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

Re: [HTCondor-users] Output of condor_q after upgrade no longer grouped by batch

That is correct.  The automatic batching of jobs by the executable name across multiple ClusterId values was unpopular with most users and so it was removed.

The batch name generated by condor_q is now based on the ClusterId or DAGManJobId attributes if there is no JobBatchName attribute in the job.

You can set the JobBatchName attribute in your jobs from the name of the executable by adding this to your submit file.

batch_name = $Fn(Executable)

Or you can switch to using a single invocation of condor_submit for jobs that you wish to be batched together, this will
result in them sharing a ClusterId attribute, and so they will be reported by condor_q as a single batch.


-----Original Message-----
From: HTCondor-users <htcondor-users-bounces@xxxxxxxxxxx> On Behalf Of david_johnson@xxxxxxxxx
Sent: Friday, January 22, 2021 1:01 PM
To: htcondor-users@xxxxxxxxxxx
Subject: [HTCondor-users] Output of condor_q after upgrade no longer grouped by batch

Had a working condor cluster running consistently stable since 2018. This week I upgraded submit host, scheduler host and compute hosts from condor-8.6.13-1.1.osg34 to condor-8.8.10-1.1.osg34. 
Since then the output of condor_q is no longer grouped in batches, now one line for each job and an ID: job number in the column for BATCH NAME.  Is there no longer a default batch name based on executable name?   Am I missing something in the config?  I have preserved the existing previous configuration, based on /etc/condor/condor_config and have not converted it to USE: roles. 

  -- ddj
Dave Johnson
HTCondor-users mailing list
To unsubscribe, send a message to htcondor-users-request@xxxxxxxxxxx with a
subject: Unsubscribe
You can also unsubscribe by visiting

The archives can be found at: