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

Re: [HTCondor-users] Create jobs with time restriction



Hi Vikrant,

I mentioned below settings in the execute node condor_config.local file. For now I tried with 60s only.

MaxRunTime = 60
RunTimeExceeded = (TotalJobRunTime > $(MaxRunTime))
PREEMPT = $(RunTimeExceeded)

The above settings preempts and after some time again re-scheduled to running.Â
I understood why it is preempting as the above settings says so, but I want the job to be completed removed and never again re-scheduled for execution.

executable usingÂ
I have a script which sleeps for 80s and I tried to kill the job as it exceeds 60s.

I tried mentioning KILL = $(RunTimeExceeded) but nothing happened instead it executes normally for 80s.

Is there any settings that can remove the job instead of re-scheduling, only if such run time exceeded happens.
And thanks for sharing the relevant link, it will help me in better understanding of such settings.

Regards,
NishitÂ





On Mon, Jul 22, 2019 at 11:08 AM Vikrant Aggarwal <ervikrant06@xxxxxxxxx> wrote:
Hello,

You have mentioned to kill the job hence following link may not apply to you but still sharing it as it's a good example of taking an action on a job running more than stipulated time.Â

https://htcondor-wiki.cs.wisc.edu/index.cgi/wiki?p=HowToAutoRetryElsewhere

Coming back to your question basically you need to use STARTD evacuate condition. You may need to set the following on your executor nodes to kill the jobs running for more than 3600s (i.e 1 hour)

Maxruntime = 3600
Runtimeexceeded = (totaljobruntime > $(maxruntime))
PREEMPT = $(RunTimeExceeded)

Hope if helps.Â

Thanks & Regards,
Vikrant Aggarwal


On Mon, Jul 22, 2019 at 10:41 AM Nishit Shah <nishitshah2017@xxxxxxxxx> wrote:
Hi,

I want to execute jobs with time limit on each job.
Like may be 1 hr (at max) so any job executing in execute node should either complete the job within one hour or it will be killed.

I tried using maxjobretirementtime = 60 in the job description file just to check whether its gets killed within 60 secs, but it is not working.

I am not sure whether the provided command is actually the one that should be used.

I want to know what should be mentioned in the job description file and also in the machine condor_config.local file if I want to set for every job running in particular execute node to be completed with a certain time frame.

I am running condor in Ubuntu 18.04 and condor version is 8.8.4

Please let me know if there are any other file I need to provide.

Regards,
Nishit
_______________________________________________
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/
_______________________________________________
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/