[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [HTCondor-users] Job stays in idle condition

There are several possibilities.  First we should figure out of the jobs are really stuck in idle, or if they are actually

rapidly cycling between running and idle.


Try running this command


condor_q -af:h  Clusterid NumJobStarts NumShadowStarts


Are the number of job starts and shadow starts really 0?


If the numbers are 0, Then try


condor_q -analyze <jobid>


where <jobid> is the id of one of the jobs that is stuck in idle  (and has 0 for the numbers)


If the numbers are not 0 not, then you should look in the ShadowLog on the submit machine to see  why the jobs are failing so quickly.




From: HTCondor-users <htcondor-users-bounces@xxxxxxxxxxx> On Behalf Of Nikolaos Tasios
Sent: Saturday, April 27, 2019 5:21 AM
To: htcondor-users@xxxxxxxxxxx
Subject: [HTCondor-users] Job stays in idle condition



 I have the following issue with the HTCondor system.

 With condor_status, i can see all the machines that are available for Condor system.  The problem arises when in the configuration file with the requirements command, i assign a specific job in a specific machine observable with condor_status.  The job stays in "idle" condition for ever.

 Do you have a clue what the problem might be?


Thank you a lot,