6.2.2 Release Update
|Article:TECH61860|||||Created: 2008-01-22|||||Updated: 2011-03-25|||||Article URL http://www.symantec.com/docs/TECH61860|
Symantec Veritas NetBackup (tm) PureDisk Remote Office Edition Release Update 6.2.2
Date: July 23, 2008
This release update provides features and fixes to the Veritas NetBackup
PureDisk Remote Office Edition 6.2 software.
WARNING: When you apply this release update, all PureDisk services stop
and start. This can cause running jobs to abort.
Install PureDisk 6.2, version 22.214.171.12462, before you apply
this release update.
You can apply this release update to 6.2, 6.2.1, and 126.96.36.199.
The release update includes the following enhancements:
* New client agent software for the following platforms:
- Windows 32-bit
- Windows 64-bit (x64)
- HP-UX PA-RISC, Itanium
- Solaris SPARC 8, 9, 10
* An additional SNMP trap feature. You can now configure SNMP traps for job
failures. When configured, these new traps report failures for all kinds of
jobs. To enable these new traps, you need to change some configuration file
settings and push out the new changes. The SNMP traps that are sent have a
single-line description that notes the job(step), agent, and data selection
NOTE: This product enhancement does NOT affect or interact with PureDisk's
mechanism for setting escalation actions for all other event types.
Escalation actions forward the content of events and policies and are
entirely separate from this SNMP trap feature. The two features should be
For information about how to enable SNMP traps for job failures, see the
PureDisk Administrator's Guide.
The following additional enhancements also pertain to SNMP traps:
- SNMP messages have now a size of maximum 781 bytes.
- SNMP messages contain not only the 'description' but also 'agentid' and
'source' of the corresponding event.
- Events generated by the content router library are reduced in size.
* When you link PureDisk to an external directory service, the procedure for
configuring communication can now be performed completely from within the
PureDisk Web UI.
(ET1236804, ET1178447, ET1182585)
* Replication now supports compression and encryption.
NOTE - The following PureDisk 6.2.2 features are not available in the PureDisk
* The additional SNMP trap feature for failed jobs.
* The Web UI-based feature for configuring communication for external
directory service authentication.
* The ability to specify compression on replication policies.
* This release update includes the product changes that were included in the
PureDisk 6.2.1 update and in the PureDisk 188.8.131.52 patch. For example, this
release update includes the documentation updates and the Web UI usability and
stability fixes included in PureDisk 6.2.1.
* The release update includes all fixes that were included in the
184.108.40.206 patch, which are as follows:
- Tech Alert 300804 (http://entsupport.symantec.com/docs/300804)
The content router fingerprint cache could become inconsistent if a
'delete segment' task was processed when the fingerprint cache was
reloaded. If, during this inconsistent cache state, the deleted
segment needed to be backed up again, the content router erroneously
assumed that the deleted segment was already protected, so PureDisk
did not store the segment again.
(Internal reference: ET1213743)
- Tech Alert 300805 (http://entsupport.symantec.com/docs/300805)
If the PureDisk metabase engine had a lot of data selections and/or
a lot of changes to the data selections, it was possible for the
pg_class table from Postgres to become incomplete. This situation
could lead to data loss.
(Internal reference: ET1217891)
- Tech Alert 300328 (http://entsupport.symantec.com/docs/300328)
After upgrading PureDisk Remote Office Edition 6.1MP1A or 6.1MP2 to
version 6.2 or 6.2.1, users might encounter problems with
upgraded data removal policies. This release update corrects the incorrectly
upgraded policies, where possible, or warns you when human
intervention is needed.
(Internal reference: ET1218995)
- Tech Alert 300951 (http://entsupport.symantec.com/docs/300951)
A potential for data loss was discovered in data removal. The data
removal policy can delete a file version erroneously if that file was
not changed for a while AND if that file had recently been deleted
from the client.
(Internal reference: ET1224522)
* The release update includes upgrades to the following third-party
- Upgraded PHP to 5.2.6
- Upgraded VRTSweb to 5.0.8
- Upgraded OpenLDAP server and client libraries to 2.3.32-0.23.5
- Upgraded OpenSSL to 0.9.8g (ET 1249068)
- Upgraded PCRE to 7.4 (ET 1176010)
- Upgraded PostGreSQL to 8.2.6 (ET 1196026)
- Upgraded LibXML to 2.6.31 (ET 1204794)
- Upgraded Apache to 2.2.8 (ET 1214045)
- Increased security on Apache and PHP with an update to the configuration
files (ET 1279548 & 1279551 )
- Improved the security settings on some PureDisk server directories
* PDgetlogs is now the only tool used to gather log files from a PureDisk
environment. The functionality of PDgetjob has been integrated and
* This release update includes the following improvements to the PureDisk Web UI:
- When a policy was applied to data selections by consecutively selecting
the 'Edit policy' window in the Workflows tab and the 'Apply policy'
option in the Data Management tab, the database records could become
corrupted and the policy could be applied to the wrong data selections.
- A problem was found when using the 'Edit Policy' task on the Web UI. If
you edited and saved a policy without viewing the 'Data Selection' tab
before saving, PureDisk might save the policy with an incorrect agent
- Fixed Web UI performance problems that were encountered when a lot of
departments were defined.
- Job logs larger than 5MB can now be displayed in the job pop-up window.
- Jobs and policy runs are now removed after running Maintenance policy.
- Job logs that contain Unicode characters no longer display a SOAP Web
service error when displayed.
- The PureDisk Web UI Dashboard and Data Mining reports no longer return
- Fixed the Java exception seen when moving a replicated agent from the
- The "Email address to use for escalation" field now accepts an email
address that contains a dash (firstname.lastname@example.org), an ip-address
(email@example.com), or a short hostname (joe@localhost).
- When you remove all data selections from a policy, PureDisk no longer runs
the policy for all data selections in the storage pool.
* Controller improvements include stability improvements and a fix to prevent
the controller monitor from hanging.
* This release update includes the following improvements to the maintenance tasks:
- A fix to the metabase garbage collection mechanism removes deleted path
objects from the metabase. Undeleted path objects caused unnecessary
metabase size increases.
- This release update improves the method used to perform content router database
maintenance. On content routers processing a large amount of data, the
high volume caused the content router's database to grow rapidly, and the
Postgres vacuum procedure required subsequent increases in the
This release update changes the way maintenance is performed, as follows:
* This release update replaces the content router vacuum step, from the Server
Maintenance workflow, with a step that copies the objects table.
Previously, the workflow vacuumed this table. The new method is much
faster in most cases, and it prevents the content router database from
* This release update corrects the metabase engine vacuum step in the Server
Maintenance policy. Now, the policy also vacuums Postgres internal
system tables, such as pg_class. This prevents corruption of these
(Tech Alert 300805, Internal reference: ET1216521)
- If a Server Maintenance job is running, and the root user types 'ps -ef'
on the console, the database password is not longer displayed.
- The Maintenance workflow has been fixed to expire all jobs and policy
runs from the database. In the previous release, some jobs and policy
runs remained unexpired after a Maintenance workflow run.
* This release update includes the following improvements to the content router:
- In a multinode environment, content routers now have a dynamically
changing, internal, retention mechanism. This enhancement prevents a
possible data loss situation that could occur if one of the content
routers in the storage pool was delayed in processing the incoming data in
- The stpack tool was removed from this release update because of its potential to
contribute to data-loss scenarios.
- For supportability reasons, the dbutil tool now supports hostname based
* This release update includes the following improvements to the client agent:
- Fixed the the Allow Browsing feature, which was broken in release update 6.2.1.
- Fixed a potential endless loop in retry mechanism of PutFiles.
- Fixed the problem, encountered in certain conditions, in which the client
agent leaked network handles when the connection to the storage pool
authority was unavailable.
- Fixed the problem in which the PHP binary keeps running on the
client system after a 'Terminate' command is sent by either the user or by
the workflow engine's internal watchdog. The output in the Agent.log file
is as follows when this problem occurs:
DEBUG: Could not initialize /proc scanner
ERROR: While initializing /proc scanner: invalid data: x,y.
- Fixed the problem in which daylight savings time caused files on FAT32
file systems and Netware shared drives to be backed up even when they had
- The Windows client includes the following fixes and changes:
* The VxFI snapshotting software is upgraded from 4.3 to 4.5.1.
* This release update includes additional logging and enhanced error reporting
* A custom DoBefore Action, vsstest, is added to investigate VSS problems.
* This release update also supports the ability to download log files via
PDmanage. This is a supportability feature.
* This release update includes the following fixes to the NetBackup export engine:
- A fix that ensures that the metadata of character devices passes correctly
and does not cause a segmentation fault in the PHP binary.
- The problem in which the NetBackup export engine always logged at the
debug level, causing the log to grow in size, has been fixed.
- The ability to restart an export policy has been fixed. It is now
possible to restart an export policy for a certain date. Previously, when
an export failed for a certain date it was impossible to run the export at
* When a full disaster recovery backup is run, PureDisk queries the content
router database for changed back-end directories. The time it takes to
complete this query has been reduced.
1. Use rcp or scp to copy the tar file, NB_PDE_6.2.2_<6-digit document ID number>.tar,
to the /root directory of the PureDisk node that hosts the storage pool
2. Log on as root to the node that hosts the storage pool authority.
3. Type cd /root to go to the /root directory.
4. Type the following command to verify the integrity of the release update:
md5sum /root/NB_PDE_6.2.2_<6-digit document ID number>.tar
This command computes the md5 checksum of the release update. The md5
checksum of the release update must match '5caba3da63b3f7c58276abad6a4f9da6'.
If you obtain a different checksum, the release update was corrupted
during download. Try to download the release update again.
5. Type the following command to extract the README file:
tar -C / -xf /root/NB_PDE_6.2.2_<6-digit document ID number>.tar ./README.txt
1. Make sure that no PureDisk jobs are currently running or are scheduled to
2. Log out from the Web UI.
3. Type the following command to unpack the release update software:
tar -C / -xf /root/NB_PDE_6.2.2_<6-digit document ID number>.tar ./opt
4. Type the following command to run and install the release update:
If the topology.ini file is encrypted, the software prompts you for the
password to decrypt this file.
The release update 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.
5. Monitor the client agent update jobs.
Take the following actions to monitor the client agent upgrade jobs:
A. Invoke the Web UI.
B. Click the Workflows tab.
C. Select the Agent update workflow.
If one of the upgrade jobs fails, PureDisk deactivates those client
agents. When you activate those agents again, PureDisk schedules a new
upgrade job. If the clients are connected to the network, PureDisk
schedules the job within 15 minutes. To determine which agents are
deactivated, perform the following steps:
A. Invoke the Web UI.
B. Click the Data management tab.
C. Select the Deactivated client agents by storage pool view, which is
the third view from the left.
D. Re-activate the deactivated Windows agents. PureDisk creates a new
update job if they are not yet on the new version.
Article URL http://www.symantec.com/docs/TECH61860