GENERAL ERROR: Seeing poor performance with NetBackup PureDisk and some jobs are failing to go active.

Article:TECH52578  |  Created: 2007-01-09  |  Updated: 2007-01-12  |  Article URL http://www.symantec.com/docs/TECH52578
Article Type
Technical Solution

Product(s)

Environment

Issue



GENERAL ERROR: Seeing poor performance with NetBackup PureDisk and some jobs are failing to go active.

Solution



Overview:
An intermittent problem has been seen where the performance degrades or jobs fail to start on the NetBackup PureDisk (tm) Storage Pool Authority (SPA).  This can be caused when the pdctrl process on the SPA becomes unresponsive.

Troubleshooting:
Run the PureDisk sessions report on the SPA to see if the pdctrl process is responding.  The following example shows the pdmanage command cannot connect.
# cd /opt/pdctrl/bin
# ./pdmanage --sessions
Mon Jul  9 2007 11:46:14.456965 WARN  (4150314672): SSL operating in untrusted mode.
Mon Jul  9 2007 11:46:14.473796 ERROR (4150314672): Active connection failed.: Connection refused

When the pdmanage -sessions command is working it will list the various Agent IDs that are connected.

Log Files: N/A

Resolution:
The formal resolution to this issue is being planned for the PureDisk 6.1 Maintenance Pack 2 (MP2) and PureDisk 6.2 releases.  These releases will include a pdctrlmon process that will monitor the status of the controller and restart it automatically.

To download the PureDisk 6.1 MP2 release, once it becomes available, visit the PureDisk Support Web site at:
 http://www.symantec.com/enterprise/support/overview.jsp?pid=52672

Workaround:
To work around this problem use the root crontab to restart the pdctrl process on the SPA.  Add the following entry to the bottom of root's crontab on the SPA to restart pdctrl once every hour.
26 * * * * /etc/init.d/puredisk restart pdctrl > /dev/null 2>&1
 







Legacy ID



289875


Article URL http://www.symantec.com/docs/TECH52578


Terms of use for this information are found in Legal Notices