Re: [HTCondor-users] Jobs scheduling with flock_to

[ervikrant06@xxxxxxxxx]

We do keep a watch on negotiation cycle. Our negotiation cycle never exceeds one min.. We do use settings likeÂCLAIM_WORKLIFE which helps to build sched cache so we expect it to do matcmaking more quickly. You mentioned at-least one negotiation cycle that's the point of worry for us like why after 40mins still the jobs not considered for matchmaking.Â

Also do you think it makes sense to open a RFE so that pools not mentioned in requirement shouldn't considered for matcmakingÂin FLOCK_TO list?Â

Â

Thanks & Regards,

Vikrant Aggarwal

On Thu, Apr 23, 2020 at 2:53 AM Gregory Thain <gthain@xxxxxxxxxxx> wrote:

On 4/21/20 12:22 PM, ervikrant06@xxxxxxxxx wrote:
> 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?


I'm afraid the answer is "it depends". With the default
FLOCK_INCREMENT, at least one negotiation cycle per pool you are talking
to, so it depends on the negotiation speed. Also note that once a
schedd has a claim to use a machine, it may be able to reuse that claim
for a subsequent job, which is very fast.


-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/