[HTCondor-users] Cancelling condor_off?

[Carsten Aulbert <carsten.aulbert@xxxxxxxxxx>]

Hi,

anticipating cooling constraints, I issued `condor_off -peaceful 
-startd` an a bunch of nodes a couple of days back.

Now, the cooling constraints were not as severe as feared and the nodes 
were left running but the startd subslots are obviously still in the 
state 'Claimed/Retiring'.

Is there a way to cancel condor_off or do I have to wait for all jobs to 
finish/the nodes to drain and issue `condor_on -startd` individually 
when done?

The problem with that approach is, that some users' jobs are running for 
a long time, sometimes even weeks and short of running condor_on via 
cron or monitoring condor_status when a node drops off it, we may simply 
miss the time to issue the command.

Cheers and thanks a lot in advance

Carsten
--
Dr. Carsten Aulbert, Max Planck Institute for Gravitational Physics,
CallinstraÃe 38, 30167 Hannover, Germany, Phone +49 511 762 17185

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature