I have had that happen too.

Any chance two people were making changes to policies at the same time?

Or just any  change to that policy that might have hosed it.

And bouncing or deactivate/reactive the policy has fixed it for me before.

I know that nbpem builds a work list of due jobs on the master server.  It builds the list of scheduled jobs from data provided by bpdbm.  When a job is due nbpem sends a job try for each due job to the nbjm process.  nbjm notifies bpjobd to add the job to the jobs database, and the job appears in the activity monitor.

The unified logs for nbpem show when jobs finish .. the unified log for nbjm shows details during the job duration ... which logs captures the initialization of a job that is due to run?

The other day I noticed that scheduled jobs were not being kicked off.  I ended up having to recycle netbackup to get things working again.  I probably could of just restart nbpem, but thats besides the point.  I saw nbpem logging data through the time frame that my jobs were not starting... but was there still a part of nbpem that was hung up?  .. I'm trying to determine when my problem began .. and how to detect when it happens in the future.

Thanks,