RELEASE NOTES

EMC® NetWorker®
Release 8.1

Release Notes
P/N 302-000-559
REV 03

October 3, 2013

These release notes contain supplemental information about the EMC NetWorker 8.1
release.

Topics include:
◆ Revision history ........................................................................................................ 2
◆ Product description................................................................................................... 3
◆ New features and changes ........................................................................................ 3
◆ Fixed problems ....................................................................................................... 14
◆ Environment and system requirements ................................................................... 15
◆ Known problems and limitations ............................................................................. 18
◆ Internationalization support.................................................................................. 175
◆ Localization support ............................................................................................. 175
◆ Technical notes ..................................................................................................... 182
◆ Documentation ..................................................................................................... 188
◆ Installation ........................................................................................................... 190
◆ Troubleshooting and getting help.......................................................................... 193

The installed NetWorker version displays as a four digit number. For example, the initial
release of NetWorker 8.1 displays as NetWorker 8.1.0.0 when installed.



All new or updated software downloads and product-specific technical documentation are
available on the EMC Online Support site at support.emc.com. From the Support by
Product pages, search for NetWorker using “Find a Product.” The product page provides
access to all product specific downloads and documentation, as well as additional
content and tools.
Revision history

Revision history
The EMC® NetWorker® 8.1 release notes was originally published in July, 2013. Table 1
on page 2 presents the revision history of this document.
Table 1 Revision history

Date Description

October 3, 2013 • Updated for the GA of NetWorker 8.1.

September 10, 2013 • Updates to graphics for Recovery in NMC, Recovery wizard
• Added “NW152978” to “Known problems and limitations”.

July 25, 2013 • DA of NetWorker 8.1

2 EMC NetWorker Release 8.1 Release Notes

 Product description

Product description
The EMC NetWorker 8.1 release provides significant functional enhancements, including
support for the following:
◆ Introduction of the NetWorker VMware Protection solution with the VMware Backup
Appliance
◆ Data Domain enhancements such as support for virtual synthetic full backups and
fibre channel communication
◆ A new Recovery wizard in the NetWorker Management Console
◆ NetWorker Snapshot Management integration

“New features and changes” on page 3 provides more information on new features and
identifies the release these features were introduced.

Before installing NetWorker release 8.1, review the NetWorker release 8.1 installation
guide, administration guide, and other documentation. “Documentation” on page 188
lists the complete NetWorker release 8.1 documentation set.

New features and changes

NetWorker 8.1
The NetWorker 8.1 software contains the following new features and enhancements. More
detailed information on these features is provided in the EMC NetWorker Administration
Guide:
◆ “NetWorker VMware Protection solution” on page 4
◆ “NetWorker Snapshot Management (NSM)” on page 5
◆ “Improved space recovery for AFTD and Data Domain devices” on page 7
◆ “Dynamic nsrmmd feature for AFTD and DD Boost devices now enabled by default” on
page 7
◆ “Enhanced AFTD Client Direct support” on page 8
◆ “Storage node load balancing” on page 8
◆ “Virtual synthetic full backups for Data Domain systems” on page 8
◆ “Data Domain Boost over fibre channel and IP network connections” on page 8
◆ “Data Domain resource type NSR Data Domain in NetWorker server” on page 9
◆ “Parallel save streams” on page 9
◆ “Automatic savegroup cloning feature updates” on page 10
◆ “Nsrclone performance improvements” on page 10
◆ “Volume-level optimized deduplication” on page 10
◆ “Recovering ACL files as non-root user” on page 10
◆ “Block-based backups on Windows” on page 11

EMC NetWorker Release 8.1 Release Notes 3

New features and changes

◆ “Recovery using the Recover wizard and NMC” on page 11

◆ “nwrecover no longer available in NetWorker 8.1” on page 13
◆ “Windows 2012 support for Cluster Shared Volumes (CSV)” on page 13
◆ “NetWorker Server Disaster Recovery” on page 13

NetWorker VMware Protection solution

NetWorker 8.1 introduces a new NetWorker-integrated VMware backup and monitoring
solution. In this solution, once you deploy an EMC Backup and Recovery appliance in the
vSphere Client, you can monitor the VMs and create and modify backup and clone
policies for the appliance by using NMC. Figure 1 on page 4 displays the default policy
that appears in NMC upon deployment of the EMC Backup and Recovery appliance.

Figure 1 VMware Protection Policies in NMC

You can then assign VMs to the backup and clone policies using the EMC Backup and
Recovery plug-in in the vSphere Web Client. When you click in EMC Backup and Recovery
in the left pane of the vSphere Web Client and connect to a backup appliance, the EMC
Backup and Recovery user interface displays, as shown in Figure 2 on page 5.

4 EMC NetWorker Release 8.1 Release Notes

 New features and changes

Figure 2 EMC Backup and Recovery user interface in the vSphere Web Client

Also, you can perform full recoveries of these backups from the vSphere Client, or
file-level recoveries from the EMC Data Protection Restore Client.
Note: NetWorker VMware Protection requires a minimum DDOS version 5.3.

NetWorker Snapshot Management (NSM)

EMC® NetWorker® 8.1 server and client software includes NetWorker Snapshot
Management (NSM) that integrates and replaces the previous stand-alone NetWorker
PowerSnap™ modules for VMAX® (Symmetrix®), VNX® Block (CLARiiON®), and
RecoverPoint™. NSM is a single, licensed, wizard-facilitated NetWorker feature that
provides an improved user experience with greater flexibility to protect production data
with less dependency on backup windows.

NSM manages snapshots, which are point-in-time (PiT) copies of production data created
by the replication capabilities of the storage array. NSM can restore the data from
snapshots and can copy snapshots to conventional storage media such as disk or tape.
NSM provides snapshot backups with the same benefits that NetWorker offers for
conventional backups such as monitoring, lifecycle policies, and granular filesystem
recoveries.

NSM simplifies the user experience through integration with NetWorker client software
and the implementation of configuration and recovery wizards.

The following new NSM features and support represent improvements over the former
PowerSnap module software:
◆ Integration with NetWorker client installation package, available under NetWorker
capacity licensing
◆ NMC Client Configuration Wizard:

EMC NetWorker Release 8.1 Release Notes 5

New features and changes

• Discovery and configuration of snapshot clients

• Step-by-step configuration of snapshot creation, snapshot lifecycle policies, and
snapshot rollover to conventional storage media such as disk or tape
◆ NMC Recovery Wizard:
• Step-by-step restore procedures
• Restore from snapshot save sets stored on a storage array and from rollover save
sets stored on conventional storage media
◆ Support for VMAX VPSnap and VNX VNX-SNAP
◆ Optional configuration on a single host without a separate mount (proxy) host
◆ Secure retention of RecoverPoint client user credentials in the server lockbox
◆ Reduction of the previous multiple log files to a single log file per client
◆ Support for VMAX thin-provisioned LUNs

Upgrading from former PowerSnap modules

Upgrade from PowerSnap to NSM replaces or removes some configurations and
dependencies that the previous PowerSnap modules used.

Snapshot policy change

An upgrade or a new installation of a NetWorker 8.1 server adds a new snapshot policy
called “RolloverOnly” whose values are identical to the previous “Serverless Backup”
snapshot policy. If the “Serverless Backup” snapshot policy is not being used, it will be
automatically deleted.

The EMC NetWorker Snapshot Management Integration Guide provides snapshot policy
details.
Backward compatibility

NSM in NetWorker 8.1 is backward compatible. The NetWorker 8.1 server can recover
backups created with PowerSnap 2.x.

The EMC NetWorker Software Compatibility Guide provides the latest software
requirements and hardware requirements.

Replaced or removed

The NSM features replace or remove the following dependencies and configurations:
◆ Dependency on EMC Professional Services for installation and configuration
◆ Dependency on the symm.res file (which now becomes optional) and VMAX disk
groups
◆ Dependency on PowerPath® for VNX
◆ Dependency on RMAPI for RecoverPoint
◆ powersnap.res file eliminated
◆ qfs driver on AIX eliminated

6 EMC NetWorker Release 8.1 Release Notes

 New features and changes

NetWorker features not supported

This release of NSM does not support the following NetWorker features:
◆ Synthetic full backup.
◆ Checkpoint restart; if used, it reruns the NSM backup from the start.
◆ Multi-Tenancy Facility (“restricted datazones”).

No longer supported

NSM drops support for some configurations that the PowerSnap modules supported,
including:
◆ Rollbacks on RecoverPoint storage arrays
◆ NMDA and NMSAP on RecoverPoint arrays
◆ Heterogenous (BDM) backups
◆ VMware backups
◆ NAS backups
◆ Veritas on Microsoft Windows
◆ 32-bit Microsoft Windows
◆ Microsoft Windows 2003
The EMC NetWorker Software Compatibility Guide provides details on versions that NSM
supports, including volume managers, NetWorker modules, and cluster environments.

Dropped application information variables

NSM drops support of the following Application Information variables that the previous
PowerSnap modules supported. The presence of these variables in NSM configurations
will cause backups to fail:
◆ NSR_IMAGE_SAVE
◆ SYMM_PROVIDER_DB
◆ SYMM_PROXY_PROVIDER_DB
◆ NSR_SNAP_TYPE=NAS

Improved space recovery for AFTD and Data Domain devices

NetWorker 8.1 features improved space recovery for AFTD and Data Domain devices,
reducing the overall processing time and media database activity that previously delayed
the availability of devices.

Dynamic nsrmmd feature for AFTD and DD Boost devices now enabled by default
The Dynamic nsrmmd feature for AFTD and DD Boost devices is now enabled by default. In
previous NetWorker versions, this attribute was disabled by default. The Networker
Performance Optimization and Planning Guide provides more details about the Dynamic
nsrmmd feature.

EMC NetWorker Release 8.1 Release Notes 7

New features and changes

Enhanced AFTD Client Direct support

Non-root and cross-platform Client Direct backups to AFTDs are now supported for
NetWorker clients on UNIX/Linux or Microsoft Windows. The AFTD can be managed by a
Linux/UNIX or a Windows storage node. The AFTD can be local or mountable on the
storage node.

Storage node load balancing

Starting in NetWorker 8.1, a new feature, named Save session distribution, has been
introduced that allows one to configure how save sessions are distributed among storage
nodes. The Save session distribution feature can be applied to all NetWorker clients
globally or to selected clients only. This feature has two options:
◆ Max sessions — Save sessions are distributed based on each storage node device's
ma sessions attribute. This is the default distribution method.
◆ Target sessions — Save sessions are distributed based on each storage node device's
target sessions attribute. This option is more likely to spread the backup load across
multiple storage nodes, while the max sessions option is more likely to concentrate
the backup load on fewer storage nodes.

Note: This feature is not available for clone and recover operations.

Virtual synthetic full backups for Data Domain systems

A virtual synthetic full (VSF) backup is the same as a synthetic full backup, except that it is
performed on a single Data Domain system. Similar to synthetic full, VSF uses full and
partial backups to create a new full backup. However, since the backup occurs on a Data
Domain system using new DDBoost APIs, the backup does not require save set data to be
sent over the wire, resulting in improved performance over synthetic full and traditional
backups.
Note: Virtual Synthetic Full requires a minimum DDOS version 5.3.

DD Boost synthetic full backups

The NetWorker synthetic full backup feature is an efficient way to create new full backups
by combining existing full and partial backups. This feature integrates the NetWorker 8.1
synthetic full backup feature and DD OS 5.3 and later virtual-synthetics. Synthetic full
backups are created directly on the DD Boost devices.

Both NetWorker and Data Domain are configured by default to enable DD Boost synthetic
full backups.

Also, the savegroup attribute “Revert to full when synthetic full fails”, which was
introduced in NetWorker 8.0 and disabled by default in the Advanced tab of the Group
Properties in NMC, is enabled by default in NetWorker 8.1.

Data Domain Boost over fibre channel and IP network connections

DD Boost devices now support data transport over both fibre channel (FC) and Ethernet IP
network connections for data backup and data recovery operations. FC connectivity for DD
Boost devices requires NetWorker 8.1 and DD OS 5.3 or later, and the FC network must be
deployed as a SAN.

8 EMC NetWorker Release 8.1 Release Notes

 New features and changes

The NetWorker server also requires Ethernet IP connections to communicate with all hosts
involved in DD Boost operations and for NetWorker clone-controlled replication
operations.

After upgrading to the supported DD OS and NetWorker versions, you can convert existing
DD Boost devices that use TCP/IP connections to use FC connections. No data is lost by
the conversion and the full DD Boost features are retained, including Client Direct
functionality for backup and restore.

Data Domain resource type NSR Data Domain in NetWorker server

In previous NetWorker releases, the Data Domain resource was registered in NMC, which
meant that information about Data Domain was stored in NMC instead of the NetWorker
server's RAP database. As a result, the NetWorker server did not have a placeholder to
keep the information like maximum sessions, current sessions, and so on. Also, in the
case of disaster recovery, both NMC and the NetWorker server had to be recovered to
retrieve the registered Data Domain information.

NetWorker 8.1 moves the Data Domain registration from the NMC database to the
NetWorker server's RAP database, improving the integration of the Data Domain system
and NetWorker. A new RAP resource type “NSR Data Domain” has been created in the
NetWorker server that contains attributes such as DD model, DD OS version, and
hostname. The RAP resource is visible in NMC, and you can edit the user-inputted
attributes using nsradmin or NMC.

DD Boost immediate and scheduled clones

Use NetWorker clone-controlled replication (CCR) operations in environments where you
want to regularly create copies of DD Boost save sets, for example, for added protection
and efficient disaster recovery. CCR clones are indexed by NetWorker retention policies.

With NetWorker 8.1, you can configure CCR to run immediately after each save set
completes, known as “immediate clone,” or you can configure CCR to run in an
independently defined maintenance window after the entire group backup completes in
the main backup window. Clone settings are configured in the NetWorker Group resource.

Immediate CCR is best used where:

◆ there are many save sets in the backup queue
◆ there are many save sets of different sizes

Monitoring and reporting of DD Boost operations

NetWorker 8.1 improves its monitoring capabilities of Data Domain systems to include
information on Data Domain identity (name, hosts, model, OS version, serial number),
SNMP community string, access credentials, capacity status information, save stream
status information, and system details. You can monitor DD Boost devices for
compression efficiencies and you can view connectivity status, operation logs, and alerts.

Parallel save streams

The parallel save streams (PSS) feature provides the ability for Client resource save sets
(UNIX and Linux) to be backed up by multiple parallel save streams to one or more
destination backup devices. PSS is disabled by default.

EMC NetWorker Release 8.1 Release Notes 9

New features and changes

Note: Both the NetWorker server and client must be at NetWorker 8.1 and later in order to
use the PSS functionality for full and incremental level backups, recovery, and cloning.

Automatic savegroup cloning feature updates

The method for setting up an automatic clone operation based on the backup group
resource has a new feature. The clone operation can now be set to begin after each save
set is backed up. Alternatively, you can still select to have the clone operation begin only
after all of the save set backups for the group have completed.

Nsrclone performance improvements

Improvements to nsrclone startup processing introduced with NetWorker 8.1 reduce the
impact of fetching volume data from the media database, resulting in faster nsrclone
processing time and loading of nsrmmdbd. Also, improvements to internal save set list
processing reduces the wait time prior to cloning, and results in less CPU load.

Volume-level optimized deduplication

Volume-level optimized deduplication is a new feature in Windows Server 2012. When
the data deduplication role is enabled, a Deduplication VSS Writer supports backup of
volumes in their optimized deduplication form. NetWorker supports backup of optimized
deduplication volumes, and restore of optimized deduplication backups to a set of
eligible restore targets.

For more information on optimized data deduplication volume backups, refer to the
appendix Additional Features of the Microsoft Windows.

Recovering ACL files as non-root user

Previously, non-root users could not browse and recover files if the save sets contained
ACL/metadata associated with root permissions. NetWorker now provides the ability to
browse and recover files with associated ACLs (Access Control Lists) in directories for
which the user is not the primary owner.

To recover files with associated ACLs, ACL passthrough must be checked in the Recover
section of the NetWorker Server Properties window, as shown in the following graphic.
The feature is enabled by default.

10 EMC NetWorker Release 8.1 Release Notes

 New features and changes

Figure 3 ACL passthrough option in NetWorker Server Properties window

If ACL passthrough is disabled, the message “Permission denied (has acl)” displays when
a non-owner attempts to browse ACL files in the directory.

Block-based backups on Windows

Block based backup is a technology where the backup application scans a volume or a
disk in a file system, and backs up all the blocks that are changed or in use in the file
system. Block based backup uses Volume Shadow Copy Service (VSS) snapshot
capability to create a consistent copy of the source volume for backup and Virtual Hard
Disk (VHD) format, which is sparse, to write data to the target device. Block based backup
supports only Advanced File Type Device (AFTD) and Data Domain (DD) as target devices.
These devices support the client direct functionality that is a mandatory requirement for
block based backup. For incremental backups, it uses the Change Block Tracking (CBT)
driver to identify the changed blocks so that only the blocks that are changed can be
backed up.
Note: This feature is supported only on Windows.

Recovery using the Recover wizard and NMC

In NetWorker releases previous to 8.1, the Networker Management Console was used
primarily for backup management, and tools such as Winworkr or nwrecover were
required for recoveries. Using these services required logging into the client to initiate a
recovery.

NetWorker 8.1 introduces new recover functionality with a Recovery wizard that allows
you to recover data to NetWorker 8.1 clients from NMC. The Recover wizard supports
browsable, save set, and directed recoveries. The Recover wizard does not support
cross-platform recoveries.

EMC NetWorker Release 8.1 Release Notes 11

New features and changes

Clicking the Recover tab in the NetWorker Administration window of NMC allows you to
run the Recover wizard from the menu by selecting Recover > New Recover.

Figure 4 Launching Recovery wizard in NMC

Use the Recovery wizard to configure scheduled and immediate recoveries of:
◆ Filesystem backup
◆ Block Based Backup (BBB)
◆ BBB cloned to tape

You can also use the Recovery wizard to configure an immediate recover of a SnapShot
Management backup.

Once recovery operations are in progress, recover activity displays in the window, as in
the following.

Figure 5 Recover activity in NMC

12 EMC NetWorker Release 8.1 Release Notes

 New features and changes

nwrecover no longer available in NetWorker 8.1

The nwrecover program has been removed from NetWorker 8.1 For recoveries, use the
command line or the NMC Recovery wizard.

Windows 2012 support for Cluster Shared Volumes (CSV)

The backup and recovery of file system data on Windows Server 2012 File Servers
configured for Windows Continuous Availability with Cluster Shared Volumes (CSV) is now
supported. Support of CSV backups include levels Full, Incremental and Synthetic-full. For
additional details, refer to the Cluster Integration Guide.

NDMP performs DSA backup by default

DSA backup (DSA=y) is now enabled by default for NDMP backups. NetWorker performs
DSA backup regardless of what you specify in the storage node list, unless the first
storage node is a NDMP jukebox.

NetWorker Server Disaster Recovery

A new command line wizard program named nsrdr automates the recovery of the
NetWorker server's media database, resource files, and client file indexes. You can run
the nsrdr program in fully interactive mode and respond to questions or you can run the
program with command line options. You can recover just the media database, or the
resource files, or the client file indexes. The NetWorker Administration Guide > Recovery
chapter > NetWorker server disaster recovery section provides more information.

AGILE mode supported on HP-UX 11iv3

NetWorker 8.1 supports use of the AGILE mode in addition to the LEGACY mode on the
HP-UX 11iv3 platform. The AGILE addressing mode is the default addressing mode on the
HP- UX 11iv3 system, and creates Device Special files (DSFs) based on the device type.

More information is provided in the NetWorker Command Reference Guide.

EMC NetWorker Release 8.1 Release Notes 13

Fixed problems

Fixed problems
This section provides details on fixed bugs for the NetWorker 8.1 release. When a
cumulative build is released, information on fixes included in the latest cumulative is
provided in the NetWorker Cumulative Hotfixes document, available on the EMC Online
Support site.

Fixed bugs in release 8.1

Table 2 on page 14 lists customer reported defects resolved in release 8.1.

Table 2 Fixed in NetWorker Release 8.1

Issue number Description

NW153004 nsrmmd deletes the active-folder on the Data Domain device when it still contains a valid
active-file from a recently started backup

NW151236 Large data discrepancy between NMC reports and mminfo output

NW144437 NDMP Monthly (Full backup) clone hangs after reading a couple of volumes

NW143324 A pool cannot be modified and returns with blank pop-up after upgrading

NW143414 nsrvadp_save.exe crashes after upgrading

NW143616 NDMP History time is very large

NW144712 Encrypted hypervisor resource password can be deciphered using nsradmin

NW145267 nsrd crashes upon startup in Linux Red Hat version 6.2

NW146617 nsrvadp_save.exe faults on LIBVMDK.dll

NW149156 The nsrndmp_save command does not add the "NDMP_TAPE_BLOCK_SIZE" to the save
set information, resulting in recovery failure

NW137746 Reports slow to load in NMC

NW139191 Configured clone jobs fail but nsrclone command works

NW139374 Unable to perform VADP incremental backups for some VMs with error “Could not obtain
the drive letter for one of the drives”

NW142578 nsrcpd core dumps upon cross-platform software distribution

NW144230 nwrecover cannot be started from NMC, returns error “Unknown host”

NW144564 Nsrvadp_save crashes on proxy during VADP backup

NW146410 Recovery fails if VM has Physical Compatibility RDM disks

14 EMC NetWorker Release 8.1 Release Notes

 Environment and system requirements

Environment and system requirements

This section describes specific environment and system requirements.

System configuration requirements for a dedicated NetWorker server

Table 3 on page 15 and Table 4 on page 15 outline the following:
◆ Minimum system configuration requirements to be met when running the NetWorker
software on a dedicated NetWorker server.
◆ Tips when setting parameters at the operating system level.

Note: If the following system requirements are not met, the performance of the NetWorker
software could be significantly affected or the NetWorker daemons might crash.

Table 3 Minimum system requirements for a dedicated NetWorker server

Minimum recommended configuration

CPU Dual Core with a minimum of 1.5 GHz speed for each CPU

RAM 2 GB

Swap 4 GB minimum (at least twice RAM)

space

Table 4 Minimum system requirements for a dedicated NetWorker server

Minimum recommended configuration

The following plimit values for all NetWorker daemons should be set to maximum as follows:

Kernel Resource Current Maximum

parameters
Time Unlimited Unlimited

file(blocks) Unlimited Unlimited

data(Kbytes) Unlimited Unlimited

stack(Kbytes) 8192 Unlimited

coredump(blocks) Unlimited Unlimited

nofiles(descriptors) 65536 65536

vmemory(Kbytes) Unlimited Unlimited

The EMC NetWorker Performance Optimization Planning Guide provides information on

tuning the operating system and is available at https://support.emc.com/.

EMC NetWorker Release 8.1 Release Notes 15

Environment and system requirements

Write access to temp folders required when using the NetWorker Software
Administration wizard or nsrpush CLI for a Windows 2008 client
Microsoft Windows 2008 restricts write access to temp folders, including the folders
defined in the Windows SYSTEM user's TEMP or TMP environment variable.Write
permissions must be enabled when performing software updates, add to repository
operations, and inventory operations using the Software Administration wizard or the
nsrpush CLI.

On the Windows 2008 client, grant the Administrator and SYSTEM users write permissions
to the temp folders defined in the SYSTEM user's TEMP and TMP environment variables.

Windows hot fix required for Vista to recover VSS System files
Due to an issue with the Microsoft Windows Vista operating system, a hot fix is required to
successfully recover VSS System files. Contact Microsoft to obtain the fix. Details can be
found at http://support.microsoft.com/kb/935606.

Install latest Microsoft VSS roll-up fix for Windows Server 2003
If running NetWorker on Windows Server 2003, install the latest Microsoft VSS roll-up fix.
To view more information on the roll-up fix and to download the package appropriate to
your platform, go to the Microsoft knowledge base article at
http://support.microsoft.com/kb/940349.

QPK1123 and PHSS_37492 patches required to run NetWorker on HP-UX RISC 11.23
The NetWorker services cannot be started on an HP-UX RISC 11.23 system until the
following patches are downloaded from the HP website:
◆ QPK1123(B.11.23.0712.070a) 1185010 Quality Pack Depot
◆ PHSS_37492

Go to http://itrc.hp.com and click patch database to obtain these patches. You must
provide an appropriate username and login password to download the patches.

Linux st kernel max devices default configuration

By default, the Linux st kernel configures a maximum of 128 SCSI tape devices. As a
result, the inquire command and the Scan for Devices operation might not detect more
than 128 tape devices.

To resolve this issue, the st module of the Linux kernel must be modified and recompiled
to increase the maximum number of allowable st devices created by the operating system
to exceed the default value. Refer to Linux documentation for details about how to
reconfigure, rebuild, and install the kernel.

Features provided with base enablers

Table 5 on page 17 lists the features that are available as part of NetWorker base
enablers.

16 EMC NetWorker Release 8.1 Release Notes

 Environment and system requirements

Table 5 Features provided with base enablers

Support by edition

Feature Power Network Workgroup Business

Number of included client connections 12 10 8 8

Parallel data streams per NetWorker server 64 32 32 32

Parallel datastreams per storage node1 64 32 N/A N/A

Number of physical devices included with base 32 16 4 4

enabler

Increase in datazone's devices, per storage node 32 16 N/A N/A

license

Maximum number of devices 512 512 4 4

1.Storage nodes available for additional cost for the Power and Network Editions only.

Additional features available with base enablers

Table 6 on page 17 lists the features that are available for support with NetWorker base
enablers.

Table 6 Additional features available with base enablers

Support by edition

Feature Power Network Workgroup Business

Additional client connections Yes Yes No No

Storage nodes Yes Yes No No

Cluster support Server and Client No Maximum 2 Cluster client

Client connections

NetWorker NDMP Client connections Yes Yes No Maximum 2 NDMP client connections

Dynamic Drive Sharing option Yes Yes N/A N/A

NetWorker DiskBackup option Yes Yes Yes Yes

NetWorker Archive Module Yes Yes No No

Autochanger Software Modules All All 1-9, 1-16, One 1-26 included in base
1-20, 1-32

NetWorker Application Modules Yes Yes Yes Yes

NetWorker Windows Server 2003 Open File Option Yes Yes Yes Yes

Open File Manager Yes Yes Yes Yes

NetWorker PowerSnap Modules Yes Yes No No

NetWorker SnapImage Module Yes Yes No Yes

VSS Support for Windows 2003 Yes Yes Yes Yes

Simple Network Management Protocol (SNMP) Yes Yes Yes Yes

VTL (Virtual Tape Library) Yes Yes Yes Yes

EMC NetWorker Release 8.1 Release Notes 17

Known problems and limitations

Known problems and limitations

This section describes known limitations found in the entire NetWorker family of releases:
◆ “NetWorker 8.1” on page 18
◆ “NetWorker 8.0 Service Pack releases” on page 40
◆ “NetWorker 8.0” on page 45
◆ “NetWorker 7.6 Service Pack 3 and later 7.6 releases” on page 54
◆ “NetWorker 7.6 Service Pack 2” on page 58
◆ “NetWorker 7.6 Service Pack 1” on page 87
◆ “NetWorker 7.6” on page 107
◆ “NetWorker releases previous to 7.6” on page 128
◆ “NetWorker releases previous to 7.5” on page 145

Unless the entry for a known limitation indicates that it is resolved for a specific release,
the limitation applies to the release in which it is identified and all subsequent releases. If
a limitation is resolved, it will also be identified in the fixed bugs table for the release in
which it is resolved.

NetWorker 8.1
Table 7 on page 18 identifies issues and limitations discovered in NetWorker 8.1.

Table 7 Limitations discovered in NetWorker release 8.1

Operating
Issue number Description of limitation system affected Product feature

Reboot of VM takes several minutes after launching the Console of the All VMware
EMC Backup and Recovery appliance in the vSphere Web Client

Maximum number of concurrent VM backups may be exceeded when you All VMware
deploy more than one EMC Backup and Recovery appliance per vCenter

VM backups and recoveries cannot be performed when EMC Backup and All VMware
Recovery appliance runs out of space (fixed in GA of NetWorker 8.1)

“NW154417” You cannot recover multiple cloned block-based backup save sets from Block-based
on page 23 tape using the NMC Recovery wizard backup, Recovery

“NW154261” EMC Data Protection Restore Client interface may become unresponsive Windows VMware
on page 23 when mounting multiple backups or leaving browser idle

“NW154061” Recovery from a DD device fails if the DD CIFS is joined with the Active Windows Recovery
on page 24 Directory domain, and the DD device is configured with other than the DD
hostname

“NW153948” nsrpsd process does not start or stop automatically when NMDA or All NSM
on page 24 NMSAP is used on the application host, and nsrpsd_stay_up file is
required to prevent nsrpsd from terminating

“NW153702” NSM for HP PA-RISC only supported via RPQ HP-UX NSM
on page 24

“NW153412” When multiple versions of the same file are selected for recovery, NMC All Recovery
on page 25 Recovery wizard only recovers last modified version

18 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Table 7 Limitations discovered in NetWorker release 8.1

Operating
Issue number Description of limitation system affected Product feature

“NW153252” Group Level browse and retention policy is not honored when a backup All CLI
on page 25 group (savegrp) is run from the command line

“NW152978” Invalid browse and retention recovery sessions after effective expiration Linux, UNIX Parallel Save
on page 25 date when Parallel Save Streams (PSS) save times span midnight Streams (PSS)

“NW152400” When using autoconfig to perform DSA backup to AFTD on Linux/UNIX Linux, UNIX CLI
on page 25 remote node you must specify the -P option

“NW152175” Virtual synthetic full backup not supported with DD Archiver DD OS All Data Domain
on page 26 5.3.0.6

“NW151894” Missing information in NMC Recover reports for VMware image-level All VMware
on page 26 recoveries

“NW151883” Unsupported block-based backup configurations All Block-based

on page 26 backup

“NW151858” Extra characters in nsrdr output after index recovery All Recovery
on page 26

“NW151810” Recovering client file indexes with the nsrdr program may incorrectly log Windows Messaging
on page 26 errors for NMM Exchange DAG client members

“NW151702” Delay in starting some VMware Backup Appliances may occur when Windows Backup
on page 27 backing up large number of VMs and using more than 2 VMware Backup Linux
Appliances per backup policy

“NW151691” Backup to DD Boost device over fibre channel fails on Windows 2008 R2 Windows Backup
on page 27

“NW151638” Launching vSphere Web Client fails when you install the Client on a All VMware
on page 27 different server than vCenter

“NW151434” Backup delays may occur when backing up large number of VMs using Windows Backup
on page 27 NetWorker VMware Protection Linux

“NW150956” Recovery fails when attempting FULLVM recovery of two or more VMs with All Recovery
on page 27 the same name

“NW150855” NMC displays wrong error message to users who do not have sufficient All Messaging
on page 28 recover privileges for Block-based backup (BBB) recoveries

“NW150593” Clone-controlled replication (CCR) hangs with message “NSR server Solaris Cloning
on page 28 busy”, log indicates there is no device available

“NW150339” NetWorker VMware backups using VMware Backup Appliance to a DD Windows Cloning
on page 28 Boost device support clone controlled replication (optimized clone) only Linux
with DDOS 5.3.0.6 and later or 5.4.0.4 and later

“NW150045” For a File Level Recovery, the rename option does not work if you recover a Windows Recovery
on page 28 volume to itself

“NW149956” NetWorker Recovery Wizard does not recover mount points under selected All Recovery
on page 29 recover point

“NW149883” NDMP single file browsable recovery fails due to “Error during File NDMP All NDMP
on page 29 Extraction”

“NW149727” The Total size and Number of files fields in the Recover page of NMC do Windows Block-based
on page 29 not display the updated values backup

EMC NetWorker Release 8.1 Release Notes 19

Known problems and limitations

Table 7 Limitations discovered in NetWorker release 8.1

Operating
Issue number Description of limitation system affected Product feature

“NW149725” Remote client backup fails with local AFTD Windows Block-based
on page 29 backup

“NW149724” Backup of the All and Disaster Recovery save sets fails when a file level Windows Block-based
on page 30 recovery is being performed on the same host backup

“NW149723” NMC does not display all the destination volumes when you perform a Windows Block-based
on page 30 block based backup recovery backup

“NW149722” Block based backup recovery fails when the Data Domain device Windows Block-based
on page 30 credentials and the Common Internet File System share credentials are backup
different

“NW149720” The size of the backed up basic volume save sets or dynamic volume save Windows Block-based
on page 30 sets is sometimes more than the size of the actual data backup

“NW149392” Virtual Machine Disk consolidation required when image-level backups of All VMware
on page 30 a client fail using VMware Backup Appliance with NetWorker 8.1

“NW149301” Backup fails with System Reserved Partition error Windows Backup
on page 31

“NW149172” Expired VMware Backup Appliance backups to a DD Boost device may not Windows Backup
on page 31 be removed from the Data Domain System for up to 48 hours

“NW148731” Networker resource group nsrexecd stuck in “Online Pending” or “Offline Windows Cluster
on page 32 Pending” state in AutoStart

“NW148548” Recover wizard does not display the NSM recover statistics All GUI
on page 32

“NW148540” Information on completed save set clones not reported when scheduled All Cloning
on page 32 clone is interrupted

“NW148514” Recover wizard does not update NSM recover statistics in the display All NSM
on page 32

“NW148373, NMC Recovery Wizard does not support save set recovery for snapshots All Recovery
NW148371”
on page 33

“NW148319” NMC shortcut appears as a text icon instead of cube icon on Windows Windows GUI
on page 33 2012 desktop

“NW148309” Cluster Shared Volumes (CSV) backup performed as level full instead of Windows Cluster
on page 33 incremental after the preferred server order list is modified

“NW148112” Clone storage node attribute not honored with a DD Boost device Linux Cloning
on page 34

“NW148067 Microsoft VSS backups may contain inconsistent data that cannot be Windows Backup
(esg135000)” restored due to a Microsoft-reported issue in the Windows Volume
on page 34 Shadow Copy framework

“NW148050” VADP snapshot not supported on VM with MS iSCSI initiator Windows VMware
on page 34

“NW148215” Clone details window does not list the status for the save sets being All Cloning
on page 35 cloned

“NW148022” Mminfo -p does not display save sets for backup type SMTAPE All CLI
on page 35

20 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Table 7 Limitations discovered in NetWorker release 8.1

Operating
Issue number Description of limitation system affected Product feature

“NW147765, Running large number of save sets in parallel may lead to backup failure UNIX Backup
NW147756” due to lack of swap space
on page 35

“NW147541” Scanning for bootstrap save set ID in NetWorker server disaster recovery Windows GUI
on page 35 wizard not supported for non-English locales

“NW146964” When using cloning mode “start on each save set completion”, only the All Cloning
on page 35 first completed save set for each job gets cloned immediately if the
user-defined save set generates more than one SSID

“NW146959” Block based backup does not support NetWorker SnapImage Module Windows Backup
on page 36 backups

“NW146802” Max-wait-time exceeded when using nsrconsolidate to rehydrate large Windows Backup
on page 36 number of Avamar backups or to synthesize large number of save sets

“NW145499” When the source directory is empty, Client Push reports transfer as All Client Push
on page 36 successful even though no files are transferred

“NW145338” Clicking back button in Software Administration wizard during upgrade Windows GUI
on page 36 returns error

“NW145286” When you specify a preferred node with the NSR_CSV_PSOL variable. the Windows CSV
on page 36 savegrp report does not report the size information for a CSV save set.

“NW145280” Software repository location cannot be changed using Software All GUI
on page 37 Administration wizard

“NW145187” Client upgrade from CLI or Software Administration wizard cannot be All CLI/GUI
on page 37 cancelled once started

“NW145068” Veritas Volume Manager on Windows not supported with NetWorker Windows Configuration
on page 37

“NW144852” VMware Backup Appliance created with higher than normal default values All Devices
on page 37 for device’s target session and max session

“NW144806” Back button does not return correct page of Software Administration All GUI
on page 37 wizard after selecting inventory option

“NW144544” nsr_end() does not report correct size for save sets larger than 2 TB All CLI
on page 37

“NW142871” Save Set ALL does not back up all volumes after image-level recovery of a Windows Backup
on page 38 block-based backup if original volume was retained

“NW141678” Limit of 8 Logical Unit Numbers (LUNs) per target in LEGACY mode on All CLI
on page 38 HP-UX 11i v3

“NW141532” Upgrade to NetWorker 8.1 on Windows 2003 x64 client fails using Windows Upgrading
on page 38 Software Distribution wizard due to missing MS Visual C++ 2005
Redistributable pkg

“NW124800” NMC Client Configuration wizard allows creation of duplicate VADP proxies Solaris GUI
on page 38

“NW121443, Recovery of application module datafiles occasionally fails All Recovery

NW105687,
LGTsc19387”
on page 38

EMC NetWorker Release 8.1 Release Notes 21

Known problems and limitations

Table 7 Limitations discovered in NetWorker release 8.1

Operating
Issue number Description of limitation system affected Product feature

“NW121940” vStorage_API: Directive drop down lists VCB instead of VADP Windows VMware
on page 39

“NW120006” Recovery of VM to an ESX server different from the originating server may Windows Recovery
on page 39 fail due to different network adapter

“NW102770, NSM in VCS cluster setup does not remove the associated snapshot Solaris NSM
LGTsc17244” sessions
on page 39

“NW102236, When MSCS cluster disk is online, rollback should exit with a proper error Windows NSM
LGTpa58948”
on page 39

“NW080016, Recovery of application module datafiles occasionally fails Windows NSM

LGTsc19387”
on page 40

Reboot of VM takes several minutes after launching the Console of the

EMC Backup and Recovery appliance in the vSphere Web Client
When you log into the vSphere Web client and launch the Console for the EMC Backup
and Recovery appliance, a delay of several minutes may occur while the VM reboots. A
message appears in the output similar to the following:
Identity added: /home/dpn/.ssh/dpnid (/home/dpn/.ssh/dpnid)

If you see this message, do not shutdown the VM, and allow time for the reboot to
complete.

Maximum number of concurrent VM backups may be exceeded when you

deploy more than one EMC Backup and Recovery appliance per vCenter
A maximum of 26 concurrent VM backup sessions per EMC Backup and Recovery
appliance is supported. Deploying more than one EMC Backup and Recovery appliance
per vCenter may result in exceeding the recommended maximum number and lead to
performance issues and errors similar to the following:
Backup CANCELLED ,wraping-up session with Server
Max retries for downloading VM .vmx file exceeded, count=3

VM backups and recoveries cannot be performed when EMC Backup and

Recovery appliance runs out of space (fixed in the GA of NetWorker 8.1)
When a deployed EMC Backup and Recovery appliance for VMware runs out of space, you
cannot perform any further backups and recoveries until you clear space from the
appliance.
Workaround

To clear space on the EMC Backup and Recovery appliance, log in to the EMC Backup and
Recovery appliance VM and perform the following:
1. Stop the EMC Backup and Recovery appliance by running the following:
dpnctl stop

22 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

2. Remove the MCS DB dump file, located at:

rm /usr/local/avamar/var/mc/server_data/mcs_data_dump.sql
3. Restart the EMC Backup and Recovery appliance by running the following:
dpnctl start
4. Verify that the server is running by using the following command:
su - admin
5. Restart the server if required:
emserver.sh -start
6. Run the clean_db script to remove entries from events and audits tables:
/usr/local/avamar/bin/clean_db.pl
7. Run the vacuum command for the Events and Audits table to reclaim space:
Psql -Uadmin -p5555 mcdb
>> vacuum full events;
>> vacuum full audits;
8. Shut down the EMC Backup and Recovery appliance:
dpnctl stop
9. Restart the appliance:
dpnctl start

Note: To ensure the EMC Backup and Recovery appliance does not run out of space after a
re-boot, run the following after rebooting the system:

su - admin
emserver.sh -start

You cannot recover multiple cloned block-based backup save sets from
tape using the NMC Recovery wizard
NW154417

When you perform a recovery of multiple cloned block-based backup save sets from a
tape by using the NMC Recovery wizard, the Select the Data to Recover page displays only
one out of the multiple save sets that were cloned. As a result, you can recover only one
save set.
Workaround

To recover all of the save sets, perform the recovery by using the CLI.

EMC Data Protection Restore Client interface may become unresponsive

when mounting multiple backups or leaving browser idle
NW154261

When you perform a VM file-level recovery using the EMC Data Protection Restore Client
interface, the browser may become unresponsive if you mount multiple backups at the
same time, or if you leave the browser idle for more than 10 minutes.

EMC NetWorker Release 8.1 Release Notes 23

Known problems and limitations

Workaround

Restart the browser on the system, and then log back into the EMC Data Protection
Restore Client. When performing the file-level restore, only mount the backup that you
want to restore files from.

Recovery from a DD device fails if the DD CIFS is joined with the Active
Directory domain, and the DD device is configured with other than the DD
hostname
NW154061

Block based backup recovery from a DD device fails if the DD CIFS is joined with the Active
Directory domain, and the DD device is configured with either the DD IP address or the DD
Fully Qualified Domain Name (FQDN).
Workaround

1. Open NMC.
2. Under the Devices tab, click Devices.
3. Right-click the DD device from which the recovery has failed, and click Unmount.
4. Right-click the same DD device, and click Properties.
The Device Properties dialog box appears.
5. In the Device access information field, change the device name from the IP address or
the FQDN to the hostname.
6. Click OK.
7. Right-click the DD device, and click Mount.
8. Perform the recovery.

nsrpsd process does not start or stop automatically when NMDA or NMSAP
is used on the application host, and nsrpsd_stay_up file is required to
prevent nsrpsd from terminating
NW153948

If you use any version of NMDA or NMSAP on the application host, then the nsrpsd
process does not start or stop automatically, and NMDA backups/restores will fail. You
must start nsrpsd manually, as done in previous PowerSnap releases.

Also, after 30 minutes of inactivity, the nsrpsd process terminates. To prevent nsrpsd
from terminating, create the file nsrpsd_stay_up in the nsr/res directory.

NSM for HP PA-RISC only supported via RPQ

NW153702

NetWorker 8.1 only supports NSM for HP PA-RISC via RPQ.

24 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

When multiple versions of the same file are selected for recovery, NMC
Recovery wizard only recovers last modified version
NW153412

When you select multiple versions of the same file for recovery using the NMC Recovery
wizard, only the most recently modified version gets recovered, even though the logs
report that all versions were recovered. Note that this issue does not occur when using
Winworkr, which can recover all versions of the same file.
Workaround

When using the NMC Recovery wizard, you must perform multiple recoveries to recover all
versions of the file.

Group Level browse and retention policy is not honored when a backup
group (savegrp) is run from the command line
NW153252

When a group backup is run from the command line, for example, savegrp -vvv -G
<Group_Name>, the browse and retention policies assigned to the group are overridden by
any browse and retention policies set for each backup client.
Workaround

Run the backup group from NMC.

Invalid browse and retention recovery sessions after effective expiration

date when Parallel Save Streams (PSS) save times span midnight
NW152978

PSS save point backup save times that span midnight by seconds or minutes can result in
both browse and retention policy expirations that differ by an entire day.

When a PSS save point backup occurs near midnight and the multiple save times span
midnight for the backup, the result can be two different browse expiration dates. The
earlier adjoining date becomes the effective browse expiration date for the save point. Do
not perform a save point browse recovery beyond this effective date. If a browse recovery
session is attempted after the effective browse expiration date, the session remains
invalid although a partial recovery of some files might occur. The same applies to PSS
retention expiration dates.
Note: The aggregate expiration date and time of a PSS save point backup is always the
earliest expiration date and time of the multiple back up media save sets.

When using autoconfig to perform DSA backup to AFTD on Linux/UNIX

remote node you must specify the -P option
NW152400

DSA backup to an AFTD device on a Linux or UNIX remote node using the autoconfig
feature fails from any server platform when you do not specify the -P option.

EMC NetWorker Release 8.1 Release Notes 25

Known problems and limitations

Workaround

Ensure that you specify -P when using autoconfig for DSA backup.

Virtual synthetic full backup not supported with DD Archiver DD OS 5.3.0.6

NW152175

NetWorker 8.1 does not support virtual synthetic full (VSF) backups to DD Boost devices
on the Data Domain Archiver tier for Data Domain system running DD OS 5.3.0.6.

Workaround

Upgrade to DD OS 5.4.

Missing information in NMC Recover reports for VMware image-level

recoveries
NW151894

The NetWorker Recover Reports in NMC for VMware image-level recoveries do not return
any information for the Client Name, Source Client, and Target Client fields. The reports
identify these fields as “unavailable.”

Unsupported block-based backup configurations

NW151883

Block-based backup is not supported on:

◆ UNIX
◆ Linux host configured as a client
◆ Linux host configured as a storage node.

Extra characters in nsrdr output after index recovery

NW151858

Extra characters appear in the nsrdr output after performing an index recovery. This issue
typically occurs when you run multiple sessions concurrently. These extra characters do
not indicate a problem with the recovery status and can be ignored.

Recovering client file indexes with the nsrdr program may incorrectly log
errors for NMM Exchange DAG client members
NW151810

If you have an NMM Exchange DAG client configured and you are recovering client file
indexes on the NetWorker server using the nsrdr program, you may see errors similar to
the following in the nsrdr.log file:
CFI recover for the client <Exch2010-1.cent.local> is failed

If these client names are Exchange DAG client members, you can ignore these errors. A
client file index is created for the Exchange DAG client only and not its members.

26 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Delay in starting some VMware Backup Appliances may occur when

backing up large number of VMs and using more than 2 VMware Backup
Appliances per backup policy
NW151702

In the NetWorker VMware Protection solution, when you back up a large number of VMs
and use more than two VMware Backup Appliances per backup policy, upon initiating the
backup from NMC a delay may occur starting some of the appliances on the backup
policy.

Backup to DD Boost device over fibre channel fails on Windows 2008 R2

NW151691

When you configure a NetWorker client on Windows 2008 R2 to backup to a DD Boost

device using fibre channel (SAN) communication, the backup fails.
Workaround

Perform the following steps:

1. Disable the EMC PowerPath Generic Purpose Extension system device.
2. Reboot the NetWorker client.

Launching vSphere Web Client fails when you install the Client on a
different server than vCenter
NW151638

You cannot launch the vSphere Web Client when you install the Client on a different server
than the vCenter server.

Backup delays may occur when backing up large number of VMs using
VMware Backup Apliance
NW151434

Backing up a large number of VMs on the VMware Backup Appliance may result in delays
of some VM backups. When this occurs, a message appears in the avnwcomm log on the
appliance indicating the server is busy.
Workaround

If delays occur due to the number of VM backups, wait until the queue clears before
starting more backups.

Recovery fails when attempting FULLVM recovery of two or more VMs with
the same name
NW150956

FULLVM (image-level) recovery fails when attempting a recovery of two or more the VMs
with the same name. A warning similar to the following appears in the Set Restore Options
page of the Restore a backup wizard in the vSphere Web Client's EMC Backup and
Recovery user interface:

EMC NetWorker Release 8.1 Release Notes 27

Known problems and limitations

Warning - Two or more of the names entered for VM restores are the same. It is
recommended that you use unique names to avoid restore collisions that result in one of
the restore operations failing.
Workaround

Rename the VMs before performing the recovery.

NMC displays wrong error message to users who do not have sufficient
recover privileges for Block-based backup (BBB) recoveries
NW150855

For BBB recoveries performed from NMC where the user does not have sufficient recover
privileges, the wrong error message displays. The displayed error message indicates:
Unable to perform mount/unmount request. Kindly configure CIFS share
if you are using a Data Domain device

The error message should indicate that the user needs to have “Recover Local data”
privileges.
Workaround

Add the logged-in user to the NetWorker User Group named Users.

Clone-controlled replication (CCR) hangs with message “NSR server

busy”, log indicates there is no device available
NW150593

When backing up VMs to a single device and concurrently cloning the same source save
set to multiple DDRs, the CCR operation hangs with the message “NSR Server busy”, and
the logs indicate that there is no device available even though backups are still running
successfully on the devices.
Workaround

If you encounter this issue, turn off Dynamic MMDs.

NetWorker VMware backups using VMware Backup Appliance support

clone controlled replication (optimized clone) to a DD device target using
DDOS 5.3.0.6 and later or DDOS 5.4.0.4 and later
NW150339

Clone controlled replication (optimized cloning) is supported for NetWorker VMware

backups that use the VMware Backup Appliance only if the target Data Domain device
uses DDOS 5.3.0.6 and later or DDOS 5.4.0.4 and later. Regular cloning from a DD Boost
device to an AFTD or tape is not affected by this limitation.

For a File Level Recovery, the rename option does not work if you recover a
volume to itself
NW150045

While performing a File Level Recovery, if you select an entire volume for recovery and the
destination is the volume itself, then despite being a duplicate, the volume is over-written
instead of being renamed and then recovered.

28 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

NetWorker Recovery Wizard does not recover mount points under selected
recover point
NW149956

When using the NetWorker Recovery Wizard, files that appear under a selected recover
point will be recovered, but mount points under the selected recover point will not be
recovered.

NDMP single file backup succeeds but browsable recovery fails

NW149883

When a single file only is included in a saveset backup and the file is then selected for
recovery when performing a browsable recovery operation, the recovery fails with the
following error message:
Error during File NDMP Extraction
Workaround

Include more than one file in an NDMP backup.

The Total size and Number of files fields in the Recover page of NMC do not
display the updated values
NW149727
After performing a block based backup recovery, the Total size and Number of files fields
in the Recover page of NMC do not display the updated values. However, for a traditional
recovery, the Total size and Number of files fields display the updated values.

Remote client backup fails with local AFTD

NW149725
While creating a local Advanced File Type Device (AFTD) device, if you do not specify the
Uniform Naming Convention (UNC) path, the AFTD device is Client Direct disabled. If you
try to perform a backup to this AFTD device from a remote block based backup client, the
backup fails.

Workaround
1. Right-click the AFTD device that you created.
2. Select Modify device properties.
3. Under the General tab, in the Device access information field, specify the UNC path.
The AFTD device is Client Direct enabled.
4. Perform the backup.

EMC NetWorker Release 8.1 Release Notes 29

Known problems and limitations

Backup of the All and Disaster Recovery save sets fails when a File Level
Recovery is being performed on the same host
NW149724
To perform a recovery, you must mount the virtual hard disk the attribute of which will then
be read-only. The read-only files cannot be backed up. So, when you are performing a File
Level Recovery on a host, if you try to back up the All save set or the Disaster Recovery
save set on the same host before the recovery is completed, the backup fails.

Workaround

Perform the backup after the recovery is completed.

NMC does not display all the destination volumes when you perform a
block based backup recovery
NW149723
NMC displays one volume only when two or more volumes have the same serial number or
GUID.

Block based backup recovery fails when the Data Domain device
credentials and the Common Internet File System share credentials are
different
NW149722
To recover data from a Data Domain (DD) device, you must enable Common Internet File
System (CIFS) share to access save sets on the DD device. You must ensure that the
credentials that you provide while creating the CIFS share are same as the DD device
credentials. Otherwise, the recovery fails.

The size of the backed up basic volume save sets or dynamic volume save
sets is sometimes more than the size of the actual data
NW149720
After you back up basic volume save sets or dynamic volume save sets such as spanned,
striped, mirrored, RAID5, and so on, the size of the backed up data is typically slightly
more than the size of the actual data. Sometimes the size of the backed up data is
approximately triple the size of the actual data.
Perform the backup after the recovery is completed.

Virtual Machine Disk consolidation required when image-level backups of

a client fail using VMware Backup Appliance with NetWorker 8.1
NW149392

An image-level backup of a client using a VMware Backup Appliance with NetWorker 8.1
may fail with an error message indicating there are too many snapshots, even when there
are no snapshots for this particular client. When this occurs, a message appears in
vCenter events for this VM indicating that Virtual Machine Disk consolidation is needed.
Workaround

Run Virtual Machine Disk consolidation to remove any redundant snapshot files.

30 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Backup fails with System Reserved Partition error

NW149301

Backups fail with a System Reserved Partition (SRP) error:

device is not ready

Windows 2008 R2 Server has 100 MB as SRP. When backing up system state, VSS
includes the SRP (used for BitLocker and Boot files) but the backup fails because the SRP
is offline. This can occur if the Windows automount capability is disabled. Although there
are circumstances where the automount capability must be disabled, it can result in the
partition being offline after reboot.

Workaround

To work around this issue, use either of the following solutions:

Solution 1

From the command prompt, run DISKPART with the following commands:
C:\>DISKPART
C:\DISKPART> List volume
C:\DISKPART> Select volume <number of 100 MB system partition>
C:\DISKPART> Online volume (if the volume is offline)

The volumes can go offline if automount is disabled while using 3rd party storage
software or if the user manually disabled the automount for the volume.

This Microsoft article provides details on preventing volumes from going offline by
checking and setting the system automount attribute, KB2419286:
http://support.microsoft.com/kb/2419286

Solution 2

From the Disk Management console:

1. Access Disk Management, open command prompt:
C:\>Diskmgmt.msc

2. Assign the drive letter to the 100 MB partition to bring the disk online:
a. Right click the 100 MB volume and select Change Drive Letter and Paths….
b. Assign a new drive letter to the volume.
Assigning the drive letter ensures that the volume will be online after a reboot.

Expired VMware Backup Appliance backups to a DD Boost device may not

be removed from the Data Domain System for up to 48 hours
NW149172

When VMware Backup Appliance backups use a DD Boost device and data on the device
is expired (such as when the device is relabeled), data in the corresponding Avamar
MTree (managed tree) on the Data Domain system may not be removed for up to 48 hours
after the data on the DD Boost device has been expired.

EMC NetWorker Release 8.1 Release Notes 31

Known problems and limitations

Note: Data from the NetWorker Mtree gets deleted as expected, when the scheduled file
system clean runs on the DDR.

The Networker resource group in Autostart remains idle

NW148731

The Networker resource group in Autostart remains idle for “Online Pending” or “Offline
Pending”.

Scripts must be manually edited to correctly configure a highly available NetWorker

server.

Workaround

To work around this issue:

1. In config, take the resource group offline and modify it in Autostart.
2. In the NetWorker resource group Settings, edit the Startup Sequence script to add the
/yes switch to stop nsrexecd:
system("net stop nsrexecd /yes")
3. In the Shutdown Sequence script, add the following lines above nsrexecd:
system("net stop gstd")
sleep(15)
exit(0)

Recover wizard does not display the NSM recover statistics

NW148548

After an NSM recovery, the NetWorker Recover Wizard does not display statistics for the
recovery operation on its main screen. The expected statistics are total size of the
recovery and the number of files recovered.

Information on completed save set clones not reported when scheduled

clone is interrupted
NW148540

When a scheduled clone job is interrupted, the scheduled clone completion email and the
Clone window in NMC indicate that no save sets were cloned, even if some save set
cloning occurred prior to the interruption.

Recover wizard does not update NSM recover statistics in the display
NW148514

Following an NSM recovery operation with the NetWorker Recover Wizard, the main wizard
screen does not update the displayed fields for total size and number of the recovered
files as expected.

32 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

NMC Recovery Wizard does not support save set recovery for snapshots
NW148373, NW148371

The NMC Recovery Wizard user interface does not support the selection of entire snapshot
save sets for the restore operations. The wizard supports only file-by-file restores and
rollbacks.
Workaround

Use the nsrsnapadmin utility (with the R option) or the nsrsnap_recover command to
restore a save set.

NMC shortcut appears as a text icon instead of cube icon on Windows

2012 desktop
NW148319

Due to a Java Web Start application defect, when the NMC server is installed on a
Windows 2012 system, the NMC shortcut appears as a text icon instead of cube icon on
the desktop.
Workaround

The appearance if the icon can be fixed by changing the icon path to
%ProgramFiles%\EMC NetWorker\Management\GST\web\console.ico.

Cluster Shared Volumes (CSV) backup performed as level full instead of

incremental after the preferred server order list is modified
NW148309

NetWorker 8.1 supports the backup of Windows Server 2012 Cluster Shared Volumes
(CSV) with an optional preferred server order list (PSOL). The PSOL designates a proxy
cluster node as a backup source. When the PSOL is modified and the backup level is set
to incremental, NetWorker performs a full level backup, and then labels the backup as an
incremental backup.
Note: The backup level in savegroup reports shows that an incremental backup was
performed, but the number of files saved is consistent with a full backup level.

This behavior occurs only during the initial backup following modification of the PSOL. All
subsequent backup levels are created as configured.

The cause of this inconsistency after a PSOL modification is due to an interaction between
the default Backup renamed directories attribute and the timestamps of the CSV mount
points on each cluster node. Inconsistent timestamps cause NetWorker to flag the mount
point folder as renamed, resulting in the save of all files and folders on the mount point.
Workaround

Disable the Backup renamed directories attribute for the NetWorker client resource
associated with the CSV save set(s). The section “Support for backing up renamed
directories” in the NetWorker Administration Guide provides more information about this
attribute.

EMC NetWorker Release 8.1 Release Notes 33

Known problems and limitations

Clone storage node attribute not honored with a DD Boost device

NW148112

When you perform a CCR, NetWorker does not honor the Clone storage node attribute.
Workaround

Unselect the Media Type Required attribute for the destination clone pool, and then
perform the clone.

Microsoft VSS backups may contain inconsistent data that cannot be

restored due to a Microsoft-reported issue in the Windows Volume
Shadow Copy framework
NW148067 (esg135000)

When a volume that is extended on 8 GB boundaries is extended by a multiple greater

than 8 GB, a NetWorker backup that uses the Windows Volume Shadow Copy feature to
create snapshots may contain inconsistent data that cannot be restored, due to a known
Microsoft issue in the Windows Volume Shadow Copy. When this occurs, there will be no
indication that inconsistent data has been backed up, and the backups will be marked as
“Completed”. This issue affects NetWorker client systems running Windows Server 2003,
Windows Vista, Windows Server 2008, Windows 7, or Windows Server 2008 R2.

Additional information about the cause of this issue is available in the Microsoft KB
article 2748349:

http://support.microsoft.com/kb/2748349
Note: This limitation is related to extending volumes on an exact 8GB boundary. There is
no issue if:
-the volume is not extended, or
-the volume is extended, but not on 8GB boundaries.

Workaround

To fix this issue:

1. On the NetWorker client, install the Microsoft hotfix identified in the Microsoft KB
article 2748349.
2. Restart the computer after applying this update.
3. Allow the next scheduled backup to run.

VADP snapshot not supported on VM with MS iSCSI initiator

NW148050

VADP snapshot (quiesced backup) operations on the VM are not supported when using
Microsoft's iSCSI initiator within the VM. If quiesced backup operations are attempted on
a VM with Microsoft iSCSI LUNs attached within the guest operating system, virtual
machines may fail with the following error:
Creating a quiesced snapshot failed because the created snapshot
operation exceeded the time limit for holding off I/O in the frozen
virtual machine."

34 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

The following knowledgebase article provides more information:

http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displa
yKC&externalId=1009073

Clone details window in NMC does not list the status for the save sets
being cloned
NW148215

The Clone details window, accessed from the Monitoring tab in NMC, is blank while
cloning is in progress. The window gets populated with failed and successful save set
information only after the cloning completes. No progress details are available.

Mminfo -p does not display save sets for backup type SMTAPE
NW148022

When a backup is performed for backup type SMTAPE, running mminfo -p does not
display the save sets. The output of mminfo -p only displays save sets with the flag cbNs,
and the flag attached to these save sets is crNs.

Running large number of save sets in parallel may lead to backup failure
due to lack of swap space
NW147765, NW147756

When performing backups of a large number of save sets in parallel, your system may run
out of swap space, leading to backup failure. Ensure that you maintain a sufficient
amount of swap space if running a large number of save sets in parallel.

Scanning for bootstrap save set ID in NetWorker server disaster recovery

wizard not supported for non-English locales
NW147541

The option to scan for a bootstrap save set ID in the NetWorker server disaster recovery
wizard is not supported for non-English locales. In this case, use the scanner command to
find the bootstrap ID.

When using Immediate Cloning, only the first completed save set for each
job gets cloned immediately if the user-defined save set generates more
than one SSID
NW146964

During a group backup with the Clones attribute in the NMC Group resource enabled, the
cloning mode “start on each save set completion” (Immediate cloning) does not work as
expected when a user-defined save set can generate more than one SSID.

For example, in a Windows filesystem backup, defining one save set as ALL generates one
SSID each for C:\ , D:\ , E:\ , SRP , DISASTER_RECOVERY:\, index and bootstrap. NetWorker
clones the first save set for DISASTER_RECOVERY:\ immediately after the save set backup
completes, but only starts the clone operation for subsequent save sets after the backup
of all the remaining save sets and the index save sets complete.

EMC NetWorker Release 8.1 Release Notes 35

Known problems and limitations

Block based backup does not support NetWorker SnapImage Module

backups
NW146959

If the NetWorker SnapImage Module has been installed on a machine, and you either
install NetWorker 8.1 or upgrade the NetWorker version to 8.1 with the block based
backup option enabled on the machine, the NetWorker SnapImage Module backups fail.

Max-wait-time exceeded when using nsrconsolidate to rehydrate large

number of Avamar backups or to synthesize many save sets
NW146802

When using nsrconsolidate to rehydrate a large number of Avamar backups or to

synthesize many save sets at the same time, NetWorker exceeds the max-wait-time,
resulting in backup failure. Messages similar to the following appear:
nsrrecopy: Aborting nsrrecopy on host hostname for client clientname
save set savesetname, due to data not received for 600 seconds
Workaround

When performing a large number of rehydrated or synthetic full backups, divide the
ssids/save set names for specification on separate nsrconsolidate commands.

When the source directory is empty, Client Push reports transfer as

successful even though no files are transferred
NW145499

When the source directory (media path) is empty, transferring files by using the Client
Push option reports the transfer as successful even though no files are transferred.
Workaround

Ensure that the source directory (media path) contains files to be transferred.

Clicking back button in Software Administration wizard during upgrade

returns error
NW145338

In the Software Administration wizard, you cannot return to the previous step after the
upgrade operation is selected. Clicking the Back button in the Software Administration
wizard during an upgrade operation returns the message “Error, failed to initialize.”

When you specify a preferred node with the NSR_CSV_PSOL variable the
NMC savegroup details dialog does not report the backup size for a CSV
save set
NW145286

When performing a CSV backup for a client that uses the NSR_CSV_PSOL variable to
define the preferred node, the NMC savegroup details dialog may not include the backup
size information for the CSV volume save set. To find missing save set size information for
a CSV, view the savegroup completion notification report or the media data save set views
(in NMC or using mminfo).

36 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Software repository location cannot be changed using Software

Administration wizard
NW145280

The location of the software repository cannot be changed using the Software
Administration wizard. To change the location of the software repository, use the
nsradmin command. The section “Changing the location of the software repository” in the
NetWorker Server Management chapter of the NetWorker Administration Guide provides
details.

Client upgrade from CLI or Software Administration wizard cannot be

cancelled once started
NW145187

A Software Distribution upgrade of a client cannot be cancelled after being started from
the command line or the Software Administration wizard. Cancelling an upgrade
operation of a client from the command line or by using the Master job cancel button in
the Software Administration wizard fails.

Veritas Volume Manager on Windows not supported with NetWorker

NW145068

NetWorker 8.1 does not support Veritas Volume Manager (VxVM) on Windows. NetWorker
fails if a backup is attempted.

VMware Backup Appliance created with higher than normal default values
for device’s target session and max session
NW144852

Registering a VMware Backup Appliance with the NetWorker server results in higher than
normal default values created for the device’s target session and max session, with Target
session=50 and max session=200. Typically, the default values for a device created in
NetWorker (for example, a Data Domain device) are target session=10 and max
session=60.

Back button does not return correct page of Software Administration

wizard after selecting inventory option
NW144806

After performing an add to/delete from repository in the Software Administration wizard
and then selecting the inventory option, clicking the back button does not return the
previous page of the wizard.
Workaround

Close the Software Administration wizard and restart the operation.

nsr_end() does not report correct size for save sets larger than 2 TB
NW144544

Calling nsr_end() to retrieve save set size after a VADP backup reports the incorrect size if
the save set is larger than 2 TB.

EMC NetWorker Release 8.1 Release Notes 37

Known problems and limitations

Save Set ALL does not back up all volumes after image-level recovery of a
block-based backup if original volume was retained
NW142871

If a block-based backup is restored as an image-level recovery and the original backed up

volume is retained, subsequent Save Set ALL backups will not back up either the original
volume or the restored volume. This occurs because the image level recovery replicates
the entire volume, including the volume serial number, which results in two volumes with
the same serial number. During backup, if a duplication is discovered in the list of
volumes, NetWorker will remove the duplicated volume from the snapshot.
Workaround

When NetWorker detects the duplicate volume, a message is logged indicating which
volume will be skipped. If the volume that is not required is removed or moved to offline
state, then the other volume is backed up.

Limit of 8 Logical Unit Numbers (LUNs) per target in LEGACY mode on

HP-UX 11i v3
NW141678

In LEGACY mode on HPUX 11iv3, the NetWorker inquire command does not display
devices with LUN greater than 7. This is due to an operating system limitation on HPUX
11iv3. LEGACY mode addressing limits are 8 LUNs per target.

Upgrade to NetWorker 8.1 on Windows 2003 x64 client fails using

Software Distribution wizard due to missing MS Visual C++ 2005
Redistributable pkg
NW141532

When upgrading to NetWorker 8.1 on a Windows 2003 x64 client using the Software
Distribution wizard, the wizard fails to install the required MS Visual C++ 2005
Redistributable package. As a result, the upgrade fails.

NMC Client Configuration wizard allows creation of duplicate VADP proxies

NW124800

The NMC Client Configuration wizard currently allows for the creation of duplicate VADP
proxies, leading to multiple duplicate proxies with the same and different mount points.
EMC recommends not creating duplicate proxies, and having one proxy client with a
single mount point.

Recovery of application module datafiles occasionally fails

NW121443, NW105687, LGTsc19387

For NetWorker application modules, conventional recovery occasionally fails with the
destination directory PATH_NOT_FOUND error. This can occur for NSM backups of multiple
datafiles configured on multiple file systems that span more than one disk. The recovery
succeeds for some datafiles, but a failover to a previous backup occurs for the remaining
datafiles.

38 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Workaround

Run nonconcurrent save streams during the recover process. To perform this operation,
add NSR_MAX_STREAMS=1 to the RMAN recover configuration script as specified by
NSR_PROXY_PFILE. This value sets parallelism to 1.

vStorage_API: Directive drop down lists VCB instead of VADP

NW121940

VADP should be listed in the directive drop down list rather than VCB.

Recovery of VM to an ESX server different from the originating server may

fail due to different network adapter
NW120006

When you recover a VM to a different ESX server and the Network adapter that the VM
uses is not the default network adapter “VM Network”, network connectivity on the
recovered VM might fail.
Workaround

To resolve this issue:

1. Edit the settings for the VM in the vSphere client.
2. Delete the existing network adapter device.
3. Create a new network adapter device and select the appropriate network.
4. Reconfigure the networking within your operating system to associate an IP address to
the new network adapter.

NSM in VCS cluster setup does not remove the associated snapshot
sessions
NW102770, LGTsc17244

NSM in a VCS cluster setup does not remove the associated snapshot sessions.
Workaround

1. Manually remove the session by using Navisphere® software.

2. Keep all cluster nodes in the same VNX storage group.

When MSCS cluster disk is online, rollback should exit with a proper error
NW102236, LGTpa58948

If you perform a rollback when the disk is online in an MSCS cluster environment, the
rollback occurs and fails with a core dump. The expected outcome is that NSM exits
gracefully with a proper error message and the rollback does not occur.

This defect can occur in the following workflow:

1. Perform a backup on the virtual host.
2. Keep the disk online in cluster group.

EMC NetWorker Release 8.1 Release Notes 39

Known problems and limitations

3. Perform a rollback.

The rollback will fail with a resulting core dump.

Recovery of application module datafiles occasionally fails

NW080016, LGTsc19387

For NetWorker application modules, conventional recovery occasionally fails with the
destination directory PATH_NOT_FOUND error for NSM backups of multiple datafiles that
were created on multiple file systems that span more than one disk. The recovery
succeeds for some datafiles, but a failover to a previous backup occurs for the remaining
datafiles.
Workaround

Run non-concurrent save streams during the recover process. To perform this operation,
add NSR_MAX_STREAMS=1 to the RMAN recover configuration script as specified by
NSR_PROXY_PFILE. This will set parallelism to 1.

NetWorker 8.0 Service Pack releases

Table 8 on page 40 identifies problem issues and limitations discovered in NetWorker 8.0
SP1 and SP2.

Table 8 Limitations discovered in NetWorker release 8.0 SP1

Operating
Issue number Description of limitation system affected Product feature

“NW150781” Error may appear when recovering named pipes on Mac-OS, though the Mac-OS Recovery
on page 41 recovery completes successfully

“NW149533” Scanner command does not support use of m and f options together for Solaris CLI
on page 41 NDMP tape device

“NW137173, Backup and recovery of RDM disks not supported Windows Backup
NW131061”
on page 41

Optimized backup of Data Deduplication volumes not supported Windows Backup

“NW145009” VSS:NSR_SYS_WRITER_WIN32_SERVCOMP_USER=yes command only Windows Compatibility

on page 42 supported for Windows Server 2012 clients on NetWorker

“NW145942” VDDK 5.0 not compatible with Windows 2012 Windows Compatibility
on page 42

“NW136850” Using Client Configuration wizard to configure NetWorker 7.6 SP3 or AIX, Linux, Compatibility
on page 42 earlier client fails when server is NetWorker 8.0 SP1 Windows

“NW144804” Software Distribution wizard inventory fails for Solaris AMD 64 clients Solaris Configuration
on page 43

“NW143033” Unable to configure StorageTek silo on Windows 2012 Windows Configuration

on page 43

“NW147400” Upgrade of Windows 2003 R2 64-bit client using Software Distribution Windows Upgrading
on page 43 wizard fails

40 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Table 8 Limitations discovered in NetWorker release 8.0 SP1

Operating
Issue number Description of limitation system affected Product feature

“NW142082” Critical volume not restored after Windows BMR to a new VM using WinPE Windows Recovery
on page 43 4.0 x64 ISO

“NW145612” NetWorker remote code execution vulnerabilities All Security

on page 44
“NW145894”
on page 44

“NW146153, Global directives not supported for VADP backup and recovery All VMware
NW145248”
on page 44

Error may appear when recovering named pipes on Mac-OS, though the
recovery completes successfully
NW150781

When recovering named pipes on a Mac-OS, an error message may appear indicating that
the catalog file will not be recovered. NetWorker does, however, recover the file and the
recovery completes successfully, so this error can be ignored.

Scanner command does not support use of m and f options together for
NDMP tape device
NW149533

When using an NDMP tape device, the scanner command does not support the use of the
m and f options together. Avoid using the f option for NDMP tape devices. When run
without the f option, scanner works correctly without errors.

Backup and recovery of RDM disks not supported

NW137173, NW131061

NetWorker does not support backup and recovery of Virtual compatible raw disks (RDM).
During NetWorker backup of a VM, no RDM related information is backed up, and no RDM
disks/data are restored upon VM recovery. If RDM disks are required, they must be
reattached after the recovery.
Note: If reattaching RDM disks after recovery, make note of all LUNs that are zoned to the
protected VMs.

Optimized backup of Data Deduplication volumes not supported

NetWorker 8.0 does not support the optimized backup of a deduplicated volume. Files are
backed up at their full non-optimized size. As a result, at restore time the required restore
capacity may be bigger than the original deduplicated volume.

To enable scheduled backups of optimized de-duplicated volumes in a non-optimized

format:
1. From the NetWorker Administration window, select Configuration.
2. Select Clients.

EMC NetWorker Release 8.1 Release Notes 41

Known problems and limitations

3. Right-click the name of the client on which you want to perform non-optimized
backups, and then select Properties.
4. In the Properties dialog box, select the Apps & Modules tab.
5. In the Save Operations attribute, type the following command:
VSS:NSR_DEDUP_NON_OPTIMIZED=yes
6. Click OK.

To enable manual backups from the NetWorker User application (winworkr):

1. In Windows, add a Windows system environment variable and give it the following
value:
NSR_DEDUP_NON_OPTIMIZED=yes
2. Start winworker and perform the backup.

For more information on the Windows Server 2012 Data Deduplication feature, go to
Backup and Restore of Data Deduplication-Enabled Volumes at
http://msdn.microsoft.com/en-us/library/windows/desktop/hh769304(v=vs.85).aspx

VSS:NSR_SYS_WRITER_WIN32_SERVCOMP_USER=yes command only

supported for Windows Server 2012 clients on NetWorker
NW145009

The NetWorker client resource Save Operations attribute command

"VSS:NSR_SYS_WRITER_WIN32_SERVCOMP_USER=yes" is not supported for a NetWorker
Windows 8 client. It is currently only supported for NetWorker Windows Server 2012
clients.

The Windows Server 2012 System Writer separates application Win32 services files by
adding a new component, Win32 Services Files. This provided NetWorker the option to
backup and recover these files as part of a volume backup or system state backup.
However, the Windows 8 client System Writer does not separate these files, and as a
result, all application service files are classified as “System Files” and treated as part of
critical system state. Without the separate classification of these files by the Windows 8
client System Writer, NetWorker must process Win32 service files on Windows 8 as
system state.

VADP proxy backup and recovery not supported on Windows 2012

NW145942

NetWorker 8.0 SP1 as VADP proxy is not supported on Windows 2012 server. As a result,
VADP proxy backup and recovery fails if attempted in this configuration.

Using Client Configuration wizard to configure NetWorker 7.6 SP3 or

earlier client fails when server is NetWorker 8.0 SP1
NW136850

After the NetWorker servers are upgraded to NetWorker 8.0 SP1, using the NMC Client
Configuration wizard to configure a NetWorker 7.6 SP3 and earlier NetWorker Client will
fail.

42 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

If the NetWorker servers are upgraded to NetWorker 8.0 SP1, ensure that the NetWorker
client is upgraded to a minimum NetWorker 7.6 SP4.

Software Distribution wizard inventory fails for Solaris AMD 64 clients

NW144804

Performing an inventory of a SolAMD64 client using the Software Distribution wizard fails
if no supported version of the media kit is present in the repository.
Workaround

Add the current version of the Media kit to the repository. This enables the Software
Distribution wizard to inventory and upgrade the latest supported products.

Example An inventory of a SolAMD64 client with a Server version that supports the SolAMD 64
platform is successful when any Media Kit (that is the same version as the server) is
added to the Software Repository.

Unable to configure StorageTek silo on Windows 2012

NW143033

A communication error during silo connectivity causes the configuration of a StorageTek

silo on Windows 2012 to fail using either the jbconfig command or NMC.
Workaround

Verify connectivity with the silo using the ACSLS Configurator tool before running jbconfig
or using NMC to configure the silo.

Upgrade of Windows 2003 R2 64-bit client using Software Distribution

wizard fails
NW147400

Upgrading a Windows 2003 R2 64-bit client from the Windows server fails when using the
Software Distribution wizard. The upgrade must be performed manually.

Critical volume not restored after Windows BMR to a new VM using WinPE
4.0 x64 ISO
NW142082

After performing a Windows BMR to a system other than the original VM using the WinPE
4.0 x64 ISO image, the system does not bring storage spaces and other critical disks
online. A critical volume is swapped, with that volume’s data restored to the wrong disk.
For example, if more than one disk is available for restore and the disks were created in an
ordered list, then the first available disk is used to restore the data.

A Microsoft support incident is currently open for this issue and the Release Notes will be
updated when the issue is resolved.

EMC NetWorker Release 8.1 Release Notes 43

Known problems and limitations

NetWorker remote code execution vulnerabilities

The following security vulnerabilities exist in NetWorker 8.0.0.5 and earlier versions.
NW145612

A buffer overflow vulnerability exists in the EMC NetWorker nsrindexd RPC service that
could potentially be exploited by a malicious user to create a denial of service condition
or execute arbitrary code.
NW145894

EMC NetWorker provides some of its services through the SunRPC remote procedure call
mechanism. One of these services (nsrindexd), which listens on a dynamic port, exposes
a SunRPC interface. A buffer overflow vulnerability exists in this service that could
potentially be exploited by a malicious user to create a denial of service condition or
execute arbitrary code on the vulnerable system in the context of the affected application,
commonly system.
Workaround

The following EMC NetWorker releases address this issue:

◆ EMC NetWorker 8.0.0.6 and later
◆ EMC NetWorker 8.0 SP1 and later

EMC strongly recommends that all customers upgrade to these releases at the earliest
opportunity. Registered EMC Online Support customers can download software from
support.emc.com by selecting Support by Product and typing NetWorker. From this page,
select Downloads for the latest software downloads, and Documentation or Advisories for
product documentation.

Global directives not supported for VADP backup and recovery

NW146153,NW145248

Global directives are not supported by NetWorker for VADP backup and recovery. Both
encryption and compression directives result in backup failure.
Note: This issue is observed only in *FULL* and ALLVMFS workflows. FLR-disabled image
backups complete successfully.

44 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

NetWorker 8.0
Table 9 on page 45 identifies problem issues and limitations discovered in NetWorker
8.0.

Table 9 Limitations discovered in NetWorker release 8.0 (page 1 of 2)

Operating
Issue number Description of limitation system affected Product feature

“NW140994” Volume skipped when All-ntfs specified as save set in Windows client Windows Backup
on page 46 resource

“NW139076” Some save sets fail when running synthetic full backup if the previous Windows Backup
on page 46 backup was aborted

“NW138855” Independent disks that are skipped during VADP backup of guest VM are Windows Backup
on page 47 restored upon recovery

“NW138260” For synthetic full backups, add root “/” when specifying a save set name UNIX Backup
on page 47 for UNIX clients

“NW137334” Write completion delay on backups performed using NetWorker server Linux Backup
on page 47 previous to 7.6.3.3 and NetWorker 8.0 storage node

“NW136696” Synthetic full operations may fail of number of concurrent operations on Solaris Backup
on page 47 storage node exceeds client parallelism on target storage node

“NW136208” Renamed files might skipped from a synthetic full backup on Windows Windows Backup
on page 48

“NW137671” Concurrent clone jobs should not exceed 35 Solaris Backup

on page 48

“NW128078” Backup fails for certain save sets when VSS is disabled Windows Backup
on page 48

“NW140815” Bootstrap data that is recovered after executing mmrecov does not All CLI
on page 49 display in mminfo output

“NW138634” Inquire command does not detect iSCSI-based VTL on Solaris 11 x64 VM Solaris CLI
on page 49

“NW130706” nsrgrpcomp does not produce group completion summary or print All CLI
on page 49 completion reports

“NW136616” Creating NetWorker 8.0 clients using Client Configuration wizard fails if All Compatibility
on page 49 NetWorker server is not version 8.0

“NW136453” Oldauth mode cannot be used for NetWorker server if using with NMC 8.0 All Compatibility
on page 49

“NW128735” NetWorker server may fail to restart on Solaris 10 platform if NMC is on Solaris Compatibility
on page 50 same host

“NW141406” RECOVER_FROM_REP_HOST variable setting not honored when using Solaris Configuration
on page 50 nsrconsolidate for rehydration of Avamar deduplicated save sets on
replication node

“NW139511” HP ONCPlus package required on HP-UX 11.31 ia64 to perform Data HP Configuration/
on page 50 Domain operations Devices and Media

“NW137717” Use GMT time zone if failure occurs when specifying Maximum time All Configuration
on page 51 values for browse, retention and expiration policies

EMC NetWorker Release 8.1 Release Notes 45

Known problems and limitations

Table 9 Limitations discovered in NetWorker release 8.0 (page 2 of 2)

Operating
Issue number Description of limitation system affected Product feature

“NW136874” Read-only device of AFTD not mounted if AFTD configured on AIX, Linux Configuration/
on page 51 pre-NetWorker 7.6 SP3 server using NetWorker 8.0 storage node Devices and Media

“NW138470” LDAP configuration wizard shows all NetWorker servers managed by NMC All Configuration
on page 51 instead of only NetWorker 8.0 servers

“NW138408” Windows 8 clients cannot be created using Client Configuration wizard Windows Configuration
on page 51

“NW138085” NSR peer information certificate does not load for NSR client All Configuration
on page 52

“LGTsc15813 Changes made to external role mapping for server do not display until Windows GUI
” on page 52 NMC is restarted

“NW138208” License conformance summary in NMC displays incorrect number of All GUI
on page 52 licenses used for NetWorker Fast Start

“NW138204” NetWorker storage node device properties continue to include Read-only Windows GUI
on page 52 devices after upgrade from NetWorker 7.6 SP2

“NW137926” nsrstage exists if recovery performed at the same time as staging and Windows Recovery
on page 52 cloning completes before recovery

“NW133500” NetWorker client (nsrexecd) does not start after recovering VSS save sets Windows Recovery
on page 53 from a previous release

“NW130140” nsrndmp_recover proceeds with assertion and fails intermittently Windows Recovery
on page 53

“NW136889” VM backup not supported for clients in suspended state Windows VMware
on page 53

Backup problems and limitations

Volume skipped when All-ntfs specified as save set in Windows client
resource
NW140994

The first volume of the NTFS volumes on the client is skipped during a savegrp probe and
savegrp scheduled backup when All-ntfs is specified as the save set in the Windows client
resource.

Some save sets fail when running synthetic full backup if the previous
backup was aborted
NW139076

When running a synthetic full backup, some save sets fail with the error “Failed to start
save session with <Server>” if the previous backup was aborted. This is due to a limit to
the number of nsrconsolidate sessions that can be run in parallel. The limit is defined as
the server's client parallelism value, divided by 2.

46 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Workaround

Under these circumstances, increase the server's client parallelism or set the client
parallelism to the max value. Recommendations for client parallelism settings are
provided in Chapter 17, “NetWorker Server Management,” of the NetWorker
Administration Guide.

Independent disks that are skipped during VADP backup of guest VM are
restored upon recovery
NW138855

During a VADP backup of a guest VM with Independent disks, the backup logs indicate
that the independent disks are not included as part of the backup and will be skipped, as
expected. However, when recovering the guest VM, these independent disks are
re-attached with the VM. These independent disks do not contain any data and are visible
as unformatted disks.

For synthetic full backups, add root “/” when specifying a save set name
for UNIX clients
NW138260

For UNIX clients, include root “/” when specifying a save set name for synthetic full
backup. Not including root “/” might cause the backup to fail and the following error
message to display:
nsrconsolidate: Saveset <save sets -name> for ssid <save set id> does
not start with '/'

For example, if /tmp is misspelled as tmp in the save set list of the Client resource, the
backup will fail.

Write completion delay on backups performed using NetWorker server

previous to 7.6.3.3 and NetWorker 8.0 storage node
NW137334

When a backup is performed using a NetWorker server previous to version 7.6.3.3 and a
NetWorker 8.0 storage node, write completion does not occur until several minutes after
the savegrp reports the backup completed successfully. If a recovery is attempted within
that time window, the recovery will not start until the write completion notice has been
written to the logs.

Synthetic full operations may fail of number of concurrent operations on

storage node exceeds client parallelism on target storage node
NW136696

If the number of concurrent synthetic full operations on a storage node exceeds the client
parallelism of the target storage node, some of the synthetic full operations may fail.
Workaround

Increase the client parallelism of the storage nodes to a larger value, or limit the number
of concurrent synthetic full operations.

EMC NetWorker Release 8.1 Release Notes 47

Known problems and limitations

Renamed files might be skipped from a synthetic full backup on Windows

NW136208

When scheduling a full backup on Windows platforms, configure the backup levels to be
equal to or greater than the previous backup level. This prevents renamed files from being
skipped when the synthetic full backup is created. For example, if mixed backup levels
are run before a synthetic full backup is run, renamed files might not be included in the
synthetic full backup since the level backup following the incremental backup might not
include the renamed files.

In this example, a synth_full backup is run after a series mixed level backups on
Windows. After each backup, user data is modified where files and directories are
renamed and added. The data does not change between the last backup and the
synth_full backup:

2733595803 full
2700041446 1
2666487094 incr
2632932738 5
2599378382 incr
2565824025 9
2532269678 full <<< synth_full

When a synthetic full backup is run in this scenario, the level 5 and level 9 backups might
not include the files that were backed up at the previous incr backup if any files were
renamed before the incr backup. In that case, a synthetic full backup that covers save sets
at level full, 5 and 9 will not include renamed file backed up at level incr.
Workaround

To prevent renamed files from being skipped from a synthetic full backup on Windows,
consider the following when scheduling a backup:
1. Configure Windows clients within a dedicated group; not mixed with UNIX clients.
2. Ensure that the backup levels in the schedule are equal to or greater than the previous
backup level.

Cloning and staging problems and limitations

Concurrent clone jobs should not exceed 35
NW137671

It is recommended that the number of clone jobs running concurrently not exceed 35.
When the number of sessions is greater than 35, performance issues may be observed. If
problems are encountered, restart the NetWorker server.

Backup fails for certain save sets when VSS is disabled

NW128078

When VSS is disabled (VSS:*=OFF) on Windows 7, Windows 2008, and later versions of
the Windows operating system, VSS backups will not occur and backing up the following
save sets for a NetWorker client resource yields these results:
◆ DISASTER_RECOVERY:\ save set: Backup fails at the beginning of backup operation.
◆ VSS SYSTEM STATE save sets: Backup fails.

48 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

◆ VSS SYSTEM STATE save sets and volume save sets: Volumes are backed up but VSS
save sets fail.
◆ ALL save set: Backups fail.

Command line problems and limitations

Bootstrap data that is recovered after executing mmrecov does not display
in mminfo output
NW140815

The bootstrap data that is recovered after executing mmrecov does not display in the
mminfo command output. The media database is backed up as part of the bootstrap, and
that save set information is saved in the online media database (the media database
recovered from a bootstrap is from a point-in-time before that same bootstrap save set
information was added to the media database).

Inquire command does not detect iSCSI-based VTL on Solaris 11 x64 VM

NW138634

Running the inquire command from a NetWorker server with an iSCSI-based VTL
configured on a Solaris 11 x64 bit VM fails to detect the drives, slots and tapes assigned
in the iSCSI VTL.

nsrgrpcomp does not produce group completion summary or print

completion reports
NW130706

Specifying the nsrsgrpcomp command with the -S option (for the group completion
summary report) or the -C option (for the print completion message) does not produce
these reports.

Compatibility problems and limitations

Creating NetWorker 8.0 clients using Client Configuration wizard fails if
NetWorker server is not version 8.0
NW136616

Creating a NetWorker 8.0 client using the NMC Client Configuration wizard fails if the
NetWorker server is a version previous to 8.0.
Workaround

Manually create the client.

Oldauth mode cannot be used for NetWorker server if using with NMC 8.0
NW136453

A NetWorker 7.5.x, 7.6.x or 8.0 server in oldauth mode cannot be used with NMC 8.0,
which by default is in nsrauth/oldauth mode.

EMC NetWorker Release 8.1 Release Notes 49

Known problems and limitations

Workaround

Set the NetWorker server to nsrauth or nsrauth/oldauth mode.

NetWorker server may fail to restart on Solaris 10 platform if NMC is on

same host
NW128735

If NMC is running on the same host as the Networker server on a Solaris 10 platform, the
nsrexecd daemon may fail to restart. If this occurs, a socket binding error displays.
Workaround

Install one of the following patches:

◆ Solaris x86: 147441-04
◆ Solaris SPARC: 147440-04

Configuration problems and limitations

RECOVER_FROM_REP_HOST variable setting not honored when using
nsrconsolidate for rehydration of Avamar deduplicated save sets on
replication node
NW141406

Setting the RECOVER_FROM_REP_HOST variable to yes on the NetWorker server and

running nsrconsolidate to perform rehydration of Avamar deduplicated save sets fails to
rehydrate the save sets located on an Avamar replication node. Instead, the
nsrconsolidate output indicates that the primary Avamar server was used as the source
for rehydration.

HP ONCPlus package required on HP-UX 11.31 ia64 and HP-UX PA-RISC to

perform Data Domain operations
NW139511

Data Domain device operations, such as labelling of a Data Domain volume, fail for
NetWorker on HP-UX 11.31 ia64 and HP-UX PA-RISC, due to a dependency on an updated
version of the HP ONCPlus (Open Network Computing) package.
Workaround

Upgrade the HPUX ONCPlus package, which contains the NFS client, to 11.31.06 or later.
ONCPlus can be downloaded from HP at:

https://h20392.www2.hp.com/portal/swdepot/displayProductInfo.do?productNumber=
ONCplus

50 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Use GMT time zone if failure occurs when specifying Maximum time values
for browse, retention and expiration policies
NW137717

When the maximum browse, retention, or expiration policy times specified in the
command line are set close to the maximum allowable in a time zone other than GMT,
NetWorker might not allow these times. If this is the case, an error occurs similar to the
following for a local time zone 4 hours east of GMT:
# save … -y "01/01/2039 03:59:59" test.txt
5803:save: invalid retention time: 01/01/2039 03:59:59
Workaround

Use the GMT time zone to specify these time values. For example, in the above instance,
the line should be specified as:
# save … -y "12/31/2038 23:59:59 GMT" test.txt

If you specified policy times using a time zone other than GMT and no error occurred, this
change is not necessary.
Note: The UNIX man page for nsr_getdate (this man page also applies to Windows)
provides more information on general date input issues under the BUGS section in the
man page.

Read-only device of AFTD not mounted if AFTD configured on

pre-NetWorker 7.6 SP3 server using NetWorker 8.0 storage node
NW136874

When an AFTD is configured on a NetWorker server previous to NetWorker 7.6 SP3 using a
NetWorker 8.0 storage node, the read-only device of the AFTD does not get mounted.
Workaround

The read-only device must be mounted manually.

LDAP configuration wizard shows all NetWorker servers managed by NMC

instead of only NetWorker 8.0 servers
NW138470

The Configure Login Authentication page of the NMC LDAP configuration wizard displays a
list of NetWorker servers for which the CST configuration can be distributed. Only
NetWorker 8.0 and later servers should appear in this list, however, the list currently
shows all NetWorker servers managed by NMC. Ensure that you select only NetWorker 8.0
and later servers.

Windows 8 clients cannot be created using Client Configuration wizard

NW138408

NetWorker 8.0 does not support creating Windows 8 clients using the NMC Client
configuration wizard. You must create these clients manually.

EMC NetWorker Release 8.1 Release Notes 51

Known problems and limitations

NSR peer information certificate does not load for NSR client
NW138085

The NSR peer information certificate file for a NSR client does not load. An error displays
indicating “operation not applicable”.
Workaround

Regenerate the NSR peer certificate and clear that client's NSR peer certificate from other
hosts.

GUI problems and limitations

Changes made to external role mapping for server do not display until
NMC is restarted
LGTsc15813

When making a change to privileges for an external user in any directory server (Active
Directory / OpenLDAP environment), NMC does not immediately display the change. A
restart of NMC is required for the change to appear.

License conformance summary in NMC displays incorrect number of

licenses used for NetWorker Fast Start
NW138208

In the License conformance summary dialog in NMC, the Number Used field under Disk
Backup reflects the number of devices in use instead of the number of licenses in use for
NetWorker Fast Start.

NetWorker storage node device properties continue to include Read-only

devices after upgrade from NetWorker 7.6 SP2
NW138204

After upgrading from NetWorker 7.6 SP2 to NetWorker 8.0, the NetWorker storage node
Device properties continue to include read-only (.RO) devices that existed previously in
the device count. If any user scripts rely on this value, these scripts may require updating.

The number of devices adjusts to the current number after shutting down and restarting
NMC.

Recovery problems and limitations

nsrstage exists if recovery performed at the same time as staging and
cloning completes before recovery
NW137926

If a recovery is performed at the same time as a staging operation, and cloning completes
before the recovery, nsrstage exits and an error displays indicating that recovering space
from the volume failed.

52 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Workaround

When the recovery completes, run the following command to reclaim space on the source
volume:
nsrstage -C -V AFTD.001

The -C option instructs nsrstage to perform a volume cleaning operation by scanning a

volume for save sets with no entries in the media database and recovering that space.
Space for recyclable and aborted save sets is also recovered from the volume, with the
save set entries removed from the media database. You can perform this operation on
disk family volumes.

NetWorker client (nsrexecd) does not start after recovering VSS save sets
from a previous release
NW133500

The NetWorker client process (nsrexecd) does not start after performing a recovery of VSS
save sets that were backed up using a NetWorker release 7.5 SP4 or earlier (for example,
if a NetWorker 7.5 SP4 VSS backup is recovered using NetWorker 8.0).
Workaround

Reinstall the previous version of NetWorker to perform the VSS recovery, and then
upgrade the NetWorker client to the latest version and perform a full backup.

nsrndmp_recover proceeds with assertion and fails intermittently

NW130140

When nsrndmp_recover proceeds with assertion, the recovery intermittently fails.

Assertions do not affect functionality.

VMware problems and limitations

VM backup not supported for clients in suspended state
NW136889

NetWorker does not support the backup of virtual machines that are in suspended state.
Backing up a VM with clients in suspended state fails and returns the following error:
Virtual machine xxxxx is currently suspended. NetWorker does not back
up the virtual machines in a suspended state.

EMC NetWorker Release 8.1 Release Notes 53

Known problems and limitations

NetWorker 7.6 Service Pack 3 and later 7.6 releases

Table 10 on page 54 identifies problem issues and limitations discovered in NetWorker
7.6 SP3, 7.6 SP4, and 7.6 SP5.

Table 10 Limitations discovered in NetWorker release 7.6 SP3 and later

Operating
Issue number Description of limitation system affected Product feature

“NW133490” VSS backup performed with NetWorker 7.6 SP2 cannot be recovered using Windows Backup
on page 54 NetWorker 7.6 SP3

“NW131610” Index and bootstrap save times changed in Savegroup completion report All Backup
on page 55 to reflect job completion time instead of session completion time

“NW113943” Running inquire command displays auto-changers but not tape drives on Solaris Command Line
on page 55 Solaris 10 and Solaris 11 SPARC platforms interface

“NW136068” Notification commands moved from /usr/ucb to /usr/bin on Solaris 11 Solaris Configuration
on page 55

“NW134987” Client resource cannot be created for a NetWorker 7.6 SP3 client using the Solaris Configuration
on page 56 NMC Client Configuration wizard when the NMC version is at NetWorker
7.6 SP2 or earlier

“NW133542” Deduplication (Avamar and Data Domain) not supported in NetWorker 7.6 AIX Deduplication
on page 56 SP3 running on AIX 7.1 and Solaris 11 platform

“NW120687” “curphyhost” cannot be selected using NMC Client Configuration wizard Windows GUI
on page 56

“NW133391, NetWorker server process (nsrd) becomes unresponsive if started on a Windows Installation
NW132813” host with Powerpath 5.3 or 5.6 installed
on page 56

“NW135238” Permissions to base directory may be changed after save set recovery if All Recovery
on page 57 NTFS directory contents include current and parent directories (. and ..)

“NW134799” Recovery of archived save sets/files dumps core if archived services is not All Recovery
on page 57 enabled

“NW133500” NetWorker client (nsrexecd) does not start after recovering VSS save sets Windows Recovery
on page 53 from a previous release

“NW133449” winworkr terminates after disaster recovery of a VM using nbd mode Windows Recovery
on page 57

“NW132405” DAR recovery does not recover ACL attributes in NetApp NDMP restore All Recovery
on page 57

“NW132961” NBDSSL transport mode not supported for VM recovery Windows VMware
on page 57

VSS backup performed with NetWorker 7.6 SP2 cannot be recovered using
NetWorker 7.6 SP3
NW133490

A VSS backup that was performed using NetWorker 7.6 SP2 cannot be recovered using
NetWorker 7.6 SP3. The VSS system file recovery fails with an error similar to the
following:
VSS BCD BOOT FILES has not been successfully recovered.

54 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Workaround

After performing the upgrade, a Full level backup of the system must be performed so that
the VSS system save sets can be recovered.
Note: Only the VSS system save sets cannot be recovered from the NetWorker 7.6 SP2
backup; all other data can be recovered using the normal recovery procedure.

Index and bootstrap save times changed in Savegroup completion report

to reflect job completion time instead of session completion time
NW131610

In NetWorker 7.6 SP3, the Savegroup completion report has been changed to report the
index and bootstrap backup times based on the start and end times of jobs (job
completion time) instead of session completion time. These completion times are
displayed for the group resource in the Status tab of the Completion section, and in the
output when savegrp is run from the command line.

The NMC Group Details window still reports the index and bootstrap backup time based
on the session completion time.

Running inquire command displays auto-changers but not tape drives on

Solaris 10 and Solaris 11 SPARC platforms
NW113943

Running the NetWorker inquire command on the Solaris 10 and Solaris 11 SPARC
platforms only displays auto-changers; tape devices are not displayed.
Workaround

Re-order items in the /etc/name_to_major file so that the st driver is listed before any *st
drivers.

Notification commands moved from /usr/ucb to /usr/bin on Solaris 11

NW136068

Notification commands such as mail and logger have been moved from /usr/ucb to
/usr/bin on Solaris 11. As a result, messages are not updated.
Workaround

Create symbolic links to point to the new location of the commands, as in the following:
# mkdir /usr/ucb
# ln -s /usr/bin/logger /usr/ucb/logger
# ln -s /usr/bin/mail /usr/ucb/mail

EMC NetWorker Release 8.1 Release Notes 55

Known problems and limitations

Client resource cannot be created for a NetWorker 7.6 SP3 client using the
NMC Client Configuration wizard when the NMC version is NetWorker 7.6
SP2 or earlier
NW134987

Creating a Client resource for a NetWorker 7.6 SP3 client using the NMC Client
Configuration wizard fails if the NMC version is 7.6 SP2 or earlier, due to differences in
certification between NetWorker 7.6 SP2 and earlier and the NetWorker 7.6 SP3 client
software.
Workaround

Perform one of the following as a workaround:

◆ Use the NMC client properties or nsradmin to create the Networker Client.
◆ Upgrade the NMC version to NetWorker 7.6 SP3.

Deduplication (Avamar and Data Domain) not supported in NetWorker 7.6

SP3 running on AIX 7.1 and Solaris 11 platforms
NW133542

Due to libraries that are not supported by the AIX 7.1 and Solaris 11 platforms,
deduplication (Avamar and Data Domain features) cannot be used in NetWorker 7.6 SP3
running on AIX 7.1 or Solaris 11 variants.

“curphyhost” cannot be selected using NMC Client Configuration wizard

NW120687

The option to select “curphyhost” as the storage node using NMC’s Client Configuration
wizard is not available. If the wizard is used on a client that has “curphyhost” defined, it
will be removed from the client definition.

You must manually create the client to specify “curphyhost” as the storage node.

NetWorker server process (nsrd) becomes unresponsive if started on a

host with Powerpath 5.3 or 5.6 installed
NW133391, NW132813

The NetWorker server process nsrd becomes unresponsive after starting NetWorker if
Powerpath version 5.3 or 5.6 is already installed. This issue does not occur with
Powerpath 5.5.
Workaround

Install Powerpath version 5.5, if both Powerpath and NetWorker need to co-exist on the
same host.

56 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Permissions to base directory may be changed after save set recovery if

NTFS directory contents include current and parent directories (. and ..)
NW135238

When performing a recovery, a list is created that includes all files that are required to
rebuild the file hierarchy. The NTFS directory contents will include references to “.” and
“..” (current and parent directories) within the file table and these files are included in the
save set recovery. As a result the Permissions to the parent and base directory (where the
data is being restored to) may be changed.
Workaround

Recover the data to a different location, and then copy the data to the desired location
and set the file attributes as required.

Recovery of archived save sets/files dumps core if archived services is not

enabled
NW134799

When performing a recovery of archived save sets/files, ensure that archived services for
the client is enabled. If archived services is not enabled, a core dump occurs.

winworkr terminates after disaster recovery of a VM using nbd mode

NW133449

When a disaster recovery of a VM is performed using the nbd transport mode in winworkr,
winworkr terminates. The VM still starts up without any impact and the VADP recovery
completes successfully.

DAR recovery does not recover ACL attributes in NetApp NDMP restore
NW132405

Direct access recovery (DAR) or file-by-file restore does not properly recover the ACL
attributes in a NetApp NDMP restore.
Workaround

When restoring ACL attributes, enable the DDAR restore and mark the entire directory
where the file is located prior to recovering.
Note: If the recovery involves multiple level backups, Networker disables the DDAR
restore, even if it was manually enabled prior to recovery. In this occurs, ACL recover may
fail. To avoid this issue, do not to restore from multiple level backups in a single recover
session, especially if ACL information is required.

VM recovery fails for NBDSSL and hotadd transport modes

NW132961

Recovery of a VM in NetWorker fails for the transport mode NBDSSL and for hotadd mode
for ESX 5.0 and with VC 5.0.

EMC NetWorker Release 8.1 Release Notes 57

Known problems and limitations

Workaround

Before performing the following steps, ensure that you delete any snapshots that are
active on the VM. Do not power on the VM until these steps have been performed.
1. Right click the VM and select Edit settings.
2. Select the virtual hard disk and select Remove but do not delete the VMDK. Click OK.
3. Return to the Edit settings menu and select Add.
4. Choose Hard Disk and use an existing virtual disk.
5. Associate the new hard disk with the VMDK file, then click OK. For example, use the
Add disk pop-up window and add the hard disks by pointing them to the correct VMDK
file in the datastore.
6. Power on the VM.

NetWorker 7.6 Service Pack 2

Table 11 on page 58 identifies problem issues and limitations discovered in NetWorker
7.6. Service Pack 2.

The known limitations are separated into the following categories:

◆ “Backup problems and limitations” on page 63
◆ “CLI problems and limitations” on page 63
◆ “Cloning and staging problems and limitations” on page 64
◆ “Configuration problems and limitations” on page 64
◆ “Devices and media problems and limitations” on page 65
◆ “GUI problems and limitations” on page 67
◆ “Messaging problems and limitations” on page 70
◆ “Restore problems and limitations” on page 71
◆ “Security problems and limitations” on page 72
◆ “Windows support problems and limitations” on page 85

Table 11 Limitations discovered in NetWorker release 7.6 Service Pack 2 (page 1 of 6)

Operating
Issue number Description of limitation system affected Product feature

“NW128451” Savegroup may fall back to “agent assistd indirect save” Windows Backup
on page 63

“NW124240” Manual backups with save command fail on Windows remote client from Windows Backup
on page 63 non-root user account

“NW124330” SFR enabled backup not supported for NTFS partitions residing on IDE All Backup
on page 63 disks

“NW122242” mminfo does not reset space usage after aborted savegroup Solaris CLI
on page 63

58 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Table 11 Limitations discovered in NetWorker release 7.6 Service Pack 2 (page 2 of 6)

Operating
Issue number Description of limitation system affected Product feature

“NW124967” mminfo -mv and NMC show incorrect file type device (FTD) volume size Windows CLI
on page 64 after stage operation

“NW123160” Recovery of a clone volume fails if the original backup was marked as All Cloning and
on page 64 suspect Staging

“NW120358” Data Domain clone job records not flagged in jobsdb All Cloning and
on page 64 Staging

“NW127039” Remote access attribute for NetWorker client resource of the NetWorker All Configuration
on page 65 server must list names of all cluster nodes to ensure lockbox functionality
on NetWorker server’s cluster node

“NW127956” Active read sessions limited to 1 when recovering data from a Data Domain Windows Devices and
on page 66 device Media

“NW128111” DD Archiver storage systems are limited to one data movement policy per All Devices and
on page 66 NetWorker datazone by default Media

“NW125290” Device Configuration wizard cannot be used to label CIFS share AFTD on a All Devices and
on page 66 Data Domain system Media

“NW124165” System error detected when labeling a Data Domain device Windows Devices and
on page 66 Media

“NW123401” Data Domain features cannot be enabled through NMC on AIX 64-bit AIX Devices and
on page 67 Media

“NW120013” AFTD percentage capacity not updated after capacity is changed and device All Devices and
on page 67 is relabeled Media

“NW125024” Selecting AFTD on first window of Device Configuration wizard returns error Linux Devices and
on page 67 Media

“NW122973” Alert not displayed in NMC on the first day of evaluation license Windows GUI
on page 67

“NW125853” Drive letter marked in red in Recover browser if drive is not on proxy server Windows GUI
on page 68

“NW125786” Scan operation fails in NetWorker User when browse sessions are open for Windows GUI
on page 68 a client while the client’s save set is being scanned into the NetWorker
server

“NW125671” NMC may not display SNMP traps when traps are sent from Data Domain All GUI
on page 68 boxes running DDR OS 5.0 and multiple NICs are configured

“NW123840” Help topic does not display for NetWorker Clone reports in NMC All GUI
on page 68

“NW124145” Non-readable characters displayed in the NetWorker servers search in Windows GUI
on page 68 Winworkr

“NW119577” Multiple browse sessions displayed in NMC during single step recovery Windows GUI
on page 68

“NW124476” NetWorker client does not recognize HomeBase path and Profiling fails if Windows Installation and
on page 69 HomeBase Agent is installed manually upgrading

“NW119766” Changes to handling of the nwclust.pl script when upgrading to NetWorker All Installation and
on page 69 7.6 SP2 in an AutoStart 5.4 and later cluster upgrading

EMC NetWorker Release 8.1 Release Notes 59

Known problems and limitations

Table 11 Limitations discovered in NetWorker release 7.6 Service Pack 2 (page 3 of 6)

Operating
Issue number Description of limitation system affected Product feature

“NW122342” Warning message appears when starting NetWorker services on UNIX UNIX Messaging
on page 70

“NW127028” Directed recovery to CIFS share on destination client is not supported All Restore
on page 71

“NW126148” Bare Metal Recovery using HomeBase and a version of NetWorker different Windows Restore
on page 71 from the one used for backup may result in dll mismatch

“NW126007” Disaster recovery logs display “recover completion time” for metadata Windows Restore
on page 72 recovery instead of indicating that recovery of metadata was successful or
identifying this as “recovery of metadata completion time”

“NW125168” Empty file gets created after recovery fails due to incorrect passphrase Windows Restore
on page 72

“NW122275” Encryption attribute not set on encrypted folder after recovery Windows Security
on page 72

“NW123237” Volumes that are BitLocker encrypted will not be encrypted after recovery Windows Security
on page 72

“NW129816” VADP only supports English Windows operating systems for vCenter/Virtual Windows VMware support
on page 72 Center and VADP proxy

“NW129735” Incorrect proxy information displayed in NMC for VM client resource after Windows VMware support
on page 73 upgrading the NetWorker server to 7.6 SP2

“NW129337” Backup of a Windows 2003 VM is not browsable Windows VMware support

on page 73

“NW129143” Recovery dumps core for traditional save set when VADP and traditional Windows VMware support
on page 74 save sets co-exist on Linux

“NW129041” NMC does not display VMs when auto-discovery is run on Linux with Windows VMware support
on page 74 vCenter 4.1 update 1

“NW129093” Unable to browse guest based backups on non NTFS filesystem (applies to All VMware support
on page 74 NetWorker 7.6.2.1 build 638 and later)

“NW127928” Recovering non-Windows VMs that have a mix of VADP image-level and UNIX, Linux VMware support
on page 75 client-based backups (fixed in NetWorker 7.6.2.1 build 638 and later)

“NW127589” Transport mode falls back to NBD mode when configured transport modes Windows VMware support
on page 76 SAN or hotadd are not available on non-NTFS VMs

“NW127651” CBT tool does not return error when attempting to use with non-NTFS VMs Windows VMware support
on page 76

“NW127446” Non-NTFS thin disk VMs get restored as thick disk during disaster recover Windows VMware support
on page 76

“NW127004” FULLVM backup of a NTFS VADP client falls back to non-NTFS mode if the Windows VMware support
on page 76 drive has an unformatted partition

“NW126608” VADP backup fails with VDDK error when specified transport mode is not Windows VMware support
on page 77 available

“NW126553” VM does not power on after disaster recovery of CBT-enabled backup Windows VMware support
on page 77

“NW126457” NetWorker User program cannot be used for VADP recovery in non-English Windows VMware support
on page 77 locale

60 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Table 11 Limitations discovered in NetWorker release 7.6 Service Pack 2 (page 4 of 6)

Operating
Issue number Description of limitation system affected Product feature

“NW126321” Guest backups of Windows 2008 R2 and Windows 2007 may fail on ESX Windows VMware support
on page 77 4.0

“NW125927” SYSTEM accounts must have Full Control permissions for all NTFS disks to Windows VMware support
on page 78 back up Windows VMs

“NW125800” Virtual compatible raw disk restored as local virtual disk Windows VMware support
on page 78

“NW125515, Using SAN transport mode results in slow performance for disaster recovery Windows VMware support
NW125130” from physical proxy of VM with thin disks
on page 78

“NW125230, Recovery of a VM using SAN or hotadd transport mode requires policy be Windows VMware support
NW125142” set to OnlineALL
on page 78

“NW125212” Savegroup output of VADP ALLVMFS backup still refers to VCB directive Windows VMware support
on page 79

“NW125179” “Release disk lease” messages appear during SAN mode backup Windows VMware support
on page 79

“NW125164” FULLVM level -0 backup must be performed after enabling CBT on the VM Windows VMware support
on page 79

“NW125084” NetWorker uninstallation on VADP proxy host Windows VMware support

on page 79

“NW124883” Application information attributes for VADP backups should be added to all Windows VMware support
on page 79 resources if multiple instances are created with the same proxy name

“NW124800” NMC Client Configuration wizard allowing duplicate proxy clients to be Solaris VMware support
on page 80 created and configured on different mount points

“NW124730” Using hotadd transport mode to access a datastore directly attached to a Windows VMware support
on page 80 remote ESX host may return VDDK error

“NW124431” Temporary mnt folder gets created when recovering data from CBT-enabled Windows VMware support
on page 80 VM backup

“NW123719” vSphere Client cannot be launched from NMC for VMware Infrastructure All VMware support
on page 80 Client 2.5

“NW123164” VADP backup may fail for VM name containing special characters Windows VMware support
on page 80

“NW123219” Progress messages display during VADP backup and recovery Windows VMware support
on page 81

“NW123142” Port ID does not get restored after disaster recovery of a VM originally Windows VMware support
on page 81 connected to dvSwitch

“NW122619” VADP backup fails if Datastore name contains the characters % & = ? All VMware support
on page 81

“NW122586” Restoring a VM back to vApp container fails with error “A specified All VMware support
on page 81 parameter was not correct”

“NW122566” Client Configuration wizard cannot be used to configure default client as Windows VMware support
on page 82 proxy if NetWorker server and proxy client are used on the same machine

EMC NetWorker Release 8.1 Release Notes 61

Known problems and limitations

Table 11 Limitations discovered in NetWorker release 7.6 Service Pack 2 (page 5 of 6)

Operating
Issue number Description of limitation system affected Product feature

“NW122389” Web browser displays a 404 Not found or similar error the first time the Windows VMware support
on page 82 vSphere client is run from NMC if the default port number of the Virtual
Center server has been changed

“NW122381” All save sets (NTFS and non-NTFS backups) marked as browsable Windows VMware support
on page 82

“NW122197” Backup of VM on ESX 3.5 fails for VMs using short name All VMware support
on page 82

“NW122090” Incorrect warning message appears in Savegroup log file during non-NTFS Windows VMware support
on page 83 backup of VM with independent persistent disk

“NW121958” VM backup fails with HTTP Error 500 in cases where VM resides in a VMFS Windows VMware support
on page 83 datastore which is shared by multiple ESX/ESXi hosts or in cases where the
VM resides in HA/DRS Cluster setup

“NW121168” Incremental backup of a VM with image-based backups All VMware support

on page 83

“NW121068” VADP_* and VCB_* application information attributes both supported for Windows VMware support
on page 83 VADP implementation

“NW120474” VM cannot be restored to folder “Discovered Virtual Machine Windows VMware support
on page 83

“NW120006” vSwitch becomes blank when performing disaster recovery of a VM to an Windows VMware support
on page 84 ESX server different than the original server

“NW119897” Improper error message appears when invalid datastore is specified during Windows VMware support
on page 84 VADP restore

“NW119729” Error messages for system related files appear during disaster recovery of a Windows VMware support
on page 84 VM on Windows 2008

“NW119040” SSID recovery of incremental backup that has not changed since the last Windows VMware support
on page 84 incremental backup results in second drive that is not needed being
recovered

“NW119468” networkr.raw output should be properly formatted Windows VMware support

on page 84

“NW119466” Windows 2008 FULLVM backup displays warning messages in savegroup Windows VMware support
on page 85 log even though backup is successful

“NW118840” Improper error message appears when disaster recovery run from Windows VMware support
on page 85 command line without -o option

“NW118828” VMs created in vCenter or folders created on proxy not cleaned up when Windows VMware support
on page 85 recovery is stopped

“NW118382” Remove snapshot operations are triggered when DR is started with Windows VMware support
on page 85 winworkr or from the command line

“NW127613” VSS SYSTEM FILESET save set may fail to back up the boot configuration Windows Windows support
on page 85 data properly

“NW127617” DISASTER_RECOVERY save set may fail to back up the BCD properly Windows Windows support
on page 86

“NW126617” Windows disaster recovery offline restore logs may be temporarily Windows Windows support
on page 87 unavailable for browsing

62 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Table 11 Limitations discovered in NetWorker release 7.6 Service Pack 2 (page 6 of 6)

Operating
Issue number Description of limitation system affected Product feature

“NW123457” Windows Disaster recovery may fail if target system has more disks than Windows Windows support
on page 87 the original system

“NW124548” Save set consolidation not supported for Data Domain devices configured Windows Windows support
on page 87 on Windows 2008

“NW121933” Resource pool path uses forward slash for path separator instead of Windows Windows support
on page 87 backward slash

Backup problems and limitations

Savegroup may fall back to “agentassistd indirect save”
NW128451

A savegroup may fall back to "agent assistd indirect save" due to an authentication issue
between the NetWorker client and server.
Workaround

Perform the following steps to resolve this issue:

1. Add the hosts file entry in the server, storage node, and client.
2. Remove the peer information on all the machines.
3. Restart NetWorker services on the client.

Manual backups using save command fail when performed on Windows

remote client from non-root user account
NW124240

Manual backups using the save command cannot be performed on Windows remote
clients from a non-root user account. The error “User does not have sufficient backup
privileges or is not administrator” appears.

SFR enabled backup not supported for NTFS partitions residing on IDE
disks
NW124330

SFR enabled backup is not supported for NTFS partitions residing on IDE disks. When an
SFR backup to IDE disks with NTFS partitions is attempted, the backup falls back to
non-NTFS mode.

CLI problems and limitations

mminfo does not reset space usage after savegroup is aborted
NW122242

After staging a file type device, both NMC and the output for mminfo -mv display the
incorrect volume size.

EMC NetWorker Release 8.1 Release Notes 63

Known problems and limitations

mminfo -mv and NMC show incorrect file type device (FTD) volume size
after stage operation
NW124967

After staging a file type device, both NMC and the output for mminfo -mv display the
incorrect volume size.

Cloning and staging problems and limitations

Recovery of a clone volume fails if the original backup was marked as
suspect
NW123160

When recovery of a clone volume is performed, if the original backup tape for the clone
was marked as suspect in NMC, the recovery fails.

Data Domain clone job records not flagged in jobsdb

NW120358

For Data Domain clone operations, the clone job record created in jobsdb does not
contain a flag to distinguish the type of clone as a Data Domain clone.

Configuration problems and limitations

Java JRE 6 Update 29 and JRE 7 Update 1 compatibility issues with NMC in
NetWorker 7.6 SP2 (fixed in NetWorker 7.6.2.7 and later)
NW134183 (esg125257)

JRE version 6 Update 29 (JRE 1.6.0_29) and JRE 7 Update 1 are not supported by NMC in
NetWorker releases previous to NetWorker 7.6.2.7. NMC will not function correctly if these
versions are installed.
Workaround

Update to NetWorker 7.6.2.7 or later.

Java JRE 6 Update 27 and JRE 7 compatibility issues with NMC in

NetWorker releases previous to 7.6.2.5 Build 681
NW128816 (esg124036)

JRE version 6 Update 27 (JRE 1.6.0_27) or later and JRE 7 are not supported by NMC in
NetWorker 7.6 SP2 Build 681 and earlier. NMC will not function correctly if these versions
are installed.
Workaround

A fix is available in NetWorker 7.6.2.7 and later. If JRE version 6 Update 27 or JRE 7 is
currently installed, update to NetWorker 7.6.2.7 or later.

64 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

HIST and UPDATE variables ignored during SMTape backup on Windows

NW129596

When backing up NetApp filers using Snapmirror-to-tape (SMTape) on Windows, the HIST
and UPDATE variables are ignored in the client resource’s Application Information
attributes.

Remote access attribute for NetWorker client resource of the NetWorker

server must list names of all cluster nodes to ensure lockbox functionality
on NetWorker server’s cluster node
NW127039

If a lockbox is created for a NetWorker server configured on a cluster, the following step
must be performed in order for the lockbox to work on the cluster node the NetWorker
server is running on:
◆ The Host section for the NetWorker client resource's remote access attribute must list
the names of all the cluster nodes. The values of the names must be in the following
form:
• For UNIX cluster nodes, use the hostname command output
• For Windows cluster nodes, use the full computer name (go to Control Panel >
System > Computer name to get the full name)

Note: The lockbox can be created prior to or after performing this step.

This configuration must be performed during the initial startup of the virtual NetWorker
server (that is, before the virtual server fails over to the other cluster). Otherwise, the
lockbox will have to be deleted and then re-created.

Devices and media problems and limitations

Tape drives configured from EDL on HP-UX are not assigned unique serial
numbers, leading to device ordering issue
LGTsc21701

When tape drives are configured from an EDL on the HP-UX platform, the tape drives are
not assigned unique serial numbers. As a result, a device ordering issue occurs when
running the jbconfig command.
Workaround

The EDL requires the following patch/update:

disklibraryupdate-dl1549upto18-SP05 and update-dl154922

Hardware Providers not supported for NetWorker client backup operations

The NetWorker client does not currently support hardware providers for backup
operations. Only software providers are currently supported.

If you are using a hardware provider, perform one of the following workarounds as
appropriate:

EMC NetWorker Release 8.1 Release Notes 65

Known problems and limitations

◆ Uninstall the hardware providers.

◆ Migrate any data that is backed up by the NetWorker client to a disk LUN (Logical Unit
Number) that is not controlled by a hardware provider.

For example, suppose you have an application module, such as NMM, to back up a SQL
server database, and you want to continue to back up the SQL databases with a hardware
provider. To accomplish this, migrate the SQL software binaries and, if applicable, any
operating system data or any regular filesystem data to a disk LUN that is not controlled
by a hardware provider such as the C:\ drive. In this way, the application module can still
use the hardware provider to backup the SQL databases, while all other data is backed up
with a software provider.

Active read sessions limited to 1 when recovering data from a Data

Domain device
NW127956

When recovering data from a Data Domain type device, the number of active read
sessions will be limited to 1, regardless of the number of recovery sessions that are
started, because the DD Boost library used with NetWorker 7.6 SP2 is not multi-threaded.

DD Archiver storage systems are limited to one data movement policy per
NetWorker datazone by default
NW128111

Data Domain (DD) Archiver data movement policies are managed on an mtree basis. You
can specify one data movement policy per mtree. All NetWorker Data Domain devices in a
datazone back up to a single mtree on a DD Archiver storage system by default.
Workaround

You can create additional mtrees for a datazone manually by using the NetWorker Device
Configuration Wizard and specifying a new top level folder in the Select Folders to use as
Devices attribute.

Device Configuration wizard cannot be used to label CIFS share AFTD on a

Data Domain system
NW125290

When a CIFS share AFTD is used on a Data Domain storage system, the device cannot be
labeled using the Device Configuration wizard.

System error detected when labeling a Data Domain device

NW124165

A System error similar to the following is detected when attempting to label a Data
Domain device:
nsrmm: No volume with name <volume> found in media database.
nsrmm: SYSTEM error: Data Domain disk

66 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

AFTD percentage capacity not updated after capacity is changed and

device is relabeled
NW120013

After changing the % capacity of an AFTD and relabelling the device, the % capacity is not
updated to the new value. Running mminfo -m displays the % capacity prior to the
relabelling.
Workaround

Delete the device and create a new device with the required capacity.

Data Domain features cannot be enabled through NMC on AIX 64-bit

NW123401

On an AIX 64-bit system, libDDBoost does not work. As a result, Data Domain features
cannot be enabled through NMC, which is 64-bit on AIX.

Selecting AFTD on first window of Device Configuration wizard returns

error
NW125024

Opening the Device Configuration wizard and selecting the AFTD radio button from the
first window returns the following error message in the Java console:
!!!!Recieved client communication error = 1

NMC then becomes unresponsive and must be restarted.

GUI problems and limitations

NMC client browser OS and JRE Requirements table: Link to download JRE
1.6 on HP-UX has changed
The NMC client web interface provides vendor specific links to download the Java Runtime
Environment (JRE) 1.6 if the JRE 1.6 is not already installed on the NMC client.

The URL for downloading JRE 1.6 on HP-UX has recently been changed by the vendor. The
correct URL is now:
https://h20392.www2.hp.com/portal/swdepot/displayProductInfo.do?productNumber=
HPUXJDKJRE60

Alert not displayed in NMC on the first day of evaluation license

NW122973

When NetWorker is installed using an evaluation license and NMC is started, NMC does
not display an alert in the Monitoring tab indicating an evaluation license is being used.
This problem only occurs on the first day of use; on subsequent days, this alert is
displayed in NMC.

EMC NetWorker Release 8.1 Release Notes 67

Known problems and limitations

Drive letter marked in red in Recover browser if drive is not on proxy server
NW125853

In the NetWorker User program’s Recover window, if a drive letter is not present on the
server, the drive is highlighted with a red question mark.

Scan operation fails in NetWorker User when browse sessions are open for
a client while the client’s save set is being scanned into the NetWorker
server
NW125786

If a client browser is open in the NetWorker User program, and index files exist for a
backup that is being scanned in for the client on a Windows NetWorker server, the scan
operation fails with errors similar to the following:
scanner: (ssid <number>) index error, store failed
scanner: error, Cannot unlink the existing key file before re-creating
it. Error 'Permission denied'.
Workaround

Close any NetWorker User browse sessions that are open for the client whose save set is
being scanned into the NetWorker server.

NMC may not display SNMP traps when traps are sent from Data Domain
boxes running DDR OS 5.0 and multiple NICs are configured
NW125671

When SNMP traps are sent from a Data Domain box running DDR OS version 5.0 and
multiple NICs are configured, SNMP picks up the IP address of the first available NIC, so
any traps sent from the DDR will be encoded with that first NIC’s IP address. If NetWorker
expects to receive SNMP traffic from another NIC’s IP, the SNMP traps will not display in
NMC.

Help topic does not display for NetWorker Clone reports in NMC
NW123840

When selecting the Reports tab in NMC and clicking the on-screen help icon, a help topic
does not display for Networker Clone reports.

Non-readable characters displayed in the NetWorker servers search in

Winworkr
NW124145

A Winworkr NetWorker servers search displays non-readable characters in the in WinPE

wizard.

Multiple browse sessions displayed in NMC during single step recovery

NW119577

During a single step recovery, multiple browse sessions will be displayed in the
NetWorker Console Monitoring window. This is expected behavior.

68 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Installation and upgrading problems and limitations

Glibc-locale-32bit package required on 64-bit SuSE Linux system to run
NMC
NW129515

If you are running NMC on a 64-bit SuSe Linux system, the following package needs to be
installed:
glibc-locale-32bit

NetWorker client does not recognize HomeBase path and Profiling fails if
HomeBase Agent is installed manually
NW124476

During the NetWorker installation, if the HomeBase Agent 6.4.1 is installed manually, the
NetWorker client does not recognize the HomeBase path after the installation. Upon
configuring the client and starting a scheduled backup, Profiling fails with the error
"nsrbmr ... error :<Couldn't spawn hba.exe>".
Workaround

Restart the NetWorker services so that nsrexecd recognizes the HomeBase Agent
installation. Profiling now completes successfully.

Changes to handling of the nwclust.pl script when upgrading to NetWorker

8.0 in an AutoStart 5.4 and later cluster
NW119766

Starting from release 5.4, AutoStartTM no longer provides the nwclust.pl script. The
NetWorker 8.0 release includes the script, but since the location of the script has
changed, the lcmap script needs to be updated to point to the new location if you are
upgrading from a Networker version previous to NetWorker 7.6 SP2.
Workaround

Reconfigure the AutoStart cluster so that the generated lcmap script points to the correct
location of the nwclust.pl script.

On UNIX:
1. Run networker.cluster -r.
2. Run networker.cluster.

On Windows:
1. Run lc_config -r.
2. Run lc_config.

More information on cluster configuration is provided in the EMC NetWorker 7.6 Service
Pack 2 Cluster Installation Guide. The following example shows the generated lcmap
script on Windows after the cluster configuration:

SET FT_DIR=C:\Program Files\EMC\AutoStart\eas54

EMC NetWorker Release 8.1 Release Notes 69

Known problems and limitations

SET FT_DOMAIN=eas54
"C:\Program Files\EMC\AutoStart\eas54\bin\ftPerl" "C:\Program
Files\EMC NetWorker\nsr\bin\nwclust.pl"

Note: When performing a new installation of NetWorker 8.0, these steps for upgrading do
not apply. Follow the instructions for performing a cluster install provided in the EMC
NetWorker 8.0 Cluster Installation Guide.

Messaging problems and limitations

Warning message appears when starting NetWorker services on UNIX
NW122342

When starting NetWorker services on a UNIX system, a warning message appears if a

random number generator (PRNGD) is not running on the host where nsrexecd is running.
Workaround

Install a random number generator if one does not already exist on the system.

To check if a PRNGD is running, run the following command on a UNIX system:

ps -ef | grep prng

An output similar to the following appears:

ganaps@bu-ffalo#bash2.03|~/cvs/nw_7_6_dev/dev/nsr/server> ps -ef |
grep prng
root 353 1 0 Sep 30 ? 4:18 /usr/local/sbin/prngd
/var/run/egd-pool

Performance problems and limitations

Latency issues with media database operations
NW121129

When a large media database is under heavy load and nsrim is running in parallel, latency
issues occur with media database operations. If several large savegroups are run in
parallel with a large number of clients and/or savesets, the backups may seem to hang or
take very long to complete. Other operations, such as mminfo or NMC/GUI commands
that perform media database queries, will also hang or take long to respond if nsrim is
crosschecking the media database at the same time.

The following common symptoms are observed:

◆ nsrim -X does not complete in a reasonable amount of time
◆ nsrmmdbd process consumes nearly 100% of the processor it is running on
◆ pstack of nsrmmdbd displays recalculate_vol_rmedia function
◆ Many save set clone instances on a volume have expired on the volume retention
date.

Workaround

To fix this issue:

70 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

1. Download and install the latest Cumulative Fixes package for NetWorker 7.5 and 7.6
from the following locations:
ftp://ftp.legato.com/pub/NetWorker/Cumulative_Hotfixes/7.5
ftp://ftp.legato.com/pub/NetWorker/Cumulative_Hotfixes/7.6
2. Create a file called mdb_no_recalc under the /nsr/debug directory.

Creating the /nsr/debug/mdb_no_recalc file and installing the hot fix stops the volume
retention recalculation during the crosscheck of the media database, thereby reducing
the overall time media database crosschecking takes to complete. This is particularly
useful for media databases that contain a large number of save set records per volume
where many of the save sets in the volume(s) are expiring on the same day or at the same
time.
Note: To check the status of a volume, the volflags reporting option in mminfo must be
used instead of volretent.

Restore problems and limitations

Directed recovery to CIFS share on destination client is not supported
NW127028

Directed recovery does not work when the directory specified to relocate the files to on the
destination client is a mapped CIFS share.

Bare Metal Recovery using HomeBase and a version of NetWorker different

from the one used for backup may result in dll mismatch
NW126148

Performing a bare metal recovery using HomeBase and a version of NetWorker that is
different from the one that was used for the backup may result in mismatched NetWorker
DLLs. When this mismatch occurs, it is during the restore of the <Nw-installdir>\EMC
NetWorker\nsr folder.

Error messages similar to the following may appear upon subsequent use of NetWorker
recover commands:
The procedure entry point setIsVcb could not be located in the dynamic
link library LIBASM.DLL
Workaround

To avoid this issue, perform one of the following, listed in the preferred order:
◆ Always use the same version of NetWorker for the backup and the HomeBase recovery
of a system.
◆ Ensure that the NetWorker installation volume (for example, C:\) is the last data drive
recovered by HomeBase, occurring after the recovery of VSS system save sets and any
other volumes. After the final recovery and restart of the system, the mismatch will be
resolved.
◆ Create a HomeBase exclude entry for the <NW-installdir>\EMC NetWorker\nsr folder.
The HomeBase documentation provides more information on where and how to
create this exclude information.

EMC NetWorker Release 8.1 Release Notes 71

Known problems and limitations

Disaster recovery logs display “recover completion time” for metadata

recovery instead of indicating that recovery of metadata was successful or
identifying this as “recovery of metadata completion time”
NW126007

In the disaster recovery logs from the NetWorker User program and the CLI, the recovery of
metadata is displayed as “recover completion time” instead of indicating that the
recovery of metadata was successful, or identifying that this is “recovery of metadata
completion time”.

Empty file gets created after recovery fails due to incorrect passphrase
NW125168

When a file-level recovery fails due to incorrect passphrase entry, an empty file gets
created in the destination specified for the recovery. This file can be deleted.

Security problems and limitations

Encryption attribute not set on encrypted folder after recovery
NW122275

If a folder has been encrypted in Windows, for example by selecting Folder Properties >
Advanced > Encrypt contents to secure data, it will be recovered as encrypted. However,
the encryption attribute will not be set on the folder. You can manually reset the
encryption attribute after the Windows disaster recovery. This is a Microsoft limitation.

Volumes that are BitLocker encrypted will not be encrypted after recovery
NW123237

Windows disaster recovery can back up critical volumes that are BitLocker encrypted.
However, the recovered volumes will not be encrypted. You can use the BitLocker icon in
the Control Panel to re-apply the volume encryption. This is a Microsoft limitation.

VADP only supports English Windows operating systems for

vCenter/Virtual Center and VADP proxy
NW129816

For NetWorker installed on Windows, VADP only supports English versions of the Windows
operating systems for the vCenter or Virtual Center machine and the VADP proxy.
Note: NetWorker does support the backup/recovery of non-English versions of guest
operating systems for the VMs.

72 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Incorrect proxy information displayed in NMC for VM client resource after

upgrading the NetWorker server to 7.6 SP2
NW129735

After upgrading the NetWorker server from 7.6 SP1 to 7.6 SP2, the VM client resource
associated with the VCB proxy does not display the correct information in NMC. For
example, if you are using both VADP and VCB proxies, VM client resources that are still
associated with VCB proxies will display the VADP proxy when viewing the VM resource in
NMC.

The correct information displays in the nsradmin output for the client resource.

Backup of a Windows 2003 VM is not browsable

NW129337

A VADP image-level backup of a Windows 2003 VM is not browsable if the VM was

originally Windows 2000 and later when the guest operating system was upgraded to
Windows 2003. The savegroup log displays the following message, even though all the
attached disks are NTFS:
nsrvadp_save: Parser failed because of a non-NTFS volume inside the VM
Workaround

To browse these VADP backups, download and install the following Windows 2003
resource kit from Microsoft into the VM:

http://www.microsoft.com/downloads/en/details.aspx?familyid=9d467a69-57ff-4ae7-9
6ee-b18c4790cffd

Once installed, perform the following steps to create a junction point:

1. Open a command prompt in the VM.
2. In the command prompt, run cd to the Tools folder in the resource kit. By default, the
Tools folder is located in C:\Program Files\Windows Resource Kits\Tools.
3. Enter mkdir "C:\windows" in the command prompt
4. Enter linkd "c:\windows" "c:\winnt" in the command prompt.
5. You can now confirm the junction point is created by running the following:
cd \
dir /a

The output should display one entry indicating Junction, as in the following:
05/28/2011 03:17 AM <JUNCTION> windows
05/28/2011 02:23 AM <DIR> WINNT

6. After these steps have been performed, run a level 0 FULLVM backup of the VM and
verify that it is browsable.

EMC NetWorker Release 8.1 Release Notes 73

Known problems and limitations

Recovery dumps core for traditional save set when VADP and traditional
save sets co-exist on Linux VM
NW129143

When a traditional backup and then a FULLVM backup of a Linux VM is performed that
includes both VADP and traditional save sets, the traditional VM save sets are not
browsable upon recovery. If the browse time of the save set is changed to point to the
traditional save sets, a core dump occurs.

NMC does not display VMs when auto-discovery is run on Linux with
vCenter 4.1 update 1
NW129041

When configuring the hypervisor resource and enabling auto-discovery in NMC on a

NetWorker server for Linux with vCenter 4.1 update 1, the VMs do not display in NMC. This
issue does not occur with vCenter 5.0.

Workaround
Update to vCenter 5.0.

Unable to browse guest based backups on non NTFS filesystem (applies to

NetWorker 7.6.2.1 build 638 and later -- fixed in NetWorker 7.6 SP3)
NW129093

This issue applies to NetWorker releases 7.6.2.1 build 638 and later and is fixed in
NetWorker 7.6 SP3 and later. Traditional guest based (client based) backups are not
browsable in the recovery GUI for VMs that are running a non NTFS filesystem and that
have a mix of VADP and guest based backups. This issue does not apply to Windows VMs
that are using NTFS. Additionally, save set recoveries are not affected and can be
performed in the usual way.

To work around the issue, run the following commands from a command line on the VADP
proxy or the VM:

To find the backup time:

mminfo -av -s networker_server -q "client=virtual_client"

To perform the recovery:

recover -t backup_time -s networker_server -c virtual_client
Example The following VM (host name mars) has a mix of both VADP and traditional guest based
backups. This example shows how to recover a traditional backup save set on the VM by
first locating the time of the backup save set using the mminfo command and then by
using that time with the recover command. The host name of the NetWorker server in this
example is jupiter.

C:\mminfo -av -s jupiter -q "client=mars"

volume type client date time size ssid fl lvl name
kuma-1 Data Domain mars 5/24/2011 10:38:39 PM 281 MB 1658578527 cb full /root
kuma-1.ROData Domain mars 5/24/2011 10:38:39 PM 281 MB 1658578527 cb full /root
kuma-6 Data Domain mars 5/24/2011 10:59:22 PM 5243 MB 1440475890 cb full FULLVM
kuma-6.RO Data Domain mars 5/24/2011 10:59:22 PM 5243 MB 1440475890 cb full FULLVM
C:\recover -t "5/24/2011 10:38:39 PM" -s jupiter -c mars

74 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Notice that in the previous example output from the mminfo command, the first two lines
listed are for a traditional backup and the last two lines are for a VADP backup, which is
denoted with the save set name, FULLVM. The NetWorker Command Reference Guide
provides more information about using the recover command to mark (select) files and to
perform the recovery.

Recovering non-Windows VMs that have a mix of VADP image-level and

client-based backups (fixed in NetWorker 7.6.2.1 build 638 and later)
NW127928

The following considerations apply to NetWorker releases previous to 7.6.2.1 build 638
when recovering non-Windows VMs that have a mix of VADP image-level and client-based
backups.

If using traditional NetWorker client-based backups along with VADP image-based

backups for the same VM client, you must first remove the indices of the previous
traditional save sets before perfoming an image-level recovery of the full virtual machine,
otherwise the image-level recovery fails. The only indices that need to be removed are
those indices of the traditional save sets whose backups were performed prior to the
VADP image-level backup that is selected for restore.

Run the following command on the NetWorker server to mark the browsable save sets
corresponding to the traditional backup as recoverable save sets.
nsrim -c client_name -N traditional_saveset_name -l

The last parameter in the command is a lower-case L.

This command removes the oldest full save and all dependant save sets from the online
index. You may need to run the command multiple times for every level FULL browsable
traditional saveset and for every traditional save set name.

After removing the indices, you can perform the image-level recovery using either the
NetWorker User program or the command line.
Example For example, a Linux client mars has a mix of both VADP image-level and traditional
backups as seen in the following output:
C:\>mminfo -avot -q "client=mars,volume=delve.001"
volumetypeclientdatetimesizessidfl lvlname
delve.001 adv_file mars 4/14/2011 9:55:55 AM 3483 MB 4154881857 cb full /usr
delve.001 adv_file mars 4/14/2011 10:01:35 PM 103 MB 3953675679 cb incr /usr
delve.001 adv_file mars 4/14/2011 10:07:10 AM 15 GB 4104550902 cb full FULLVM
delve.001 adv_file mars 4/14/2011 2:55:31 PM 3481 MB 4003904887 cb full /usr
delve.001 adv_file mars 4/14/2011 3:03:18 PM 103 MB 3903242058 cb incr /usr
delve.001 adv_file mars 4/14/2011 3:28:30 PM 15 GB 3852911942 cb full FULLVM

If you want to recover the latest image-level backup (in the above example,
SSID=3852911942), first remove all the indices of browsable save sets that are from the
previous traditional backups.

In this case, because there are two instances of browsable level FULL of the save set name
/usr that need to be removed, the following command must be run twice on the
NetWorker server:
nsrim -c mars -N /usr -l

If you want to recover from the second last image-level backup, (for example, from
SSID=4104550902), first remove all the indices of browsable save sets which are from
the previous traditional backups.

EMC NetWorker Release 8.1 Release Notes 75

Known problems and limitations

In this case, since there is one instance of browsable level FULL for the save set name /usr
that needs to be removed, the following command must be run once on the NetWorker
server:
nsrim -c mars -N /usr -l

Note: Browsable recovery of the traditional backup save sets will no longer be possible
once the respective indices are removed. If the traditional backup indices are still needed,
they can be restored once the image-level recovery is complete by running the following
command on the NetWorker server:
scanner -c client_name -i device_path

For example: scanner -c mars -i c:\device2

Transport mode falls back to NBD mode when configured transport modes
SAN or hotadd are not available on non-NTFS VMs
NW127589

When the transport mode is configured to SAN or hotadd for the proxy on a non-NTFS VM
and these transport modes are not available, VDDK falls back to NBD mode, and no
message is logged to indicate that the transport mode has been changed to NBD.

CBT tool does not return error when attempting to use with non-NTFS VMs
NW127651

The Changed Block Tracking (CBT) tool can only be used for incremental backups of NTFS
VMs. However, the tool cannot determine whether the operating system in use is NTFS or
non-NTFS. As a result, no error message is returned when CBT is enabled or disabled for
non-NTFS VMs.

Non-NTFS thin disk VMs get restored as thick disk during disaster recover
NW127446

When NetWorker backs up a non-NTFS thin disk VM, both allocated sectors and
unallocated sectors get backed up. As a result, upon disaster recovery of the thin disk VM,
the thin disk is fully allocated, and is restored as thick disk.

FULLVM backup of a NTFS VADP client falls back to non-NTFS mode if the
drive has an unformatted partition
NW127004

When performing a FULLVM backup of a NTFS VADP client, if the drive has an unformatted
partition, the virtual machine falls back to non-NTFS mode and the index is not included in
the backup.

The following errors appear in the savegroup output:

nsrvadp_save: Parser failed because of a non-NTFS volume inside the VM.
nsrvadp_save: Starting save operation of the non-NTFS Image.

76 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

VADP backup fails with VDDK error when specified transport mode is not
available
NW126608

If the transport mode specified for VADP backup is not available at the time of backup (for
example, if SAN mode is selected but the SAN connection becomes unavailable during
the backup), then the backup fails with the following messages:
nsrvadp_save: The following internal error occurred: An error was
returned from nsrvddk.exe: Error opening disk
VDDK Error: You do not have access rights to this file

VM does not power on after disaster recovery of CBT-enabled backup

NW126553

Disaster recovery of a CBT-enabled backup completes successfully, however, the VM fails

to power on after the recovery, displaying the following error:
the parent virtual disk has been modified since the child was created
Workaround

Ignore the error, then run the disaster recovery again from the same backup. When the
recovery completes successfully, manually power on the VM.

NetWorker User program cannot be used for VADP recovery in non-English

locale
NW126457

The NetWorker User program cannot be used for VADP recovery operations when
launched from a non-English Windows machine.

Guest backups of Windows 2008 R2 and Windows 2007 may fail on ESX
4.0
NW126321

The VMware tools installer contains a program, comreg.exe, which has a Windows 2008
R2 bug that prevents it from registering the VMware Snapshot Provider with VSS. Guest
backup of a Windows Server 2008 R2 or Windows 2007 guest may fail for certain versions
of ESX 4.0.0 due to this issue.

Instructions for taking a manual snapshot with vCenter are provided at the following
location:
http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displa
yKC&externalId=1022720

To resolve this issue, upgrade to ESX 4.0 update 2 or ESX 4.1, or to upgrade your ESX
4.0.0 server with a VMware patch, go to the following link:
http://kb.vmware.com/selfservice/microsites/search.do?cmd=displayKC&externalId=10
13127

EMC NetWorker Release 8.1 Release Notes 77

Known problems and limitations

SYSTEM accounts must have Full Control permissions for all NTFS disks to
back up Windows VMs
NW125927

For Windows VMs, image backups with file-level recovery enabled may not back up a
folder if the folder does not have Full Control permissions for the SYSTEM user account.
Workaround

To work around this problem, ensure that all NTFS disks have Full Control permissions for
the SYSTEM user account. To set these permissions:
1. Right-click an NTFS volume.
2. Select Properties.
3. Go to the Security tab and ensure SYSTEM account is listed and has been given Full
Control permissions for the entire volume.

Virtual compatible raw disk restored as local virtual disk (fixed in

NetWorker 8.0 SP1)
NW125800

When an image-level recovery of a VM is performed from what was initially a Virtual

compatible raw disk (RDM) backup, the RDM disk is restored as a local virtual disk.

Using SAN transport mode results in slow performance for disaster

recovery from physical proxy of VM with thin disks
NW125515, NW125130

When performing a disaster recovery from a physical proxy of a VM with thin virtual disks,
the SAN transport mode should not be selected. Due to a VMware limitation, performance
becomes very slow when using SAN mode, and the vCenter server receives many requests
to "Clear lazy zero" and "Allocate blocks".

It is recommended to perform one of the following:

◆ Recover with a virtual proxy using the hotadd transport mode
◆ Recover from the physical proxy using NBD or NBDSSL mode

Recovery of a VM using SAN or hotadd transport mode requires policy be

set to OnlineALL
NW125230, NW125142

When recovering a VM using either the SAN or hotadd transport mode, the following
one-time configuration must be performed on the proxy host before initiating the
recovery:
1. Open a command prompt on the proxy host.
2. Run the following command:
DISKPART
3. Enter SAN and check for the SAN policy.

78 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

4. If the policy indicates offline, enable the policy by entering the following:
SAN POLICY=OnlineALL

You can now initiate the VM recovery using SAN or hotadd mode.
Note: If recovery is initiated from a Windows machine other than the proxy, these steps
need to be performed on the machine where the recovery is initiated.

Savegroup output of VADP ALLVMFS backup still refers to VCB directive

NW125212

When performing an ALLVMFS backup, the savegroup output for the backup refers to VCB,
indicating the “Directive is not VCB Directive for client <client name>”. The VCB directive is
no longer valid and can be ignored.

“Release disk lease” messages appear during SAN mode backup

NW125179

During a SAN mode backup, several "Release disk lease" messages are logged under
vCenter tasks, and appear after the “Create virtual machine snapshot” message. The
“Release disk lease” messages can be ignored.

FULLVM level -0 backup must be performed after enabling CBT on the VM

NW125164

After enabling Changed block tracking (CBT) on the VM, run a FULLVM level -0 backup.
Also, power on the system once for the change to take effect before running backups.

NetWorker uninstallation on VADP proxy host

NW125084

When the NetWorker client is uninstalled from the VADP proxy host, a prompt to reboot
the system appears. If No is selected, and the client is reinstalled without rebooting the
system, the VADP backups will fail with the following errors:
nsrvadp_save: Failed to start vStorage API driver service.
nsrvadp_save: Failed to Initialise LibVADP Context.
Workaround

If the NetWorker client is to be reinstalled on the same VADP proxy host, then a system
reboot is required before performing a VADP backup or restore again.

Application information attributes for VADP backups should be added to

all resources if multiple instances are created with the same proxy name
NW124883

If multiple instances are created with the same proxy name, then the application
information attributes for the VADP backup should be added for all resources. For
example, if Save Set ALL is created as the proxy and the proxy client has 3 instances (1 for
ALLVMFS, 1 for FULLVM, and 1 with Save Set ALL), add VADP_DISABLE_FLR=YES in the
application for each instance, then perform a full virtual machine backup.

EMC NetWorker Release 8.1 Release Notes 79

Known problems and limitations

NMC Client Configuration wizard allowing duplicate proxy clients to be

created and configured on different mount points
NW124800

Incorrect behavior in NMC allows duplicate proxy clients pointing to the same vCenter
server to be created using the NMC Client Backup Configuration wizard, and also allows
these duplicate clients to be configured on different mount points.

Using hotadd transport mode to access a datastore directly attached to a

remote ESX host may return VDDK error
NW124730

During VM backup or recovery, the following error message my appear if the transport
mode hotadd is used to access a datastore that is directly attached to a remote ESX host:
VDDK Error: You do not have access rights to this file
Workaround

Perform one of the following:

◆ Use a proxy on the same ESX host as the DAS datastore, or
◆ Include nbd in the transport mode specification of the proxy

Temporary mnt folder gets created when recovering data from

CBT-enabled VM backup
NW124431

When a VM file-level backup is performed with CBT enabled, during recovery a temporary
folder (mnt) is created along with the restored data. This folder can be deleted.

vSphere Client cannot be launched from NMC for VMware Infrastructure

Client 2.5
NW123719

When the vSphere Client is launched from NMC for VMware Infrastructure Client 2.5, login
fails with the message “The requested operation is not implemented by the server”.

VADP backup may fail for VM name containing special characters

NW123164

If the name of the virtual machine to be backed up using VADP contains special
characters, the backup may fail. For example, issues with the VM name have been
encountered when the following special characters are used:

`^~=;! /([]{}@$\&#%+

The following links provide information on special characters to avoid when specifying the
virtual machine name:

http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displa
yKC&externalId=1026469

80 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

http://communities.vmware.com/message/1597730

Progress messages display during VADP backup and recovery

NW123219

During VADP backup and recovery, several progress messages display related to
snapshot creation or deletion, or registration of the virtual machine. For example, a
backup displays the following messages in the savegroup logs:
Creating snapshot for 'vm-423' ...
Task is -1% complete
Task is 95% complete
Task is -1% complete
SnapShot created Successfully for VM

These messages can be ignored.

Port ID does not get restored after disaster recovery of a VM originally

connected to dvSwitch
NW123142

After performing a disaster recovery of a VM which was originally connected to dvSwitch,

the port ID does not get restored as part of the dvSwitch configuration. As a result, the
restored VM loses network connectivity.
Workaround

Prior to powering on the VM after the disaster recovery, assign a port ID that is not being
used by any other VM.

VADP backup fails if Datastore name contains the characters % & = ?

NW122619

A VADP-based backup of a VM residing on a datastore fails if the datastore name contains

the characters % & = or ?

Do not use these characters in the datastore name.

Restoring a VM back to vApp container fails with error “A specified

parameter was not correct”
NW122586

When restoring a VM to the VMware vApp container on which it originally resided, the
restore fails with the following error:
An error occurred for task 'RegisterVM_Task'.
Details: A specified parameter was not correct.
Workaround

Restore the VM to the root of the VC/ESX host, and then move the VM to the vApp
container.

EMC NetWorker Release 8.1 Release Notes 81

Known problems and limitations

Client Configuration wizard cannot be used to configure default client as

proxy if NetWorker server and proxy client are used on the same machine
NW122566

When the NetWorker server and proxy client are to be used on the same machine and the
NetWorker server software is installed, there is no option to specify the client (which is
created by default after the installation) as a proxy client using the Client Configuration
wizard.
Workaround

In order to configure the client as a proxy client:

◆ Create a new client and configure this client as the proxy resource, then delete the
default client resource (multiple instances of the proxy client are not supported), or
◆ Modify the client resource manually in the Application Information attributes of the
proxy.

Web browser displays a 404 Not found or similar error the first time the
vSphere client is run from NMC if the default port number of the Virtual
Center server has been changed
NW122389

If the default http port of the Virtual Center server has been changed from port 80 to a
different port number through the Virtual Center GUI (Administration > VirtualCenter
Management Server Configuration > Web Service > HTTP), then the browser page that
appears the first time the vSphere client is run from NMC on that machine will display a
404 Not found or similar error.
Workaround

To fix this issue, append the port number (preceded by a colon) to the URL in your browser
and then refresh the browser. The correct web page then displays.

All save sets (NTFS and non-NTFS backups) marked as browsable

NW122381

All save sets are being marked as browsable, regardless of backup type (for example,
NTFS or non-NTFS VM backup), even though the non-NTFS backups are not browsable.

Backup of VM on ESX 3.5 fails for VMs using short name

NW122197

VMware appends a dot (.) to all virtual machines with a short name. As a result, VC cannot
find the exact VM, and backup of the VM fails with the following error:
nsrvadp_save: Unable to find the VM with IP / Name:

This problem only occurs on ESX version 3.5.

82 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Incorrect warning message appears in Savegroup log file during non-NTFS

backup of VM with independent persistent disk
NW122090

During a non-NTFS type backup of a VM with an independent persistent disk, a warning

message appears in the savegroup log file, indicating that the size shrunk while saving.
This warning message is incorrect and can be ignored

VM backup fails with HTTP Error 500 in cases where VM resides in a VMFS
datastore which is shared by multiple ESX/ESXi hosts or in cases where
the VM resides in HA/DRS Cluster setup
NW121958

When performing a backup of a VM which resides on VMFS datastore which is shared by

multiple ESX/ESXi hosts or is in a HA/DRS Cluster setup, the backup fails with an HTTP
Error 500 indicating the configuration file could not be downloaded.
Workaround

If the VC and ESX/ESXi version is 4.0, upgrade both the VC and ESX/ESXi server to 4.0
update 2 or later.

If the VC and ESX/ESXi version is 4.1, upgrade both the VC and ESX/ESXi server to 4.1
Update 1 or later.

Incremental backup of a VM with image-based backups

NW121168

For image-based backups, performing an incremental backup of a VM is not supported

after a hardware change, OS patch update, Service Pack update, drivers update or other
similar change has been made at the operating system or hardware level on the VM.

Perform a full image-level backup after every change made at the operating system and
hardware level on the VM.

VADP_* and VCB_* application information attributes both supported for

VADP implementation
NW121068

The new variables introduced with NetWorker 7.6 SP2 for VADP are not compatible with
previous NetWorker releases. Therefore, the VCB_* appinfo variable that was introduced
with NetWorker 7.6 is still supported with NetWorker 7.6 SP2.

VM cannot be restored to folder “Discovered Virtual Machine”

NW120474

There is currently no option to restore the VM to the folder Discovered Virtual Machine,
which is the default folder created under each ESX server.

EMC NetWorker Release 8.1 Release Notes 83

Known problems and limitations

vSwitch becomes blank when performing disaster recovery of a VM to an

ESX server different than the original server
NW120006

If a disaster recovery of a VM is performed to a different ESX server, and this ESX server
has a virtual switch (vSwitch) that is different than the original ESX server, the recovery is
performed with a blank vSwitch.
Workaround

Manually change the vSwitch to the one used with the current ESX server.

Improper error message appears when invalid datastore is specified

during VADP restore
NW119897

If an invalid datastore is specified during VADP restore, a proper error message indicating
an invalid datastore is not provided. Instead, a message similar to "Invalid datastore
format '<given datastore name>' may display.

Error messages for system related files appear during disaster recovery of
a VM on Windows 2008
NW119729

During disaster recovery of a VM on a Windows 2008 system, several error messages

similar to the following appear for system related files:
recover: Error in read of reference_C

These messages do not affect the recovery and can safely be ignored.

SSID recovery of incremental backup that has not changed since the last
incremental backup results in second drive that is not needed being
recovered
NW119040

When an incremental backup of a VM with two drives is run and there have been no
changes made since the last incremental backup of the drives, during SSID recovery the
second drive is unnecessarily recovered. Clicking on this second drive returns an error
message.

networkr.raw output should be properly formatted

NW119468

When disaster recovery is performed using the NetWorker User program’s Recover
window, messages in the networkr.raw file output are not properly formatted.

84 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Windows 2008 FULLVM backup displays warning messages in savegroup

log even though backup is successful
NW119466

A FULLVM backup of a Windows 2008 32-bit or 64-bit VM from a Windows 2008 32-bit
proxy displays warning messages in the savegroup log, even though the backup
completes successfully.

Improper error message appears when disaster recovery run from

command line without -o option
NW118840

A proper error message does not display when a disaster recovery is performed from the
command line without the -o option. Instead, the error indicates that “The system cannot
find the path specified”.

VMs created in vCenter or folders created on proxy not cleaned up when

recovery is stopped
NW118828

When a VM recovery operation in progress is stopped by the user or interrupted due to a

network disconnection, VMs that were created in vCenter or folders created on the proxy
are not cleaned up. The VM/folders must be manually deleted.

Remove snapshot operations are triggered when DR is started with

winworkr or from the command line
NW118382

When a disaster recovery is started from the NetWorker User program (winworkr) or from
the command line, Remove snapshot operations get triggered. The operation appears in
the Recent tasks pane of the vSphere Client GUI.

Windows support problems and limitations

VSS SYSTEM FILESET save set may fail to back up the boot configuration
data properly
NW127613

VSS SYSTEM FILESET save set backup may fail to back up the boot configuration data
(BCD) properly. If VSS SYSTEM FILESET encounters this problem, the backup fails and
error messages similar to the following appear in the log file:
VSS SYSTEM FILESET:\ VSS BCD BOOT FILES - ERROR: VSS failed to get
mapping info for E:\Boot\, vsserror=0x1: Incorrect function
VSS SYSTEM FILESET:\ VSS BCD BOOT FILES - ERROR: Aborting backup of
saveset VSS SYSTEM FILESET: because of the error with writer VSS BCD
BOOT FILES.

This problem occurs if NetWorker cannot correctly identify the partition (volume) that
contains the BCD.

EMC NetWorker Release 8.1 Release Notes 85

Known problems and limitations

Workaround

The inability to identify the BCD partition can be related to a system with multiple
partitions labeled as “active” partitions by the Windows operating system. If there is only
a single active volume, the problem should not occur.

Using Windows Disk Management, check the status information of the volumes. If there
are multiple “active” volumes, then determine if more than one partition requires that
status; if possible, remove the “active” designation from all partitions except your active
system volume (the volume that contains the bootmgr and the Boot directory with the
BCD file for the running instance of the operating system).
IMPORTANT
Ensure that caution is used if removing the “active” designation, since changing an active
partition can make your system unbootable. The system may not start up if the partition
marked as active does not contain the Windows boot files (or the boot files for another
operating system). If it is not possible to determine the correct active partition or to reduce
the active partition designation to only one, contact EMC NetWorker Support for
assistance.

DISASTER_RECOVERY save set may fail to back up the BCD properly

NW127617

DISASTER_RECOVERY save set backup may fail to back up the boot configuration data
(BCD) properly. If DISASTER_RECOVERY encounters this problem, the backup fails and an
error message similar to the following appears in the log file:
VSS SYSTEM BOOT: No such file or directory

This problem occurs if NetWorker cannot correctly identify the partition (volume) that
contains the BCD.
Workaround

The inability to identify the BCD partition can be related to a system with multiple
partitions labeled as “active” partitions by the Windows operating system. If there is only
a single active volume, the problem should not occur.

Using Windows Disk Management, check the status information of the volumes. If there
are multiple “active” volumes, then determine if more than one partition requires that
status; if possible, remove the “active” designation from all partitions except your active
system volume.
IMPORTANT
Ensure that caution is used if removing the “active” designation, since changing an active
partition can make your system unbootable. The system may not start up if the partition
marked as active does not contain the Windows boot files (or the boot files for another
operating system).
If it is not possible to determine the correct active partition or to reduce the active
partition designation to only one volume, contact EMC NetWorker Support for assistance.

You may be able to determine the correct active partition or reduce the active partition
designation to only one volume by manually mounting the BCD partition and then
performing a DISASTER_RECOVERY backup.

86 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Windows disaster recovery offline restore logs may be temporarily

unavailable for browsing
NW126617

By default, the Windows disaster recovery WinPE image is set to Pacific Standard Time
(PST). You may temporarily be unable to browse and recover the Offline Restore Logs save
set if your local time is earlier than PST time. For example, if your local time is PST plus two
hours, then the Offline Restore Log save set would not be available for recovery by
browsing until the two hours have elapsed.

Windows Disaster recovery may fail if target system has more disks than
the original system
NW123457

When performing a Windows disaster recovery, if the target system for recovery has more
disks than the system originally used for backup, the recovery may fail.
Workaround

From the System Recovery Summary window, click Options and insert -p in the Additional
Options: field.

Save set consolidation not supported for Data Domain devices configured
on Windows 2008
NW124548

Save set consolidation is not supported on a Windows 2008 system that is configured
with Data Domain devices.

Resource pool path uses forward slash for path separator instead of
backward slash
NW121933

When specifying the resource pool path for a disaster recovery, the path uses a forward
slash as the path separator instead of a backward slash. For example, if there is a
resource pool hierarchy where resource pool2 exists within resource pool1, you must use
a forward slash as path separator (for example, resource pool1/resource pool2).

NetWorker 7.6 Service Pack 1

Table on page 88 identifies problem issues and limitations discovered in NetWorker 7.6.
Service Pack 1.

The known limitations are separated into the following categories:

◆ “Backup problems and limitations” on page 90
◆ “CLI problems and limitations” on page 93
◆ “Cloning and Staging problems and limitations” on page 94
◆ “Configuration problems and limitations” on page 98
◆ “Devices and media problems and limitations” on page 100

EMC NetWorker Release 8.1 Release Notes 87

Known problems and limitations

◆ “GUI problems and limitations” on page 101

◆ “Installation and upgrading problems and limitations” on page 103
◆ “Restore problems and limitations” on page 103
◆ “VMware Consolidated Backup (VCB) problems and limitations” on page 105
◆ “VMware Consolidated Backup (VCB) problems and limitations” on page 105

Known limitations related to internationalization support in NetWorker 7.6 are provided in

the section “Internationalization support” on page 175.

Table 12 Limitations discovered in NetWorker release 7.6 Service Pack 1 (page 1 of 3)

Operating Product
Issue number Description of limitation system affected feature

“NW121944” NDMP DSA backup of NAS filer or SnapImage client fails with error “could not All Backup
on page 90 establish jobid with server”

“NW120918” File-by-file recovery of a DSA backup from Data Domain device fails All Backup
on page 91

“NW120421” For BMR-enabled Windows NetWorker clients, generating profile information Windows Backup
on page 91 using HomeBase Agent 6.4 fails

“NW120031” Glibc 2.3.4 or later required on Linux host in order to run as storage node for Linux Backup
on page 92 Data Domain device

“NW117892” nsrmmd may consume large amount of memory when backing up to Data All Backup
on page 92 Domain device by way of the storage node

“NW114817” Savegroup with customized backups script fails UNIX Backup

on page 92

“NW104883” Cloud backup devices and partial save sets All Backup
on page 92

“NW121894” nsrclone with -d option ignored when used to specify destination storage node All CLI
on page 93 for writing save sets or when target clone pool has media type required=DD

“NW118223” Scanner -i does not rebuild media indices All CLI

on page 93

“NW118064” AFTD capacity displayed in kb instead of % in output for mminfo -m All CLI
on page 93

“NW114937” nsrls -m option may not list media database statistics Linux CLI
on page 94

“NW119530” Path-to-tape: when spanned save set is cloned from media type with smaller All Cloning
on page 94 block size to media type with larger block size, recovery from media type with and
larger block size fails Staging

“NW119156” Cloning fails when Data Domain or NDMP save set of a client is encountered and All Cloning
on page 94 there is no client resource on the NetWorker server and
Staging

“NW118255” Device status continues to display as “active” after cloning operation terminated WIndows Cloning
on page 95 and
Staging

“NW118893” Clone controlled replication with nsrclone may revert to a regular clone if the All Cloning
on page 95 NetWorker server’s primary host name is not specified and
Staging

88 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Table 12 Limitations discovered in NetWorker release 7.6 Service Pack 1 (page 2 of 3)

Operating Product
Issue number Description of limitation system affected feature

“NW117750” nsrclone/nsrstage performs regular clone for source Data Domain volume on Linux Cloning
on page 96 remote storage node if client resource was not created in NetWorker server and
Staging

“NW117470” Concurrent stage with stage or clone from the same Data Domain device or AFTD Linux Cloning
on page 96 is not supported and
Staging

“NW117179” A regular volume clone operation hangs if the source volume is on a remote Linux Cloning
on page 96 storage node and it is unmounted initially but is mounted after the clone and
attempts to start Staging

“NW117137” Clone controlled replication (optimized clone) for Data Domain device is only Solaris, Cloning
on page 97 performed when DD clone volume is mounted and available for writing Windows and
Staging

“NW116696” When cloning a mix of save sets from different source devices such as Data Windows Cloning
on page 97 Domain, AFTD, and NDMP devices, the same clone target volume may not be and
used for all save sets Staging

“NW115417” Clone jobs may time out and not complete if 30 or more are scheduled to occur Windows Cloning
on page 97 at the same time and
Staging

“LGTsc31596 Path-to-tape cloning of save sets spanning across multiple volumes with All Cloning
” on page 98 different block sizes fails and
Staging

“NW123374” NetWorker does not support IPMP feature on Solaris 10 Solaris Configurati
on page 98 on

“NW120856” Data Domain host name must be specified in Device access information All Configurati
on page 98 attribute when configuring Data Domain devices on

“NW120373” NetWorker and device operations can fail when TCP Chimney is enabled for NICs Windows Configurati
on page 98 on Windows Server 2003 or Windows 2008 R2 on

“NW117057” Failure reported during disaster recovery of VSS SYSTEM BOOT; set Windows Configurati
on page 99 NSR_RECOV_TEMP_CLEANUP variable on

“NW114428” Incremental backup failure may occur due to number of VSS shadow copies; set Windows Configurati
on page 99 NSR_VSS_WAIT_INTERVALS variable on

“NW120540” Labelling a Data Domain device to a new pool when the device contains data All Devices
on page 100 belonging to a different pool can result in data loss and Media

“NW118590” Labelling a Data Domain device fails after converting the device from AFTD to All Devices
on page 101 Data Domain and Media

“NW120536” Username/password validation may fail using NMC New Device wizard for file UNIX GUI
on page 101 system browsing if storage node is UNIX

“NW120324” Data Domain systems do not display in NMC Enterprise window or NetWorker All GUI
on page 102 Administration Devices window upon reinstalling NMC after complete uninstall

“NW118886” Monitoring of SNMP events for Data Domain not available on HP-UX HP-UX GUI
on page 102

“NW115920” NSR clone resources that are created with the nsradmin program cannot be Windows GUI
on page 102 edited as scheduled resources in the GUI

EMC NetWorker Release 8.1 Release Notes 89

Known problems and limitations

Table 12 Limitations discovered in NetWorker release 7.6 Service Pack 1 (page 3 of 3)

Operating Product
Issue number Description of limitation system affected feature

“NW124547” HomeBase Agent 6.2.1 does not automatically install as part of NetWorker Client Linux Installation
on page 103 installation on Red Hat Enterprise Linux 5.5 and later or
Upgrading

“NW119766” nwclust.pl script must be copied to AutoStart binary installation directory if All Installation
on page 103 using NetWorker in AutoStart 5.4 or later cluster environment or
Upgrading

“NW119115” Remote cluster services not started after Authoritative Restore of MSCS cluster Windows Restore
on page 104 database

“NW116575” Bootmgr file deleted upon recovery of C:\directory with VSS System save set Windows Restore
on page 104 selected

“NW115372” NDMP: recover or nsrndmp_recover displays debug messages during DSA Windows Restore
on page 105 recover on Windows even though recover is successful

“NW118106” VCB topology map shows Helper VM for hotadd mode is associated with same Windows VCB
on page 105 group used for hotadd proxy

“NW117233” Single Step recovery of VM within a vApp fails with P2V error Windows VCB
on page 105

“NW117714” Single Step recovery of Windows 2008 x64 VM fails with P2V error Windows VCB
on page 106

“NW118197” PVSCSI controller changes to BusLogic controller, leading to boot failure for x64 Windows VCB
on page 106 VMs and performance issues for x86 VMs

“NW118200” SSR disaster recovery of SuSE Linux Enterprise Server 11 VM fails with P2V error Windows VCB
on page 106

“NW118486” Vmxnet3 adapter changes to e1000 adapter during VM recovery Windows VCB
on page 106

“NW113740” NetWorker cannot use mapped drive for VCB mount point Windows VCB
on page 107

Backup problems and limitations

NDMP DSA backup of NAS filer or SnapImage client fails with error “could
not establish jobid with server”
NW121944

When performing an NDMP DSA backup of a NAS filer or SnapImage client, either by
running savegrp from the command line or from NMC (manual or scheduled savegroup),
the backup fails with the error “could not establish jobid with server”; nsrdsa_save core
dumps on UNIX servers and generates a memory dump on Windows servers.
Workaround

A hotfix is available for NetWorker 7.6 SP1. Perform the following steps to install the
hotfix.

On UNIX:
1. Go to ftp.legato.com/pub/NetWorker/Updates/NW121944/7.6.1/7.6.1/and
download the appropriate fix for the NetWorker server platform.

90 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

2. Stop the NetWorker daemons when no backup or recovery operations are in progress
(from the command line, run nsr_shutdown).
3. Rename the current nsrdsa_save file (located in /usr/sbin by default), making note of
the file permissions.
4. Copy the downloaded nsrdsa_save binary to the location of the original nsrdsa_save
file. If required, set the file permissions on the new nsrndsa_save binary to match
those of the original.
5. Start the NetWorker daemons on the NetWorker server (from the command line,
networker start).

On Windows:
1. Go to ftp.legato.com/pub/NetWorker/Updates/NW121944/7.6.1/7.6.1/ and
download the appropriate fix for the NetWorker server platform.
2. Stop the NetWorker Remote Exec Services when no backup or recovery operations are
in progress.
3. Rename the current nsrdsa_save.exe file (located in C:\Programs Files\EMC
NetWorker\nsr\bin by default).
4. Copy the downloaded nsrdsa_save.exe binary to the location of the original
nsrdsa_save.exe file.
5. Start the NetWorker Backup and Recovery service. This will also start the NetWorker
Remote Exec service.

File-by-file recovery of a DSA backup from Data Domain device fails

NW120918

Performing a file-by-file recovery of a DSA backup from a Data Domain device fails with an
error similar to the following:
nsrndmp_recover: NDMP Service Warning: These files were not restored
Workaround

Change the recover mode from client DFA to mmd by creating an empty file named
nodirectfile in the following location on UNIX or Windows:
◆ UNIX: /nsr/debug/nodirectfile
◆ Windows: %NW INSTALL Directory%\nsr\debug\nodirectfile

For BMR-enabled Windows NetWorker clients, generating profile

information using HomeBase Agent 6.4 fails
NW120421

For Windows NetWorker clients, if HomeBase Agent version 6.4 has been installed and
you have enabled BMR profiling on the NetWorker client, profiling generation fails with an
error similar to the following:
HBA: ‘hba.bat’ is not recognized as an internal or external command

EMC NetWorker Release 8.1 Release Notes 91

Known problems and limitations

Workaround

Manually create a hba.bat file in the same directory where hba.exe is located. The file
should contain the following lines:
@echo off
hba.exe %*

Glibc 2.3.4 or later required on Linux host in order to run as storage node
for Data Domain device
NW120031

A glibc 2.3.4 or later library is required by libDDBoost.so to initialize the DDCL library on
supported Linux systems such as SuSE Linux version 9. If a version earlier than glibc 2.3.4
is installed on the Linux system, the system will not be able to run as a storage node for
the Data Domain device.

nsrmmd may consume large amount of memory when backing up to Data

Domain device by way of the storage node
NW117892

When backing up to a Data Domain device by way of the storage node, nsrmmd consumes
a large amount of memory and does not release the memory, even when the backup is no
longer running.

Savegroup with customized backups script fails

NW114817

When a savegroup is run with a customized backup script, the savegroup fails.
Workaround

On HP-UX, instead of using the posix shell (/bin/sh) for customized backup scripts to be
automatically started by the savegroup, use the korn shell (/bin/ksh).

Cloud backup devices and partial save sets

NW104883

By default, the CheckPoint restart feature does not support cloud backup devices
because partial save sets are not retained on cloud backup devices.
Workaround

When the cloud backup device is used as a backup device for a Checkpoint restart, on the
Server Properties menu, enable the Keep Incomplete Backups attribute. If the Keep
Incomplete Backups attribute is not enabled, the NetWorker software will not keep the
partial save sets.

92 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

CLI problems and limitations

Running nsrclone with -d option ignored when used to specify destination
storage node for writing save sets or when target clone pool has media
type required=DD
NW121894

The nsrclone command is ignored when used with the -d option to set the destination
storage node for writing save sets, or to set the destination storage node if the target
clone pool has its selection criteria set to media type required=Data Domain.

For example, when running nsrclone -vvv -d scip2b131 -b DDclone -S <ssid> where the
server is scip2d14 and the target pool is DDclone, if the target pool’s selection critiera is
media type required=Data Domain, save sets are written to a volume in DDclone on the
server scip2d14 instead of the storage node scip2b131, as was specified in the command
line.
Workaround

Use the Devices tab of NMC to specify the destination storage node at Devices > Storage
Nodes , then right-click the storage node and select Properties. The properties for Clone
storage nodes are specified under Configuration.

Scanner -i does not rebuild media indices

NW118223

Running scanner -i <device> does not rebuild the media indices from the volumes read.
Workaround

To rebuild the media indices:

1. Prior to running scanner -i, run scanner -m <device>.
2. Run scanner -i <device>.
3. Synchronize the media database and indices by running nsrim -X.

AFTD capacity displayed in kb instead of % in output for mminfo -m

NW118064

When mminfo -m is run and the AFTD percentage capacity restriction is in use, AFTD
capacity does not display as a percentage in the output. Instead, the output displays
device capacity in kb.

For example, if AFTD capacity was set to 98% as in the following example, an output
similar to the following appears:
bash-2.05# mminfo -m
state volume written (%) expires read mounts capacity
pep.talisman.com.003 0 KB 0% undef 0 KB 2 98 KB

EMC NetWorker Release 8.1 Release Notes 93

Known problems and limitations

nsrls -m option may not list media database statistics

NW114937

When using the nsrls -m option to print media database information, statistics of the
NetWorker media database may not be listed.
Workaround

If nsrls -m does not list the media database sizing status, stop and restart the NetWorker
services and then run nsrls -m again.

Cloning and Staging problems and limitations

Path-to-tape cloning only supported for media types with the same device
block size and device file size
NW119530

When a save set is cloned from a media type with a block size that is different from the
destination media type, recovery from the clone volume to the destination volume fails
with an error message similar to the following:
recover: Unable to read checksum from save stream
recover: Encountered an error recovering <filename>
recover: Error encountered by NSR server `scip2d.lss.emc.com': Bad or
missing record in save set <save set ID>, lost <# bytes> bytes
starting at offset <# bytes>.
Workaround

Clone the path-to-tape save set to a media type that has the same device block size and
device file size.

To check the block size of a labelled volume, open the Device Properties dialog box and
check the volume block size in the Volume tab.

If the block sizes are different, do the following:

1. Go to the Device Properties > Advanced tab and change the Device Block Size attribute
so that the source and destination volume block sizes match.
2. Relabel the volumes so that the new block sizes take effect.

IMPORTANT
If you need to configure the device file size of a device, do so with caution. If the device file
size and the actual file size do not match, you may not be able to recover the resulting
clone image.

Cloning fails when Data Domain or NDMP save set of a client is

encountered and there is no client resource on the NetWorker server
NW119156

A clone operation fails when a Data Domain or NDMP save set of a client is encountered
and there is no resource defined for that client on the NetWorker server.

94 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Workaround

Create a client resource on the NetWorker server.

Device status continues to display as “active” after cloning operation

terminated
NW118255

When a cloning operation is terminated by way of the nsrmmd process being stopped, the
status of the device used for the operation continues to display as active in NMC.
Workaround

Refreshing the NMC browser results in the correct status being displayed.

Clone controlled replication with nsrclone may revert to a regular clone if

the NetWorker server’s primary host name is not specified
NW118893

When performing clone controlled replication by using the nsrclone command, ensure
that any NetWorker server alias names are set up correctly. Otherwise, the clone
controlled replication operation may revert to a regular clone operation. This situation can
occur when invoking the nsrclone command on a host that is not the NetWorker server.

For example, suppose that the following command is invoked on a remote storage node,
that is, a storage node that is not on the NetWorker server host:
nsrclone -s mars -vvv -S saveset_id

where mars is an alias for the NetWorker server’s primary hostname, ersat2d049. In this
example, a regular clone operation may be performed instead of a clone controlled
replication operation.
Workaround

To avoid this situation, perform one of two things:

◆ Invoke the nsrclone command with the primary hostname of the NetWorker server. For
the previous example, the command would be:
nsrclone -s ersat2d049 -vvv -S saveset_ID
◆ Ensure that the alias hostname for the NetWorker server is specified in the Aliases
attribute of the NetWorker server’s client resource. For the previous example, you
would add the name mars to the Aliases attribute of the NetWorker client resource
named ersat2d049. When this is done, you can invoke the nsrclone command as
follows:
nsrclone -s mars -vvv -S saveset_id

EMC NetWorker Release 8.1 Release Notes 95

Known problems and limitations

nsrclone/nsrstage performs regular clone for source Data Domain volume

on remote storage node if client resource was not created in NetWorker
server
NW117750

If a client resource for the remote storage node is not created in the NetWorker server,
running nsrclone/nsrstage always performs a regular clone for the source Data Domain
volume on the remote storage node.
Workaround

In order for the NetWorker server to manage and monitor clone operations, the NetWorker
storage nodes, at both the source and target locations, must be clients of the same
NetWorker server. A clone may not be created in a different datazone.

Create or add a client resource for the remote storage node in the NetWorker server.

Concurrent stage with stage or clone from the same Data Domain device or
AFTD is not supported
NW117470

Performing a concurrent stage with the stage or clone from the read-only and read-write
volume of the same Data Domain device or AFTD is not supported. If this is attempted, an
ansrd core dump may occur.

A clone operation hangs if the source volume is on a remote storage node

and is umounted initially but is mounted after the clone attempts to start
NW117179

If the source volume is on a remote storage node and is unmounted, starting a clone
operation will not complete successfully even if the source volume is mounted after the
clone operation attempts to start. The clone program nsrclone will hang with the following
message:
Server <server_name> busy, wait 30 seconds and retry

This issue does not occur if the storage node is on the NetWorker server, that is, when the
storage node is not remote. However, if you do not set Media type required to Data
Domain in the target Backup Clone pool, a regular clone operation will be performed even
if the source is a Data Domain volume.
Workaround

Ensure the source volumes to clone are mounted and the target volume is available for
writing before initiating the clone operation, then set Media type required to Data Domain
in the Backup Clone type pool for the clone's target device, making sure to specify that
this pool is the target for the clone operation.

96 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Clone controlled replication (optimized clone) for a Data Domain Device is

performed only when the Data Domain clone volume is mounted and
available for writing
NW117137

When performing a clone controlled replication operation (optimized clone) from one
Data Domain device to another Data Domain device, ensure that the Data Domain clone
volume is mounted prior to initiating the clone operation. If the Data Domain clone
volume is not mounted and the clone operation is initiated, then a regular clone operation
will be triggered instead of a clone controlled replication operation.
Workaround

There is a new pool attribute, Media type required, that when set will trigger an optimized
clone on the target Data Domain device:
1. Set up a clone pool and assign only Data Domain devices to this pool.
2. Set the Media type required attribute in the Pool resource to Data Domain.

Note: It is especially important to set this new attribute when using clone controlled
replication over WAN, so that the clone operation never falls back to regular cloning over a
long distance network.

When cloning a mix of save sets from different source devices such as
Data Domain, AFTD, and NDMP devices, the same clone target volume may
not be used for all save sets
NW116696

Clone operations that mix save sets from different source devices such as Data Domain
devices, AFTD devices, or NDMP devices, may be written to different target volumes.
Although this behavior is by design, you may prefer to write all save sets in the clone
operation to the same clone volume.
Workaround

If the clone operation includes save sets from different devices and you want all save sets
to be written to the same volume, include only one volume in the clone target pool.

Clone jobs may time out and not complete if 30 or more are scheduled to
occur at the same time
NW115417

Scheduling multiple clone jobs (30 or more) to occur at the same time may result in some
clone jobs timing out and not completing.
Workaround

Do not schedule more than 30 clone jobs to occur at the same time.

EMC NetWorker Release 8.1 Release Notes 97

Known problems and limitations

Path-to-tape cloning of save sets spanning across multiple volumes with

different block sizes fails
LGTsc31596

NDMP path-to-tape cloning fails if the source save set resides on multiple volumes with
different block sizes. For example, when backup is performed using one LTO4 and one
LTO2 volume, the cloning fails with the following error message:
"Error - Current NDMP session's block size 65536 does not match with
new volume's block size 131072".

If performing path-to-tape cloning, save sets should be backed up on similar volumes.

Configuration problems and limitations

NetWorker does not support IPMP feature on Solaris 10
NW123374

When running NetWorker on the Solaris 10 operating system, EMC recommends not
enabling the IPMP feature, which is disabled by default. When IPMP is enabled on a host,
NetWorker may not be able resolve the host, resulting in server-side operations failing to
execute, or executing on the wrong host.

Data Domain host name must be specified in Device access information

attribute when configuring Data Domain devices
NW120856

When configuring Data Domain devices, do not specify an IP address in the Device
attribute Device access information when multiple NICs are enabled on the Data Domain
system. Instead, specify the Data Domain system host name for this attribute.
Note: A Data Domain license is required for each IP address used. It is recommended to
use only one NIC per NetWorker datazone.

NetWorker and device operations can fail when TCP Chimney is enabled
for NICs on Windows Server 2003 or Windows 2008 R2
NW120373

RPC errors can occur with some combinations of TCP Chimney and the NIC driver when
TCP Chimney is enabled for NICs in a Windows Server 2003 or Windows 2008 R2
environment, leading to failed connections. The TCP Chimney feature was introduced in
the Windows Server 2003 Scalable Networking Pack, and is enabled by default for that
release.

This issue can cause a failure of the following NetWorker and device operations:
◆ Scheduled and manual savegroups can fail, with manual savegroups reporting the
error “Lost connection to server, exiting”.
◆ Device connections can fail with the error “Lost media database connection”.
◆ Device operations can fail with the error “RPC send operation failed”, along with a
more specific explanation such as “Broken pipe” or “Connection reset by peer”,
among others.

98 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Workaround

To work around this issue:

1. Disable TCP chimney by running the following command:
netsh int tcp set global chimney=disabled
2. Restart NetWorker services.

More information related to this issue is provided in the following knowledgebase

articles:
◆ http://support.microsoft.com/kb/942861
◆ http://support.microsoft.com/kb/945977
◆ http://social.technet.microsoft.com/forums/en-US/winservergen/thread/8a72f5d6-6f0
9-498a-b1a8-bcede2bfeecd
◆ http://communities.vmware.com/thread/91454?tstart=0&start=195

Failure reported during disaster recovery of VSS SYSTEM BOOT; set

NSR_RECOV_TEMP_CLEANUP variable
NW117057

When performing a disaster recovery in multiple Windows platforms and copying the
registry, a failure may be reported during the recovery of VSS SYSTEM BOOT due to the
size of the PendingRenameFileOperations registry value, which is populated during the
disaster recovery. The error message indicates a lack of system resources.

If this error appears, it is recommended to restart the disaster recovery after setting the
environment variable NSR_RECOV_TEMP_CLEANUP to an appropriate value (for example,
1) in the system space. Setting this variable ensures that the above error does not appear,
and that the recovery and subsequent cleanup of the temporary recover files after reboot
occur without this interruption.

Incremental backup failure may occur due to number of VSS shadow

copies; set NSR_VSS_WAIT_INTERVALS variable
NW114428

Incremental backups may fail due to the number of VSS shadow copies that are created
and released by Windows when a backup is performed. A new environment variable,
NSR_VSS_WAIT_INTERVALS, has been introduced to resolve the issue if a failure occurs.
The syntax of the variable is as follows:
NSR_VSS_WAIT_INTERVALS=<Mutex wait in mins>,<Snap attempt wait time in
seconds>

where <Mutex wait in mins> is the timeout interval that a save instance waits until
requesting a snapshot to the VSS framework. The recommended value in case of a failure
is 30.

<Snap attempt wait time in seconds> is the time interval that occurs between
successive snapshot requests being sent to the VSS framework. The recommended value
in case of a failure is 5.

If the problem persists, higher values for both the parameters can be set.

EMC NetWorker Release 8.1 Release Notes 99

Known problems and limitations

Devices and media problems and limitations

NetWorker may become unresponsive when performing label operations
with NetWorker Data Domain Deduplication Devices
NW122429

NetWorker may fail and become unresponsive when attempting to label Data Domain
devices. This situation can occur if the NetWorker Data Domain device resource is created
on a host that is running a Pentium 3 or lower processor. The first symptom is most likely
to be an error message generated during the attempt to label the device. The message
looks similar to the following:
nsrmm: error, Cannot find an available nsrmmd service for operation on
device <device_name>.
Workaround

If the NetWorker storage node host is running on a supported Intel 32-bit or 64-bit
platform, ensure that the processor is a Pentium 4 or later.

Labelling a Data Domain device to a new pool when the device contains
data belonging to a different pool can result in data loss
NW120540

Labelling a Data Domain device to a customized pool when the device does not belong to
a Data Domain pool (ddpool) will label the volume but will not mount the volume. If this
device contains data that belongs to a different pool prior to labelling the device to the
new pool, the data that was saved to the previous pool will be lost.
Workaround

To prevent data loss, ensure that the data previously backed up on the device is already
cloned or archived to another destination prior to configuring and labelling the device to a
different pool.

To label and mount the device to a customized Data Domain pool:

1. Unmount the device, and disconnect the device from its current pool if it is connected
to one.
2. From NMC, go to Media > Media Pool, then select the pool name for the customized
Data Domain pool.
3. Go to Properties > Selection Criteria > Devices, and enable the checkboxes in order to
select from the list of devices the device to be used for this pool. Click OK.
4. Label that device to the customized Data Domain pool from step two (the default
setting is Mount after Labeling).

The device will be labeled and mounted successfully.

100 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Labelling a Data Domain device fails after converting the device from AFTD
to Data Domain
NW118590

Labelling a Data Domain device fails with the following error message after the device has
been converted from an AFTD to a Data Domain device:
labelling operation failed on rd=devicename.RO:volume error number is
a read only attribute

This error occurs due to a difference in permissions. When the device is used as an AFTD,
the owner of the files and directories is root, whereas the owner of the Data Domain
device will be the user account configured in the device resource.
Workaround

Perform the following to remove the volume from the media database and change
permissions:
1. To remove the volume from the media database, run the following
nsrmm -dy <volume name on aftd>
2. Change permissions to 777 by running the following:
\rm -rf <aftd_path>/*
chmod 777 <aftd_path>

The device can now be labelled successfully.

GUI problems and limitations

Username/password validation may fail using NMC New Device wizard for
file system browsing if storage node is UNIX
NW120536

When using the NMC New Device wizard to configure an AFTD, username/password
validation for browsing the file system may fail if the storage node is a UNIX host. This
failure occurs if the system is missing the Pluggable Authentication Modules (PAM)
library, or when the rule in the pam.conf file (/etc/pam.conf) for OTHER service is set to
deny.
Workaround

If validation fails using the New Device wizard on a UNIX storage node, install the PAM
package appropriate to your environment if it is not already installed, and then modify the
pam.conf file so that the rule for OTHER service is not set to deny. Refer to your operating
system's documentation for more information.

EMC NetWorker Release 8.1 Release Notes 101

Known problems and limitations

Data Domain systems do not display in NMC Enterprise window or

NetWorker Administration Devices window upon reinstalling NMC after a
complete uninstall
NW120324

When NMC is uninstalled and the NMC database is removed, upon reinstalling NMC and
adding the NetWorker server that has Data Domain devices configured, the Data Domain
system does not appear in the NMC Enterprise window or the NetWorker Administration
Devices window.
Workaround

To add one of these hosts to the Enterprise:

1. From the Console window, click Enterprise.
2. In the left pane, right-click Enterprise, then select New > Host. The Add New Host
wizard appears.
3. Enter a hostname, IP address, DNS name, or WINS name in the Host Name attribute,
then click Next.

Note: Host names and aliases cannot exceed 80 characters.

4. Select the server type and click Next.

5. Follow the instructions for configuring the selected host type, then click Finish.

After Data Domain Systems are added to the Enterprise, they will appear in NMC
Enterprise and the NetWorker Administration Devices window under the Data Domain
Systems folder for corresponding NetWorker servers.

Monitoring of SNMP events for Data Domain not available on HP-UX

NW118886

HP-UX operating systems do not monitor SNMP events for Data Domain systems.

NSR clone resources that are created with the nsradmin program cannot
be edited as scheduled resources in the GUI
NW115920

NSR clone resources that are created with the nsradmin command line program cannot be
edited as scheduled clone resources in the Console GUI. Scheduled clone resources can
be selected for editing in the Console by opening the NetWorker Administration interface
and then selecting Configuration > Clones.
Workaround

Perform one of the following:

◆ Create scheduled clone resources in the GUI. Scheduled clone resources can be
created and edited in the GUI by opening the NetWorker Administration interface and
then by selecting Configuration > Clones.

102 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

◆ If you must create a NSR clone resources with the nsradmin program, create a
corresponding NSR task resource with the nsradmin program. Together, these
resources will enable you to edit the clone item as a scheduled resource in the GUI.
The corresponding NSR task resource must have its name and action attributes
specified as follows:
• name: clone.nsrclone_resource_name;
• action: "NSR clone:nsrclone_resource_name";
For example, if the NSR clone resource was named TestClone1, the name and action
attributes of the NSR task resource would be:
• name: clone.TestClone1
• action: NSR clone: TestClone1
These entries are case-sensitive.

Installation and upgrading problems and limitations

HomeBase Agent 6.2.1 does not automatically install as part of NetWorker
Client installation on Red Hat Enterprise Linux 5.5 and later
NW124547

The NetWorker 7.6 Service Pack 1 Client installation does not automatically install
HomeBase Agent version 6.2.1 on the Red Hat Enterprise Linux version 5.5 and later
platform. However, the NetWorker Client Installer does extract the HomeBase package in
the /opt/homebase folder.
Workaround

Manually run the /opt/homebase/setup-homebase.sh script to install the HomeBase

Agent.

nwclust.pl script must be copied to AutoStart binary installation directory

if using NetWorker in AutoStart 5.4 or later cluster environment
NW119766

Starting from release 5.4, AutoStart no longer provides the nwclust.pl script. When
configuring NetWorker in an AutoStart 5.4 or later cluster environment, contact support to
obtain the nwclust.pl script, then copy this script to the AutoStart binary installation
directory (for example, /opt/EMCas54/bin).

Restore problems and limitations

NetWorker/HomeBase does not support disaster recovery for Active
Directory and DFS on Windows 2008
NW123851

When BMR is enabled for NetWorker in a Windows 2008 environment, disaster recovery is
not supported for Active directory/DFS deployment. A full system backup of the
HomeBase profiles is successful, but recovery of DFS writers fails.

EMC NetWorker Release 8.1 Release Notes 103

Known problems and limitations

After disaster recovery, remote desktop connection to the recovered

system fails with authentication error
NW124657

After performing a disaster recovery with BMR-enabled, attempting to connect to the

system using the remote desktop connection fails with the error “An authentication error
has occurred.”
Workaround

Change the remote desktop connection file (Default.rdp) by adding the following line to
the file:
enablecredsspsupport:i:0

Remote cluster services not started after Authoritative Restore of MSCS

cluster database
NW119115

After performing an authoritative restore of the MSCS cluster database on Windows Server
2008 R2, the remote cluster services are not started. An error message similar to the
following appears:
recover: Unable to start Cluster Service on all the remote nodes
Workaround

Restart the cluster services on the remote cluster nodes manually.

Bootmgr file deleted upon recovery of C:\directory with VSS System save
set selected
NW116575

If the bootmgr file is located under the C:\ directory, a backup of folders under “C :\” with
“VSS SYSTEM FILESET: \” will also back up the bootmgr file as part of the file system.
During recovery, if you select the entire C:\ directory and the Overwrite option is selected,
the original bootmgr file gets deleted and the following errors appear:
52973: Winworkr: Didn't recover requested file C:\bootmgr
52973: Winworkr: Didn't recover requested file C:\config.sys

This leads to corruption of the machine. Upon rebooting the machine, a message appears
indicating "bootmgr file is missing, press Ctrl+Alt+Del to restart".
Note: This issue is only seen when a backup is performed using NMC (scheduled backups)
and the machine has its bootmgr file in C:\. This does not occur with client-initiated
Winworkr backups. Note also that a machine running Windows 2008 R2 can be configured
to hold the boot related files in a separate volume.

Workaround

During recovery, mark the backed up folders present in the C:\ directory one-by-one; do
not include the bootmgr and Config.sys files.

104 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Note: If you select the entire C:\ directory and then try to unmark only bootmgr and
config.sys, the recovery will not work.

NDMP: recover or nsrndmp_recover displays debug messages during DSA

recover on Windows even though recover is successful
NW115372

When performing a DSA recover on a Windows system using recover or nsrndmp_recover,

debug messages appear, even though the recovery completes successfully. These
messages can be ignored.

VMware Consolidated Backup (VCB) problems and limitations

Single step recovery using winworkr fails when VMware Standalone
Converter 4.0.x and later is installed
NW122363

Performing a Single Step recovery from the proxy host using the NetWorker User program
(winworkr) fails when the installed version of VMware Standalone Converter is 4.0.x and
later.
Workaround

Install VMware Standalone Converter version 3.0.3 to the proxy host.

Hot-add operation does not work when Windows proxy is a virtual

machine residing on an ESX 4.1 host
NW121774

Due to a VMware limitation, performing a VCB backup on NetWorker with

VCB_TRANSPORT_MODE=hotadd fails when the Windows proxy is a virtual machine
running on an ESX 4.1 host.

VCB topology map shows Helper VM for hotadd mode is associated with
same group used for hotadd proxy
NW118106

The Topology map shows that the helper VM used for hotadd mode is associated with the
same group that is used for the hotadd proxy VM.

Single Step recovery of VM within a vApp fails with P2V error

NW117233

Performing a Single Step recovery of a VM within a vApp fails with the following error:
winworkr: Restore failed with error P2VError
UNKNOWN_METHOD_FAULT(vmodl.fault.InvalidType)
Workaround

Perform disaster recovery to a staging location and then use the latest VMware
Standalone Converter 4.x to perform a restore to ESX/VC.

EMC NetWorker Release 8.1 Release Notes 105

Known problems and limitations

Single Step recovery of Windows 2008 x64 VM fails with P2V error
NW117714

Performing a Single Step recovery of a Windows 2008 x64 VM initiated from the
NetWorker server fails with the following error:
Check dest params in progress...
...failed with error P2VError
UNKNOWN_METHOD_FAULT(sysimage.fault.OsVersionNotFound)
Workaround

Perform disaster recovery to a staging location and then use the latest VMware
Standalone Converter 4.x to perform a restore to ESX/VC.

PVSCSI controller changes to BusLogic controller, leading to boot failure

for x64 VMs and performance issues for x86 VMs
NW118197

When performing a Single Step recovery of a VM, the Paravirtual SCSI controller on
Windows 2003 VMs gets changed to a BusLogic Parallel controller, resulting in decreased
performance for Windows x86 VMs and boot failures for Windows x64 VMs. This issue is
only seen if using SSR with VMware Converter 3.0.3.
Workaround

Perform disaster recovery to a staging location and then use the latest VMware
Standalone Converter 4.x to perform a restore to ESX/VC.

SSR disaster recovery of SuSE Linux Enterprise Server 11 VM fails with

P2V error
NW118200

Performing an SSR disaster recovery of an SLES Enterprise Server 11 VM fails with the
following error:
winworkr: Restore failed with error P2VError UFAD_SYSTEM_ERROR(Failed
to find conversion from sles11 to type:
vim.vm.GuestOsDescriptor.GuestOsIdentifier)
Workaround

Perform disaster recovery to a staging location and then use the latest VMware
Standalone Converter 4.x to perform a restore to ESX/VC.

Vmxnet3 adapter changes to e1000 adapter during VM recovery

NW118486

When performing a Single Step recovery of a VM, a vmxnet3 network adapter gets
changed to an e1000 network adapter.
Workaround

Perform disaster recovery to a staging location and then use the latest VMware
Standalone Converter 4.x to perform a restore to ESX/VC.

106 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

NetWorker cannot use mapped drive for VCB mount point

NW113740

Attempting to use the VCB mount point on a CIFS share (mapped drive) fails with an error
indicating that the device is not a writeable volume.
Workaround

Back up to the VCB mount point on the local disk instead of the CIFS share.

NetWorker 7.6
Table 13 on page 107 identifies problem issues and limitations discovered in NetWorker
7.6.

The known limitations are separated into the following categories:

◆ “Backup problems and limitations” on page 111
◆ “Cloning and Staging problems and limitations” on page 114
◆ “CLI problems and limitations” on page 116
◆ “Cloud device problems and limitations” on page 116
◆ “Configuration problems and limitations” on page 116
◆ “Installation and upgrading problems and limitations” on page 117
◆ “Messaging problems and limitations” on page 118
◆ “Performance problems and limitations” on page 119
◆ “Restore problems and limitations” on page 120
◆ “VMware Consolidated Backup limitations and descriptions” on page 121
◆ “General problems and limitations” on page 128

Known limitations related to internationalization support in NetWorker 7.6 are provided in

the section “Internationalization support” on page 175

Table 13 Limitations discovered in NetWorker release 7.6 (page 1 of 5)

Operating Product
Issue number Description of limitation system affected feature

“LGTsc30233” AIX deduplication client fails with AVCTL error due to case-sensitivity AIX Backup
on page 111

“LGTsc31729” Savegroup fails when backing up VSS save sets with deduplication enabled Windows Backup
on page 111

“LGTsc31933” Multiple passphrases cannot be used for cloud-based encrypted backup and Windows Backup
on page 112 recovery

“LGTsc31401” Selecting “Discover” option in Software Administration wizard causes nsrcpd Solaris Backup
on page 112 process to hang

“LGTsc28262” NetWorker Server may take a long time to restart if size of media management Windows Backup
on page 112 database is very large

EMC NetWorker Release 8.1 Release Notes 107

Known problems and limitations

Table 13 Limitations discovered in NetWorker release 7.6 (page 2 of 5)

Operating Product
Issue number Description of limitation system affected feature

“LGTsc08220” Limits in NetWorker support for NTFS Windows Backup

on page 113

“LGTpa88903 Cannot backup two NetWorker clients with the same hostname but different fully All Backup
” on page 113 qualified domain names (FQDNs)

“LGTsc32100” DFS backup failure under certain conditions if VSS is enabled Windows Backup
on page 113

“LGTsc19011” Savegrp may fail if a client has more than 1116 save sets AIX, Solaris Backup
on page 113

“LGTsc29210” Backups may fail if server parallelism is set to maximum value All Backup
on page 113

“NW105699” NDMP path-to-tape cloning fails to clone save sets spanning across two volumes Linux Cloning
on page 114

“NW105306” Status of aborted clone is not shown under Show Manual Clone History in NMC Solaris Cloning
on page 114

“NW105684” Path-to-tape cloning check fails if spanning of first fragment of the source image Linux Cloning
on page 114 and destination spanning occur simultaneously during cloning

“LGTsc31283” Running nsrclone -J <storage node> hangs for any source device type All Cloning
on page 114

“LGTsc29276” Path-to-tape cloning requires a separate volume to write the end notes Linux Cloning
on page 115

“LGTsc31929” Cloning of Path-To-Tape instance from a tape of different block or device file size All Cloning
on page 115 than that of the original source tape not supported

“LGTsc29190” Using nsrclone to clone an NDMP clone hangs if source volume is not mounted Solaris Cloning
on page 115

“LGTsc32017” nsrlic output may not list all connected clients to NMDA UNIX enabler UNIX CLI
on page 116

“NW105390” Directed recover initiated from server may fail with machines in different domains Windows CLI
on page 116 from CLI on Windows

“NW104933” Cloud device created with target session attribute of 1 even if another value is Windows Cloud device
on page 116 specified

“LGTsc32171” NDMP target session value changed to 32 when evaluation mode expires Linx Configuration
on page 116

“LGTsc29940” Device file size value cannot be reset to default unless nsrmmd or NetWorker Solaris Configuration
on page 116 services are restarted

“LGTsc28693” Quantum DXi: May not be able to configure library from a particular node in a DDS Solaris Configuration
on page 117 configuration

“LGTsc30786” Default client parallelism should be set to 4 except for the NetWorker server All Configuration
on page 117

“NW110327” nsrjobd running after uninstallation of Sun StorageTek EBS on Windows Windows Installation
on page 118

“LGTsc32944” Library disabled after upgrading from NetWorker 7.4 release to NetWorker 7.6 AIX Upgrading
on page 118

108 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Table 13 Limitations discovered in NetWorker release 7.6 (page 3 of 5)

Operating Product
Issue number Description of limitation system affected feature

“NW105501” Usage error does not appear when nsrda binaries run from command line Solaris Messaging
on page 118

“NW105414” Logging and warning messages not sufficient when reaching soft runtime limit All Messaging
on page 118

“LGTsc29953” Running ndmpsup with full disk specified for output location does not return error WIndows Messaging
on page 119

“NW105832” No warning message appears when staging is started for disabled destination Windows Messaging
on page 119 device

“NW105777” Save output error and parse error messages appear in Savegroup log and NMC WIndows Messaging
on page 119 Group Details upon backup with directives turned on

“LGTsc29791” NMC may consume increasing amounts of memory on Windows Windows Performance
on page 119

“LGTsc24128” HP TruCluster with DNS enabled may result in slow connections to the NetWorker All Performance
on page 119 server

“LGTsc29958” Empty directory for NDMP is not recovered during Direct Recover Linux Restore
on page 120

“LGTsc33060” Volume mount point is recovered as a folder Windows Restore

on page 120

“LGTsc30685” Recovery reports are not generated immediately after a successful recover Solaris Restore
on page 120

“LGTsc17739” Changing the alias order in the /etc/hosts file causes NetWorker to rename the HP-UX Restore
on page 120 NetWorker server and create a new client resource with a different clientid

“LGTsc00742” The recover command may fail if the NetWorker server’s short name cannot be All Restore
on page 120 resolved from the client

“LGTsc24329” Client file index recovery fails for NDMP client on Windows Windows Restore
on page 121

“LGTsc25674” Windows file permissions may be changed after a full recovery Windows Restore
on page 121

“LGTsc31671” Unable to recover save sets on NDMP path-to-tape clone volume that has been Solaris Restore
on page 121 deleted and then restored using scanner command

“LGTsc30685” Recovery reports are not generated immediately after a successful recover Solaris Restore
on page 120

“LGTsc00742” The recover command may fail if the NetWorker server’s short name cannot be All Restore
on page 120 resolved from the client

“LGTsc24329” Client file index recovery fails for NDMP client on Windows Windows Restore
on page 121

“LGTsc25674” Windows file permissions may be changed after a full recovery Windows Restore
on page 121

“LGTsc31671” Unable to recover save sets on NDMP path-to-tape clone volume that has been Solaris Restore
on page 121 deleted and then restored using scanner command

“LGTsc32763” RECOVER_FULL_PATHS variable for NetApp filer recovery not supported UNIX Restore
on page 121

EMC NetWorker Release 8.1 Release Notes 109

Known problems and limitations

Table 13 Limitations discovered in NetWorker release 7.6 (page 4 of 5)

Operating Product
Issue number Description of limitation system affected feature

“NW114517” Index browsing (file selection recovery) may not be supported when using Windows VCB
on page 121 VMware clones and the hotadd transport mode

“NW105846, Index browsing fails when traditional backups are performed pior to VCB/VADP Windows VCB
NW105379” backups
on page 122

“NW105962” Single step recovery of VM on Virtual Center or ESX server does not work if Windows VCB
on page 122 non-default port number used

“NW105583” Backup with VCB Directive applied to Windows 2000 VM does not skip system32 Windows VCB
on page 123 folder under C:\WINNT

“NW126643, Multiple “directory is not empty” messages appear during full image recovery or Windows VCB
NW105298” drive letter recovery
on page 123

“NW105265” Single-file recovery is not supported for VMs containing non-ASCII filenames Windows VCB
on page 123

“NW105098” Performance is slow on Windows VMs when multiple disaster recovery sessions Windows VCB
on page 123 are performed simultaneously

“NW105095” Stopping the savegroup does not stop the VCB mounter operation WIndows VCB
on page 123

“LGTsc32604” Single-step recovery has limitation to restore VMs with thin disks to ESX 4.0 Windows VCB
on page 124 Server.

“LGTsc32380” File-level recovery of an encrypted file is not supported Windows VCB

on page 124

“LGTsc30075” Recovery of a VM host with multiple drives displays "Directory not empty" in the Windows VCB
on page 125 recover GUI

“LGTsc30810” Additional system generated or user files are recovered during save set recovery Windows VCB
on page 125 from an incremental backup

“LGTsc30813” Restored virtual machine starts in forceful powered off state during FULLVM Windows VCB
on page 125 restore

“LGTsc31164” nsr resource takes default values if invalid application information inputs are Windows VCB
on page 125 provided for the proxy server

“LGTsc30927” GPT disks are not supported for single file recovery (SFR) Windows VCB
on page 126

“LGTsc31190” Single file recovery (SFR) cannot be performed on disks with no drive letter Windows VCB
on page 126

“LGTsc31499” Single file recovery (SFR) is not supported for VM's configured with dynamic disks Windows VCB
on page 126

“LGTsc31545” Memory footprint for nsrvcb_save is huge during VMware backups Windows VCB
on page 126

“LGTsc31562” VM should be in the "powered on" state for the first backup Windows VCB
on page 127

“LGTsc31575” Shared folder settings are not restored from a file-level restore on the VM Windows VCB
on page 127

110 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Table 13 Limitations discovered in NetWorker release 7.6 (page 5 of 5)

Operating Product
Issue number Description of limitation system affected feature

“LGTsc31651” Any system changes to the VM requires a full backup Windows VCB
on page 127

“LGTsc31735” Save set consolidation is not supported for VCB backups Windows VCB
on page 127

“LGTsc31737” No directives are supported for VCB backup when save set is “*FULL*” and the Windows VCB
on page 127 level is “FULL”

“LGTsc31971” Backup of a powered off VM is only possible if the VM was turned off gracefully Windows VCB
on page 127

“NW110377” Secure Storage Library not initialized during Lockbox entry creation due to HP-UX General
on page 128 missing library (fixed for the 7.6 Service Pack 1 release) problems

“NW105048” NetWorker does not save or recover DFS empty directories Windows General
on page 128 problems

“LGTsc32722” Archived data not properly indexed on MAC-OS MAC General

on page 128 problems

“LGTsc30717” NetWorker services might not start due to lock files All General
on page 128 problems

“LGTsc32410” Compression asm does not work with other asms Windows General
on page 128 problems

Backup problems and limitations

AIX deduplication client fails with AVCTL error due to case-sensitivity
LGTsc30233

Deduplication backup of a NetWorker client on AIX may fail with an AVCTL error. This is
because client names or domains on the Avamar server are not case-sensitive at the
creation time, but are case-sensitive when logging on to a particular domain account.
Workaround

Keep the same naming convention of client name or domain (case-sensitive) after the first
client record creation.

Savegroup fails when backing up VSS save sets with deduplication

enabled
LGTsc31729

When backing up the VSS save sets on Windows 2008 with deduplication enabled, the
savegroup fails with the following error in NMC and the savegroup log upon completion:
vm20-win-6:VSS OTHER:\ 57803:save: Saving with De-Duplication enabled
vm20-win-6:VSS OTHER:\ VSS OTHER: No writers found for this save set:
VSS OTHER.

EMC NetWorker Release 8.1 Release Notes 111

Known problems and limitations

Multiple passphrases cannot be used for cloud-based encrypted backup

and recovery
LGTsc31933

Multiple passphrases cannot be used for cloud-based encryption and decryption, due to
the difficulty of specifying a passphrase for decryption that is different from the one that
was used for encrypting that data. The passphrase used when backing up encrypted data
must be the same one used when performing a recovery and decrypting the data.

The recover -p option, which provides a way to specify a passphrase different from the
one currently configured on the NetWorker server, cannot be used for encrypted cloud
backups.

The following limitations also apply to passphrase use in NetWorker:

◆ A cloud device can only support one single passphrase for both backup and recovery
when encryption is enabled.
◆ Changing the datazone passphrase impacts all newly initiated backup and recovery
operations and should be avoided.
◆ If a backup using an old passphrase needs to be recovered, that old passphrase must
be restored on the NetWorker server during the recovery process.

Selecting “Discover” option in Software Administration wizard causes

nsrcpd process to hang
LGTsc31401

When you perform “Add products to the repository” using the Software Administration
wizard in NMC and select Discover after successful completion of the task, the nsrcpd
process becomes unresponsive and the application hangs.
Workaround

To avoid this issue, perform one of the following:

◆ Kill the nsrcpd process, then restart the daemons. Selecting the Discover option in
NMC now completes successfully.
◆ Run the command line utility nsrpush to perform the operation instead of using NMC.

NetWorker Server may take a long time to restart if size of media

management database is very large
LGTsc28262

If the NetWorker media management database is very large, the NetWorker server may
take a long time to establish client connections when it is restarted. The reason is that a
consistency check of the media management database is triggered when the server is
restarted.

To reduce the size of the media management database, run the nsrim -C command. Be
aware that this command may take a long time to run and that the NetWorker server will
be unavailable during this time. Run the command when the NetWorker server is not busy.

More information about reducing the media management database is provided in the
NetWorker Administration Guide.

112 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Limits in NetWorker support for NTFS

LGTsc08220

The NetWorker software supports the backup and recovery of NTFS using Win32
API-defined subsystem. Besides Win32 Subsystem, NTFS also supports other subsystems
like Portable Operating System Interface (POSIX). The Win32 subsystem supports certain
file naming conventions, and does not have ability to support file naming conventions of
other subsystems like POSIX, such as the ability to differentiate between filenames that
are identical except for uppercase and lowercase characters (for example, temp.txt and
TEMP.txt) and the ability to have multiple dots (.) in filenames.

POSIX applications that access NTFS file systems may implement file naming convention
differently than the Win32 subsystem. It is possible that certain file naming features may
be present that are not part of the Win32 subsystem. If this is the case, the NetWorker
software will not correctly back up these files.

Cannot backup two NetWorker clients with the same hostname but
different fully qualified domain names (FQDNs)
LGTpa88903

NetWorker cannot cope with two different machines that have the same hostname but
different FQDNs. For example, if two divisions of a company have a machine named
"mailserver.abc.com" and "mailserver.xyz.com," they will be unable to back up the
second host.

DFS backup failure under certain conditions if VSS is enabled

LGTsc32100

If VSS is enabled, a DFS backup or a VSS SYSTEM BOOT backup may fail under the
following conditions:
◆ The namespace folder is not a shared folder target.
◆ Replication is configured for the namespace.
◆ Files exist in the namespace folder and have replicated to the member server.

Workaround

To avoid this issue, create a namespace in the folder target or do not enable replication.

Savegrp may fail if a client has more than 1116 save sets
LGTsc19011

A savegrp operation may fail when client is configured with more than 1116 save set files.
The following message may appear:
Probe job had unrecoverable failure(s), please refer to daemon.raw for
further details

Backups may fail if server parallelism is set to maximum value

LGTsc29210

Index backups may fail with the following error message if the maximum server
parallelism is set to 400 and there were 400 simultaneous sessions running:

EMC NetWorker Release 8.1 Release Notes 113

Known problems and limitations

server-name:index 1 retry attempted

server-name:index P?: No such file or directory
Workaround

Reduce the server parallelism value.

Cloning and Staging problems and limitations

NDMP path-to-tape cloning fails to clone save sets spanning across two
volumes
NW105699

NDMP path-to-tape cloning of a volume fails to clone save sets that span across two
volumes. However, when that save set is cloned individually, cloning is successful.

Status of aborted clone is not shown under Show Manual Clone History in
NMC
NW105306

When the nsrndmp_clone operation is aborted from the command line, Show Manual
Clone History in NMC does not display the status.
Path-to-tape cloning check fails if spanning of first fragment of the source image and
destination spanning occur simultaneously during cloning

NW105684

The path-to-tape cloning check, which checks the consistency of the source tape file size
while cloning is in progress, fails if spanning of the first fragment of the source image and
destination spanning occur at the same time during cloning. A series of nsrndmp_clone
error messages similar to the following will display, indicating the tape file size error and
identifying which save set failed cloning.
nsrndmp_clone: Tape server paused: reached the end of file
nsrndmp_clone: Error - Incorrect tape file size 1073741824
nsrndmp_clone: Tape server halted: The backup is aborted by operator.
nsrndmp_clone: Failed to clone /Builds (ssid 3119103906)
nsrndmp_clone: Failed to clone the following save sets: 3119103906
Workaround

Manually clone the failed save set to another destination volume that has enough free
space to accommodate the save set.

Running nsrclone -J <storage node> hangs for any source device type
LGTsc31283

If nsrclone is specified with -J <storage node> for any source device type and that storage
node is not permitted for use based on the affinity lists, nsrclone will loop indefinitely and
a failure level error appears.

114 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Path-to-tape cloning requires a separate volume to write the end notes

LGTsc29276

During path-to-tape cloning, mminfo does not display end notes as part of the clone
image.

For example, configure NetWorker Server, Client, or Storage node for path-to-tape.
Configure the source and target jukeboxes with the same type of device and default size.
Manually save multiple save sets to the source volume so that one of the volume spans to
another volume. Using the following command, execute path-to-tape cloning for all the
save sets including the spanned save sets in the same sequence as they are backed up:
nsrndmp_clone -b "Default Clone"-J <clone storage node> -p -S <ssid>

When the save sets span across two volumes and writes only end notes in the third
volume, then the clone completion message and the mminfo command displays only the
first two volumes and skips the third volume as it does not contain any save sets.
Note: The third volume is not required in order to complete the restore.

Cloning of Path-To-Tape instance from a tape of different block or device

file size than that of the original source tape not supported
LGTsc31929

Cloning of path-to-tape (opaque) clone is not supported for destination tapes with
different block or device file size from that of the original source.
Example Create a backup save set (Clone instance 0) on a LTO-3 tape with a block size of 128KB
and a device file size of 4 GB. Using path-to-tape, create a clone save set (Clone instance
1) to another LTO-1 tape with a block size of 64KB and a device file size of 1 GB. When
attempting to create a new instance from Clone instance 1, cloning fails with the following
error:
nsrndmp_clone: Incorrect tape file size <current_tape_file_size>
nsrndmp_clone: Set device file size to <recommended_value> in source
device properties
Workaround

Change the source device file to the value suggested by the nsrndmp_clone message.
Once cloning is complete, change the source device file to its original value.

Using nsrclone to clone an NDMP clone hangs if source volume is not

mounted
LGTsc29190

If you perform an NDMP clone of a save set and then use nsrclone to clone the NDMP
clone, the operation will hang if the source volume is not mounted.
Workaround

Mount the source volume manually before starting the nsrclone operation.

EMC NetWorker Release 8.1 Release Notes 115

Known problems and limitations

CLI problems and limitations

nsrlic output may not list all connected clients to NMDA UNIX enabler
LGTsc32017

The output produced by running nsrlic -v may not list all Virtual clients as connected
clients to the NMDA UNIX enabler.
Workaround

Add NMDA UNIX enabler, then backup the virtual client for the Solaris host first.

Directed recover initiated from server may fail with machines in different
domains from CLI on Windows
NW105390

Directed recovery may fail when performing the directed recover on the Windows
NetWorker server to the remote client when the NetWorker server and client are in
different domains. In this case, perform the directed recovery from the NetWorker User
program.

Cloud device problems and limitations

Cloud device created with target session attribute of 1 even if another
value is specified
NW104933

When a cloud device is created using nsradmin or NMC with a target sessions attribute
other than 1, the NSR device resource is always created with a target sessions attribute of
1 rather than the specified value.

Configuration problems and limitations

NDMP target session value changed to 32 when evaluation mode expires
LGTsc32171

The NDMP device target session attribute is changed to 32 after the expiration of
evaluation mode.
Workaround

Manually change the Target session attribute from 32 to 1 and then enable the NDMP
drives.

Device file size value cannot be reset to default unless nsrmmd or

NetWorker services are restarted
LGTsc29940

There is no way to reset the value of "device file size" in NMC to the default value without
restarting nsrmmd or restarting the NetWorker services.

116 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Quantum DXi: May not be able to configure library from a particular node
in a DDS configuration
LGTsc28693

When virtual devices, tape libraries and drives exported from the DXi host are configured
in a DDS (Dynamic Drive Sharing) configuration and one of the DXi libraries is deleted
from the DDS configuration and subsequently rescanned and reconfigured, the operation
may fail with the error message "Storage node xxxxx is not part of storage node list
associated with NSR unconfigured library". This can occur due to a DXi SCSI
reserve/release issue after a new install or server/storage node reboot.
Workaround

Perform one of the following:

◆ Reboot the DXi hosts, then perform a re-scan from the storage node that encountered
the initial failure. The reboot of the DXi forces a release of the SCSI reservation.
◆ Configure the failed DXi library from another storage node in the DDS configuration.

Default client parallelism should be set to 4 except for the NetWorker

server
LGTsc30786

The client parallelism attribute for a NetWorker client is set to 12 by default. This value
can cause problems with interoperability features such as Avamar, EMC Celerra backups,
and VSS backups.
Workaround

Set the client parallelism attribute on NetWorker clients to 4. However, keep the
NetWorker server client parallelism attribute set to 12.

Installation and upgrading problems and limitations

Upgrade of NMC for NetWorker 7.4 Service Pack 5 to release 7.6 may result
in core dump on AIX
NW113316

When performing an upgrade from NetWorker 7.4 Service Pack 5 to release 7.6 on AIX,
uninstallation of the previous NMC may result in a core dump of the dbsrv9 process, with
a core file created in the <NMC install dir>/cores/gstd directory. This has no impact on
Console functionality after the upgrade, and can be ignored.

NETWORKER.cfg file must be backed up prior to upgrading from any

NetWorker version 7.5 Service Pack 1 or earlier
NW110457
When upgrading from any version of NetWorker 7.5 Service Pack 1 or earlier, the
NETWORKER.cfg file must be backed up on each client and used to replace the new
NETWORKER.cfg file created after the upgrade, in order to maintain user preferences.

EMC NetWorker Release 8.1 Release Notes 117

Known problems and limitations

nsrjobd running after uninstallation of NetWorker on Windows

NW110327

The nsrjobd process continues to run after NetWorker has been uninstalled on Windows.
As a result, the NetWorker Installation folder cannot be deleted.
Workaround

Using Task Manager, kill the nsrjobd process manually and then delete the NetWorker
installation folder.

Library disabled after upgrading from NetWorker 7.4 to NetWorker 7.6

LGTsc32944

After upgrading from a NetWorker 7.4 release to NetWorker 7.6, configured libraries may
not become available and, after several unsuccessful connection attempts, may be
disabled. This may be due to a Control Port change.
Workaround

To work around this issue:

1. Disable the library.
2. Re-scan the Storage node where the failure occurred.
3. Re-enable the library.

The Library now appears in the Ready State.

Messaging problems and limitations

This section details the problems and limitations related to error messages.

Usage error does not appear when nsrdsa binaries run from command line
NW105501

nsrdsa binaries are not meant to be run from the command line. However, when this is
attempted, NetWorker hangs and does not display an error notifying the user of the
incorrect usage.

Logging and warning messages not sufficient when reaching soft runtime
limit
NW105414

When a soft runtime limit is reached for any operation, the logging and warning messages
do not provide any information about the limit being exceeded.

118 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Running ndmpsup with full disk specified for output location does not
return error
LGTsc29953

When ndmpsup is run (ndmpsup -c client -o <disk>) and the location specified for the
output file is a disk that is full, no error message appears indicating that the disk is full,
and only a partial output file is created.

No warning message appears when staging is started for disabled

destination device
NW105832

No warning or error message appears when staging is started and the destination device
is disabled. An error message should display indicating that NetWorker is waiting for 1
writable device.
Workaround

Check the device configuration prior to staging and enable the device if necessary.

Save output error and parse error messages appear in Savegroup log and
NMC Group Details upon backup with directives turned on
NW105777

When backing up a client with directives turned on, the savegroup passes successfully.
However, parse error messages and save output errors appear in the Savegroup log file
and NMC Group Details window, such as "bad directory specification for /nsr/tmp".

Performance problems and limitations

NMC may consume increasing amounts of memory on Windows
LGTsc29791

Over a period of time, NMC may consume large amounts of memory that could result in
the system becoming unresponsive.

HP TruCluster with DNS enabled may result in slow connections to the

NetWorker server
LGTsc24128

If DNS is enabled on TruCluster, connections to the NetWorker Server are delayed about
30 seconds each. Also the initial start of the NetWorker server may take longer.
Workaround

Disable DNS on the TruCluster server.

EMC NetWorker Release 8.1 Release Notes 119

Known problems and limitations

Restore problems and limitations

Empty directory for NDMP is not recovered during Directed recover
LGTsc29958

Performing a Directed recover of an empty directory for NDMP does not recover the empty
directory, even though the operation is reported as successful.

Volume mount point is recovered as a folder

LGTsc33060

When performing a mount point recovery, the volume mount point is recovered as a folder
instead of a mount point.
Workaround

To recover the data within the mount point:

1. Manually create the mount point (if it does not exist already).
2. Use winworkr to recover the data under the mount point.

This procedure is documented in the NetWorker Administration Guide.

Recovery reports are not generated immediately after a successful recover

LGTsc30685

After a successful recover, a subsequent savegroup completion is required to generate

the recovery report as the index is updated only after the completion of a save group.

Changing the alias order in the /etc/hosts file causes NetWorker to

rename the NetWorker server and create a new client resource with a
different clientid (fixed in NetWorker 7.6 SP3)
LGTsc17739

The NetWorker server always takes the first alias specified in the /etc/hosts file as the
NetWorker server name. If the first alias in /etc/hosts is different than the current
NetWorker server name client resource, the next time the NetWorker daemons start,
NetWorker will rename the NetWorker server to the first alias listed in /etc/hosts, and
assign a different clientid.
Workaround

To avoid this issue, when adding or modifying the /etc/hosts file, ensure the first alias
exactly matches the current NetWorker server name client resource.

The recover command may fail if the NetWorker server’s short name
cannot be resolved from the client
LGTsc00742

The recover operation may fail if the short name of the NetWorker server cannot be
resolved from the client performing the recover.

120 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Client file index recovery fails for NDMP client on Windows

LGTsc24329

Recovery of the client file index fails when running nsrck -L7 for an NDMP client that was
running on Windows.

Windows file permissions may be changed after a full recovery

LGTsc25674

If all of the system files are recovered and “overwrite existing files” was selected, some
system configuration permissions and security settings may be changed.

Unable to recover save sets on NDMP path-to-tape clone volume that has
been deleted and then restored using scanner command
LGTsc31671

You may be unable to recover save sets from a volume that has been restored using the
scanner command in the following situation:
1. Perform a multiplexed backup.
2. Perform an NDMP Path-to-tape clone of the volume.
3. Delete the volume containing the clone instance.
4. Run the scanner command on the deleted volume.
5. Try to recover a save set.

RECOVER_FULL_PATHS variable for NetApp filer recovery not supported

LGTsc32763

When performing a DAR recovery for a NetApp filer, the NetApp specific environment
setting "RECOVER_FULL_PATHS" is not supported for NetWorker release 7.6. Setting or
unsetting the NetWorker environment variable "NSR_NDMP_RECOVER_NO_FULL_PATHS"
no longer has any effect on this NetApp specific environment setting (for example,
RECOVER_FULL_PATHS) for the DAR recovery.

VMware Consolidated Backup limitations and descriptions

Note: NetWorker 7.6 supports only Microsoft Windows 2003 (32-bit or 64-bit) OS on the
VCB proxy.

Index browsing (file selection recovery) may not be supported when using
VMware clones and the hotadd transport mode
NW114517

Due to a VMware limitation, when the VCB transport mode is set to hotadd
(VCB_TRANSPORT_MODE=hotadd), index recovery is not supported in the following
cases:
◆ If either the virtual machine to be backed up or the VCB proxy virtual machine are
VMware clones of one another.

EMC NetWorker Release 8.1 Release Notes 121

Known problems and limitations

◆ If both the virtual machine to be backed up and the VCB proxy virtual machine were
cloned from the same parent virtual machine.

Index browsing fails when traditional backups are performed prior to

VCB/VADP backups
NW105846, NW105379

If there are traditional NetWorker client-based backups already existing for a VM client
and VCB/VADP backups of the same client are performed, then a browsable recovery may
not display the entire index of the client.

The following VCB/VADP backups are affected:

◆ *FULL* (with FLR enabled)
◆ ALLVMFS

Workaround

Perform one of the following:

◆ If you know the exact path of the file/folder to be recovered, then in the command line
interface of the recovery, browse to the path of the particular folder to view the index.
For example, if C:\test-data\D1 is the folder to be recovered, then run recover>cd
C:\test-data\D1 from the command line to view the index and recover its contents.
◆ Run the nsrim command on the NetWorker server to mark the browsable save sets
corresponding to the traditional backup as recoverable save sets in order to view the
full index of the client.
For example, if the client 10.31.79.12 already has a traditional backup for the save set
C:\F1, then run the following command to delete the indexes of such traditional
savesets:
nsrim -c 10.31.79.12 -N C:\F1 -l

The last parameter in this command is a lower-case L

Note: This command removes the oldest full save and all dependant save sets from
the online index. You may need to run this command multiple times for every level full
browsable traditional save set present.
Browsable recovery of the traditional backup save sets will no longer be possible once
the indexes are removed. However, you can still perform a save set recovery for the
traditional backups whose indexes are removed.

Single step recovery of VM on Virtual Center or ESX server does not work if
non-default port number used
NW105962

Single step recovery of a VM cannot be performed on the Virtual Center or ESX server if a
non-default port number (a port number other than 443) is used.

122 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Workaround

Recover the VM image to a staging location and use VMware Converter to export the VM
image to the ESX server.

Backup with VCB Directive applied to Windows 2000 VM does not skip
system32 folder under C:\WINNT
NW105583

When VCB Directive is applied to a Windows 2000 VM, the system32 folder under
C:\WINNT is not skipped during backup. File-level full and incremental backups with VCB
Directive enabled should skip the system and system32 folders in the Windows install
folder.

Multiple “directory is not empty” messages appear during full image

recovery or drive letter recovery
NW126643, NW105298

When using the recover command or the NetWorker User program (winworkr), save set
recovery of full image or browsable recovery of an entire drive letter displays multiple
incorrect error messages similar to the following:
cannot write to <directory_location>: The directory is not empty.

The recovery completes successfully and these messages can be ignored.

Single-file recovery is not supported for VMs containing non-ASCII

filenames
NW105265

If VMs contain filenames with non-ASCII characters, single-file recovery is not possible
and the full-image backup implementation is used for backup.
Workaround

Perform the ALLVMFS backup and recover.

Performance is slow on Windows VMs when multiple disaster recovery

sessions are performed simultaneously
NW105098

When performing disaster recovery of multiple Windows VMs simultaneously from the
same proxy machine, the recovery time increases due to VMware VDDK limitation.
Workaround

Perform disaster recovery of multiple VMs in a sequential manner instead of performing

recovery simultaneously.

Stopping the savegroup does not stop the VCB mounter operation
NW105095

When you stop the savegroup, the VCB mounter does not stop until the operation is
completed.

EMC NetWorker Release 8.1 Release Notes 123

Known problems and limitations

Workaround

Stop the VCB mounter operation manually.

Savegroup log error message for Hypervisor user without VCB privileges is
not descriptive
LGTsc30913

The error message that appears if a Hypervisor user attempts a VCB operation but does
not have VCB privileges does not fully describe the problem. The message indicates:
nsrvcb_save: Error: Other error encountered: vim.fault.NoPermission
Only full level VCB backups can be performed when Backup renamed
directories attribute is enabled

Only full level VCB backups can be performed when Backup renamed
directories attribute is enabled
LGTsc30843

When enabling the Backup renamed directories attribute for a VM client instance, you can
only perform full level VCB backups. You cannot perform incremental and differential
file-level backups with the attribute enabled.

Single-step recovery has limitation to restore VMs with thin disks to ESX
4.0 Server
LGTsc32604

Due to the VMware Converter 3.0.3 limitation, single-step recovery always restores the VM
with thick disks irrespective of whether the VM was configured with thin disks before.
Workaround

To restore the VM with thin disks to ESX 4.0, first perform recovery to a staging location
and then use the VMware Converter 4.x to perform disaster recovery (DR) with thin disks.

File-level recovery of an encrypted file is not supported

LGTsc32380

Due to Microsoft Windows limitation, browsable recovery of a NTFS encrypted file to a

local file system on a proxy machine is not supported.
Workaround

Share a CIFS folder from the VM and perform a recovery of the encrypted file from the VCB
proxy to the CIFS share directly or install NetWorker Client on the VM to recover encrypted
file to a local file system of the VM.

Limitations when performing file-level recovery of image-level backup

The following considerations apply when performing a file-level recovery of an image-level
backup:
◆ Support for file-level recovery from image-based backup is available only on Windows.
◆ File-level recovery is supported only on VMs having Windows OS with NTFS5 file
system.

124 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

◆ File-level incremental backup of a VM is not supported after a hardware change in the

VM. Perform a full image-level backup after every change in the VM hardware.
◆ In case of a remote VCB proxy client, single step recover requires the members of the
VCB proxy clients administrator group to be part of the remote access list of the VM
clients or should have the "Remote access all clients" privilege.

Recovery of a VM host with multiple drives displays "Directory not empty"

in the recover GUI
LGTsc30075

During a ALLVMFS backup on a VM host with multiple drives, select all the drives to be
recovered in the recovery GUI and start the recovery to the relocated target location.
Recovery is completed successfully to the target location with "Directory not empty" error
at the end of each selected drive.
Workaround

Ignore the error. The recovery completes successfully.

Additional system generated or user files are recovered during save set
recovery from an incremental backup
LGTsc30810

During VCB incremental backup, additional files modified since the last backup, including
system generated files or user files, are backed up.
Workaround

Ensure sufficient disk space is available for recovery. Ignore the additional files that are
recovered.

Restored virtual machine starts in forceful powered off state during

FULLVM restore
LGTsc30813

During a FULLVM restore using the GUI or Command Line Interface, the restored virtual
machine will start in forceful powered off state due to VCB snapshot limitations.

nsr resource takes default values if invalid application information inputs

are provided for the proxy server
LGTsc31164

When wrong inputs are provided in the Application Information tab for the proxy server,
nsr resource will take the default values by default. For example, if
VCB_TRANSPOSRT_MODE=nbd and the user types VCB_TRANSPORTMODE=nbd, then the
nsr resource takes the default mode of transport VCB_TRANSPORT_MODE=san.
Workaround

Ensure to provide the correct values.

EMC NetWorker Release 8.1 Release Notes 125

Known problems and limitations

GPT disks are not supported for file-level recovery

LGTsc30927

During a image-based backup, GPT disks are not supported for file-level recovery. If the
VM has any GPT disk then the backup will always be image-based backup without support
for file-level recovery.
Workaround

Ensure there are no GPT disks.

Single file recoverycannot be performed on disks with no drive letter

LGTsc31190

When a disk is formatted with NTFS partition without a drive letter, file systems which are
candidates for single file recovery (SFR) backup will adopt FULLVM method of backup.
Workaround

Ensure each disk partition is associated with a drive letter.

Single file recovery is not supported for VM's configured with dynamic
disks
LGTsc31499

When VM's are configured with dynamic disks, SFR is not supported.
Workaround

Ensure there are no dynamic disks.

VCB implementation is not supported on VM’s configured on or with RDM

physical disks
LGTsc31500

VCB (FULLVM or SFR) implementation is not supported when VM's are configured on RDM
physical disks or disks which are attached to RDM physical disks.

Memory footprint for nsrvcb_save is huge during VMware backups

LGTsc31545

When VCB backups are performed, nsrvcb_save binary consumes huge amount of
memory. For example, to back up a VM with 200 GB vmdk, nsrvcb_save binary can
consume up to 1.15 GB of RAM.

Therefore, it is recommended to provision a minimum of 1 GB or more memory space for

nsrvcb_save binary.
Note: For every VM backup, one nsrvcb_save binary is spawned. Memory consumed by the
nsrvcb_save binary should be released once the VM backup is complete.

126 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

VM should be in the “powered on” state for the first backup

LGTsc31562

When running the first backup, the VM should be in the powered on state irrespective of
whether the VCB_VM_LOOKUP_METHOD is based on the IP address or the display name.

Shared folder settings are not restored from a file-level restore on the VM
LGTsc31575

When a file-level restore is performed from a full image or incremental or differential

backup of a shared folder on the VM, the share settings on the folder are not restored.
Workaround

Manually reset the share settings for the folder.

Any system changes to the VM requires a full backup

LGTsc31651

When you add or delete hard disks, install or uninstall applications, or update OS patches
to the VM, perform a full level backup manually.

Save set consolidation is not supported for VCB backups

LGTsc31735

For any VCB backup, either full image or file-level, save set consolidation is not
supported.

No directives are supported for VCB backup when save set is “*FULL*” and
the level is full
LGTsc31737

During VCB backups, when the save set is “*FULL*” and the level is full, no directives are
supported. For example, directives such as skip are not supported for VCB image-level
backup because disaster recovery is not possible after using this directive.

Backup of a powered off VM is only possible if the VM was turned off

gracefully
LGTsc31971

Before performing a VCB backup with file-level recovery for a VM in powered off state,
complete the shutdown process of the VM gracefully.

EMC NetWorker Release 8.1 Release Notes 127

Known problems and limitations

General problems and limitations

Secure Storage Library not initialized during Lockbox entry creation due to
missing library (fixed for the 7.6 SP1 release)
NW110377

When you create a lockbox entry (using either nsradmin, NMC, or the client configuration
wizard) on HP-UX ia64, lockbox entry creation fails with an error indicating that the Secure
Storage Library was not initialized. This is due to a missing library, libccme.sl.

A hot fix containing the missing library libccme.sl has been provided in
ftp://ftp.legato.com/pub/NetWorker/Updates/NW110377. Instructions for installing the
hot fix are provided in the ReadMe file at
ftp://ftp.legato.com/pub/NetWorker/Updates/NW110377/README.

NetWorker does not save or recover DFS empty directories

NW105048

DFS empty directories are not backed up or recovered by NetWorker.

Archived data not properly indexed on MAC-OS

LGTsc32722

When data is archived using nsrarchive on a MAC-OS platform, NetWorker adds a “/” to
the name of the save set in the index. For example, the path
/Volumes/sat-tree/scip2c217/backup/qgO becomes
/Volumes/sat-tree/scip2c217/backup/qgO/, and the data can only be retrieved if “/” is
added to the end of the name.

NetWorker services might not start due to lock files

LGTsc30717

If the nsrmmgd process is terminated and a message similar to the following appears,
ensure that all nsrmmgd processes have been shutdown. The NetWorker software keeps
the locks in place until all instances of nsrmmgd are gone.
nsrmmgd SYSTEM error: Lock file `C:\Program Files\EMC
NetWorker\nsr\tmp\nsrmmgd.lck' cannot be locked.

Compression asm does not work with other asms

LGTsc32410

The compression asm does not work in conjunction with any other asms.

NetWorker releases previous to 7.6

Table 14 on page 129 identifies problem issues and limitations discovered in NetWorker
release 7.5.

The known limitations are separated into the following categories:

◆ “Backup problems and limitations descriptions” on page 131
◆ “CLI problems and limitations descriptions” on page 133

128 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

◆ “Configuration problems and limitations descriptions” on page 133

◆ “Devices and media problems and limitations descriptions” on page 136
◆ “GUI problems and limitations descriptions” on page 137
◆ “Installation problems and limitations descriptions” on page 138
◆ “Licensing problems and limitations descriptions” on page 138
◆ “Localization problems and limitations descriptions” on page 139
◆ “Messaging problems and limitations descriptions” on page 139
◆ “Performance problems and limitations descriptions” on page 141
◆ “Restore problems and limitations descriptions” on page 142
◆ “Security problems and limitations descriptions” on page 144
◆ “Upgrading problems and limitations descriptions” on page 144

Table 14 Limitations discovered in NetWorker releases previous to 7.6 (page 1 of 3)

Operating system Product

Issue number Description of limitation affected feature

“LGTsc25711” Running nsravtar directly from the command line might fail AIX CLI
on page 133

“LGTsc26202” Cannot configure Windows Firewall Support in the NetWorker Installation Windows Configurati
on page 133 wizard on

“LGTpa94658” NMC cannot connect to GSTDB if EMC AlphaStor or EMC DPA is installed All Configurati
on page 134 on

“LGTsc28059” Multiplexing must be disabled for save sets to be cloned from the DXi/DL3D All Devices
on page 136 using NDMP path-to-tape and Media

“LGTsc27571” Tapes left in "Ejecting" state after labelling large number of volumes in a Windows Devices
on page 136 Windows virtual environment and Media

“LGTsc27106” Querying large numbers of save sets in the NetWorker user interface may cause Windows GUI
on page 137 a Java heap space error

“LGTsc26673” Inconsistent license information displayed for NDMP Workstation Client Windows Licensing
on page 138

“LGTsc26980” Non-ASCII hostnames are not supported by NetWorker Windows Localizatio

on page 180 n

“LGTsc31111” Savegroup Completion report output contains extra, unnecessary information All Messaging
on page 139

“LGTsc26965” Single step recovery of a full virtual machine: some error messages are cyptic Windows Messaging
on page 140

“LGTsc27063” Single step recovery of a full virtual machine: some messages are non-standard Windows Messaging
on page 140

“LGTsc24358” When upgrading to NetWorker release 7.5 on Windows, reboot NMC Windows Upgrading
on page 144

“LGTsc24329” Recovering client file index fails if NDMP attribute is selected on client resource Windows Restore
on page 142

EMC NetWorker Release 8.1 Release Notes 129

Known problems and limitations

Table 14 Limitations discovered in NetWorker releases previous to 7.6 (page 2 of 3)

Operating system Product

Issue number Description of limitation affected feature

“LGTsc27447” Single step recovery of a full virtual machine: VCB restore must use Virtual Windows Restore
on page 142 Center IP/name and password/username combination if Virtual Center used for
backup

“LGTsc26268” Single step recovery of a full virtual machine: Validation of the VC/ESX server Windows Restore
on page 142 credentials are validated once the NetWorker recover complete

“LGTsc22164” Software distribution feature cannot be used to perform upgrade on some Linux, Solaris Backup
on page 131 platforms

“LGTsc24483” Low system memory may cause unpredictable results Solaris Backup
on page 131

“LGTsc15176” VCB backup of multiple virtual machines simultaneously may fail Backup
on page 131

“LGTsc11447” Hosts file entries for IPv6 must contain localhost before aliases AIX Backup
on page 132

“LGTsc11274” Backup fails when save set list has a very large number of save sets defined Solaris Backup
on page 132

“LGTsc21070” Save jobs may terminate prematurely or later than desired if NetWorker server All Backup
on page 132 and clients time settings differ

“LGTsc14530” VCB fails with custom pre-freeze script error All Backup
on page 132

“LGTsc23714” Savegroup initiated with “@” symbol at beginning of name does not trigger All CLI
on page 133 desired group

“LGTsc10723” AFTD direct file recover option may yield incomplete volume information Solaris CLI
on page 133

“LGTsc23081” NetWorker Console unable to generate reports in pure IPv6 environment HP-UX Configurati
on page 135 on

“LGTsc18543” Relabelling volume after deleting remote client resource does not purge file Windows Configurati
on page 135 index on

“LGTsc11811” Do not perform client backups using temporary IPv6 addresses Linux, Windows Configurati
on page 135 on

“LGTsc17146” Raw disk backup and restore not supported on NetWorker with Microsoft Windows Devices
on page 137 Windows 2008 and Media

“LGTsc07321” Inquire command does not detect tape drive on HP-UX HP-UX Devices
on page 137 and Media

“LGTsc12061” Cannot dismiss events in the events table until the event is resolved All GUI
on page 138

“LGTsc08415” A user cannot be a member of more than 512 groups Windows GUI
on page 138

“LGTsc17805” NetWorker installation directory path longer than 160 characters returns error, Windows Installation
on page 138 but installation still completes successfully

“LGTsc26210” License Conformance Summary displays incorrect base license type Solaris Licensing
on page 138

“LGTsc21657” In a non-English environment, the character encoding used for the NMC client Windows Localizatio
on page 179 and NetWorker client must be the same n

130 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Table 14 Limitations discovered in NetWorker releases previous to 7.6 (page 3 of 3)

Operating system Product

Issue number Description of limitation affected feature

“LGTsc26288” JRE 1.6 users cannot export reports as PDF documents for non-English locales All Localizatio
on page 180 on AIX and HPUX n

“LGTsc22735” mminfo query ignores “!” on non-boolean fields Windows Messaging

on page 141

“LGTsc21253” nsrjobd may consume large amount of memory depending on backup Windows Performanc
on page 141 environment e

“LGTsc19690” NetWorker programs may experience delays on SuSE Linux Enterprise Server Linux Performanc
on page 141 version 9.x using IPv6 e

“LGTsc12672” nsrck -L7 -t must be performed with additional steps in order to recover client All Restore
on page 143 index with save sets passed browse period

“LGTsc20723” Ctime attribute not modified on AIX version 5.3 and later, Mac OS X version AIX, Mac OS X Restore
on page 144 10.4 and later

“LGTsc21704” Authentication error appears when NetWorker connecting to client Solaris Security
on page 144

“LGTsc24076” New client's global attributes do not inherit existing client values after Windows Upgrading
on page 144 upgrading NMC

Backup problems and limitations descriptions

Software distribution feature cannot be used to perform upgrade on some
platforms
LGTsc22164
You cannot perform a NetWorker upgrade using the software distribution feature on the
linux_ia64, linux_ppc64, and solaris_amd64 client platforms.

Low system memory may cause unpredictable results

LGTsc24483
If your NetWorker server runs low on system memory, unpredictable behavior could result.
“Environment and system requirements” on page 15 provides information on minimum
recommended system requirements.

VCB backup of multiple virtual machines simultaneously may fail

LGTsc15176
Due to a limitation in VMware Consolidated Backup (VCB) version 1.1, a VCB backup of
multiple virtual machines simultaneously may fail with an exit code of -2146234327.
Workaround

There are two possible workarounds for this limitation:

1. Upgrade to VMware Consolidated Backup version 1.5.
2. Update the following settings in the config.js file used by VCB on the VCB proxy host:
a. Set MAX_RETRIES to a number greater than 0.

EMC NetWorker Release 8.1 Release Notes 131

Known problems and limitations

b. Set PREEXISTING_MOUNTPOINT to delete to attempt to remove the pre-existing

mount point.

Hosts file entries for IPv6 must contain localhost before aliases
LGTsc11447
If an entry for the IPv6 localhost address ::1 is added to the /etc/hosts file, localhost must
be entered before aliases such that the order is <address> <localhost> <alias>. For example,
if the alias was "loopback", the entry for this address should read:

::1 localhost loopback

Backup fails when save set list has a very large number of save sets
defined
LGTsc11274
When a very large number of save sets are defined in the client's save set attribute,
backups may fail with the message:
SYSTEM error: Not enough space

This error may occur when the total length of all save sets included in the save set list
exceeds 25 KB.
Workaround

Create one or more Client resources for the client and divide the save sets between the
two clients. The NetWorker Administration Guide provides information about creating
multiple client resources for the same client.

Save jobs may terminate prematurely or later than desired if NetWorker

server and clients time settings differ
LGTsc21070
The NSR_UNRESPONSIVE_JOB_TIMEOUT environment variable can be used to
automatically terminate jobs that have no activity for a predetermined period of time
regardless of their state. If this environment variable is used and there is a discrepancy
between client and server times, save jobs may be terminated prematurely or later than
desired.
Workaround

Periodically synchronize the time between the NetWorker server and clients.

VCB fails with custom pre-freeze script error

LGTsc14530
Due to a known limitation with VMware, VCB backup of a Solaris virtual machine may fail
due with the error of “Custom pre-freeze script failed,” even though no customization is
made to the pre-freeze script.

This limitation has been fixed in ESX 3.5 Update 2.

132 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

CLI problems and limitations descriptions

Running nsravtar directly from the command line might fail
LGTsc25711
Running nsravtar directly from the command line might fail if the following environment
variable was not updated:

LD_LIBRARY_PATH
Workaround

Add /usr/lib/nsr to the LD_LIBRARY_PATH environment variable.

Savegroup initiated with “@” symbol at beginning of name does not

trigger desired group
LGTsc23714
When a savegroup with the "@" symbol at the beginning of the name is initiated from the
command line or from NMC, the default group will be triggered instead of the specified
savegroup. The "@" symbol is being interpreted as a special flag/option, not as part of
the savegroup name.
Workaround

Do not use the "@" symbol at the beginning of a savegroup name.

AFTD direct file recover option may yield incomplete volume information
LGTsc10723
Recovering data from AFTD volumes using direct files recovery method will not update the
"read" information in the volume record. As a result, the mminfo command will incorrectly
display the "read" data as 0.

For example, the command, mminfo -a -r "written, read, volume" might yield results
similar to the following:
written read volume
99 KB 0 KB trout.legato.com.002
0 KB 0 KB trout.legato.com.002.RO

Configuration problems and limitations descriptions

Cannot configure Windows Firewall Support in the NetWorker Installation
wizard
LGTsc26202
The option to configure Windows Firewall support does not appear in the NetWorker
Installation wizard under the following conditions:
◆ The Windows Firewall Service is not turned on.
◆ Windows Firewall is not supported for a particular Windows operating system.

EMC NetWorker Release 8.1 Release Notes 133

Known problems and limitations

The Windows Firewall Service is turned on and enabled by default on the following
Windows operating systems:
◆ Windows XP SP2 and later
◆ Windows 2008

Note: Windows Firewall is not supported on Windows 2000.

Workaround

The Windows Firewall service is turned off by default on Windows Server 2003.

To turn on the firewall service on:

1. Ensure that Service Pack 1 or later has been installed. This Service Pack is required for
Windows Firewall support.
2. Start the Windows Firewall/ICS service.

NMC cannot connect to GSTDB if EMC AlphaStor or EMC DPA is installed

LGTpa94658
By default, the NetWorker Console database uses port 2638 for TCP/IP communications.
Other applications may also use this port or, as is the case with EMC AlphaStor or EMC
DPA installed with default options, may have their own instance of the iAnywhere
database installed, which can create conflicts if both applications are installed on the
same host.
Workaround

To change the service port used by the NetWorker Console database:

1. Stop the GST Service.
2. Stop any other products that are using the iAnywhere database, for example EMC
AlphaStor or EMC DPA.
3. Ensure that no dbsrv9 processes are running. If the dbsrv9 processes are running,
stop them.
4. Open a terminal or command prompt window.
5. Depending on the shell you are using, use the appropriate command (for example.
setenv for csh, export for sh) to update the library path environment variable in the
terminal window to the following locations:
• Solaris: /opt/LGTOnmc/bin:/opt/LGTOnmc/sybasa/lib
• Linux/AIX/HPUX: /opt/lgtonmc/bin:/opt/lgtonmc/sybasa/lib
• Microsoft Windows (assumes default installation location):
C:\Program Files\EMC NetWorker\Management\GST\sybase\bin
The environment variable to set varies by platform, for example:
• Solaris/Linux: LD_LIBRARY_PATH
• AIX: LIBPATH

134 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

• HP-UX: SHLIB_PATH
6. Edit the gstd.conf file to add or change the following line:
db_svc_port=port_number
For example:
db_svc_port=2639
The gstd.conf is located in the following locations:
• Solaris: /opt/LGTOnmc/etc
• Linux/AIX/HP-UX: /opt/lgtonmc/etc
• Microsoft Windows: C:\Program Files\EMC Networker\Management\GST\etc
7. Run the gstconfig command to update the port value in the NetWorker Console. The
gstconfig command is located in the following locations:
• Solaris: /opt/LGTOnmc/bin
• Linux/AIX/HP-UX: /opt/lgtonmc/bin
• Windows: C:\Program Files\EMC NetWorker\Management\GST\bin
8. Close the terminal or command prompt window, then restart the GST Service.

NetWorker Console unable to generate reports in pure IPv6 environment

LGTsc23081
The NetWorker Console is unable to generate reports when deployed in a pure IPv6
environment due to a Sybase iAnywhere 9 limitation.

Relabelling volume after deleting remote client resource does not result in
purge of file index
LGTsc18543
The file index of a remote client cannot be purged by relabelling a volume after deleting
the remote client resource.

The file index of the remote client can be purged (for the volume ID) by relabelling a
volume when the remote client is a current valid client resource.
Workaround

Run nsrck -L 3 from the command line to purge the file index.

Do not perform client backups using temporary IPv6 addresses

LGTsc11811
Temporary or randomly generated IPv6 TCP/IP addresses are not supported in NetWorker.
If the address for a client is not stored in DNS or in a hosts file and has not been added to
the client resource, NetWorker will be unable to back up the client.

EMC NetWorker Release 8.1 Release Notes 135

Known problems and limitations

Devices and media problems and limitations descriptions

Multiplexing must be disabled for save sets to be cloned from the
DXi/DL3D using NDMP path-to-tape
LGTsc28059
Multiplexing of save sets is not supported for save streams that are to be cloned from the
Dxi/DL3D with the NDMP path-to-tape feature.
Note: Only volumes created without multiplexing can be used with the NDMP path-to-tape
feature.

Workaround

Device properties must be set to disable multiplexing capability.

To disable multiplexing of save sets:

1. From NMC on the NetWorker server, select Devices.
2. For each configured target device on the DL3D or Dxi:
a. Select Device > Properties.
b. Click the Configuration tab.
c. Set Target sessions to 1.
d. Set Max sessions to 1.
e. Click OK.

Note: Backup performance may be affected by disabling multiplexing of save sets.

Tapes left in "Ejecting" state after labelling large number of volumes in a

Windows virtual environment
LGTsc27571
After labelling a large number of virtual volumes over a large number of virtual drives in a
Windows 2008 environment, some tapes may be left in the "Ejecting" state without being
removed from the drive.
Workaround

Select Diagnostic Mode within NMC (View > Diagnostic Mode), then perform the following:
1. In the Devices pane, highlight the affected library.
2. Right-click the library and select Properties.
3. Click the Advanced tab.
4. In the Jukebox features selection box, select the autoeject feature.
5. Click the Timers tab.
6. Set the load sleep, unload sleep, and eject sleep values to 0.

136 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

7. Click Save to save the changes.

Note: You may be required to disable and re-enable the jukebox for changes to take effect.

Raw disk backup and restore not supported on NetWorker with Microsoft
Windows 2008
LGTsc17146
NetWorker does not support backup and restore of raw disks with Windows Server 2008.

Inquire command does not detect tape drive on HP-UX

LGTsc07321
When you attach a tape drive to the HP-UX 11i V2 64-bit host and run the inquire
command, the tape drive is not detected, even though the device was configured,
labelled and mounted and a save was successful.
Workaround

Identify the drive path in the /dev/rmt folder, and using this path configure the device, as
usual.

Whenever a new device is attached to the system, ensure that the cached file
/tmp/lgto_scsi_devlist is updated. Remove this temp file and then run the inquire
command, which will rebuild the file.

GUI problems and limitations descriptions

Querying large numbers of save sets in the NetWorker user interface may
cause a Java heap space error
LGTsc27106
Querying large numbers of save sets in the NetWorker user interface may fail with a Java
heap space error.
Workaround

Increase the Java heap size used by the NMC application:

1. On the Console server host, open the gconsole.jnlp file in a text editor. The
gconsole.jnlp file is located in:
<Console_install_dir>\web
2. Increase the default max-heap-size value from 700MB to 1400MB. For example,
<resources>
<j2se version="1.6+" initial-heap-size="64M"
max-heap-size="1400M"/>

Note: To provide meaningful query results and to reduce the chance of encountering this
error, narrow the save set search criteria by specifying selection parameters.

EMC NetWorker Release 8.1 Release Notes 137

Known problems and limitations

Cannot dismiss events in the events table until the event is resolved
LGTsc12061
Events cannot be dismissed or hidden in the events table until the event is resolved or
has finished.

A user cannot be a member of more than 512 groups

LGTsc08415
A user cannot be a member of more than 512 groups when running any operation with
NetWorker. If the user is a member of more than 512 groups, the following message
appears:
Maximum number of groups exceeded, some groups may be dropped from the
credential. Number of groups the user belongs to: XXX, maximum number of groups
supported: 512

Installation problems and limitations descriptions

NetWorker installation directory path longer than 160 characters returns
error, but installation still completes successfully
LGTsc17805
In Windows 2003, when a NetWorker installation directory path is longer than 160
characters, an error message appears during installation. However, if you click OK in the
error message dialog box, the installation completes successfully. The NetWorker
software is then installed in a new directory where the name is the first 160 characters of
the folder selected during installation.

The longest pathname currently supported by NetWorker is 160 characters.

Licensing problems and limitations descriptions

Inconsistent license information displayed for NDMP Workstation Client
LGTsc26673
If an IP address is specified in the NDMP Array name attribute of the NetWorker client
resource, the license information displayed for the WORKSTATION CLIENT TYPE may be
incorrect. NetWorker client license information can be displayed with the nsrlic –v
command. In this case, the license information displayed for the NDMP CLIENT LICENSES
is correct. However, the information displayed for the WORKSTATION CLIENT TYPE may be
incorrect.

License Conformance Summary displays incorrect base license type

LGTsc26210
The License Conformance Summary, accessed through the Configuration tab of the
NetWorker Administration window, incorrectly displays the base license type as "Eval"
when it should be "Data Backup Utility".

138 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Localization problems and limitations descriptions

More information on localization important notes and tips is provided in
“Internationalization support” on page 175.

Non-ASCII hostnames are not supported by NetWorker

LGTsc26980
NetWorker does not support hosts that have non-ASCII characters in the hostname.

In a non-English environment, the character encoding used for the NMC

client and NetWorker client must be the same
LGTsc21657
In a non-English environment, characters do not display correctly if the character
encoding is different on the NMC client than the files on the NetWorker host that are being
browsed.
Workaround

In a non-English environment, ensure that the NMC client uses the same character
encoding as the files on the NetWorker client host that are to be browsed. For example, if
the NMC client uses the zh_CN.EUC locale it will not properly display files that were
created using the zh_CN.UTF-8 locale.

The character encoding used for the NMC client and NetWorker client must be the same to
ensure the proper display of characters. In this example, the NMC client should be started
using the zh_CN.UTF-8 locale.

JRE 1.6 users cannot export reports as PDF documents for non-English
locales on AIX and HPUX
LGTsc26288
The NetWorker option to export reports in Acrobat PDF format is not supported on AIX and
HPUX hosts that are running JRE 1.6 in a non-English locale.

Messaging problems and limitations descriptions

Savegroup completion report output contains additional, unnecessary
information
LGTsc31111
Upon completion of a NetWorker 7.2.x client savegroup backup on NetWorker servers 7.3
and later, the following additional information is added to the output of the savegroup
completion report. This additional information is unnecessary and can be ignored:
type: NSR client description;
pools supported: Yes;
browse time supported: Yes;
multiple balanced streams supported: Yes;
remote user: <NetWare_User>;
arch: NetWare IA32;
client OS type: NetWareOS;
CPU type: Intel Pentium;

EMC NetWorker Release 8.1 Release Notes 139

Known problems and limitations

CPUs: 2;
kernel arch: Intel Pentium;
machine type: server;
MB used: 57221;
NetWorker version: nwnc_7_2;
OS: NetWare IA32 6;
version: nwnc_7_2;
save set: path="<Save_Set_Name>:", level=full, diskno=0,
max_sessions=1,
stype=save ;
parallelism: 8
V <NetWare_client_Name>: <NetWare_save_set_name>: level=6, 8360 MB
00:09:34 21096 files
<NetWorker_Server_Name>: index:<NetWare_client_Name> level=6, 47 MB
00:00:03 56 files

Single step recovery of a full virtual machine: some error messages are
cyptic
LGTsc26965
During the single step recovery of a full VMware virtual machine, some error messages
may appear to be cryptic. These error messages are generated by the Vmware converter
product. The error messages may refer to the underlying function name rather than the
cause or required solution. The following is an example of one such error message:
70047:winworkr: Retore failed with error P2VError
IMPORT_DEST_INSUFFICIENT_SPACE()

The previous error message was generated because there was insufficient space in the
staging location that was specified for the single step recovery.

Single step recovery of a full virtual machine: some messages are

non-standard
LGTsc27063
During the single step recovery of a full VMware virtual machine, some messages may
appear to be non-standard. These messages are generated by the VMware converter
product. The following is an example of such messages:
[2009-02-17 14:27:37.380 'Libs' 4380 info] HOSTINFO: Seeing Intel CPU,
numCoresPerCPU 4 numThreadsPerCore 1.
[2009-02-17 14:27:37.380 'Libs' 4380 info] HOSTINFO: This machine has
2 physical CPUS, 8 total cores, and 8 logical CPUs.
[2009-02-17 14:27:37.802 'Libs' 4380 info] System libeay32.dll library
is older than our library (90709F < 9070AF)
[2009-02-17 14:27:49.850 'Libs' 4380 warning]
SSLVerifyCertAgainstSystemStore: Subject mismatch: VMware vs
10.31.236.60
[2009-02-17 14:27:49.850 'Libs' 4380 warning]
SSLVerifyCertAgainstSystemStore: The remote host certificate has these
problems:
* The host name used for the connection does not match the subject name
on the host certificate
* A certificate in the host's chain is based on an untrusted root.
[2009-02-17 14:27:49.850 'Libs' 4380 warning] SSLVerifyIsEnabled:
failed to open the product registry key. Assuming verification is
disabled. LastError = 0
[2009-02-17 14:27:49.850 'Libs' 4380 warning]
SSLVerifyCertAgainstSystemStore: Certificate verification is disabled,
so connection will proceed despite the error p2vTool version 3.0.3
build-89816

140 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

mminfo query ignores “!” on non-boolean fields

LGTsc22735
Performing an mminfo query with “!” (for example, if you run mminfo -q "!client=hydra")
should exclude all save sets for the client hydra. However, mminfo ignores the “!” and
returns all save sets for the specified client.

Performance problems and limitations descriptions

nsrjobd may consume large amount of memory depending on backup
environment
LGTsc21253
The nsrjobd daemon runs on the NetWorker server and is responsible for monitoring
NetWorker activity during a backup or recovery operation. Be aware that, depending on
the size of your backup environment, nsrjobd can require large amounts of RAM.

NetWorker programs may experience delays on SuSE Linux Enterprise

Server version 9.x using IPv6
LGTsc19690
NetWorker programs may experience a start up delay on a SuSE Linux Enterprise Server
(SLES) version 9.x using IPv6 in either a single stack or dual stack (IPv4/IPv6)
environment. This situation may occur if the Scope:Link IPv6 address of the Linux host
network interface is not included in the /etc/hosts file.

To determine the Scope:Link address of the host network interface, run the ipconfig
command from the host’s command line. In the following example, the host named geo1
has a network interface named eth0, which has a Scope:Link address of
fe80::2c0:4ff:fe68:c24/64:
geo1> ifconfig
eth0 Link encap:Ethernet HWaddr 00:C0:4F:68:C2:4F
inet addr:192.168.0.1 Bcast:10.5.163.255 Mask:255.255.254.0
inet6 addr: 3ffe:80c0:22c:74:2c0:4ff:fe68:c24/64 Scope:Global
inet6 addr: fe80::2c0:4ff:fe68:c24/64 Scope:Link
inet6 addr: 2001:abcd:0:1001:2c0:4ff:fe68:c24/64 Scope:Global
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:137451625 errors:0 dropped:0 overruns:0 frame:0
TX packets:13840698 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:1834082191 (1749.1 Mb) TX bytes:69705018 (66.4 Mb)
Interrupt:19 Base address:0xdc00
Workaround

Add the Scope:Link address of the host network interfaces to the /etc/hosts file on the
affected NetWorker client. The following example shows how an entry in the /etc/hosts
file might appear:
# Scope:Link address
fe80::2c0:4ff:fe68:c24 geo1.fss.dcp.com geo1

EMC NetWorker Release 8.1 Release Notes 141

Known problems and limitations

Restore problems and limitations descriptions

Recovering client file index fails if NDMP attribute is selected on client
resource
LGTsc24329
Recovering the client file index for a NetWorker client will fail if the NDMP attribute is set
on the client resource.
Workaround

Before using the nsrck -L7 command to recover a client file index, ensure that the NDMP
attribute is not selected in the corresponding NetWorker client resource. After the client
file index is recovered, you can once again select the NDMP attribute on the client
resource.

Single step recovery of a full virtual machine: VCB restore must use Virtual
Center IP/name and password/username combination if Virtual Center
used for backup
LGTsc27447
When a VMware Consolidated Backup (VCB) is performed using Virtual Center (for
example, if the config.js file contains Virtual Center as the Host and its credentials), then
the restore must be performed using the Virtual center IP (or its fully qualified domain
name) with its credentials. Similarly, this process should be followed when performing a
VCB restore using the ESX server, otherwise the restore will fail with an error message
indicating "unknown path."

Single step recovery of a full virtual machine: Validation of the VC/ESX

server credentials are validated once the NetWorker recover completes
LGTsc26268
The VC/ESX server credentials are validated once the NetWorker recover operation
completes:

If wrong user credentials are entered, the following might occur:

◆ The restore operation fails after the completion of the NetWorker recover operation.
◆ The user must wait until the NetWorker software completes the recover operation
before being notified that a wrong credential was entered.
◆ For a large save set, the NetWorker recover operation will take a longer time to
complete depending upon the size of the save set. Notification of wrong ESX/VC
credentials happens after the NetWorker recover operation completes.

Workaround

Ensure that you enter the proper user credentials to avoid long response time for the
following operations:
◆ VC/ESX server credentials validation. This task is performed in the later stages of a
VCB restore operation.

142 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

◆ Before performing a VMware Consolidated Backup (VCB) restore operation. The

validation of wrong credentials might cause longer response times.

Single step recovery of a full virtual machine: cannot specify CIFS path for
staging location
LGTsc27559
When performing a single step restore of a full VMware virtual machine, the staging
location you specify cannot be a Common Internet File System (CIFS) path. The recovery
will fail with an error message similar to one of the following:

Error 1:
Recovering files of client 'langre-rh-1.takman.com' from
server '11.32.64.115'.
Total disk space available for recover is 0 KB.
Total disk space needed for recover is 2304 MB.
67854:winworkr: Failed to recover save sets: Recover failed due to low
disk space!..

Error 2:
Recovering files of client 'langre-rh-1.takman.com' from
server '11.32.64.115'.
70483:winworkr: Staging location Z:\rack2 is in use by other process.
Select new staging location or delete the staging location Z:\rack2
manually.
Restore of virtual machine failed!...

nsrck -L7 -t must be performed with additional steps in order to recover

client index with save sets passed browse period
LGTsc12672
The nsrck -L7 -t command can no longer be used solely to perform a recovery of a client
index where save sets have passed the period identified in the browse policy.

The following operations must be performed in addition to running the nsrck command on
the save sets in order to recover the index information:
1. Identify a save set containing the index for the desired time/client.
2. Adjust the retention of the identified client file index to a date in the future.
3. Change the mode to notrecyclable for the identified client file index.
4. Identify the save sets required to browse (for example, the full backup and all the
following incremental backups prior to the desired date).
5. Set a new retention period for each of the individual save sets to a date in the future.
6. Change the mode to notrecyclable for each of the individual save sets.
7. Run the nsrck -L7 -t command.
8. Browse the save sets and perform the recover as usual.

EMC NetWorker Release 8.1 Release Notes 143

Known problems and limitations

Ctime attribute does not get modified on AIX version 5.3 and later, Mac OS
X version 10.4 and later
LGTsc20723
When a file is renamed on AIX version 5.3 and later or Mac OS X 10.4 and later, the ctime
attribute is not modified. The ctime attribute is the timestamp NetWorker uses to
determine whether a file has been modified when performing a non-full backup. If
renaming files on AIX version 5.3 and later or Mac OS X 10.4 and later, perform a full
backup to ensure recovery of the renamed files.

Security problems and limitations descriptions

Authentication error appears when NetWorker connecting to client
LGTsc21704
When trying to connect to a client to browse the client's file system, NetWorker displays
the error message "Authentication failed or no valid authentication methods supported by
both client and server".
Workaround

If this error message appears, search the daemon log file for more information and
instructions.

Upgrading problems and limitations descriptions

When upgrading to NetWorker release 7.5 on Windows, you might need to
reboot the system
LGTsc24358
When upgrading to NetWorker release 7.5 on Windows from a previous release of
NetWorker software, you might need to reboot the system.

New client's global attributes do not inherit existing client values after
upgrading NMC
LGTsc24076
After upgrading to the latest version of NMC and launching the Console for the first time,
the Configuration window appears, prompting you to set the configuration for the Console
and to add NetWorker servers. In the Set Database Backup Server page, the checkbox
Create client resource on this server is selected by default. If you click Finish with this
checkbox selected, the NetWorker software disables the client’s global attributes archive
services, hard links, and short filenames if enabled prior to the upgrade when the new
client resource is created.
Workaround

Clear the checkbox Create client resource on this server.

144 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

NetWorker releases previous to 7.5

Table 15 on page 145 identifies problem issues and limitations discovered previously to
NetWorker release 7.5

The known limitations are separated into the following categories:

◆ “Archiving problems and limitations descriptions” on page 149
◆ “Backup problems and limitations descriptions” on page 150
◆ “CLI problems and limitations descriptions” on page 154
◆ “Cloning and Staging problems and limitations descriptions” on page 157
◆ “Compatibility problems and limitations descriptions” on page 157
◆ “Configuration problems and limitations descriptions” on page 158
◆ “Devices and Media problems and limitations descriptions” on page 159
◆ “GUI problems and limitations descriptions” on page 162
◆ “Installation problems and limitations descriptions” on page 163
◆ “Licensing problems and limitations descriptions” on page 165
◆ “Localization problems and limitations descriptions” on page 165
◆ “Messaging problems and limitations descriptions” on page 166
◆ “NetWorker support for Novell’s OES Linux limitations” on page 167
◆ “NetWorker Management Console problems and limitations descriptions” on
page 170
◆ “Restore problems and limitations descriptions” on page 170
◆ “Upgrading problems and limitations descriptions” on page 174

Table 15 Limitations discovered in NetWorker releases previous to 7.5 (page 1 of 5)

Operating system Product

Issue number Description of limitation affected feature

“LGTsc05166” DiskXtender Data Manager file systems get archived on NetWorker. Windows archiving
on page 149

“LGTsc20517” For each NetWorker client that is BMR-enabled, set the save set attribute in the Linux, Windows Backup
on page 150 client resource to ALL. Solaris

“LGTsc19514” Delay in cancelling savegroups in NMC for BMR-enabled clients. Linux, Windows, Backup
on page 150 Solaris

“LGTsc16575” A backup might fail for CIFS shares on Microsoft Windows hosts. Windows Backup
on page 150

“LGTsc17703” How to back up a NetWorker Windows client that has multiple domains that are Windows Backup
on page 151 part of both an AD domain, and a DNS domain.

“LGTsc10292” Incorrectly configured NPS role results in failed backup of VSS SYSTEM Windows Backup
on page 151 SERVICES of Windows Server 2008.

“LGTsc11914” Active Directory Lightweight Directory Services backup fails for Windows 2008 Windows Backup
on page 151 client.

EMC NetWorker Release 8.1 Release Notes 145

Known problems and limitations

Table 15 Limitations discovered in NetWorker releases previous to 7.5 (page 2 of 5)

Operating system Product

Issue number Description of limitation affected feature

“LGTsc14956” Not all drives in a VCB file level backup are browsable. Windows Backup
on page 152

“LGTsc15236” Incorrect error during recovery of VCB backups. Windows Backup

on page 152

“LGTsc12970” Backups of /dev directory fail in HP-UX 11i v3 on PA_RISC and IA64. AIX, Solaris, Backup
on page 153 HP-UX, Tru64
UNIX

“LGTsc06312” The nsr_render_log program does not accept input from STDIN. AIX, Solaris, Backup
on page 153 HP-UX, Tru64
UNIX

“LGTsc09257” A user cannot be a member of more than 512 groups. All Backup
on page 153

“LGTpa91475” The savepnpc command fails if the precommand and post command contains All Backup
on page 153 any non-ASCII characters.

“LGTsc00850” Restrictions associated with the ZFS file system. AIX, Solaris, Backup
on page 154 HP-UX, Tru64
UNIX

“LGTsc19916” The contents of the /opt, /var, /etc directories are not excluded when the -B Linux, Windows, CLI
on page 154 option is used. Solaris

“LGTsc09483” Not all save sets listed after running mminfo command with dedupe option. Solaris CLI
on page 154

“LGTsc19402” The -e exclude file option behaves differently for files and directories. Linux, Windows, CLI
on page 155 Solaris

“LGTsc19084” The recover -e exclude file option does not support NDMP file systems. All CLI
on page 155

“LGTsc19405” No filename checks for remote clients when using the --e exclude file option Linux, Windows, CLI
on page 156 with a directed recovery operation. Solaris

“LGTsc10665” Cannot view log files created with the nsr_render_log command. All CLI
on page 156

“LGTsc06809” Clients running pre-7.4 releases cannot use the mminfo -N command to query All CLI
on page 156 save set names longer than 255 bytes.

“LGTpa95406” The jbverify command is not supported on Solaris 10. AIX, Solaris, CLI
on page 156 HP-UX, Tru64
UNIX

“LGTsc15337” Marking volumes as Recyclable might slow system performance. All Cloning and
on page 157 Staging

“LGTsc19401” Removable Storage Manager (RSM) not supported. All Compatibili

on page 157 ty

“LGTsc20557” The Homebase Agent is not supported in a Cluster environment. Linux, Windows, Compatibili
on page 157 Solaris ty

“LGTsc14700” Support for Windows Change Journal Manager with Microsoft Windows 2008. Windows Compatibili
on page 158 ty

“LGTsc18397” Cannot set nsrdir parameter for virtual server using remote Microsoft Windows Configurati
on page 158 Management Console (MMC). on

146 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Table 15 Limitations discovered in NetWorker releases previous to 7.5 (page 3 of 5)

Operating system Product

Issue number Description of limitation affected feature

“LGTsc06585” Running the Linux 64 bit package with the 2.6.x kernel may result in crashes. Linux Configurati
on page 158 on

“LGTsc19241” For Microsoft Windows Server 2008 raw devices are not saved under the virtual Windows Devices
on page 159 client's index. and Media

“LGTsc19676” Broker connects to first available storage node if no device or mmd available on Solaris Devices
on page 159 first host. and Media

“LGTsc11158” Cannot create an advanced file-type device on a remote storage node device. All Devices
on page 160 and Media

“LGTsc10950” VMware Consolidated Backup 1.0.x is not supported on Windows Server 2003, Windows Devices
on page 160 Enterprise Edition (x64) Service Pack 2. and Media

“LGTsc06358” NetWorker becomes unresponsive when attempting an invalid clone operation. All Devices
on page 160 and Media

“LGTsc01923” Cannot query UNIX non-ASCII save sets in NMC. AIX, Solaris, Devices
on page 160 HP-UX, Tru64 and Media
UNIX

“LGTpa93001” Virtual tapes are listed as "full" after mounting. Windows Devices
on page 160 and Media

“LGTpa88188” Scan for Devices configures all devices as a regular storage node device. All Devices
on page 161 and Media

“LGTsc01446” The Virtual Jukebox attribute for a CDL virtual tape library is not automatically All Devices
on page 161 set after upgrading to release 7.4. and Media

“LGTsc10295” dedup recover sessions do not display in the NMC GUI. All GUI
on page 162

“LGTsc08756” Text searched for in the NetWorker Console Help program does not get Solaris GUI
on page 162 highlighted correctly if using JRE 1.5.x for Asian languages.

“LGTsc00365” On HP-UX the NetWorker Server window does not appear. All GUI
on page 162

“LGTsc01115” Cannot copy and paste Japanese characters using NMC. All GUI
on page 162

“LGTpa95162” Newly created groups are not displayed by the NetWorker Console GUI after Linux, AIX, GUI
on page 162 moving system date back in time. Solaris, HP-UX,
Tru64 UNIX

“LGTpa96168” Cannot launch NetWorker Console Help program on Solaris 10 with a JRE earlier AIX, Solaris, GUI
on page 163 than version 1.5.0_09. HP-UX, Tru64
UNIX

“LGTsc16831” For Microsoft Windows, ensure that the HomeBase Agent software installed in Windows Installation
on page 163 the same installation path as the NetWorker software.

“LGTsc18641” Before restoring the base profile, manually edit the fields in the Linux, Windows, Installation
on page 164 recovery-configuration.xml file. Solaris

“LGTsc18925” Unable to launch the NetWorker Dashboard software with NetWorker release Windows Installation
on page 164 7.4 Service Pack 3.

“LGTsc19388” NetWorker and the HomeBase Agent (BMR) licenses work independently. Linux, Windows, Licensing
on page 165 Solaris

EMC NetWorker Release 8.1 Release Notes 147

Known problems and limitations

Table 15 Limitations discovered in NetWorker releases previous to 7.5 (page 4 of 5)

Operating system Product

Issue number Description of limitation affected feature

“LGTsc05879” NMC version 3.4 and NWD version 1.0 cannot be installed on the same host. All Localizatio
on page 163 n

“LGTsc06280” Inventory operation in Software Administration wizard delays for unreachable AIX, Solaris, Localizatio
on page 161 clients. HP-UX, Tru64 n
UNIX

“LGTsc05166” DiskXtender Data Manager file systems get archived on NetWorker. Windows Localizatio
on page 149 n

“LGTsc02862” The NMC Japanese Online Help displays incorrect characters on Linux systems. Linux Localizatio
on page 180 n

“LGTsc05339” Recovering a large number of files may take a long time in the French locales on Solaris Localizatio
on page 180 Solaris. n

“LGTsc03894” Garbled characters may appear in the NetWorker Console GUI font list on Solaris Localizatio
on page 180 Solaris. n

“LGTsc02814” Problem with highlighted text in the NetWorker Console Help program after Solaris Localizatio
on page 180 performing a search using JRE 1.5.x for Asian languages. n

“LGTpa88887” Entering non-ASCII characters in NetWorker user interfaces. All Localizatio

on page 181 n

“LGTpa92833” Non-ASCII save set names are displayed incorrectly in nsradmin visual mode on Linux Localizatio
on page 181 Linux. n

“LGTsc16423” nsrbmr does not give proper error messages under certain conditions. Linux, Windows Messaging
on page 166 Solaris

“LGTsc17737” Error message does not display for annotations with similar characters at end. Windows Messaging
on page 166

“LGTsc08054” Erroneous error message may be generated by Linux operating systems when Linux Messaging
on page 166 performing an I/O to a volume.

“LGTsc08978” New pdksh package required on SuSE 10 x86. Linux Messaging

on page 167

“LGTsc03478” Incorrect error message reported when starting the NetWorker console on Red AIX, Solaris, Messaging
on page 167 Hat 64-bit Enterprise Server 4. HP-UX, Tru64
UNIX

“LGTsc06288” Error states you need to perform an uninstall when updating from NetWorker All Messaging
on page 167 7.2.2 32-bit to NetWorker 7.5 64-bit.

“LGTsc02903” NetWorker Management Console may become unresponsive when opened AIX, Solaris, NMC
on page 170 using Exceed. HP-UX, Tru64
UNIX, Linux

“LGTsc19364” Last-modified date changed for files recovered to Sun Cluster global file Solaris Restore
on page 170 system.

“LGTsc10796” Do not recover identical save sets with different ssids to the same target AIX, Solaris, Restore
on page 170 directory. Windows, Tru64
UNIX, Linux,
HP-UX

“LGTsc14577” NetWorker interactive recover does not recover the registry and the com+ regdb Windows Restore
on page 171 writers.

148 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Table 15 Limitations discovered in NetWorker releases previous to 7.5 (page 5 of 5)

Operating system Product

Issue number Description of limitation affected feature

“LGTsc11618” Restoring full VM image on ESX server fails. Linux Restore

on page 171

“LGTsc11734” Unable to restore files encrypted with both Microsoft Windows Encrypting File Windows Restore
on page 171 System & AES encryption.

“LGTsc05990” FSRM Disk Quota may not be restored to previous settings after recovery. All Restore
on page 172

“LGTsc20558” After a BMR recovery operation the root password might change. Linux, Windows, Restore
on page 172 Solaris

“LGTpa95900” On NetApps files, filenames ending with special characters might be renamed. Windows Restore
on page 172

“LGTpa94966” The winworkr program fails to retrieve successfully archived files if a slash (\) Windows Restore
on page 173 character was entered in the annotation string.

“LGTpa83927” Directed recover using the CLI fails with permission errors on Windows. Windows Restore
on page 173

“LGTsc00167” The winworkr program will not relocate to a partition not existing on the Windows Restore
on page 173 initiating host if the NetWorker server is running release 7.4 and the client is
running release 7.2.

“LGTsc05053” Performing a save set query can take a long time if the query parameter in the All Restore
on page 174 Query Save Set tab is set to "Save Time" and From and To calendars for 3 or
more days.

“LGTpa96554” For recoveries operations using Celerra filers, misleading error messages might All Restore
on page 174 display.

“LGTsc08958” JRE version mismatch causes authentication failure, stops GSTD. Linux Upgrading
on page 174

“LGTpa95019” When upgrading from release 7.3.1 the Virtual jukeboxes attribute will not be Windows Upgrading
on page 174 set correctly.

“LGTsc01587” Group details window is empty after upgrading from NetWorker release 7.2.2. All Upgrading
on page 175

Archiving problems and limitations descriptions

DiskXtender Data Manager file systems get archived on NetWorker
LGTsc05166

On the Data Manager file system for Windows, when you perform a client or server
initiated archive of the file system containing the file stubs (or if you archive the file stubs
themselves), file systems managed by the DiskXtender Data Manager are archived by
NetWorker, and no error message is displayed. With DiskXtender running on UNIX,
NetWorker properly fails the archiving and reports the error.

There are several problems that may result from the archive:
◆ If an archive is scheduled with grooming, this may cause the file stubs to be deleted.
◆ If an archive retrieve is performed, the existing file stubs may be overwritten.

EMC NetWorker Release 8.1 Release Notes 149

Known problems and limitations

◆ If the DX schedule is running after grooming, the file on the media server may be
deleted.

Backup problems and limitations descriptions

For each NetWorker client that is BMR-enabled, set the save set attribute
in the client resource to ALL
LGTsc20517

If the extended profile is not backed up as part of the NetWorker data set, a recovery or
migration of the source to the target system cannot be performed.

For each NetWorker client that is BMR-enabled, set the save set attribute in the client
resource to ALL. This ensures that the extended profile is part of NetWorker data backup.

The location of extended profile differs for Microsoft Windows and UNIX:
◆ Microsoft Windows: NetWorker_install_path\HomeBaseAgent\history
◆ UNIX: /opt/homebase-agent/history

Note: Where NetWorker_install_path is the location of the NetWorker installation files.

Delay in cancelling savegroups in NMC for BMR-enabled clients

LGTsc19514

Stopping a savegroup from NMC when there are BMR clients enabled for profiling does
not result in the savegroup being cancelled immediately. The clients with BMR-enabled
must wait until the profiling jobs complete before the savegroup is cancelled. Once the
profiling completes, the savegroup is cancelled and subsequent savegroups for the
clients will not be triggered.

A backup might fail for CIFS shares on Microsoft Windows hosts

LGTsc16575

A backup might fail for CIFS shares on a Microsoft Windows host. When nsrexecd is
started as a service (context account), it cannot obtain the CIFS path because of access
restrictions to its security credentials from the NetWorker client (nsrexecd). By default,
nsrexecd starts as a service under the SYSTEM account that does not have the correct
permissions to access the CIFS share.

A message similar to the following appears:

save: \\10.31.73.80\bluenose\sam\t1.txt: "No such file or directory"
Workaround

To work around this issue:

1. Create an account on the NetWorker client machine with the same username and
password as an account with permissions to the CIFS shares.
2. Perform one of the following backup operations:
• Back up the CIFS shares save set:

150 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

a. Edit the Client resource for the NetWorker client that is backing up the CIFS
share:
- For the Remote User attribute, type the username from step one.
- For the Password attribute, type the password from step one.
b. Use the savegrp command to backup the CIFS shares save set from the
command line. For example:
# savegrp client_name -G group_name
• Back up the CIFS shares:
a. Ensure that the NetWorker client is connected to the CIFS share. For example:
Run a command shell that:
- Has a mapped share with net use \\IP_address\share.
-Is connected with the username and password that has access to
the CIFS share.
b. Use the Save - L command to back up the CIFS shares. For example:
# Save -L

How to back up a NetWorker Windows client that has multiple domains

that are part of both an AD domain, and a DNS domain
LGTsc17703

To back up a NetWorker Windows client that has multiple domains that are part of both an
AD domain and a DNS domain, perform one of the following steps:
◆ Define the AD domain name, which is the Full Computer Name, in the NetWorker
server’s /etc/hosts file.
◆ Define the AD domain, which is the Full Computer Name, in the Active Directory DNS.
Also, on the NetWorker server, open the Client resource and add the Full Computer
Name in the alias list.

Incorrectly configured NPS role results in failed backup of VSS SYSTEM

SERVICES of Windows Server 2008
LGTsc10292

Due to a problem identified by Microsoft with Windows Server 2008, an incorrectly

configured Network Policy Server (NPS) results in the writer incorrectly reporting files to
the NetWorker software. As a result, the backup of the VSS SYSTEM SERVICES fails.

The Microsoft defect number for this issue is 872233.

Active Directory Lightweight Directory Services backup fails for Windows

2008 client
LGTsc11914

When running a scheduled backup of a save group which includes a Windows 2008 client
configured with Active Directory Lightweight Directory Services (AD LDS), the backup may
fail with the following error:

EMC NetWorker Release 8.1 Release Notes 151

Known problems and limitations

VSS USER DATA:\ VSS USER DATA: ERROR: VSS failed to process snapshot,
error=0x800423f4. The VSS writer operation failed because of an error
that might recur if another shadow copy is created.

There will also be ADAM Writer errors in the Application Event Log that correspond to the
time of the save of the ADAM component.
Workaround

To avoid this problem, increase the Client Retries attribute on the Advanced tab of the
Group resource. Increasing the value to 2 or 3 will resolve the problem.

Not all drives in a VCB file level backup are browsable

LGTsc14956

When performing a virtual machine restore from a file level VCB backup, only the drive
letters that exist on the proxy host will be seen in the Recover window. For example, if the
proxy host has only a C:\ drive and the virtual machine has a C:\, D:\ and E:\ drive, then
during the restore of the virtual machine, only the C:\ drive of the Virtual machine will be
displayed.
Workaround

To resolve this problem, perform one of the following:

◆ From the NetWorker User program:
1. Select Change Selection from the File menu.
2. Type the drive that you want to browse.

Note: This workaround will need to be performed each time you change the browse
time.

◆ From the recover command:

1. Change to the root of the current directory:
recover> cd /
2. Change to the drive that you want to browse, for example:
recover> cd e:
3. Verify that you can now browse the drive:
recover> dir

Incorrect error during recovery of VCB backups

LGTsc15236

During a file-based recovery of a VCB backup, when a drive letter (rather than a filename
or directory) is identified for recovery, one of several error messages similar to the
following is displayed:
Failed to write to directory_location The directory is not empty.

These errors occur when using either the recover command or the NetWorker User
Program (winworkr).

152 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Workaround

These error messages are incorrect and should be ignored.

Backups of /dev directory fail in HP-UX 11i v3 on PA_RISC and IA64

LGTsc12970

If you back up the /dev directory using a local storage node in an HP-UX 11i v3
environment on PA_RISC and IA64, the backup fails and the system freezes.
Workaround

To work around this issue:

◆ Do not back up the /dev directory. Do not specify Save Set ALL in the client's Save Set
attribute, and do not specify save sets that include the /dev directory in the backup.
◆ Use a directive to exclude the /dev directory from the backup.
◆ Use a remote storage node.

The nsr_render_log does not accept input from STDIN

LGTsc06312

The nsr_render_log does not accept input from STDIN. The following error message is
displayed:
Please provide the input log filename with the path.

After the error message the usage information is listed.

Workaround

Type the following command:

tail -f /nsr/logs/daemon.raw | nsr_render_log -

A user cannot be a member of more than 512 groups

LGTsc09257

A user cannot be a member of more than 512 groups when running any operation with
NetWorker software. If the user is a member of more than 512 groups, the following
message appears:
Maximum number of groups exceeded, some groups may be dropped from the
credential. Number of groups the user belongs to: XXX, maximum
number of groups supported: 512

The savepnpc command fails if the precommand and post command

contains any non-ASCII characters
LGTpa91475

The savepnpc command fails if the precommand and post command contains any
non-ASCII characters. When editing the <group-name>.res file for the savepnpc command,
the use of UTF-8 BOM characters is unsupported. Some text editors, such as Windows
Notepad, include UTF-8 BOM at the beginning of the file. You need to create a UTF-8
encoded file without the UTF-8 BOM.

EMC NetWorker Release 8.1 Release Notes 153

Known problems and limitations

If you use a text editor that includes UTF-8 BOM characters, use a hex editor to remove the
first 3 bytes in the UTF-8 encoded file. The UTF-8 BOM character is 3 bytes long and the
value is 0xef 0xbb 0xbf.

Restrictions associated with the ZFS file system

LGTsc00850

The following are restrictions associated with the ZFS file system:
◆ Only a root user with full access to ZFS directories may recover files. ZFS files can be
restored to a UFS file system. When restoring ZFS files to a UFS file system, only the
permission information is retained, the access control entries are not retained. If a
non-root user attempts to recover a ZFS file, the recover operation will core dump.
◆ ZFS snapshots and the files in ZFS directories are not backed up or restored when
restoring the original files. File systems must be explicitly specified in the client's save
set attribute.
◆ Backup and recovery of raw partitions on Solaris ZFS file systems is unsupported.

CLI problems and limitations descriptions

The contents of the /opt, /var, /etc directories are not excluded when the
-B option is used
LGTsc19916

During the HomeBase base profile recovery, the exclude.NETWORKER file is created in
<hombaseagent>/install path. The -B option uses this exclude file and excludes the
specified files and directories.

However, the exclude.NETWORKER file does not exclude files and directories under the
following directories. Data from these directories are not recovered from the source to the
target system since they might cause issues during the recovery and migration
operations:
◆ /opt
◆ /var
◆ /etc

To recover application files installed under /var or /opt, you must manually recover the
data to the appropriate directories for the application. To recover configuration files or
service binaries under /etc, you must manually recover those files and directories.

Also, when a file system recover is performed using the recover -B option, if the extended
profile is excluded the profile needs to be recovered separately. The profile by default
resides on the client in <installdir>\HomeBaseAgent\history\.

Not all save sets listed after running mminfo command with dedupe
option
LGTsc09483

When you run the mminfo command with the -q dedupe option to view save sets created
using deduplication, the command does not list the following types of save sets:

154 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

◆ Empty save sets

◆ Save sets in which nothing was backed up as the result of skip directives

Such save sets are treated as regular save sets, not as deduplicated save sets.

The recover -e exclude file option behaves differently for files and
directories
LGTsc19402

The recover -e exclude file option behaves differently for files and directories.

If the content of the exclude file has /etc/a*, the following occurs:
◆ Filenames that start with the character a are excluded.
◆ Directory names that start with the character a are excluded only when the directories
are empty.
◆ The NetWorker recover operation looks for the entries which start with the character a:
• If it is the end node in the tree, then the recover operation will exclude that entry.
• If it is not end node in the tree, then the NetWorker software will recover the
contents inside the tree.

Examples

If the content of exclude file has /etc/a*, type the following in the exclude file:
◆ To exclude all contents under a directory:
directory_name/*
◆ To exclude all files and directories inside the '/':
/*
◆ To avoid recovering the '/':
/**
◆ To exclude all files starting with the character a inside '/':
/a*

This excludes all of the files starting with the character a inside '/'. It also excludes the
empty directories starting with the character a.
◆ To exclude all files and empty directories inside the '/':
/?*

The recover -e <exclude file> option does not support NDMP file systems
LGTsc19084

Do not use the -e exclude file option to exclude NDMP file system files and directories
during a recover operation.

If the -e exclude file option is used with the recover command during a NDMP file system
recover operation, the specified files listed with the -e exclude file option will not be
excluded from the recovery and the following message will appear:

EMC NetWorker Release 8.1 Release Notes 155

Known problems and limitations

Total files excluded in the exclusion list is ‘0’

No filename checks for remote clients when using the -e exclude file
option with a directed recovery operation
LGTsc19405

The NetWorker software does not validate specified exclude filenames when the -e
exclude file option is used with a remote client during a directed recovery operation. The
validation will not be performed because the specified exclude file is present on the
target client when the recovery is invoked.

If the -e exclude file option is used with a local recover operation, the NetWorker software
validates the specified exclude filename. If the file does not exist, an error message
appears. This validation is limited to only local recover operations.

Cannot view log files created on a Windows on UNIX with the

nsr_render_log command
LGTsc10665

When log files created on Windows operating systems are viewed on a UNIX operating
system using the nsr_render_log command, nsr_render_log core dumps. View the log
files created on a Windows operating system using a Windows operating system.

Clients running pre-7.4 releases cannot use the mminfo -N command to

query save set names longer than 255 bytes
LGTsc06809

Clients running pre-7.4 releases cannot use the mminfo -N command to query save set
names longer than 255 bytes. This issue is caused by the enhancement in release 7.4 for
supporting longer save set names for to up to 1024 bytes (was 255 bytes in release 7.3)
to accommodate non-ASCII save set names, which will take more bytes.
Workaround

There are two workarounds for this issue if you are using a pre-7.3 client:
◆ Do not specify -N option on pre-7.4 mminfo to query save set names longer than 255
bytes. For example, the mminfo -avot command can be issued, but will only display
the first 255 bytes of save set name.
◆ Use the mminfo -N command with a 7.4 client to view save set names longer than 255
bytes.

The jbverify command is not supported on Solaris 10

LGTpa95406

The jbverify command is not supported on Solaris 10. The NetWorker software has an
autoconfiguration option provided that ensures a jukebox is configured correctly.

156 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Cloning and Staging problems and limitations descriptions

Marking volumes as recyclable might slow system performance
LGTsc15337

The following factors might slow system performance when marking a volume as
recyclable.

For each volume that is to be marked as recyclable:

1. The number of save set clone instances that reside on that particular volume and are
required to be marked as expired.
2. The number of clones or clone instances for each of the save set clone instances on
that particular volume, that reside on other volumes.
The number of clones that are created for a save set, equal the number of save set
clone instances. Each save set clone instance has the same ssid and unique clone ID.
Each save set clone instance resides on a separate volume. The first instance of
backup also has a clone ID.
Thus, the volume being marked as recyclable might contain save sets that have
multiple clone instances that reside on other volumes. These clone instances need to
be checked before the save sets are marked as expired. A save set, identified by the
ssid, only expires when all of the clone instances, which are identified by a clone ID,
expire.

To mark a volume as recyclable:

◆ From the server’s Administration window, click Media:
a. In the navigation tree, select Volumes. The Volumes detail table appears.
b. Right-click a volume in the Volumes detail table, and select Recycle. The Recycle
window appears. It names the selected volume.
c. Select the recycle policy: Auto (default) or Manual.
d. Click OK.
◆ From the command prompt, type the following:
nsrmm -o recyclable volume

Compatibility problems and limitations descriptions

Removable Storage Manager not supported
LGTsc19401

The NetWorker software does not support the Removable Storage Manager (RSM). You
cannot use the NetWorker software to allocate one or more volumes in an RSM library.

The Homebase Agent is not supported in a Cluster environment

LGTsc20557

For NetWorker release 7.5, the Homebase Agent is not supported in a Cluster
environment.

EMC NetWorker Release 8.1 Release Notes 157

Known problems and limitations

Support for Windows Change Journal Manager with Microsoft Windows 2008

LGTsc14700

When VSS is used, the Microsoft Change Journal is not used. Microsoft Windows 2008 is
VSS only, so Windows 2008 does not use the Windows Change Journal.

Configuration problems and limitations descriptions

Cannot set nsrdir parameter for virtual server using remote Microsoft
Management Console
LGTsc18397

When configuring a virtual NetWorker server on Windows Server 2008 Core using a remote
Microsoft Management Console (MMC), the nsrdir parameter for the NetWorker server
service cannot be set because the service's Properties tab is not visible.
Workaround

It is necessary to register the NetWorker server resource type on the remote machine
running MMC in order to view the service's Properties tab to change the settings. To
register the resource type:
1. Install the NetWorker server software.
2. Run regcnsrd -r on the remote machine. You can now use a remote MMC to set the
parameters.

Note: Besides MMC, you can also use the CLI cluster command on the cluster to set the
nsrdir value. For example: cluster res networker /priv: NsrDir="I:\nsr"

Running the Linux 64-bit package with the 2.6.x kernel may result in
crashes
LGTsc06585

A compatibility issue with libc and pthreads libraries on Linux 2.6 kernel-based
environments causes problems when running the 64-bit package for Linux on platforms
built on the 2.6.x kernel (for example, RedHat AS 4 and SuSE SLES 9). Although operation
may be normal at first, continued use of the 64-bit package with a 2.6.x kernel may result
in crashes of various programs.

Use the 32-bit package when running with a 2.6.x kernel. Running a 32-bit package
instead of the 64-bit package should have no operational impact. The 64-bit package can
still be used for kernel version 2.4 Linux variants such as RedHat AS 3 and SuSE SLES 8.

158 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Devices and Media problems and limitations descriptions

For Microsoft Windows Server 2008 raw devices are not saved under the
virtual client's index
LGTsc19241

When performing a savegroup of save set "All" for the virtual client, raw device are
excluded from the backup list, even when directives are used. When a backup of the raw
device is performed from the command line, the raw device is saved under the physical
host's index, rather than the virtual client's.
Workaround

To fix the problem and back up the raw devices under the physical node’s index:
1. Edit the client resource for the virtual client:
a. Update the save set list to include the raw device, for example:
\\.\M:
b. For the Backup Command attribute, set the following value to force the save sets
go to the correct index:
save -c virtual_client_name
c. Create a directive using rawasm, for example:
rawasm: \\.\M:
2. On all nodes in the cluster, create the pathownerignore file. Ensure that the
pathownerignore file is created in the NetWorker installation directory, for example:
c:\Program Files\EMC NetWorker\nsr\bin

Broker connects to first available storage node if no device or mmd

available on first host
LGTsc19676

If there is no device or mmd available on the first host in the affinity list, the broker
continues through the affinity list until the first available storage node is located. Since
the broker is not designed for file and AFTD devices, it does not recognize that the volume
and the device are inseparable (for example, the broker could request an AFTD volume to
be mounted on a different host). This is only a problem with stand-alone devices, because
volumes are not associated with storage nodes.
Workaround

To work around the problem, assign different devices to different pools. If the volume is
not in a jukebox, the server has no way of knowing which volume can be mounted on
which storage node.

EMC NetWorker Release 8.1 Release Notes 159

Known problems and limitations

Cannot create an advanced file-type device on a remote storage node

device
LGTsc11158

If the nsrmmd daemon has not started on the storage node, creation of an advanced
file-type device (AFTD) on the storage node will fail because the nsrmmd daemon cannot
validate the AFTD path.
Workaround

If this issue is encountered, either:

◆ Select No to the Verify path prompt when the AFTD is being created, then manually
label the device.
◆ Attempt to create the AFTD again.

VMware Consolidated Backup 1.0.x is not supported on Windows Server

2003, Enterprise Edition (x64) Service Pack 2
LGTsc10950

VMware Consolidated Backup 1.0.x is not supported on Windows Server 2003, Enterprise
Edition (x64) Service Pack 2.

NetWorker becomes unresponsive when attempting an invalid clone

operation
LGTsc06358

You cannot put multiple instances of the same clone onto a single volume. NetWorker
should prevent the operation and provide an error message. Instead, if this operation is
attempted, NetWorker becomes unresponsive.
Workaround

Ensure that multiple instances of a clone are placed on different volumes.

Cannot query UNIX non-ASCII save sets in NMC

LGTsc01923

Because of differences between the encoding used to create non-ASCII save sets on UNIX
platforms and the encoding used by NMC to query the media database, using the Query
Save Set window to query for non-ASCII save sets that were created on UNIX platforms will
result in no matches being found.

Virtual tapes are listed as full after mounting

LGTpa93001

In a Windows environment when using virtual IBM tape drives, virtual tapes are listed as
"full" even though 0 KB have been written to the tape. This can be observed after labelling
and mounting the tape.

160 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Workaround

Change the system configuration to use an equivalent tape drive from a manufacturer
other than IBM. For example, type HP LTO-3, instead of IBM LTO-3.

Scan for Devices configures all devices as a regular storage node device
LGTpa88188

When adding a new device to a jukebox, the scan operation detects a device, but
incorrectly displays it as a storage node device. Attempting to configure the device as
standalone device fails with the following error message:
cannot create the device, storage node enabler is required
Workaround

1. Using the Console, create a new device. For example:

rd=Storage_Node:/dev/rmt/...
2. Specify the device is a Dedicated Storage Node by using the Configuration tab.
3. Use jbedit command to add a new drive to this device. The jbedit man page contains
more information regarding the use of the jbedit command.

The Virtual Jukebox attribute for a CDL Virtual Tape Library is not
automatically set after upgrading
LGTsc01446

If the jbconfig command was used to configure a CDL VTL in the 7.3.2 release, the Virtual
Jukebox attribute is not automatically set after upgrading to the current release.
Workaround

After upgrading to the current release, initiate a Scan for Devices operation using the
NetWorker Console to scan only the NetWorker storage node on which the VTL is
configured. The NetWorker software will update all VTL related attributes utilizing the new
VTL license.

This behavior does not occur when using the auto-configuration option.

Inventory operation in Software Administration wizard delays for

unreachable clients
LGTsc06280

When starting an inventory operation using the Software Administration wizard, there is a
delay of 7 to 8 minutes for each client configured on the server that is unreachable (for
example, if a NSR Installed Software resource entry does not exist) and has not previously
been inventoried. Once started, you cannot cancel the inventory operation and must wait
until the Select Clients for Inventory window displays to continue the operation.
Workaround

Ensure there are no unreachable clients on the server. Also, if you want to perform an
inventory operation but do not need to inventory specific clients, inventory all clients
using the nsrpush command with the -all option; for example, nsrpush –i –all.

EMC NetWorker Release 8.1 Release Notes 161

Known problems and limitations

GUI problems and limitations descriptions

Deduplication recover sessions do not appear in NMC
LGTsc10295

Deduplication recover sessions do not appear in the Monitoring window of NMC.

However, deduplication save sessions appear.

To view the Monitoring window:

1. From the Administration window, click Monitoring.
2. Click Session.

Text searched for in the NetWorker Console Help program does not get
highlighted correctly if using JRE 1.5.x for Asian languages
LGTsc08756

Due to a known limitation with JRE 1.5.x (Sun bug 6375606), when performing a search
for text in the NetWorker Console Help program, the search is completed but the keyword
is not highlighted correctly. Update the JRE version to 1.6 or later.

On HP-UX the NetWorker Server window does not appear

LGTsc00365

On HP-UX, the following error message appears if in NMC the Setup > Setup System
Options attribute is selected and the NetWorker server window does not appear:
"Unable to connect to server: Failed to contact using UDP ping."
Workaround

1. Open the NMC window and select Setup > Setup System Options.
2. Ensure that RPC ping using UDP when connecting to NetWorker is cleared.

Cannot copy and paste Japanese characters by using NMC

LGTsc01115

On UNIX platforms, cannot copy and paste Japanese characters with Ctrl+C and Ctrl+V
using NMC.

Newly created groups are not displayed by the NetWorker Console after
moving system date back in time
LGTpa95162

If the system date is moved to a past date and subsequently moved back to the current
date, newly created groups are not displayed in the NetWorker Console.
Workaround

Select the Archive Request tab from within the Monitoring task window and the Groups
table will refresh.

162 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

Cannot launch NetWorker Console Help program on Solaris 10 with a JRE

earlier than version 1.5.0_09
LGTpa96168

The NetWorker Console Help program will not launch if the JRE version is earlier than
1.5.0_09.

NetWorker release 7.6 and later requires JRE version 1.6 or later.

NMC version 3.4 and NWD version 1.0 cannot be installed on the same
host
LGTsc05879

If NetWorker Dashboard (NWD) 1.0 is installed on the NMC Server 3.4 host, NMC fails to
download, and a Java Web Start error appears. NMC 3.4 and NWD 1.0 cannot function
together due to database security and Java certificate issues, and the NMC GUI will not
launch if NWD is installed on the same host as the NMC Server.
Workaround

Uninstall NWD before updating to NMC 3.4 for NetWorker release 7.4.

Installation problems and limitations descriptions

For Microsoft Windows, ensure that the HomeBase Agent software is
installed in the same installation path as the NetWorker software
LGTsc16831

For Microsoft Windows, ensure that the HomeBase Agent software is installed in the same
installation path as the NetWorker software.

It is not recommended to manually install the HomeBase Agent software into a location
that was not specified during the NetWorker software installation process.

On Microsoft Windows, the HomeBase Agent installation path is at the same level as the
default or user-defined installation directory.

For example:
C:\Program Files\EMC NetWorker\nsr
C:\Program Files\EMC NetWorker\HomeBaseAgent
Example

For example, if the NetWorker software is installed under:

◆ C:\Program Files\EMC NetWorker which is the default location for the NetWorker
software, then the HomeBase Agent software is installed under the following location:
C:\Program Files\EMC NetWorker\HomebaseAgent
◆ C:\EMC NetWorker which is an example of a non-default installation location for the
NetWorker software, then the HomeBase Agent software is installed under the
following location: C:\EMC NetWorker\HomebaseAgent

EMC NetWorker Release 8.1 Release Notes 163

Known problems and limitations

Before restoring the base profile, manually edit the fields in the
recovery-configuration.xml file
LGTsc18641

If the HomeBase Agent is installed with the NetWorker client package, the base profile for
the HomeBase Agent might not be updated.

Before you restore the base profile, you must first update the recovery-configuration.xml
file or the recovery operation might fail.
Workaround

To work around this issue, manually edit the fields in the following file before performing
a recovery of the base profile:
install_dir/etc/config/recovery/custom/recovery-configuration.xml

Note: Where install_dir is the location of the Homebase Agent installation directory.

Unable to launch the NetWorker Dashboard software with NetWorker

installed
LGTsc18925

If the NetWorker Dashboard software is installed on a computer with the NetWorker

software already installed, the NetWorker Dashboard software will not start.

The following errors might occur:

◆ At the end of the NWD server installation process, a warning message might appear
stating that the NWD server could not start.
◆ After starting the NWD Server process, either manually through Microsoft Windows
Services or automatically on system start up, the NWD server process might stop
shortly afterwards.
◆ When launching the NWD GUI, a message might appear stating that the NWD server
process is not running.

Workaround

To work around this issue:

1. Open and edit the serverproperties.xml file. The file is located in:
NWD_Install_Dir\classes\serverproperties.xml

Note: Where NWD_Install_Dir is the installation directory where the NWD server
software was installed.

2. Locate the text specifying the dbport value. For example:

<param name="dbport">2638;DoBroadcast=NONE;HOST=myhost</param>
3. Edit the text string so that it contains only the port information. For example:
<param name="dbport">2638</param>
4. Save the serverproperties.xml file.

164 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

5. Restart the NWD server process.

Licensing problems and limitations descriptions

This section details the problems and limitations related to licensing.

NetWorker and the HomeBase Agent (BMR) licenses work independently

LGTsc19388

The software licenses for the NetWorker software and the HomeBase Agent (Bare Metal
Recovery) software work independently.
Example 1

If the Bare Metal Recovery (BMR) license is disabled and the base enabler license for the
NetWorker server where the client is configured is enabled, the following occurs when a
savegroup is run:
◆ The BMR profiling fails
◆ The NetWorker backup succeeds

Example 2

If the Bare Metal Recovery (BMR) license is enabled and the base enabler license for the
NetWorker server where the client is configured is disabled, the following occurs when a
savegroup is run:
◆ The BMR profiling succeeds
◆ The NetWorker backup fails

Localization problems and limitations descriptions

For more information on localization important notes and tips, see “Internationalization
support” on page 175.

The NMC Japanese Online Help displays incorrect characters on Linux

systems
LGTsc02862

The NMC Japanese Online Help displays incorrect characters on Linux systems due to
limitations with the Javahelp software. These limitations affect the software’s ability to
display fonts in the help viewer content pane.

Due to this limitation in Javahelp the only character encoding that appears is the system
default; not the font defined by the user.

Non-ASCII save set names are displayed incorrectly in nsradmin visual

mode on Linux
LGTpa92833

Non-ASCII save set names are displayed incorrectly in nsradmin visual mode on Linux.

EMC NetWorker Release 8.1 Release Notes 165

Known problems and limitations

Workaround

To work around this limitation, do one of the following:

◆ Use the print nsr client command in nsradmin window to view the non-ASCII save set.
◆ Use the Console GUI on the Linux client to view the non-ASCII save set.

Messaging problems and limitations descriptions

nsrbmr does not give proper error messages under certain conditions
LGTsc16423

An appropriate error message is not displayed when both BMR profiling status and
savegrp status fail due to one of the following:
◆ The Homebase Agent is uninstalled and a savegrp with BMR-enabled client is
performed.
◆ A communication issue occurs between nwserv and nwclnt.

An error message appears but BMR profiling still completes with a status of “pass” in the
following scenarios:
◆ The Homebase Agent services are stopped during or after profiling. In this case,
nsrbmr completes without any errors.
◆ Incorrect options are specified in the BMR options field under client properties. A
command usage error appears, but profiling status indicates “pass”.
◆ After installation, when the first savegrp is performed with BMR-enabled client, the
HomeBase server name is specified in the server properties and the BMR options field
is left blank.

Once the savegrp is completed for the BMR-enabled client, it is recommended to check
the status of the BMR profile completion to ensure that BMR profiling has completed
successfully.

Error message does not display for annotations with similar characters at
the end
LGTsc17737

Nsrretrieve does not display an error message when a retrieve is performed using regular
expressions for two annotations with similar characters at the end.

Erroneous error message may be generated by Linux operating systems

when performing an I/O to a volume
LGTsc08054

The following erroneous error message may be generated by Linux operating systems
when performing an I/O to a volume:
kernel: program nsrmmd is using deprecated SCSI ioctl, please convert
it to SG_IO.

Ignore this message. In upcoming NetWorker releases, this warning from the kernel will
stop.

166 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

New pdksh package required on SuSE 10 x86

LGTsc08978

A failed dependencies error occurs on the SuSE 10 x86 platform if the required version of
the pdksh package is not installed. The following is displayed:
error: Failed dependencies: /bin/ksh is needed by lgtoclnt-7.4-1
Workaround

Install pdksh-5.2.14-801.i586.rpm, which can be downloaded from the SuSE/Novell

website.

You could alternatively install the packages with the --nodeps option.

Incorrect error message reported when starting the NetWorker console on

Red Hat 64-bit Enterprise Server 4
LGTsc03478

On a Red Hat 64-bit Enterprise Server 4, the preinstalled Mozilla browser incorrectly
reports that the JRE is not installed. The NetWorker console uses Javascript to detect the
JRE installation and a warning message appears reporting that the correct JRE is not
installed on the client machine.

If the correct JRE is already installed, ignore the message. To start the NetWorker Console,
click the Start link in the line "If you have installed JRE 1.6 or later, then click here to start
NetWorker Management Console" from the Mozilla browser.

Error states you need to perform an uninstall when updating from

NetWorker 7.2.2 32-bit to NetWorker 7.5 64-bit
LGTsc06288

When updating from NetWorker release 7.2.2 32-bit on x64 to NetWorker 7.5 64-bit, an
error message is displayed stating that the 32-bit NetWorker software is already installed
and should be uninstalled.
Workaround

It is not necessary to do a complete uninstall of NetWorker release 7.2.2 to perform this

update. Perform a partial uninstall of NetWorker release 7.2.2 32-bit, then install
NetWorker release 7.5 64-bit to the same location that the 7.2.2 32-bit version was
installed.
Note: The 32-bit version of NetWorker release 7.3.x is the only version of 32-bit NetWorker
software that is supported on ntx64. NetWorker release 7.4 and later does not allow 32-bit
NetWorker to be installed on ntx64.

NetWorker support for Novell’s OES Linux limitations

Limitations to support for Novell’s OES Linux with NetWorker are as follows:
◆ “Metadata modifier field not recovered” on page 168
◆ “Creating eDirectory backup files for NetWorker backup and recovery” on page 168
◆ “Creating GroupWise backup files for NetWorker backup and recovery” on page 169

EMC NetWorker Release 8.1 Release Notes 167

Known problems and limitations

IMPORTANT
NSS+eDirectory backups and GroupWise backups are two unrelated procedures and
should be considered separately.

Metadata modifier field not recovered

When you perform backup and recovery by using the NetWorker software, it does not
recover the metadata modifier field for NSS files or directories.

Creating eDirectory backup files for NetWorker backup and recovery

eDirectory is a database of Network resources that contains extra information related to
NSS objects. However, since eDirectory is not part of the file system, when you perform
backup and recovery using the NetWorker software, NetWorker does not recover
eDirectory data as it was backed up.
Workaround

NetWorker can perform a parallel backup and recover of eDirectory files using
eMBox/DSBK (eMBox is a GUI version of DSBK). DSBK is a command line tool used to
back up eDirectory as a complete database, saving all the information on a file that is
created when running the pre-and-post script. The file created by DSBK is saved by
NetWorker along with the backup of the file system.

The following is an example of how to create a pre-and-post script to backup and recover
eDirectory.
Note: If performing a manual (non-scheduled) backup, DSBK should always be run before
the NSS file system backup. If performing a scheduled backup, initiate DSBK by a
pre-command script using savepnpc, and ensure that the savegrp includes the exported
database on the file system. DSBK should always create the eDirectory backup file before
save runs, so NetWorker can back up both the file system and the DSBK file. If performing
a recovery, the same order applies. Recover eDirectory using DSBK, and then recover the
NSS file system.

To back up and recover eDirectory by using DSBK:

1. In the /etc/dsbk.conf file, set the following value:
/root/dsbk.command
The output for the command is written to eDirectory’s log file.
2. Run the following command:
dsbk backup -f /media/nss/NSS1/ndsbk -l /media/nss/NSS1/nds.log
Two files are created; ndsbk and nds.log. The ndsbk file contains eDirectory
information that can be backed up using the NetWorker software. The nds.log file
contains messages for the successful creation or errors.
3. Using the NetWorker software, recover the ndsbk file.
Once the file has been recovered, you can recover eDirectory.
4. To recover eDirectory, run the following command:

168 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

dsbk restore -f /media/nss/NSS1/ndsbk -l /media/nss/NSS1/nds.log -r

-a -o

Creating GroupWise backup files for NetWorker backup and recovery

When you perform backup and recovery using the NetWorker software, you need to create
a pre-and-post script to backup GroupWise so that NetWorker can recover GroupWise
data as it was backed up.

The following is an example of how to create a pre-and-post script to backup GroupWise:

1. Create a backup directory (for example, backup).
2. Run this command to copy the Post office (-p) to the backup directory:
./dbcopy –m –p –v /gw /backup
3. Run this command to copy the Domain (-d) to the backup directory:
./dbcopy –m –d -v /gw /backup
You can now use the NetWorker software to back up the backup directory.
4. Run the following command to stop GroupWise:
./rcgrpwise stop
5. Remove GroupWise and the backup directory.
6. Run the following command to start GroupWise:
./rcgrpwise start

The following is an example of how to create pre-and-post script to recover GroupWise:

1. Use NetWorker to recover the backup directory.
2. Run the following command to stop GroupWise:
./rcgrpwise stop
3. Run the following command to copy the Domain (-d) from the backup directory to the
GroupWise directory:
./dbcopy –m –d –v /backup /gw
4. Run the following command to copy the Post office (-p) from the backup directory to
the GroupWise directory:
./dbcopy –m –p –v /backup /gw
5. Run the following command to start GroupWise:
./rcgrpwise start
6. Run the following command to view the status of GroupWise and confirm that the
recovery was successful:
./rcgrpwise status

EMC NetWorker Release 8.1 Release Notes 169

Known problems and limitations

NetWorker Management Console problems and limitations descriptions

NetWorker Management Console may become unresponsive when opened
using Exceed
LGTsc02903

When using Exceed to launch the NetWorker Management Console from a UNIX or Linux
system, NMC may become unresponsive.
Workaround

Enable the XTEST (X11R6) variable in the Exceed XConfig program on the Windows host.

Restore problems and limitations descriptions

Last-modified date changed for files recovered to Sun Cluster global file
system
LGTsc19364

The last-modified date for files recovered to the SUN Cluster global file system may be
displayed as the current time instead of the last date the files were changed, due to an
issue with Sun Cluster version 3.2.
Workaround

Disable pxfs_fastwrite by performing the following:

1. On all cluster nodes, run:
# echo "pxfs_fastwrite_enabled/W 0" |mdb -kw
2. Unmount, then mount the global file systems for the change to appear.

To re-enable fastwrite:
1. Set the variable pxfs_fastwrite_enabled to 1.
2. Unmount, then mount the global file systems.

More information on this issue is provided at the following website:

http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6540206

Do not recover identical save sets with different ssids to the same target
directory
LGTsc10796

Do not perform a save set recovery from the command prompt, with multiple save set IDs
that all point to the same file/directory path and will be recovered into the original
location.

Data corruption might occur when the same dir/file path is backed up multiple times. This
creates multiple save sets, each with a different save set ID. For example, ssid1 and ssid2
are save set IDs created for the same file/directory path, at different save or backup times.

170 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

When the recover command or the nsrretrieve command is run to retrieve more than one
instance of the save set, each with a different save set ID, into the original location, the
recovered instance of the directory/file from ssid1 might be overwritten by the recovered
data of the same directory/file from ssid2.
Example

Data corruption might occur in the following scenario:

recover -s server -c client -S ssid1 -S ssid2
nsrretrieve -s server -S ssid1 -S ssid2

where ssid1 and ssid2, are save set IDs of the same file/directory.

NetWorker interactive recover does not recover the registry and the com+
regdb writers
LGTsc14577

You cannot perform a system recover of the registry and com+ regdb writers by using the
interactive mode of the recover command.
Workaround

Perform a system recovery using the noninteractive mode of the recover command, or use
the NetWorker User program.

Restoring full VM image on ESX server fails

LGTsc11618

When attempting to restore a full VM image directly onto the ESX server, the recovery fails.
Workaround

Recover to the proxy host and use VirtualCenter tools (VMware Converter).

Unable to restore files encrypted with both Microsoft Windows Encrypting

File System & AES encryption
LGTsc11734

When AES encryption is applied to a file that is also encrypted using the Microsoft
Windows Encrypting File System (EFS), the backup will be reported as successful.
However, recovery of the file fails and the following message is written to the NetWorker
log file:
recover: Error recovering <filename>. The RPC call completed before
all pipes were processed.
Workaround

Do not use AES encryption when backing up files that are encrypted by using EFS.

EMC NetWorker Release 8.1 Release Notes 171

Known problems and limitations

FSRM Disk Quota may not be restored to previous settings after recovery
LGTsc05990

The NetWorker software might not restore the previous FSRM Disk Quota configuration
after the Disk Quota recovery. Some of the changes made to the setup values after saving
the FSRM Disk Quota may remain after the recovery operation. However, if the Disk Quota
is completely deleted, it can be restored to the previous configuration.

After a BMR recovery operation the root password might change

LGTsc20558

After BMR recovery (base profile) operation, the root password might change. For
example:
◆ On UNIX, the root password might be reset to risckey.
◆ On Microsoft Windows a new admin username and password is created. You can log in
as administrator with the profiled servers password or as the indigostone user. For
example:
• Admin username: indigostone
• Admin username password: H0meBase

On NetApps files, filenames ending with special characters might be

renamed
LGTpa95900

On NetApps filers, the recovery of filenames ending with the special character sequence
~n (where n is a number) might be renamed with ~1 appended at the end of the name. For
example, a file named C~1 might be restored with the name C~1~1.

This behavior might occur when the DIRECT=Y application parameter has been defined in
the NetApps client instance.
Workaround

To perform a file level recovery of a file with a special character sequence, perform one of
the following:
◆ Set the following environment variable prior to performing a file level recovery:
NSR_NDMP_RECOVER_NO_DAR=y
◆ Redefine the NetApps client application parameters:
DIRECT=Y.

Note: This will not address the recovery of legacy backups performed prior to the DIRECT=Y
parameter removal. In this case, you must set the NSR_NDMP_RECOVER_NO_DAR=y
environment variable before performing the recovery operation.

172 EMC NetWorker Release 8.1 Release Notes

 Known problems and limitations

The winworkr program fails to retrieve successfully archived files if a slash

(\) character was entered in the annotation string
LGTpa94966

The winworkr program fails to retrieve successfully archived files if a slash (\) character
was entered in the annotation string.
Workaround

Enter a double slash (\\) into the annotation string and the files will be archived and
retrieve operations perform successfully.

Directed recover fails with permission errors on Windows

LGTpa83927

A directed recover operation fails when using the command line interface (CLI) and the
winworkr program. Permission errors display if the NetWorker server and the client where
the files are to be recovered to are running Windows.
Workaround

To perform a directed recover, one of the two following conditions must be met:
◆ If the NetWorker server and target recover client are in the same domain, start the
NetWorker server (nsrd) as a domain user that is in the Windows Administrators group
on the NetWorker server machine.
◆ If neither machine is in a domain, or they are not in the same domain:
• Ensure that the user:
– Exists on both machines.
– Has the same password on both machines.
– Is in the Windows Administrators group on the NetWorker server machine.
• Start the NetWorker server (nsrd) as the local user existing on both systems with
the same password.

The winworkr program will not relocate to a partition not existing on the
initiating host if the NetWorker server is running release 7.4 and the client
is running release 7.2
LGTsc00167

If the NetWorker server is running release 7.4, and the 7.2 release is installed on a client,
a directed recover will not relocate to a partition that does not exist on the host initiating
the recover operation. An error message is displayed indicating that it is an invalid
directory.
Workaround

Upgrade the client to release 7.4 or create the appropriate directory on the host initiating
the recovery.

EMC NetWorker Release 8.1 Release Notes 173

Known problems and limitations

Performing a save set query can take a long time if the query parameter in
the Query Save Set tab is set to "Save Time" and From and To calendars for
3 or more days
LGTsc05053

In a large scale NetWorker environment, performing a save set query can take a long time
if the query parameter in the Query Save Set tab is set to Save Time and 3 or more days in
the From and To calendar.

An Operation in Progress window appears with the following message:

Getting save set information from NetWorker server <sever name>

This dialog box will remain till the save set query is complete blocking you from
performing any other operations on the NetWorker server.
Workaround

Perform the save set query using the mminfo command from the command line.

For recoveries operations using Celerra filers, misleading error messages

might display
LGTpa96554

On Celerra filers, the recovery of a backup containing a recursive directory can result in
the generation of log messages:
NDMP Service Debug: Too much retry on header research

This log message can be intermittent in nature as subsequent recoveries of the same back
might not result in the generation of a log message.

This log message does not affect the recovery and can be ignored.

Upgrading problems and limitations descriptions

JRE version mismatch causes authentication failure, stops GSTD
LGTsc08958

After updating the NetWorker release, if the JRE version installed on the system is lower
than the JRE version required by NetWorker, NMC cannot be launched and the GSTD
process stops with an "authentication failure" error message.
Workaround

Update to JRE version 1.6 or later.

When upgrading from release 7.3.1 the Virtual Jukeboxes attribute will not
be set correctly
LGTpa95019

When upgrading from the 7.3.1 release, the Virtual Jukeboxes attribute will not be set if
you previously configured a Virtual Tape Library (VTL) using the 7.3.1 release. After
upgrading to 7.5, the Virtual Jukeboxes attribute uses the normal jukebox license instead

174 EMC NetWorker Release 8.1 Release Notes

 Internationalization support

of a VTL license. The VTL still functions normally using the normal jukebox license. The
Virtual Jukeboxes attribute will function normally if upgrading to the 7.5 release from
7.3.2.

Group details window is empty after upgrading from NetWorker release

7.2.2
LGTsc01587

After upgrading to NetWorker release 7.4 from release 7.2.2, savegroups details run prior
to the upgrade do not appear in the Group Details window.

Internationalization support
These sections describe important notes and tips pertaining to the internationallization
support for the NetWorker release.

Additional information on support for internationalization can also be found in the

NetWorker Administration Guide in the NetWorker Server Management chapter.

Locale and Code Set Support

NetWorker software does not support locales (defined by your operating system) or code
sets that remap characters having special meaning for filesystems. Depending on the
filesystem, these special characters can include the forward slash (/), the backslash (\),
the colon (:), or the period(.). De_DE.646 is an example of one unsupported locale.

NetWorker software might function normally in such an environment, but might not
function normally if the locale is changed (the previously existing indexes can become
invalid).

Localization support
These sections describe important notes and tips pertaining to localized NetWorker
releases.

Important notes and tips

The NetWorker 8.1 software supports language packs, which can be installed as part of
the NetWorker installation process, or can be installed separately after the NetWorker
software has been installed. Note, however, that any program screens or messages that
have changed since NetWorker 8.0 are not localized.

The following sections contain important notes and tips, as well as limitations, pertaining
to the internationalized NetWorker software:
◆ “Localized software contains some English Text” on page 176
◆ “Java Web Start cache path and non-English characters” on page 176
◆ “Locale settings with NDMP” on page 177
◆ “Locale settings with NDMP” on page 177
◆ “Display of an unsupported character in the current locale” on page 177

EMC NetWorker Release 8.1 Release Notes 175

Localization support

◆ “Man page locales” on page 177

◆ “Supported Locales” on page 177
◆ “Changing the Locale in the NMC GUI” on page 178
◆ “Scheduled backup or Archive Requests of non-ASCII files or directories” on page 179
◆ “Maximum NetWorker supported path” on page 179
◆ “In non-English environment, character encoding used for NMC client and NetWorker
client must be the same” on page 179
◆ “The NMC Japanese Online Help displays incorrect characters on Linux systems” on
page 180
◆ “Non-ASCII hostnames are not supported by NetWorker” on page 180
◆ “JRE 1.6 users cannot export reports as PDF documents for non-English locales on AIX
and HPUX” on page 180
◆ “JRE 1.6 users cannot export reports as PDF documents for non-English locales on AIX
and HPUX” on page 180
◆ “Recovering a large number of files may take a long time in the French locales on
Solaris” on page 180
◆ “Recovering a large number of files may take a long time in the French locales on
Solaris” on page 180
◆ “Garbled characters may appear in the NetWorker Console GUI font list on Solaris” on
page 180
◆ “Problem with highlighted text in the NetWorker Console Help program after
performing a search using JRE 1.5.x for Asian languages” on page 180
◆ “Entering non-ASCII characters in NetWorker user interfaces” on page 181
◆ “Non-ASCII save set names are displayed incorrectly in nsradmin visual mode on
Linux” on page 181
◆ “The XAPPLRESDIR environment variable must be set to operate the NetWorker
Console on HP-UX” on page 181

Localized software contains some English Text

Messages and strings that were added after the NetWorker 7.5 Service Pack 1 release
have not been localized. Some strings from the operating system have also been left
intentionally unlocalized. The nsrwatch and nsradmin programs are not localized. The
English language components do not affect the functionality of the software.

The only RAP value that supports non-ASCII characters is the Save Set attribute of the
Client and Archive Request resources.

Java Web Start cache path and non-English characters

If the path for the Java Web Start cache contains non-English characters, this will cause
the NetWorker Console to fail to launch. The resolution to this problem is to change the
Java Web Start cache path to a path that contains no non-English characters.

176 EMC NetWorker Release 8.1 Release Notes

 Localization support

The Java Web Start cache path is changed in the Java Web Start Console. Different JRE
versions have different names for the Java Web Start Console; consult java.sun.com for
details.

Locale settings with NDMP

When running NDMP backups, the locale setting has to be consistent in your
environment. All UNIX operating system locale settings on the filer (including UTF-8) must
be the same and the NMC client can only be run on an UNIX client set to the exact same
locale setting as the filer.

Backup and recovery operations can be run on any locale, but if you try and browse on a
locale that is different from the original locale the filenames will show up as random
characters.

Display of an unsupported character in the current locale

If the NetWorker software encounters a character that is unsupported in the current
locale, it replaces the character with a '?'.

Man page locales

Man pages are displayed based on the locale setting for a specific language. The
following EUC locales are supported:
◆ Chinese: EUC-CN
◆ French: ISO8859-15
◆ Korean: EUC-KR
◆ Japanese: EUC-JP

If the locale is not set to a specific language matching an installed language pack, the
man pages will be displayed in English.

Supported Locales
Table 16 on page 177 lists the supported locales.

Table 16 Supported Locales (page 1 of 2)

Operating system

Language Windows Solaris HPUX AIX Linux

English English OS Default OS Default OS Default OS Default

Locale Locale Locale Locale

French French UTF-8 UTF-8 UTF-8 UTF-8

(France) ISO8859-1 ISO8859-1 ISO8859-1 ISO8859-1
ISO8859-15 ISO8859-15 ISO8859-15 ISO8859-15

Japanese Japanese UTF-8 UTF-8 UTF-8 UTF-8

(Japan) EUC-JP EUC-JP EUC-JP EUC-JP
S-JIS S-JIS

EMC NetWorker Release 8.1 Release Notes 177

Localization support

Table 16 Supported Locales (page 2 of 2)

Operating system

Language Windows Solaris HPUX AIX Linux

Chinese Chinese UTF-8 UTF-8 UTF-8 UTF-8

(China) EUC-CN EUC-CN EUC-CN EUC-CN
GB18030 GB18030
GBK GBK
BIG5 BIG5

Korean Korean UTF-8 UTF-8 UTF-8 UTF-8

(Korea)

Note: Localization is not supported on the Mac OS, Tru 64 or SGI platforms.

Changing the Locale in the NMC GUI

There are three conditions for the displayed textual elements (messages, dates, time and
numbers) in the NMC GUI:
1. If there is an appropriate NetWorker language pack installed on the NMC server, all
textual elements will be rendered to the current user locale.
2. If no NetWorker language pack is installed on the NMC server, all textual elements will
be rendered to English.
3. If the locale is neither supported by the NetWorker software nor has an appropriate
NetWorker language pack installed, all textual elements will be rendered to English,
except dates, times and numbers which will appear in the current user locale that is
supported by the installed JRE on the user host.

The NMC GUI must be restarted to apply any change to the locale.

Example 1
There is a French language pack installed on the NMC server and the user locale is French.
The user logs in to the NMC server. In the NMC GUI, all textual elements will be in French.

Example 2
The installed JRE supports French and the user locale is French. There is no French
(France) language pack installed on the NMC server. In the NMC GUI, all texts and
messages will be in English, except dates, time and numbers will be in French.

It is the user’s responsibility to change the locale and apply the change correctly to the
operating system. For example, changing and applying the locale to Windows server 2003
and Solaris 5.9 are as follows:

Windows server 2003

a. Click Start.
b. Select Control panel > Regional and Language Options > Regional Options.
c. Select the language.
d. Select the location.

178 EMC NetWorker Release 8.1 Release Notes

 Localization support

e. Click Ok.

Solaris 5.9
a. Logout to the Welcome dialog box.
b. Click Options > Language.
c. Select the language.
d. Log in.

Note: Date and times may be still in North American format. Not all date and time
displayed are internationalized. This is a known deficiency that will be addressed in a
future release.

Scheduled backup or Archive Requests of non-ASCII files or directories

The value of the Save Operations attribute in Clients or Archive Requests resource should
be set to:
◆ NetWorker UNIX clients at release 7.4 or later: I18N:mode=nativepath.
◆ NetWorker UNIX clients at a release level prior to 7.4: I18N:mode=utf8path.
◆ Microsoft Windows clients: I18N:mode=utf8path

If the Client Configuration wizard is used to create the Clients resource, the Save
Operations attribute will be automatically filled-in based on the client platform when
non-ASCII save sets are specified.

Maximum NetWorker supported path

The maximum length of the NetWorker supported path has increased to 12 KB. The
number of characters supported in the path is dependent on the language of the
characters and any specific operating system limitations.

Non-English characters require more bytes than English characters. Ensure that the
filepath and directory names remain within the limits imposed by the operating system
and the NetWorker software.

In non-English environment, character encoding used for NMC client and NetWorker client must be
the same
LGTsc21657
In a non-English environment, characters do not display correctly if the character
encoding is different on the NMC client than the files on the NetWorker host that are being
browsed.
Workaround

In a non-English environment, ensure that the NMC client uses the same character
encoding as the files on the NetWorker client host that are to be browsed.

EMC NetWorker Release 8.1 Release Notes 179

Localization support

For example, if the NMC client uses the zh_CN.EUC locale it will not properly display files
that were created using the zh_CN.UTF-8 locale. The character encoding used for the NMC
client and NetWorker client must be the same to ensure the proper display of characters.
In this example, the NMC client should be started using the zh_CN.UTF-8 locale.

The NMC Japanese Online Help displays incorrect characters on Linux systems
LGTsc02862
The NMC Japanese Online Help displays incorrect characters on Linux systems due to
limitations with the Javahelp software. These limitations affect the software’s ability to
display fonts in the help viewer content pane.

Due to this limitation in Javahelp the only character encoding that appears is the system
default; not the font defined by the user.

Non-ASCII hostnames are not supported by NetWorker

LGTsc26980
NetWorker does not support hosts that have non-ASCII characters in the hostname.

JRE 1.6 users cannot export reports as PDF documents for non-English locales on AIX and HPUX
LGTsc26288
The NetWorker option to export reports in Acrobat PDF format is not supported on AIX and
HPUX hosts that are running JRE 1.6 in a non-English locale.

Recovering a large number of files may take a long time in the French locales on Solaris
LGTsc05339
In the French locale on Solaris, a degradation in performance may be seen when
recovering a large number of files (greater than 100,000).

To improve performance, expand the command dialog box to reveal the complete
pathnames of the files being recovered in the output field.

Garbled characters may appear in the NetWorker Console GUI font list on Solaris
LGTsc03894
Garbled characters may appear in the NetWorker Console GUI font list for font names if
the fonts do not have English names, or the localized names are not recognized by the
JRE.

Problem with highlighted text in the NetWorker Console Help program after performing a search
using JRE 1.5.x for Asian languages
LGTsc02814
Due to a known limitation with JRE 1.5.x (Sun bug 6375606), text that is highlighted in the
NetWorker Console Help program after a search has been performed will not be
highlighted correctly. Update the JRE version to 1.6 or later.

180 EMC NetWorker Release 8.1 Release Notes

 Localization support

Entering non-ASCII characters in NetWorker user interfaces

LGTpa88887
Non-ASCII characters are supported only for the Save Set attribute in Client and Archive
Request resources. However, user interfaces such as NMC do not prevent the user from
entering non-ASCII characters for other attributes in NetWorker resources.

Non-ASCII save set names are displayed incorrectly in nsradmin visual mode on Linux
LGTpa92833
Non-ASCII save set names are displayed incorrectly in nsradmin visual mode on Linux.
Workaround

To work around this limitation, perform one of the following:

◆ Use the print nsr client command in nsradmin window to view the non-ASCII save set.
◆ Use the Console GUI on the Linux client to view the non-ASCII save set.

The XAPPLRESDIR environment variable must be set to operate the NetWorker Console on HP-UX
LGTpa79450
To operate the NetWorker Console on HP-UX that is running a non-English locale, the
XAPPLRESDIR environment variable must be set as follows:
XAPPLRESDIR=/usr/lib/X11/app-defaults

EMC NetWorker Release 8.1 Release Notes 181

Technical notes

Technical notes
This section describes important notes and tips for using the NetWorker software.

NetWorker cloud backup option and network connectivity

Cloud backups are highly dependent on the network connection that is used to access the
service. Any disruption in connectivity or a slowdown in network access speed may
adversely affect cloud backups or recoveries.

Additionally, if you define large Network Write Sizes for the cloud device, make sure to set
the Send/Receive Timeout attribute proportionally high to avoid read/write timeouts.
Optimal values for send/receive timeouts vary depending on the network speed and
bandwidth.

Open File Management on Microsoft Windows

NetWorker clients will automatically use VSS for file system backups, avoiding the need
for Open File Manager. A license is not required when used in conjunction with a
NetWorker server. NetWorker clients prior to release 7.4 Service Pack 2 also automatically
use VSS for file systems backups, whether or not the NetWorker VSS Client for Microsoft
Windows is installed. However, those clients will require a NetWorker VSS Client
Connection license.

Table 17 Recommended management of open files

Windows 2000 and XP Windows 2003 Windows 2008 and Vista

No VSS is available in the operating Use VSS to back up open files. Use VSS to back up open files.
system. For all versions of NetWorker, • If both client and server are using release • If both client and server are Windows
use Open File Manager to back up open 7.4 Service Pack 2 or later, no VSS Client 2008, no VSS Client Connection
files. Connection license is required. license is required.
• If client or server are at using a release • If the client is Windows Vista and the
prior to 7.4 Service Pack 2, a VSS Client server is using release 7.4 Service Pack
Connection license is required. 2 or later, no VSS Client Connection
license is required.
• If a Windows 2008 or Vista client is
used with a server using a release prior
to 7.4 Service Pack 2, a VSS Client
Connection license is required.

EMC NetWorker License Manager not supported on Solaris Opteron platform

The 7.6 NetWorker release does not support the EMC NetWorker License Manager on the
Solaris Opteron platform.

NetWorker Management Console does not support Microsoft Internet Explorer

version 7.0
Microsoft Internet Explorer version 7.0 is not supported by NMC on any Windows
platforms except Microsoft Vista.

182 EMC NetWorker Release 8.1 Release Notes

 Technical notes

A server does not support a NetWorker client running release 7.2.x or earlier
A NetWorker server does not support a NetWorker client running release 7.2.x or earlier.

Single clients or archive requests resource for UNIX non-ASCII paths support only
one locale
A single Clients or Archive Requests resource supports non-ASCII UNIX paths belonging to
only one locale. If you have paths in multiple locales, you must create multiple Clients or
Archive Requests resource. A Clients or Archive Requests resource supports paths only
from a single locale on UNIX.
Note: This limitation does not apply to Microsoft Windows paths.

VMware qualification
NetWorker release 8.0 and later is qualified with VMware. The NetWorker Administration
Guide and the EMC NetWorker Release 8.0 VMware Integration Guide provide more
information on support for VMware.

Symbolic links are not restored during DAR recovery with NetApp
During a DAR recovery, symbolic links for files, directories, and other specific files, such
as device files or named pipes, cannot be recovered. To recover these files, use the
NetApp restore command with the -x option. The Network Appliance documentation has
more information about the NetApp restore command.

SYSTEM save set archive unsupported on Microsoft Windows

Archiving of SYSTEM or VSS SYSTEM save sets is not currently supported.

NetWorker connections through a firewall

The NSR_KEEPALIVE_WAIT variable sets the timeout limit that the nsrexecd daemon uses
to keep messages active once a connection to the NetWorker server has been established
through a firewall. The period that nsrexecd will send keep-alive messages to nsrexec is
adjustable by the NSR_KEEPALIVE_WAIT environment variable. Set this environment
variable to the desired number of seconds between keep-alive wait messages. If the
NSR_KEEPALIVE_WAIT variable is not set or is set to an invalid value, (0, a negative
number, or a nonnumeric string) then no keep-alive message is sent.

Pause recommended between file creation and backup with EMC IP4700
If a level 1 to 9 backup is run on an EMC IP4700 filer within five minutes of creating a file,
more files than expected may be saved. For example, if a level 1 backup is run, followed
by a level 2 backup, and both of these backups complete within five minutes of the file
being created, the newly created file might appear on both the level 1 and level 2
backups, even though the files should only be added to the level 1 backup. To avoid this
problem, wait at least five minutes after creating a file to run a backup.

EMC NetWorker Release 8.1 Release Notes 183

Technical notes

Bus reset can rewind tape on Microsoft Windows

Tape devices shared by more than one computer can experience unpredictable bus resets
from any of the computers. These reset commands can cause a tape on a shared bus
(such as SCSI or Fibre Channel) to rewind. The results can include:
◆ Tapes that are prematurely treated as full.
◆ Corrupted data on tapes.

System configurations that do not properly protect tape devices shared by more than one
computer can experience these bus resets. Some switching hardware can be configured
to protect tape devices from resets. Certain operating systems include built-in protection
(that can be turned on by the user) against stray bus resets.

To determine whether the switch or operating system includes such protection, and to
learn how to use it, refer to the manufacturer’s documentation or contact the
manufacturer.
Note: Whatever solution you select must block the reset command from the tape drives,
but must not block it from certain cluster-controlled disks. The reset is a necessary part of
the disk arbitration process in some cluster environments. To determine whether this
applies in your environment, refer to the cluster documentation.

The NetWorker software does not support configuring a tape device in a shared SCSI or
Fibre Channel environment without using either a hardware switch or an appropriate
operating system solution to handle the bus reset issue.

Microsoft does not support attaching any tape drive to any shared SCSI bus that also
hosts cluster-controlled disks because of this issue.

Older versions of Intel unsupported on NetWorker software on Linux

To take advantage of IA-32 586 and 686 optimizations in the compiler, as well as the new
instructions provided on these architectures, the NetWorker release 7.6 for Linux does not
support older versions of Intel, such as 486.
Note: NetWorker release 6.1.x for Linux fully supports older versions of Intel architecture,
including 386 and 486.

NetWorker features that are unsupported for Windows NT 4.0, Windows 2000, and
Windows Server 2003
These NetWorker features are unsupported on Windows:
◆ Directed recovery of Encrypting File Systems (EFS)
◆ Directed recovery of SYSTEM or VSS SYSTEM save sets
◆ Verified (-V) backup of the EFS
◆ Verified (-V) backup of SYSTEM or VSS SYSTEM save sets
◆ Archiving of SYSTEM or VSS SYSTEM save sets

184 EMC NetWorker Release 8.1 Release Notes

 Technical notes

Considerations when using an advanced file type device

The AFTD device can be deployed in varying environments with local disks, NFS and CIFS
mounted/mapped disks. Operation of this feature is affected by the configuration. Ensure
that the AFTD is fully operational in the production environment before you deploy it as
part of regularly scheduled operations.

As part of the validation process, include these tests:

◆ Backup
◆ Recover
◆ Staging
◆ Cloning
◆ Maximum file-size compatibility between the operating system and a disk device
◆ Device behavior when the disk is full

Some versions of NFS or CIFS drop data blocks when a file system becomes full. Use
versions of NFS, CIFS, and operating systems that fully interoperate and handle a full file
system in a robust manner.

On some disk devices, the volume labelling process may take longer than expected. This
extended labelling time depends on the type of disk device being used and does not
indicate a limitation of the NetWorker software.

The upper limits of save set size depend on the upper limits supported by the operating
system or the file size specified by the disk device vendor.

Configuration options to send write error alert if NFS mount of AFTD device
fails silently while writing to AFTD
It is possible for the NFS mount of an AFTD device to fail silently during large writes to the
AFTD, which can result in write operations in progress continuing on the underlying mount
point in the local file system. To prevent this from occurring, use one of the following
configurations which will send a write error at the point in time when the NFS mount fails:
◆ Ensure the local mount point permissions do not have local write permissions.
◆ Use the overlay mount option for operating systems that support this option (for
example, -O on Solaris), so the local file system mount point is considered read-only.
◆ Create a sub-directory structure for the AFTD below the top-level mounted directory.
This will result in the path becoming invalid if the NFS mount fails, producing a write
error.

Disable a driver before configuring an a media library if the driver was automatically
installed on Microsoft Windows 2000 or Server 2003
On a Windows 2000 or Server 2003 operating system, if a driver is automatically installed
for a media library, first disable the driver before configuring a media library using the
NetWorker software.

This does not apply to media library configured as RSM autochangers. Media libraries
configured as RSM works with media library driver enabled.

EMC NetWorker Release 8.1 Release Notes 185

Technical notes

Enabling NetWare 4.22 clients on Windows

Support for a NetWare client is controlled by the environment variable
NSR_SUPPORT_NetWare_4X. By default, the supported is disabled. To enable support for
NetWare clients set the server system environment variable NSR_SUPPORT_NetWare_4X
to 1:
1. Open Control Panel and select System.
2. In System Properties attribute, select the Advanced tab.
3. Select the Environment Variables option. A new window will appear called System
Variables.
4. Click New.
a. Type NSR_SUPPORT_NetWare_4X for the Variable Name.
b. Type 1 for the Variable Value.
5. Reboot the computer.
6. Open a command prompt window and type this command:
set NSR_SUPPORT_NetWare_4X=1

When NetWare support is enabled and a NetWare client establishes connection with the
nsrd process, a message indicating “Support for NetWare clients is enabled” appears in
the daemon log file
IMPORTANT
Authentication methods in the NetWorker 7.3.2 and later server have been changed to
provide greater security. The 4.x client is no longer fully compatible with this version of the
server. Upgrade to the 7.2.1 NetWorker client for NetWare. For NetWare 5.1 systems, apply
LGTpa86701. Netware 5.0 and older systems are not compatible with the NetWorker 7.2.1
client for NetWare.

Manual backups from the client interface using the NetWare Console can be successfully
completed. Note that manual backups will only use media from the default pool.

Adjusting client parallelism to decrease VSS backup failures on Microsoft Windows

If VSS is licensed and enabled and timeout failures occur when backing up save sets, try
decreasing the value of the client Parallelism setting.

During a VSS backup, a snapshot is taken of each specified save set. The client
Parallelism setting determines how many snapshots are taken concurrently. Since
snapshots consume system resources, the more snapshots that are taken concurrently,
the more likely it is that a snapshot will fail.

After a failed attempt, NetWorker software waits ten minutes before attempting another
snapshot, as recommended by Microsoft. After three unsuccessful snapshot attempts,
the backup of the save set fails.

When backing up a large number of save sets, decreasing the value of the client
Parallelism helps to ensure successful snapshots.

186 EMC NetWorker Release 8.1 Release Notes

 Technical notes

Using the NSR_NDMP_RESTORE_LIMIT environment variable on Microsoft Windows

The NSR_NDMP_RESTORE_LIMIT environment variable is used to limit memory
consumption during recoveries involving a large number of index entries (millions). This is
specifically useful if there is not enough swap space or memory for the number of index
entries selected for the recovery. If the variable is not set, the recovery might fail with an
“out of memory” error message.

To avoid a failure:
1. In the NetWorker startup script, set the NSR_NDMP_RESTORE_LIMIT environment
variable to an appropriate value.
2. The value of the NSR_NDMP_RESTORE_LIMIT environment variable determines the
maximum number of entries that the recover program can allocate memory to. For
example, if the total number of entries is 3 million, then the
NSR_NDMP_RESTORE_LIMIT can be set to 50,000 or 1,000,000, but less than
3,000,000.
3. In the NetWorker Administration window, select Recover.
4. Select the files to recover, and click OK.

The recovery is divided into multiple recoveries, where each has the
NSR_NDMP_RESTORE_LIMIT entries successfully recovered in their respective sessions.

Manually removing data left behind from a partial uninstall of NetWorker software on
Microsoft Windows
When performing a partial uninstall of the NetWorker software by using the Add/Remove
Programs option in the Windows Control Panel, if Change is selected, certain folders, files,
and registry keys remain on the system. This occurs when the Remove NetWorker
Metadata checkbox was left unselected.

To remove this data:

1. Open Windows Explorer, and delete %SystemDrive%\Program Files\EMC
NetWorker\nsr.
2. Open the Windows Registry Editor, and delete
<\\HKEY_CURRENT_USER\Software\EMC NetWorker>.

If Remove is selected, the checkbox will not appear and a partial uninstall is performed.

Client retries setting with VMware Consolidated Backups

When using NetWorker software with VMware Consolidated Backups (VCB), the Client
Retries attribute for Group resources should be set to 0. If Client Retries is set to any value
higher than 0 and the backup fails, the NetWorker software will retry the backup
regardless of the state of the VCB backup itself. This can result in the VMware utility
generating errors such as "mount point already exists" or "backup snapshot already
exists."

EMC NetWorker Release 8.1 Release Notes 187

Documentation

Change to save set expiration time

When a save set reaches its scheduled browse or retention policy deadline (the day when
the save set expires or becomes no longer browsable), the time of expiration on that day
is 23:59:59 (11:59:59 P.M.). For NetWorker releases 7.2.x and earlier, the time of
expiration for the save set matches the time of day the save set was originally created.

For example, if a save set was backed up on May 1, 2005 at 1:00 P.M. and the browse or
retention policy is set to 5 years, with NetWorker releases 7.2.x and earlier, the save set
would have expired on May 1, 2010 at 1:00 P.M. With NetWorker releases 7.3.x and later,
the save set will expire on May 1, 2010 at 11:59:59 P.M.

Documentation
This section describes the documentation for the following NetWorker releases:
◆ “NetWorker product documentation” on page 188
◆ “NetWorker related documentation” on page 189

NetWorker product documentation

This section describes the additional documentation and information products that are
available with NetWorker.

EMC NetWorker Installation Guide

Provides instructions for installing or updating the NetWorker software for clients, console
and server on all supported platforms.

EMC NetWorker Cluster Integration Guide

Contains information related to integration of the NetWorker software on cluster server
and clients.

EMC NetWorker Administration Guide

Describes how configure and maintain the NetWorker software.

EMC NetWorker Release Notes

Contains information on new features and changes, fixed problems, known limitations,
environment and system requirements for the latest NetWorker software release.

EMC NetWorker Licensing Guide

Provides information about licensing NetWorker products and features.

NetWorker Error Message Guide

Provides information on common NetWorker error messages.

NetWorker Performance Optimization Planning Guide

Contains basic performance planning, tuning, and optimization information for NetWorker
environments.

188 EMC NetWorker Release 8.1 Release Notes

 Documentation

NetWorker Command Reference Guide

Provides reference information for NetWorker commands and options.

NetWorker Management Console Online Help

Describes the day-to-day administration tasks performed in the NetWorker Management
Console and the NetWorker Administration window. To view Help, click Help in the main
menu.

NetWorker Avamar Integration Guide

Provides planning and configuration information on the use of Avamar deduplication
nodes within an integrated NetWorker backup and storage management environment.

NetWorker Cloning Integration Guide

Provides planning, practices, and configuration information for using the NetWorker,
NMM, and NMDA cloning feature

NetWorker Data Domain Deduplication Devices Integration Guide

Provides planning and configuration information on the use of Data Domain devices for
data deduplication backup and storage in a NetWorker environment.

NetWorker Snapshot Management Integration Guide

Provides the ability to catalog and manage snapshot copies of production data that are
created by using mirror technologies on EMC storage arrays.

NetWorker VMware Integration Guide

Provides planning and configuration information on the use of VMware within a
NetWorker backup and storage management environment.

NetWorker User Online Help

The NetWorker User program is the Windows client interface. Describes how to use the
NetWorker User program which is the Windows client interface connect to a NetWorker
server to back up, recover, archive, and retrieve files over a network.

NetWorker related documentation

For more information about NetWorker software, refer to this documentation:

EMC NetWorker Software Compatibility Guide

A list of supported client, server, and storage node operating systems for the following
software products: AlphaStor, ArchiveXtender, DiskXtender for Unix/Linux, DiskXtender
for Windows, Backup Advisor, AutoStart, AutoStart SE, RepliStor, NetWorker, and
NetWorker Modules and Options.

NetWorker Procedure Generator

The NetWorker Procedure Generator (NPG) is a stand-alone Windows application used to
generate precise user driven steps for tasks such as installation, upgrading, migrating,
and disaster recovery, carried out by customers, Support and the field. With the NPG,

EMC NetWorker Release 8.1 Release Notes 189

Installation

each procedure is tailored and generated based on user-selectable prompts. This

generated procedure gathers the most critical parts of NetWorker product guides and
combines experts' advice into a single document with a standardized format.

Technical Notes and White Papers

Provides an in-depth technical perspective of a product or products as applied to critical
business issues or requirements. Technical Notes and White paper types include
technology and business considerations, applied technologies, detailed reviews, and
best practices planning.

Documentation corrections and additions

The following section provides information on updates to the core NetWorker
documentation set that currently do not appear in those guides.

NetWorker Cluster Installation Guide

The NetWorker 8.0 SP1 Cluster Installation Guide currently provides information on
configuring a highly available HA) NMC server. However, support for a highly available
NMC server is reviewed by request only. Please contact EMC sales to initiate a Request for
Product Qualification (RPQ).

Installation
This section provides information on installing and updating the NetWorker software.
More information on installation and updating procedures is provided in the EMC
NetWorker Release 8.1 Installation Guide:
◆ “NetWorker client location and space requirements” on page 190
◆ “VMware tools required when performing backups on VM” on page 192
◆ “Update enablers” on page 192
◆ “Java not included with the NetWorker software” on page 192
◆ “Installing the NetWorker client in a failover zone” on page 192
◆ “Reconfiguring Console authentication for LDAP if NMC fails to start after update” on
page 193

NetWorker client location and space requirements

This section specifies the location and space requirements for the NetWorker client
software for NetWorker release 8.1

190 EMC NetWorker Release 8.1 Release Notes

 Installation

Microsoft Windows
Table 18 on page 191 specifies the location and space requirements for the NetWorker
client software in a Microsoft Windows environment.

Table 18 Microsoft Windows location and space requirements

Space

NetWorker files Location x86 x64 ia64

NetWorker client files and EMC NetWorker\nsr\ 110 MB 110 MB 110 MB

NetWorker catalog

EMC HomeBase Agent EMC NetWorker\ 400 MB 400 MB Not

binary HomebBaseAgent\ applicable

Temporary space required %TEMP% 400 MB 400 MB Not

for EMC HomeBase Agent applicable

Linux
Table 19 on page 191 specifies the location and space requirements for the NetWorker
client software in a Linux environment.

Table 19 Linux location and space requirements

Space

NetWorker files Location x86 x64 ia64

NetWorker client files and /usr/bin 110 MB 110 MB 110 MB

NetWorker catalog /usr/sbin
/usr/lib/nsr
/opt/nsr

EMC HomeBase Agent binary /opt/homebase-agent/ 400 MB 400 MB not

applicabl
e

Temporary space required for /tmp 400 MB 400 MB not

EMC HomeBase Agent applicabl
e

Solaris location and space requirements

Table 20 on page 191 specifies the location and space requirements for the NetWorker
client software in a Solaris environment.

Table 20 Solaris location and space requirements (page 1 of 2)

Space

NetWorker files Location Sparc x86 amd64

NetWorker client files and /usr/bin 200 MB 200 MB 200 MB

NetWorker catalog /usr/sbin
/usr/lib/nsr
/opt/nsr

EMC NetWorker Release 8.1 Release Notes 191

Installation

Table 20 Solaris location and space requirements (page 2 of 2)

Space

NetWorker files Location Sparc x86 amd64

EMC HomeBase Agent binary /opt/homebase-agent/ 440 MB not not

applicabl applicabl
e e

Temporary space required for /tmp 440 MB not not

EMC HomeBase Agent applicabl applicabl
e e

VMware tools required when performing backups on VM

When performing backup operations on a VM, VMware tools must be installed in order for
the backups to complete.

Update enablers
An update enabler is necessary to move from an existing major NetWorker release to a
newer major NetWorker release, for example, from NetWorker 8.0 to NetWorker 8.1.
Updating within a minor release, for example, from NetWorker 7.6 SP2 to NetWorker 7.6
SP3, does not require an update enabler.

With NetWorker 8.0 and later, if an update enabler is required, the NetWorker software
automatically adds the required update enabler code to its configuration at the
completion of a successful installation. The update enabler expires after 45 days. You
must contact EMC Licensing within 45 days to permanently authorize the update enabler.
The NetWorker server generates an alert and displays the alert in NMC (and in nsrwatch)
to notify that this enabler needs to be authorized.

If the auth (authorization) code for the update enabler code is not applied within 45 days,
the NetWorker server software will be disabled. Typing the auth code enables the software
even if the update enabler code has expired.
Note: If the auth (authorization) code for the update enabler code is not applied within 45
days, the NetWorker server software will be disabled. Typing the auth code enables the
software even if the update enabler code has expired.

Java not included with the NetWorker software

Java is not included with the NetWorker install. When installing the Console server
software, a minimum JRE version of 1.6 is required, depending on the operating system.
Java is required to run the Console GUI, but is optional for the NMC server.

If you do not have the required Java version installed, go to the Java website to download
and install the appropriate JRE version. The NMC console cannot be started until the
appropriate JRE version is installed.

Installing the NetWorker client in a failover zone

The NetWorker software supports installation of the NetWorker client in a failover zone.
More information on failover zone setup is provided in the Sun Cluster Data Service for
Solaris Containers Guide, available at:

192 EMC NetWorker Release 8.1 Release Notes

 Troubleshooting and getting help

http://dlc.sun.com/pdf/819-3069/819-3069.pdf
Note: A NetWorker server or storage node cannot be installed in a failover zone. Also, an
active/active configuration is not supported.

To make the NetWorker client highly available, use the following steps:
1. After the zone has been set up as a failover zone with the SUNWsczone tool sczbt,
install the NetWorker Client.
2. When setting up the sczsh_config configuration file, ensure the following parameters
are specified with these values:
"ServiceStartCommand="/etc/init.d/networker start"
"ServiceStopCommand="/usr/sbin/nsr_shutdown"
"ServiceProbeCommand="/usr/sbin/nsrzoneprobe"
3. Run sczsh_register to register.

Note: You must have Solaris 10 or later installed to use any zone functionality, and the
SUN cluster must be installed in order to use the failover functionality.

Reconfiguring Console authentication for LDAP if NMC fails to start after update
If a NetWorker 7.5.x Console server is configured to use LDAP for authentication and you
are updating to release 8.0, in some cases NMC may fail to start after the update. If this
occurs, the gstd.raw file in the Console's logs directory contains the message “acm:
External directory library initialization failed”.
Workaround

Reset the Console authentication configuration and reconfigure LDAP by performing the
following steps:
1. Ensure that the gstd service is not running. If gstd is running, stop the service.
2. Go to <NMC install directory>/cst.
3. Delete the files Config.xml, csp.clb, csp.clb.bak and upgrade_cst.tag, if present.
4. Copy Config.xml.template to Config.xml.
5. Start the Console.
The Console starts in native authentication mode.
6. Log in as Console's 'administrator' user with the password that was last set for this
user before switching to LDAP mode.
7. Go to Setup > Configure Login Authentication and configure LDAP again.

Troubleshooting and getting help

EMC support, product, and licensing information can be obtained as follows.

Product information — For documentation, release notes, software updates, or for

information about EMC products, licensing, and service, go to the EMC Online Support
website (registration required) at:

EMC NetWorker Release 8.1 Release Notes 193

Troubleshooting and getting help

https://support.emc.com/

Technical support — For technical support, go to EMC Customer Service on the EMC
Online Support website. To open a service request through the EMC Online Support
website, you must have a valid support agreement. Please contact your EMC sales
representative for details about obtaining a valid support agreement or to answer any
questions about your account.

Online communities — Visit EMC Community Network at https://community.EMC.com for

peer contacts, conversations, and content on product support and solutions. Interactively
engage onlinewith customers, partners, and certified professionals for all EMC products.

Copyright © 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.

For the most up-to-date regulatory document for your product line, go to the EMC Online Support website.

For the most up-to-date listing of EMC product names, see EMC Corporation Trademarks on EMC.com.

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

194 EMC NetWorker Release 8.1 Release Notes