Re: [HTCondor-users] Resource consumption for completed/held jobs?

[Greg Thain <gthain@xxxxxxxxxxx>]

HI Max:

As you thought, held or completed jobs mainly consume memory in the schedd.  If the total memory footprint of the schedd isn't too big for your system, they shouldn't be causing any other slowdowns.  Excessive held or completed jobs in the queue may cause additional traffic for people running condor_q -all.  Personally, I like to get my jobs out of the held state, just because it makes it hard to find my "real" jobs.

-greg

On 2/2/21 4:30 AM, Fischer, Max (SCC) wrote:

Hi all,

we are having a large number of completed/held jobs due to a remote submission tool only fetching few of its pilots, and most jobs only being removed later by our own cleanup rules. The condor_q shows about 90% of jobs as completed or held.
Since this is unexpected usage for us, we are trying to gauge whether this is something to be worried about or not.

Is there any way to estimate how much impact this has on Condor itself? Do the completed/hold jobs basically just take up their ClassAd in memory? Since the jobs are still in the queue, does this affect the performance of the Schedd, Negotiator or other daemons?

Cheers,
Max

_______________________________________________
HTCondor-users mailing list
To unsubscribe, send a message to htcondor-users-request@xxxxxxxxxxx with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/mailman/listinfo/htcondor-users

The archives can be found at:
https://lists.cs.wisc.edu/archive/htcondor-users/