|
Hi Maria,
The change is actually changing the
HISTORY_HELPER_MAX_HISTORY which is on the Schedd side of things. So, the v24.0.7-1 Schedds would still be using the old 10k limit.
Cheers,
Cole Bollig
From: HTCondor-users <htcondor-users-bounces@xxxxxxxxxxx> on behalf of Maria Alandes Pradillo via HTCondor-users <htcondor-users@xxxxxxxxxxx>
Sent: Wednesday, April 22, 2026 5:03 AM To: HTCondor-Users Mail List <htcondor-users@xxxxxxxxxxx> Cc: Maria Alandes Pradillo <Maria.Alandes.Pradillo@xxxxxxx> Subject: [HTCondor-users] condor_adstash: default history limit Dear all,
We have some condor_adstash nodes running condor 24.12.16 ingesting data to Opensearch. I see [HTCONDOR-3215] Increase default history help limit - Jira was fixed in 24.12.4. However, I still see some lost jobs very likely due to the 10k limit (not 100% sure as condor_adstash logs do not mention anything). Our schedds are running condor 24.0.7-1 and we are wondering whether the limit is hit on the schedd side or on the node where we invoke condor_adstash?
It would be great to have some warning message in the adstash logs in these cases, or more in general, to indicate the total number of jobs ingested per cycle.
Thanks a lot, Maria |