How to resolve the Issues of nsrjobd

100x,

I am not quite sure what you mean here? The hot fix can be provided to you by EMC Support for installation. It is included in both 7.4.4 and 7.5 builds (as it was added to the RTM (Release to Market software).

As for jobsdb you should look at the size and retention variables within the NetWorker server properties. The size limit is a threshold, not a hard limit used for the housekeeping so you could find that the jobsdb is larger than this size and this is normal (but not preferred for housekeeping). The retention is the number of days we keep data within the database before deleting. This data is used by NMC to gather the statistics/reports etc so in normal running you should not need to retain for a long time. It is considered best practice to have the retention time set to 3 days (default 7) as this allows you a weekend for there to be an issue with your NMC server (to fix so it gathers the info from the jobsdb) without loosing any statistical data for reports.

It is important that the housekeeping is done via time rather than space as time based housekeeping is much more efficient. You can check what type you are doing by searching the daemon.log file for "purge" - the message will tell you if it is time or size being used as the trigger. If it is size you can increase the size parameter, so time based housekeeping is used.

I hope this is of use.

Regards,

Bill Mason

100x,

This was a known issue at 7.3.3 version of the software which was resolved with a hot fix. The hot fix should be present in 7.3.4 and 7.4.1 (and higher versions of NetWorker). If you are running 7.3.3 then you need to get a newer version of the software and upgrade (7.3.x is no longer supported therefore you should already be on a higher version unless you have paid for extended support). If this is your situation you can contact support who will issue the hot fix to you (via Support call), please quote the fix reference LGTsc09339 along with the OS version you are running when raising the call.

If you are already on a higher version it may be a different issue therefore the best way to resolve is to raise a support call with EMC to troubleshoot and resolve.

Regards,

Bill Mason

My current version is 7.4.2.Build.431,Still have this problem.

Regards.

I would suggest to upgrade to 7.4 SP5 or better 7.5 Sp2 first for improvements to nsrjobd is an ongoing process.

If this doesn't help I also would follow Bill Mason's recommendation to open a support case.

Just a hint: if you see savegroup running but no active sessions from the clients just try to increase the client parallelism of the affected clients (including client resource of NW server) by at least 4 or 5. If then sessions are started you face an issue with nsrjobd.

An other thing is the size of jobsdb: I would suggest to size this at least to 40MB otherwise nsrjobd may be purging old jobs for the size limit is exceeded anf that process is not very fast and may block otehr operations.

hi,guenterH;

How to Increase  the size of jobsdb.

Log shows

:17997 04/26/10 08:43:37  2 0 0 3029334944 10837 0 mdbackup nsrjobd jobsdb size at 6001568 exceeded high size watermark.  It needs to clean up old records

100x,

The size and retention of data for the jobsdb is within the properties of the NetWorker server (right click Networker server, select Properties and then look on the configuration tab). Within operations you will see this. below is a post I made a few days ago for another query which will also give more explanation. The default size is 40MB - you can increase this based on the housekeeping you are doing currently - we want to see time based housekeeping as it is more efficient.

As for jobsdb you should look at the size and retention variables within the NetWorker server properties. The size limit is a threshold, not a hard limit used for the housekeeping so you could find that the jobsdb is larger than this size and this is normal (but not preferred for housekeeping). The retention is the number of days we keep data within the database before deleting. This data is used by NMC to gather the statistics/reports etc so in normal running you should not need to retain for a long time. It is considered best practice to have the retention time set to 3 days (default 7) as this allows you a weekend for there to be an issue with your NMC server (to fix so it gathers the info from the jobsdb) without loosing any statistical data for reports.

It is important that the housekeeping is done via time rather than space as time based housekeeping is much more efficient. You can check what type you are doing by searching the daemon.log file for "purge" - the message will tell you if it is time or size being used as the trigger. If it is size you can increase the size parameter, so time based housekeeping is used.

I hope this is of use.

Regards,

Bill Mason