Showing results for 
Search instead for 
Did you mean: 

procedure for restarting services of a production NBU


We are lately having issues with nbpem memory leaking on NBU on HP-UX and have to restart NBU services quiet often. Since the NBU is in prodcution services, we are missing some backups here and there. Is there any way we can make the situation better?

Here is the way we usually do when we need to restart our NBU.

1. suspend scheduling #nbpemreq -suspend_scheduling

2. suspend all active/queued jobs #bpdbjobs -suspend type=all

3. stop service #netbackup stop

4. kill all pending jobs #bp.kill_all

5. start services #netbackup start

6. resume suspend jobs #bpdbjobs -resume

is there anything we are missing? thanks!


Not applicable

I would say that would cover everything in respect to stopping services and restart.

I remember having nbpem on HPUX 11.23 core dumping regularly back in the 6.x days. I feel your pain. Nothing seems to have changed.

BUT I would consider these:

Log a support request with symantec to try and work to getting a fix  for the issue. ie EEB


Consider upgrading the master to 7.5 (at least) but working with support to see if they recommend this also, if it is clean from the nbpem issue

You might have also seen this forum post:

But then again there is this EEB also in 7.5.x

We eventually migrated to Linux and be done with HP-UX as our master.

Never looked back.