we have NetBackup 6.5.4 on Solaris SPARC, OpsCenter Analytics 7.1.0 on RedHat on VmWare and OpsCenter Agent 7.1.0 on Solaris SPARC, which is also a media server.
In the Data Colection Status on OpsCenter web console I see, that the Subjobs data type fails with status Client timeout reached. In processes on Opscenter server I see, that it repeatedly quickly runs the bplist command on about 10 clients (always the same clients of total of over 300).
I increased the nbu.scl.agent.jobProxyTimeout in scl.conf, but it did not solve the problem.
The next problem is, that the Audit Data type collection was never started.
Does anybody has an idea, what could be wrong?
Audit Trail Collection is new to OpsCenter 7.1 and Requires NetBackup to be at 7.1 or higher.
Audit Trails will not collect on NBU 6.x Masters.
Regarding your SubJob Data Collection Timing out... are you still experiencing this issue?
I would recommend opening a case if your are still having issues with subjobs as log analysis maybe required to troubleshoot that issue.
However, do you notice anything in the agent logs on the Media Server?
You may be able to verify if problem relates to timeouts or if there is another exception / failure occurring during data collection.
In the Agent logs I found this messages:
ProcessErrorConsumer bplistEXIT STATUS 227: no entity was found,1,(0|)
I used the /usr/ucb/ps to get full command with arguments:
/usr/openv/netbackup/bin/bplist -B -S mastername -C clientname -k policyname -R -X -s 1321315200
This does not work neither from commandline, unless I append a "/" for filename as last argument.
Please confirm which version of OpsCenter is running.
Also, confirm any EEB information
Any plans on upgrading OpsCenter / Server & Agent to 22.214.171.124 ?
We plan to upgrade NetBackup and OpsCenter to the latest version next year, but if there is OpsCenter 126.96.36.199, I will try to upgrade it immediatelly.
Here is version we have:
HARDWARE SunOS sparc
VERSION OpsCenter Agent 7.1
BUILD NUMBER 20110203
RELEASE DATE Fri Feb 04 08:23:30 IST 2011
HARDWARE LinuxR_x86 x86_64
VERSION OpsCenter Server 7.1
BUILD NUMBER 20110203
RELEASE DATE Fri Feb 04 08:25:07 IST 2011
There is no EEB.
I upgraded OpsCenter Server and Agent to version 188.8.131.52 but the problem remained.
In the processes on Agent I see that the bplist command has now more parameters. It is strange, that start (-s) and end (-e) time are the same:
/usr/openv/netbackup/bin/bplist -B -S master -C clientname -k policyname -R -X -s 1315630823 -e 1315630823 -l -b -Listseconds -unix_files -t 0 "/*"
I wonder if I am the only one user with this problem.
SPARC Agent Package 7102:
OpsCtr_Agent_184.108.40.206_solaris.tar provides fixes OpsCenter/OpsCenter Analytics 7.1 Agent on Solaris SPARC.
Linux Server Packages 7102:
OpsCtr_Server_220.127.116.11_linuxR_x86.tar provides fixes for OpsCenter/OpsCenter Analytics 7.1 Server RedHat Enterprise Linux.
ViewBuilder Update: (if used)
OpsCtr_ViewBuilder_18.104.22.168_winnt.x86.zip provides fixes for OpsCenter/OpsCenter ViewBuilder 7.1 on Windows.
I have never seen that behavior before, but to be honest, I've never looked that granular at the command issued...
Perhaps there is some request hung on the master...
Check your /usr/openv/var directory for left over nbproxy_*.ior or ior.mgr files.
Look for files with very old dates on them... perhaps even dates since your last NBU cycle.
If you see these, you will want to clean them out.