setting for Path
Latency Monitoring is on.
In order to turn on or enable path monitoring, use the command powermt set
path_latency_monitor=on. The next step is to set the threshold using powermt
set path_latency_threshold=n. In this example, the latency threshold is set to
three seconds. Then use the powermt display options output to view the path
latency state and latency threshold to ensure they were entered correctly.
These commands are often scripted to temporarily change the thresholds to higher amounts
while a batch job is running or for a number of other reasons.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 85
 In this example, the administrator is monitoring the threshold using powermt display
latency. The results show the current and maximum latency for all devices.
To monitor latency over a period of time you can use the ‘every’ option of the display
command. For example, powermt display latency every=5.
The powermt display latency output includes the following informational columns:
The Current column reports the time in microseconds it took the last I/O on the path to
complete.
The Max column reports the time in microseconds it took the longest I/O on the path to
complete. It is noted on the slide how to do microsecond conversion.
You can use the Current and Max values to determine the appropriate threshold value for
your environment.
In this example, the Max column for the Symmetrix devices are well above the current
values, suggesting that the current latency is low and the maximum latency is roughly .6
seconds. In the previous slide, we set the threshold to three seconds, so we are well within
limits and don’t need to increase our threshold. Also, note that the latency on the path
through FA 15cA is larger than the other ports. This may point to an issue where the path to
the device is longer, has more hops, or uses an older slower switch port.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 86
 For more detailed information, you can specify a particular device or all devices when
monitoring Path Latency. In this example, the output of powermt display latency
dev=11 is shown from a different system than we were looking at before. We can see here
that there is very high latency through FA 15cA and FA 2cA. FA 2cA has hit 21.4 seconds. This
is a point where the user would do some troubleshooting to determine where the bottleneck
is.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 87
 The objectives for this lesson are shown here. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 88
 PowerPath logs error messages to the system log. Netware messages are logged to the
Logger screen and to the System Console.
Each error message includes a platform-specific prefix referencing PowerPath. Note that
Windows systems do not use a prefix.
The error message also includes the message type and a character string indicating the type
of array to which the message applies. Message types are Panic, Warning, Error, Info, or
Debug.
Please refer to the PowerPath CLI and System Messages Reference manual for information
about the error messages for
Each PowerPath family product.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 89
 This example is the output of a powermt display command on a Solaris machine showing
that all paths to devices from HBA 2304 have failed. Notice that all 42 devices are dead and
there are 42 errors. The path to the devices from the other HBA is still available as shown by
the optimal status.
On the bottom of the slide is a portion of the Solaris messages file (/var/adm/messages)
showing the messages that PowerPath wrote to the file when the path to the devices was
lost. We can see from this output that the problem occurred at 11:05:37 on Jan 7th. We can
derive that all paths through HBA 2304 have died to the Symmetrix with the serial number
ending in 308. This error log will detail all of the devices affected.
We can get the same information from a powermt display dev=all, but the error log is
helpful for Administrators. In most cases, an Administrator would have a tool that monitors
the error log and sends a notification when a path is dropped. This way the administrator can
take action to ensure that no data is lost and no applications go offline due to path failures.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 90
 This slide is an example of a log entry showing that the latency threshold has been exceeded.
This is good as an early warning system for your environment. If thresholds are being
routinely exceeded, then an administrator can be notified, troubleshoot the paths, and take
steps to avoid a future failure.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 91
 The objectives for this lesson are shown here. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 92
 Use powermt set periodic_autorestore=on to enable PowerPath autorestore
functionality. When autorestore is enabled, PowerPath periodically tests dead paths. If the
paths pass the test, PowerPath restores them to service. PowerPath then can use the
restored path for load-balancing and failover. Even when periodic autorestore is off, path
testing continues to be done on all paths to the devices, not just the paths that are dead.
When autorestore is off, PowerPath does not restore the path automatically.
Autorestore is enabled when PowerPath is installed.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 93
 The process of finding and resolving a path to a disk failure is shown on this slide. Use the
powermt display CLI command to determine if there are any dead paths to devices.
The output for step 1 shows that there is a single dead path on hardware path 2305. Note
that the I/O path summary is degraded. Also, the dead column and error column has a value
of 1.
In step 2, the powermt display dev=all command is run to determine which device
is having a problem. In the example, it is Symmetrix device 11 as shown. We see that it has a
dead path through HBA 2304.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 94
 In Step 3, the user determines the cause of the failure.
The administrator must determine the reason and follow the appropriate procedure to
correct the problem. If the problem is due to a configuration change, then reconfigure
PowerPath to make the change permanent. If the problem is due to a hardware problem,
then fix the problem and restore manually or wait for the autorestore process to restore the
path.
Step 4 can be done automatically by the autorestore feature, or manually if autorestore is
disabled or time is a factor. To manually restore the path, run the powermt restore
command. This command will scan through all devices and retest any dead paths to see if
they have been repaired. If they have been repaired, then PowerPath restores them to an
alive state.
If there are many devices visible to your host, you have the option of running powermt
restore hba=2305 dev=11. This only touches the device specified and does not scan for dead
devices. This example shows the path to dev 11 though HBA 2305. This saves computing
time on the host in large environments.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 95
 Shown here is a screenshot of the PowerPath Administrator GUI showing the dead paths to
device 14.
Notice in the scope pane that device 14 has a red X. The X indicates that all paths to the
device are dead. If the device is selected in the scope pane, the results pane also reports all
the paths to the device as dead.
Also, note that the adapters and array in the scope pane have a red slash (/). The slash states
that one or more, but not all, paths to the storage array and adapters have failed or are in a
degraded state.
Note that the Port Management, EMC PowerPath, and the PowerPath Administrator icon in
the taskbar also have a red slash indicating that one or more paths are in a degraded state.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 96
 This example shows how to restore a path using the PowerPath Administrator GUI. To restore
a path, right-click the EMC PowerPathAdmin node in the scope panel. Next, choose All Tasks
from the drop-down window. Choose Restore All Devices to restore the path.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 97
 PowerPath installation creates two log files by default. The first is the installation log, like the
one in this example. The second is the post-installation check log which we have seen
previously. These logs let the user know whether or not the install was successful and can be
invaluable when troubleshooting.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 98
 In addition to logs that can be used for troubleshooting, there is a VBScript available that can
get all pertinent information about the host and put it into an XML file. The name of this
script is PP_HealthCheck_Tool.vbs. This tool can be used to detect install/upgrade failures
and capture the state of the host at that instance.
This script is not packaged with PowerPath, but will be available through EMC Customer
Support. The VBScript generates an easily readable XML file named PowerPath.xml found
under C:\. This report can be attached along with other EMC grab reports to get to the root
cause of an issue faster.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 99
 The objectives for this lesson are shown here. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 100
 PowerPath Configuration Checker (PPCC) is a software program that verifies that a host is
configured to the hardware and software required for PowerPath multipathing features
(failover and load-balancing functions, licensing, and policies) as specified in the EMC
Support Matrix and in the EMC Simple Support Matrix.
PPCC can assist the PowerPath user in the successful deployment of the PowerPath solution
prior to and after a PowerPath installation. It also provides efficiency in that it allows the
host administrators to independently effect their own planning and installation on hosts
where PowerPath is not yet installed. It also allows administrators to more easily prepare for
upgrading an existing installation and engage in troubleshooting. For example, PPCC
supports administration after configuration changes are made on a host that includes
PowerPath, such as hardware changes or replacements or the installation or de-installation
of software.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 101
 PPCC is intended to support the planning and upgrading of a PowerPath host environment. It
can also serve as a diagnostic tool when host system configuration changes affect PowerPath
operation. Using PPCC is an easy and efficient method to identify what changes should be
made to the host’s PowerPath configuration to ensure continued support for failover and
load balancing.
PPCC can assist the administrator in three major areas that can cause a change in the host-
to-PowerPath relationship:
Planning -This task applies to a host on which PowerPath has never been installed or is not
currently installed. PPCC can identify the software that needs to be installed to support a
specific version of PowerPath.
Upgrade -This task applies to a host on which some version of PowerPath is installed. An
upgrade (or downgrade) to a different version is required. PPCC can identify components of a
configuration that need to change when a different version of PowerPath is to be installed.
Diagnostic -This task applies to a host on which some version of PowerPath is installed or on
which configuration changes have been made to PowerPath, to the host OS, and/or to other
software on the host. Also, if PowerPath does not appear to be operating correctly, running
EMC Reports along with using PPCC can assist with configuration problem analysis. Please
see the PowerPath configuration Checker User Guide for more information.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 102
 PPCC can be downloaded from the EMC Online Support website. It is packaged as a ZIP file
containing the complete PPCC product. You can uncompress and run the installer from any
folder of a system running Windows 7, Windows 2003, Windows 2008, or Windows XP. PPCC
writes its output report and log files to the installation folder by default.
PPCC works by analyzing data collected by EMC Reports for Windows or the latest Grab
version for VMware vSphere and UNIX systems. EMC Reports and EMC Grab are widely used
diagnostic utilities that collect host configuration data, system logs, and other host
information used for EMC software problem analysis and resolution. PPCC verifies data in the
EMC Reports or Grab output file against an internal database of valid PowerPath
configurations. It is because PPCC analyzes the input file using its internal configuration
database that the user must be sure to use the latest version of PPCC available.
In this example, the EMC Grab utility was run on a Solaris host that does not yet have
PowerPath installed. The output file was then copied to a Windows host with PPCC installed
and PPCC was run against the report file. Note that the PPCC command has the “-V 5.5.0”
parameter added to the end of the command line. When PowerPath is not currently installed
on the host, PPCC must be informed of the PowerPath version anticipated to be installed
after using PPCC.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 103
 After analysis, PPCC generates an HTML results file and automatically opens a web browser
to display it. The information is presented in the form of a configuration report that validates
the host environment and warns about potential configuration problems. With this
information, the administrator can then install or update any missing or outdated software as
required. EMC Reports or Grab should be run and analyzed by PPCC again after each
remedial action has been taken.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 104
 Here is a partial list of the conditions that PPCC checks.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 105
 These are the key points covered in this module. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 106
 The objectives for this module are shown here. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 107
 These are the objective for this lesson. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 108
 This example shows an Open Systems HA cluster with PowerPath installed. Both nodes of the
cluster have shared access to the same devices in the storage array. To prevent data
corruption, device access is managed by the cluster software. The example shows a typical
PowerPath/cluster implementation in which each node of the cluster has multiple paths to
the devices that are shared by the cluster.
Open Systems clustering technology manages application availability by detecting failures. It
restarts high availability applications on a surviving cluster node. The deployment of
PowerPath in the cluster eliminates the application downtime due to a failure of one path to
the devices. PowerPath detects the path failure and uses alternate paths to deliver the I/O
rather than failing over the cluster to the standby node. Cluster failover is not desired
because it takes time to perform the failover and restart the application. The result is
application outage until the failover and restart operation is complete. With PowerPath
installed, cluster failover is limited to situations where the primary node goes down or all
paths to the devices are lost.
Another advantage of using PowerPath in a clustered environment is that its load-balancing
ability can help the customer maximize performance and get the greatest value from their
cluster investment. PowerPath balances the I/O across all paths to the devices from the
primary node of the cluster.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 109
 This example shows an Open Systems HA cluster with PowerPath installed. Both nodes of the
cluster have shared access to the same devices in the storage array. To prevent data
corruption, device access is managed by the cluster software. The example shows a typical
PowerPath/cluster implementation in which each node of the cluster has multiple paths to
the devices that are shared by the cluster.
Open Systems clustering technology manages application availability by detecting failures. It
restarts high availability applications on a surviving cluster node. The deployment of
PowerPath in the cluster eliminates the application downtime due to a failure of one path to
the devices. PowerPath detects the path failure and uses alternate paths to deliver the I/O
rather than failing over the cluster to the standby node. Cluster failover is not desired
because it takes time to perform the failover and restart the application. The result is
application outage until the failover and restart operation is complete. With PowerPath
installed, cluster failover is limited to situations where the primary node goes down or all
paths to the devices are lost.
Another advantage of using PowerPath in a clustered environment is that its load-balancing
ability can help the customer maximize performance and get the greatest value from their
cluster investment. PowerPath balances the I/O across all paths to the devices from the
primary node of the cluster.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 110
 Dependent write operations preserve the integrity of data that is being written to disk,
ensuring that dependent writes are applied in the intended sequence of the application. The
typical sequence of write operations for a database update transaction proceeds in this way:
A write operation is sent to the database transaction log indicating that a database data
update is about to take place. A second write operation follows which updates the database
itself. A third write operation then updates the database log indicating that the database
update has completed successfully. With this technique, the database ensures correct
ordering of these writes by waiting for each step to complete before starting the next.
With EMC TimeFinder, the Enginuity Consistency Assist feature can be used to perform
consistent splits between source and target device pairs. Consistent split helps to avoid
inconsistencies and restart problems that can occur if database-related devices are split
without first quiescing the database. When using consistent split on a group of devices, the
database writes are held at the storage level momentarily while the foreground split occurs,
maintaining dependent-write order consistency on the target devices comprising the group.
SRDF/Consistency Group (SRDF/CG) is an SRDF feature designed to ensure the dependent-
write consistency of the data distributed across multiple devices. The basic building block of
the SRDF/CG product are consistency groups. A consistency group is a set of SRDF devices
that may reside on multiple Symmetrix systems and are enabled for database consistency.
SRDF devices that belong to the same consistency group act in unison to preserve
dependent-write consistency of a database distributed across multiple devices within the
consistency group.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 111
 When databases become very large (larger than one terabyte), the time and resources
required for host-based software to protect or run decision support queries on these
databases becomes a critical factor. The time required to quiesce or shutdown the database
for offline operations is no longer acceptable for most database applications. When you need
to perform such database operations, the use of TimeFinder or SRDF consistent split allows
you to split off a consistent, DBMS-restartable copy of the database in seconds with no
interruption to online service.
This is done by Either ECA or PowerPath knowing when a split will happen and holding I/O
from the host until the split is completed. Doing this ensures that the copy of the data being
split off is sequential and restartable, and thereby a “gold copy.”
PowerPath suspends I/O to the devices at the host level. Split is then executed, and
PowerPath resumes I/O to the original devices.
Enginuity Consistency Assist (ECA) will hold I/O at the Symmetrix level within cache, which
eliminates the need for PowerPath. Consistency is maintained across multiple servers.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 112
 When databases become very large (larger than one terabyte), the time and resources
required for host-based software to protect or run decision support queries on these
databases becomes a critical factor. The time required to quiesce or shutdown the database
for offline operations is no longer acceptable for most database applications. When you need
to perform such database operations, the use of TimeFinder or SRDF consistent split allows
you to split off a consistent, DBMS-restartable copy of the database in seconds with no
interruption to online service.
This is done by Either ECA or PowerPath knowing when a split will happen and holding I/O
from the host until the split is completed. Doing this ensures that the copy of the data being
split off is sequential and restartable and thereby a “gold copy.”
PowerPath suspends I/O to the devices at the host level. Split is then executed, and
PowerPath resumes I/O to the original devices.
Enginuity Consistency Assist (ECA) will hold I/O at the Symmetrix level within cache, which
eliminates the need for PowerPath. Consistency is maintained across multiple servers.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 113
 EMC Solutions Enabler is host-based storage management software that provides an API-
shared library and a special command set that interfaces with the SYMCLI. Using command-
line entries as well as scripts, it interfaces with devices and data objects in the Symmetrix
array complex to perform control operations on them. For Solutions Enabler versions prior to
version 6.1, consistency protection for SRDF devices required PowerPath support. Creating
an RDF type composite group and specifying it for PowerPath places the group in the host’s
PowerPath database, and gives the group the capability of being enabled for consistency
protection.
PowerPath-enabled RDF consistency group operations do not currently support concurrent
RDF devices.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 114
 The objectives for this lesson are shown here, please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 115
 Systems Management Server (SMS) is software created by Microsoft to manage Windows
Systems. This software provides management tasks such as remote control, patch
management, software distribution, operating system deployment, and hardware and
software inventory. It can also be used to manage PowerPath hosts.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 116
 These are the high level steps for configuring SMS with PowerPath. Please read the
Microsoft SMS product documentation for detailed instructions on configuration steps.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 117
 Microsoft Operations Manager (MOM) and System Center Operations Manager (SCOM) are
cross-platform data center management systems for Windows Server and UNIX-based
operating systems. Each has a single GUI that shows state, health and performance
information of monitored host systems. It also provides alerts generated according to
selected availability, performance, configuration and security criterion. They both are based
on a server/agent model.
The PowerPath Management Pack interfaces with MOM and SCOM to monitor the
installation and running operation of PowerPath on hosts across the network. The
management pack defines the services and components to be monitored and also monitors
PowerPath-managed paths and storage devices, and triggers alerts and SNMP traps based on
pre-defined criteria. Apart from monitoring the PowerPath and the Operations Manager
alerts, the management pack also monitors connectivity. When a path or volume is
unavailable for a specified period, the management pack receives SNMP traps informing it of
the disconnect. The Operations Manager alerts are viewed from the System Center console
and the SNMP traps are viewed from a site SNMP manager.
Note that in January 2010, Microsoft ended mainstream support for MOM 2005 and as such,
SCOM 2007 and 2012 replace MOM 2005.
For more information, please see the EMC PowerPath Management Pack for Microsoft
Operations Manager document for the version of PowerPath you are using.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 118
 SCOM and MOM communicate with agent software that is installed on the PowerPath hosts.
Agents on the hosts watch the Windows System Event log for PowerPath events related to
dead paths and volumes. Based on a predefined criterion, the agent generates alerts and
SNMP traps with the Path/Volume information. The alerts are then sent to the SCOM or
MOM management server. These alerts can be viewed through the PowerPath Alerts
Window.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 119
 The PowerPath Management Pack for Windows is a package designed and developed to help
system administrators monitor PowerPath installations. The management pack is separate
and is not included with PowerPath. It is available on PowerLink for download. It includes
PowerPath event monitoring features and rules. There are two different packages available.
One is for MOM 2005 and the other is for SCOM 2007.
The 2005 management pack is not compatible with a MOM 2005 SP1 database. The
packages must be imported into the MOM or SCOM server. Once imported, it pushes the
PowerPath intelligence into the managed host. It is then able to monitor PowerPath devices
that reside on managed hosts. The MOM agent monitors the PowerPath dead
paths/volumes and generates MOM alerts and SNMP traps when the paths/volumes do not
come up within a configurable time period. The default time period is one minute for a dead
path and five minutes for a dead volume. The time period can be altered by editing a config
file named pp_delay.config.
There are four different types of PowerPath event rules. These rules are listed here. Rules
can be disabled so that the events associated with them are not monitored. PowerPath
events being monitored are captured in the Events View. The events can be sorted by time.
MOM and SCOM can be configured to send out emails to the system administrator when
events occur. The OMG2007_OpsGuide.doc from Microsoft provides more details how this
action is configured.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 120
 The hosts to be managed must first be discovered. For SCOM 2007, the Discovery Wizard
can discover the hosts in the environment provided that an account with administrative
privileges is given to the Wizard. This account is used to scan and install agents on the
computers to be managed. Once the agent software is installed, hosts will become managed
objects within SCOM. MOM 2005 is slightly different because discovery rules must first be
created and run to discover the hosts to be managed. Once the hosts have been discovered,
agent software can be installed by right-clicking the unmanaged host and selecting Install
Agent. After the agent’s software is installed, the EMC PowerPath Management Pack can be
imported into SCOM or MOM for use with the PowerPath hosts. The agent/server
communication channel uses the TCP/IP protocol. The communication channel between the
agent and the server is secure, authenticated, and encrypted.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 121
 The PowerPath Management Pack is imported by right clicking Management Packs and
selecting Import Management Packs. A window will appear for the user to enter the path to
the management pack. The Management Pack can be selected by browsing to the file
(SCOM) or directory (MOM) and clicking open. Once the management pack is imported, it
will appear under the Management Packs window. This is a screen showing how to import
the management pack into SCOM 2007. The procedure for an installing agent software using
MOM 2005 is slightly different.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 122
 The SNMP service must be installed and running on the monitored Windows hosts. It is not
necessary to install SNMP on the MOM or SCOM server. SNMP should only be installed on
the MOM and SCOM server to receive SNMP traps from the monitored hosts. SNMP is
installed using the Add/Remove programs section of Windows.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 123
 The SNMP configuration utility is available with the PowerPath Management Pack. It is
installed on the hosts managed by both MOM 2005 and SCOM 2007. In order to install the
SNMP Configuration utility, both the SNMP service and the SNMPTRAP service must be
running on the managed host. This is the same host that PowerPath is installed on.
The name of the configuration utility is:
EMCPower.SNMP.Configuration.platform.5.3.Build#.exe. Upon execution, it locates the
folder that PowerPath is installed in. It then creates a folder named MOM at the same level
of the PowerPath folder. It then creates a subfolder named SNMPExtension and deploys the
PPSNMPExtension.dll, PPSNMPTrap.exe, and pp_delay.config file. The pp_delay.config file
dictates the behavior of the management packs. It contains the acceptable delay (in
seconds) for the management packs to wait for a dead path or volume to come up before
sending out a MOM alert or SNMP trap. The configuration adds in SNMP trap generation for
PowerPath dead path/volume events.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 124
 The EMC PowerPath license and version can be retrieved from monitored PowerPath hosts
within both SCOM 2007 and MOM 2005. The EMC PowerPath Management Pack must be
deployed in both cases to utilize the PowerPath tasks.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 125
 These are the key points covered in this module. Please take a moment to review them.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 126
 These are the key points covered in this course. This concludes the training.

Copyright © 2013 EMC Corporation. Do not Copy All Rights Reserved. PowerPath Configuration and Administration 127