On Oct 10, 2017, at 1:53 PM, Todd Tannenbaum <tannenba@xxxxxxxxxxx> wrote:
On 10/10/2017 10:41 AM, Mary Romelfanger wrote:
Todd,
âwhich typically shuts things down within a few seconds, killing the job
with SIGKILL if needed. This should be what happens when you do a
'systemctl stop condor'.â
This is what I have been doingâ and the processes do go away within seconds. The shutdown is successful.
What I am seeing though is that now if I do a condor_status (or condor_status âany or condor_status âmaster) after I do that stop condor.. the cores/machine still show in that list for many minutes after it is stopped (10 or 15 minutes).
Mary
Hi Mary,
I reproduced your the problem you describe. The issue is with the HTCondor RPM configuration of systemd, so I fear the above will occur with any version of HTCondor running on a Linux using systemd.
There is a ticket about this at
https://htcondor-wiki.cs.wisc.edu/index.cgi/tktview?tn=6096
As you can see in the ticket, I just committed a patch to fix the problem, so all future releases of HTCondor should not suffer the same problem. In the "Remarks" section of the ticket I described a workaround you can use until HTCondor v8.6.7 is released at the end of October.
Hope the above helps and thank you for reporting the issue,
regards,
Todd
_______________________________________________
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/