Hotfix NB_PDE_126.96.36.199_347536.tar provides critical fixes and minor enhancements to Symantec Veritas NetBackup (tm) PureDisk 6.6
Date: April 07, 2010
This Hotfix provides features and fixes to the Symantec NetBackup
PureDisk 6.6 software.
* PRODUCT FIXES
* PRODUCT ENHANCEMENTS
* KNOWN ISSUES
* VULNERABILITIES - RESOLVED
* DATALOSS ISSUES - RESOLVED
* PREINSTALLATION STEPS
* INSTALLATION INSTRUCTIONS
* UPGRADING PDDO AGENTS ON MEDIA SERVER CLIENTS
This Hotfix can be installed on PureDisk 6.6, PureDisk 188.8.131.52,
and PureDisk 184.108.40.206.
For detailed information about the PureDisk 220.127.116.11 Hotfix or the
PureDisk 18.104.22.168 Hotfix, refer to the Related Documents section at the
bottom of this article, or visit the following site:
PureDisk automatically creates upgrade jobs for the agent software on
backup and restore clients. The upgrade does not create agent upgrade
jobs for PDDO agents on NetBackup media server clients. To upgrade media
server clients, follow the procedure "Upgrading PDDO client agents on
media servers" in this file.
This patch contains the following product fixes:
* Etrack 1949511: Routing tables used by PDDO replication jobs are not
cleaned up from /Storage/var/rt.
After a PDDO or a PureDisk replication job completes, a large number of
.recommended and .current routing tables from the replication target
storage pool remained in /Storage/var/rt.
To resolve the issue, functions were added so that old routing tables
are removed from /Storage/var/rt after successful conclusion of the
replication job or by a maintenance workflow.
* Etrack 1949513: Performance degradation due to static content router
When processing a large number of objects, performance slows down and
impacts queue processing and rerouting due to static content router
With this patch, the content router cache "bits" setting is
automatically adjusted to improve performance with large numbers of
* Etrack 1949515: Tune CR queue processing
This fix reduces queue processing duration by up to 20% for large
* Etrack 1949517: Minimum operation window for rerouting: rerouting
should be able to run for a defined window without being interrupted by
For performance reasons, queue processing and rerouting do not run
simultaneously. Rerouting is halted whenever queue processing starts.
In the case where queue processing takes more than 12 hours, rerouting
will never have the ability to restart.
This patch introduces a new configuration setting, MinRerouteDuration,
which sets the minimum reroute duration in hours. The default value is
6 hours, and the maximum value is 24 hours. Now after rerouting has run
for the specified duration, rerouting gives precedence to pending queue
processing. This guarantees that rerouting can make progress on a
system that needs more free space for incoming backups even if the
content router is continuously requested to process its queue.
* Etrack 1949504: Rerouting running at 15 MB/s and causing high I/O
wait states on storage pools.
This patch provides improved performance by avoiding disk seek
operations and reading ahead. It also corrects rerouting progress
information in storaged.log.
* Etrack 1949512: Rerouting should not start/resume when CR cache load
is not complete
Rerouting started before the cache reload completed. The results were
slow rerouting and cache reloading not completing for days due to the
high number of queries to the content router database.
This patch resolve this problem by ensuring that the cache load is
completed before rerouting starts.
* Etrack 1952055: 22.214.171.124 upgrade fails when SSL certs are not created
The 126.96.36.199 update will fail to create the SSL certificates needed by
the Postgresql database on non-SPA nodes in multinode systems. This
generates an error and does not allow the database to start.
With this patch, the SSL certificates are created on the system, which
allows the databases to start properly.
* Etrack 1952130: During 188.8.131.52 install: wrong php.ini on remote MBE
when calling actions/doafter/UpdateModesOnPddoDs.php
When running the 184.108.40.206 update, a query for PDDO data selection fails.
The query is part of a php script which needs to be called earlier in
the update process for remote nodes.
This patch contains a correctly sequenced script.
* Etrack 1949710: Avoid pdweb segfault by rearranging php.ini module
The sequence of module loading in php.ini prompted the following
messages in /Storage/log/pdweb-error.log:
... child pid 10635 exit signal Segmentation fault (11)
This fix modifies php.ini so that it loads properly and the error does
* Etrack 1955926: Error Key not found in INI file /etc/pdregistry.cfg,
This error occurred where a content router was not present on the
Storage Pool Authority (SPA) node.
This fix corrects the configuration file path so that this error does
not occur when a content router is not installed on SPA node.
* Etrack 1974222: Very large POLists causing the C: drive to fill-up on
clients after a failed backup
If the backup of a Windows client fails just after all the data has
been stored on the PureDisk server, the next job step (Get Recovery
POList) generates a polist.po file under [install path]\var\srd\, which
fills up the C: drive of the client system.
With this patch, the recovery Polist is renamed before processing to
prevent the situation.
With this patch, Symantec supports increased capacities for PureDisk
storage pools that are configured for the exclusive use as PureDisk
Deduplication Option (PDDO) storage pools. The increased capacity can be
used only if the storage pool is used exclusively by NetBackup as a
deduplicated storage unit.
Capacities for PureDisk Remote Office only configurations, which back up
only files, directories, and databases, remain the same as the previous
In addition, Symantec does not support the capacity increases in this
patch for mixed-use storage pools that are configured for both PureDisk
Remote Office backups (files, directories, and databases) and for the
NetBackup Deduplication Option.
The following table summarizes the content router and storage pool
capacities supported with PureDisk 220.127.116.11.
| Configuration | Content router | Total storage pool |
| | capacity | capacity |
| PureDisk Remote Office | 16 TB per CR | 100 TB |
| Edition only | | |
| PureDisk Remote Office | 16 TB per CR | 100 TB |
| Edition with PDDO | | |
| PDDO only | 32TB per CR | 192 TB |
| | | |
For detailed information about content router and storage pool
capacities, refer to the PureDisk 18.104.22.168 Addendum. The document is
available at the following site:
Note: The capacity information in the PureDisk 22.214.171.124 Addendum
supersedes the capacity information that is stated in the PureDisk
Administrative Web UI Help.
For information about how to increase a content router's storage
capacity, see the VxFS Storage Foundation documentation.
The following are the known issues associated with this patch:
* Etrack 1997927: PDDO data removal stalled blocking incoming PDDO
If a PDDO data removal job stalls due to rerouting, subsequent
PDDO backup jobs stall or cannot start.
Workaround: To allow PDDO backup jobs to start while rerouting is
running, you must manually stop all PDDO data removal jobs and
disable the policies for PDDO data removal.
* Etrack 1987533: Replicating all files between SPAs creates redundant
data selections and clients
Policies that replicate data between two storage pool authorities
(SPAs) are creating redundant data selections and clients when the data
selections are defined at the root level, that is, all files are
included in the data selection.
Workaround: Do not select all files from the root level when defining
data selections for replication policies.
* Etrack 1975237: Win-2k3 R2 PD agent service does not get started
after agent system restart
Service Veritas NetBackup PureDisk Client Agent does not start
automatically when the agent machine is restarted on Windows 2003 R2.
This occurs for an agent upgraded from PureDisk 6.6 to PureDisk 126.96.36.199
and also for a newly added agent on PureDisk 188.8.131.52.
Workaround: Add the IP address of the client agent to
c:\WINDOWS\system32\drivers\etc\hosts. To do this, open the hosts file
in a text editor and follow the instructions in the file from Microsoft
for adding entries.
* Etrack 1974968: Disaster recovery restore fails after the PureDisk
184.108.40.206 patch is applied because a newer version of the Java Runtime
Environment is installed. The error message is as follows:
| Reinstalling Configuration |
+ Checking the content of 'topology.ini'...
+ Performing installation environment sanity check ...
While checking rpm dependencies, the following problems were found:
- Package jre-1.6.0_17-fcs (which is newer than jre-1.6.0_14-fcs)
is already installed.
Workaround: Uninstall the Java Runtime Environment package on all nodes
before running the DR_Restore_all.sh script using the following command:
rpm -e jre-1.6.0_17-fcs
* Etrack 1922902: At the end of a disaster recovery restore of PureDisk
220.127.116.11, the following message appears on screen:
./DR_Restore_all.sh: line 14: 17828 Segmentation fault
Workaround: This message does not affect the restore or system
functionality and can be ignored.
VULNERABILITIES - RESOLVED
No additional vulnerabilities were resolved in PureDisk 18.104.22.168.
DATALOSS ISSUES - RESOLVED
The following data loss issues are resolved in this Hotfix:
* Etrack 1949644: Zero sized SO found in CRDB following rerouting.
If the topology (content router sizes or numbers of content routers) is
adjusted before rerouting has completed successfully, content routers
may reroute segments multiple times. If a segment is rerouted multiple
times, it may be incorrectly replaced by a temporary segment record,
and the associated segment content may be lost.
This patch provides enhanced verification of the segments that are
stored on the content router during rerouting. The destination content
router now properly recognizes and processes temporary segment records.
* Etrack 1960853: Error status code in ABORT message incorrectly
replaced by CR_OK
In some cases, the error status code in ABORT messages were incorrectly
replaced by "OK". This implied that backup data was stored
successfully, even if an error occurred on the content router.
With this patch, the status code is saved before the message containing
the status code is overwritten.
This section describes the preinstallation steps for the Hotfix.
Downloading and extracting the patch
1. Use scp to copy the tar file, NB_PDE_22.214.171.124_347536.tar to the /root
directory of the PureDisk node that hosts the storage pool authority.
2. Log on as root to the node that hosts the storage pool authority.
3. Type the following command to verify the integrity of the
This command computes the md5 checksum of the Hotfix. The md5
checksum of the Hotfix must match 'd20a67f228a9deb0428ad94447425dbf'.
If you obtain a different checksum, the Hotfix was corrupted
during download. Try to download the Hotfix again.
4. Type the following command to extract the README file:
tar -C / -xf /root/NB_PDE_126.96.36.199_347536.tar ./NB_PDE_188.8.131.52.README
This section describes how to install the software.
Windows 2000 users should be aware that Windows 2000 servers must be
rebooted after upgrading their servers using this hotfix. Please plan the
deployment of this package appropriately to eliminate any unscheduled
down time in your environment.
Installing the patch
1. Make sure that no PureDisk jobs are currently running or are
scheduled to be run.
2. Log out from the Web UI.
3. (Conditional) Freeze the PureDisk service groups for the clustered
Perform this step if the storage pool is installed with VCS cluster
software. Use the Cluster Manager Java Console, and freeze all the
For information about how to freeze and unfreeze clustered storage
pools, see the Symantec NetBackup PureDisk Storage Pool Installation
4. Type the following command to unpack the Hotfix software:
tar -C / -xf /root/NB_PDE_184.108.40.206.tar ./opt
5. Type the following command to run and install the Hotfix:
If the topology.ini file is encrypted, the software prompts you for
the password to decrypt this file.
The Hotfix automatically pushes the software to all nodes in
the storage pool and to all clients.
At the end of a successful installation, the software prompts you to
encrypt the topology.ini file.
6. (Conditional) Unfreeze the PureDisk service groups for the
clustered PureDisk server.
Perform this step if the storage pool is installed with VCS cluster
The Veritas Cluster Server (VCS) software might detect some faults
during the upgrade process. If any upgrade actions generate a VCS
fault, use the Cluster Manager Java Console to clear the fault and
probe that resource group before you unfreeze the cluster.
7. Monitor the client agent update jobs
Take the following actions to monitor the client agent upgrade jobs:
* Start the Web UI.
* Click Monitor > Jobs.
* Set the "View jobs by" dropdown to "Policy types".
* Select the "Agent Update" workflow under "Miscellaneous
If one of the upgrade jobs fails, PureDisk deactivates those client
agents. You need to activate these agents again and to select the
client to upgrade to. Follow the steps below:
* Start the Web UI.
* Click Manage > Agents.
* In the left pane, select the deactivated agent you want to
reactivate. Select the storage pool level if you want to activate
all agents in the storage pool.
* In the right pane, select "Activate Agent(s)".
* A new agent update job is scheduled for those agents not
* Monitor these upgrade jobs again. Start them over if they fail
8. (Conditional) Upgrade the PDDO agent software on PDDO media server
Perform the procedure called "UPGRADING PDDO AGENTS ON MEDIA SERVER
CLIENTS", which follows. The release update includes upgraded PDDO
agents for Linux, Solaris, and Windows clients.
UPGRADING PDDO AGENTS ON MEDIA SERVER CLIENTS
This section describes how to upgrade PDDO agents on media server clients.
The upgrade software removes the old PDDO agent software automatically.
For more information about how to install the PDDO agent, see the Symantec
NetBackup PureDisk Deduplication Option Guide.
1. Make sure that the storage pool software for the PDDO clients has
2. Refresh the PureDisk landing page.
The landing page can be found on:
For <your-puredisk-spa>, type the hostname or the IP address of your
PureDisk storage pool authority.
Be aware that this Hotfix includes both the PureDisk 6.6 base
release agents and revised agents for Linux, Solaris, and Windows
clients. The revised agents include updates for PDDO agent
installation only. A later step in this procedure instructs you to
download the revised agents if you want to upgrade the PDDO agent on a
3. Log on to each PDDO client.
4. From the PDDO media server client, download the new PDDO agent
The landing page includes more than one software version of the PDDO
agent, so take care to download the most recent agent.
5. (Conditional) Freeze the PureDisk service groups for the clustered
6. Stop all NetBackup services on the NetBackup media server.
7. Install the agent software that is included with this
* On Windows systems, double-click the PureDisk agent icon. That
starts the Windows Installation Wizard .
* On Linux or Solaris systems, you can use either the attended or
the unattended installation method. The installer prompts you to
confirm the upgrade. For example, type the following command to
upgrade the Solaris 10 agent:
# sh pdagent-Solaris_10_sparc-220.127.116.11475.run
8. (Conditional) Unfreeze the PureDisk service groups for the
clustered PDDO server.
9. Start all the NetBackup services on the NetBackup media server.