[torqueusers] [Moabusers] how to use GEVENTCFG to offline nodes (fwd)

Eva Hocks hocks at sdsc.edu
Wed Mar 5 16:51:09 MST 2014

sorry for cross posting, the moab user list seems to be dead :(

I am looking for an example for moab to automatically disable
nodes which show errors like

pbs_server: LOG_ERROR::Batch protocol error (15033)

What I am missing in the moab documentation is when using node
categorization to record, track, and report on per node
and cluster level status

GEVENTCFG[BatchFailure] action=notify,record,disable,reserve rearm=01:00:00

is there anything else I have to configure to achive the desired action,
that is to offline failing nodes?

Thanks for any advise

More information about the torqueusers mailing list