Veritas NetBackup (tm) PureDisk Remote Office Edition 6 Patch 02 - Symantec Corporation - April 7, 2006


Veritas NetBackup (tm) PureDisk Remote Office Edition 6 Patch 02 - Symantec Corporation - April 7, 2006


Veritas NetBackup (tm) PureDisk Remote Office Edition 6 Patch 02.  This patch is cumulative and replaces Patch 01 (NB_PDE_60_P01) .

Patch NB_PDE_60_P02_282729 Installation Instructions

Symantec Corporation
April 7, 2006

Table of Contents
1 Patch description
2 Installing the patch
3 Problems fixed by this patch
4 Encrypting topology.ini

1 Patch description
Symantec recommends that you install the PureDisk patch NB_PDE_60_P02_282729 to address a number of issues that were identified during extensive tests with PureDisk.  Patch NB_PDE_60_P02_282729 is a cumulative patch. Therefore, you can install this patch without installing the previous patch NB_PDE_60_P01_282217 first.

Apply this patch to the storage pool authority (SPA) node only. The patch is automatically distributed and installed on all other nodes in the storage pool. Apply the patch after you complete the installation of your PureDisk environment.  Do not apply the patch before the installation of your PureDisk environment is completely finished.  That is, do not apply the patch before has completed successfully. When you add additional PureDisk nodes to a patched storage pool, the patch is pushed automatically to the new nodes.

During installation of the patch on an already deployed system, some services are stopped and restarted. This can cause running jobs to fail. Symantec recommends that you keep system activity to a minimum during the patch installation, for example, by disabling backup policies, and so on.

The patch creates the following files during the installation:

On the SPA:
/Storage/log/patches/NB_PDE_60_P02_<timestamp>.log: Log file of the installation.
/etc/puredisk/NB_PDE_60_P02_<timestamp>: An empty file created upon successful completion of the installation. This marker file enables you to verify that the SPA node is patched successfully.

On all other nodes:
/Storage/log/patches/backup-patch-<timestamp>.tgz: Backup of the files replaced by the patch.
/Storage/log/patches/patch-<timestamp>.log: Log file of the installation.
/etc/puredisk/patch-<timestamp>: An empty file created upon successful completion of the installation. This marker file enables you to verify that the node is patched successfully.

When the patch has been applied, the next run of your replication policies will perform a full replication. This is necessary to ensure the consistency of the replicated data selections.

2 Installing the patch
Download the patch using the download link.  Use rcp or scp to copy the downloaded file NB_PDE_60_P02_282729.tgz to the /root directory of the PureDisk node that is hosting the SPA. Install the patch as follows:

1. Log on to the node hosting the SPA as user root.
2. Type cd /root to go to the /root directory.
3. Verify the integrity of the patch:
Type md5sum /root/NB_PDE_60_P02_282729.tgz to compute the md5 checksum of the patch file.
The md5 checksum of the patch file must match the following:   48d7b43effd8ef4c22bbaefbebef0bcf
            If you obtain a different checksum, the patch was corrupted during download. Try to download the patch again.

4. To unpack the patch, type:
tar -C / -xzf /root/NB_PDE_60_P02_282729.tgz ./opt
5. Verify that your topology.ini file is correct and not encrypted.  
6. Type: /opt/pdinstall/   to run and install the patch. The patch is pushed automatically to all nodes in the storage pool.
7. After successful installation of the patch, you must update all client agents in the storage pool. Use the Update agents button in the Data management tab of the PureDisk Web user interface (UI) to update the agents.  Refer to the Administrator's Guide for more information about this topic.  A link to the Administrator's Guide can be found at the bottom of the TechNote, found below, in the Related Documents section.

3 Problems fixed by this patch
This patch fixes the following problems:

1. MBimport step in replication policies can fail due to memory exhaustion.
2. File backup errors are reported in the Error tab of the Job details window at the first backup but not at consecutive backups.
3. Unable to install Linux PureDisk agent with an install user (= user with install permission and member of Users user group).
4. Checking Volume Shadow Copy Service (VSS) snapshots during ScanFileSystem step in backup workflow fails on foreign-language Windows systems. As a consequence, backups using VSS fail on these systems.
5. Job queue is blocked if an agent cannot retrieve the next step in a job within 30 seconds. When this happens, the SPA assumes that the agent is working on that job step, but in reality, it is still waiting for that step.
6. Metabase engine can run out of memory during replication jobs.
7. Selecting a replicated data selection in the data selections tab of a policy leads to all data selections being selected.
8. Selection of Month(s) option in Metadata filter for data removal policies causes error.
9. Agent name, department name, and other fields missing in CSV reports retrieved over Web services.
10. Controller terminates without logging a message.
11. Internal system users are visible in the Permissions window.
12. Server agents wait a long time before checking for a new job step.
13. Permissions not applied to job details.
14. Default timeout on Backup Start step in backup workflow too large.
15. Encryption / decryption script and the log gathering script are missing in the install directory.
16. Watchdog cannot not kill MBimport step.
17. Find files fails on large data selections.
18. Content router can generate "tlog processing out of sync" error under heavy load.
19. Removal policies with more than one filter rule not working properly.
20. Problem with processing batches during replication jobs.
21. Workflow engine consumes too much memory.
22. Rebooting an agent while a job is running causes the job to stay in the running state until the watchdog kills it.
23. Time-out for MBimport step in replication jobs needs to be increased.
24. When a backup job is killed gracefully during the plscanfs (file system scan) step, the next backup reports success while in reality it did not backup anything. This leads to data loss.
25. When agent encounters problems, it reports a fatal error instead of a readable error message.
26. Replication jobs can skip batches of data, thus can lead to data loss.
27. Replication jobs are sometimes reported as successful, while in reality the replication failed.

The patch also updates the online help.  The online help file appears when you click the Help tab in the PureDisk Web UI.

4 Encrypting topology.ini
For security reasons, Symantec recommends that you encrypt the topology.ini file that is in the /opt/pdinstall directory of each PureDisk node. The patch installs an encryption/decryption script that can be used for this purpose. To encrypt the file, go to the /opt/pdinstall directory and type:

./ -e

The script prompts you for a password to protect the file.

To decrypt the file, type:

./ -d

Encrypt the file after you complete the installation of a PureDisk node. When you change the topology of the storage pool (for example, when you add a content router or metabase engine), you must first decrypt the topology.ini file before proceeding.  Encrypt the file again after completing the update.

Terms of use for this information are found in Legal Notices.



Did this article answer your question or resolve your issue?


Did this article save you the trouble of contacting technical support?


How can we make this article more helpful?

Email Address (Optional)