Mailing List ArchivesAuthenticated access |
![[Computer Systems Lab]](http://www.cs.wisc.edu/pics/csl_logo.gif){kind=logo}
|
Mailing List ArchivesAuthenticated access |
|
|
Hi Angel,
I recently stumbled upon the segmentation fault issue recently and made a patch with some other code that is slated for release in V10.3. I did not realize that this was in the stable series of the code so I will make sure to get the bug fix into the stable
series also. Luckily, the segmentation fault occurs at the end of the tools code so you still get all the job information. The reason for the fault is the freeing of constraint expressions for choosing which job ads to output, and specifying a clusterid or
clusterid.procid turns the passed values into a constarint _expression_ under the hood.
Regarding passing the scanlimit flag, it does only apply on the local machine. If you want to increase the amount of history ads queried during a remote history then you need to add the config knob HISTORY_HELPER_MAX_HISTORY = Number (defaults to 10000) in
the remote schedd's config.
Cheers,
Cole Bollig
From: HTCondor-users <htcondor-users-bounces@xxxxxxxxxxx> on behalf of Angel de Vicente <angel.vicente.garrido@xxxxxxxxx>
Sent: Friday, January 27, 2023 3:12 AM To: htcondor-users@xxxxxxxxxxx <htcondor-users@xxxxxxxxxxx> Subject: [HTCondor-users] Segmentation fault running condor_history Hello,
in our HTCondor Pool (Ubuntu 20.04, CondorVersion: 10.0.1 + Ubuntu 18.04, CondorVersion: 9.12.0) I've found that using the condor_history command to query a remote scheduler for a particular job gives me segmentation fault. To be clear, both these commands run without problems: + condor_history (from machine cruise) + condor_history -name cruise (from another machine) But if I want the details of a given job, running from another machine gives me segmentation fault, i.e.: + condor_history 61.4540 (from machine cruise: OK) + condor_history -name 61.4540 (from another machine: it returns the details of the job apparently OK, but then it ends with "Segmentation fault (core dumped)") I'm not sure if this is related to the following, but I also found that condor_history from the remote machine seems to have a limit of 10000 jobs: + condor_history | wc -l (from machine cruise: 11013 jobs) + condor_history -name cruise | wc -l (from another machine: 10000 jobs) By the way, I try the option -scanlimit, but it seems to only work when running condor_history in the local machine. When querying a remote scheduler, it didn't have any impact (-scanlimit 500 or -scanlimit 11000 always returned 10000 jobs history). [By the way, https://github.com/htcondor/htcondor doesn't accept issues, so is the mailing list the best place nowadays to report issues like these, or is there some other preferred channel?] Cheers, -- Ángel de Vicente -- (GPG: 0x64D9FDAE7CD5E939) Research Software Engineer (Supercomputing and BigData) Instituto de Astrofísica de Canarias (https://www.iac.es/en) _______________________________________________ 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/ |