Re: [HTCondor-users] Jobs scheduling with flock_to

[ervikrant06@xxxxxxxxx]

Thanks. I have another question related to this issue: How much time job usually spend on a single pool before moving to next one in flock_to list?Â

Noticed that not all jobs in a batch are showing equal amount of queue wait time. Some jobs stay in queue for approx 1 hour and some started runningÂwith in 2mins this happens when slots are available to accomodate jobs. If all the jobs are supposed to follow the sequence of pools in flock_to list then they should show approx equal amount of queue wait time?Â

Without FLOCK_INCREMENT setting: (pools A,B) these were not supposed to run the job (as per requirements) were showing jobs for approx 45 mins tried for negotiation? Same results were seen in Pool C which is supposed to run the job. It's confusing. This batch( 7051853 )ÂÂshown the queue wait time of 53 minutes.Â

Â

Pool A

grep 'Request 7051853.00035: autocluster 538679 (request count 1 of 1450)' /var/tmp/NegotiatorLog_21stApril2020 | sed -n '1p;$p'
04/21/20 03:57:25     Request 7051853.00035: autocluster 538679 (request count 1 of 1450)
04/21/20 04:42:20     Request 7051853.00035: autocluster 538679 (request count 1 of 1450)

Pool B

$  grep 'Request 7051853.00035: autocluster 538679 (request count 1 of 1450)' /var/tmp/NegotiatorLog_21stApril2020 | sed -n '1p;$p'
04/21/20 03:57:25     Request 7051853.00035: autocluster 538679 (request count 1 of 1450)
04/21/20 04:42:21     Request 7051853.00035: autocluster 538679 (request count 1 of 1450)

Pool C

grep ' Request 7051853.00035: autocluster 538679 (request count 1 of 1450)' /var/log/condor/NegotiatorLog | sed -n '1p;$p'
04/21/20 03:57:21     Request 7051853.00035: autocluster 538679 (request count 1 of 1450)
04/21/20 04:42:25     Request 7051853.00035: autocluster 538679 (request count 1 of 1450)

Â

Thanks & Regards,

Vikrant Aggarwal

On Mon, Apr 20, 2020 at 10:23 PM Gregory Thain <gthain@xxxxxxxxxxx> wrote:

On 4/20/20 11:10 AM, ervikrant06@xxxxxxxxx wrote:
> Hi Greg,
>
> Thanks for your reply.
>
> Running 8.5.8 (dev) release which I know is very old. I was under the
> impression that if pool requirement mentioned then it shouldn't follow
> the sequence of flocking list and directly should jump to the pool
> mentioned in requirement.


This is not the case. The jobs flock in the FLOCK_TO order, no matter
what their job requirements are. I'd suggest setting FLOCK_INCREMENT,
and if you have the ability, upgrading condor versions.


-greg

_______________________________________________
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/