[slurm-users] Fairshare factor not reflected in job priority

[Marcin Stolarek]

Sorry for spamming, I've found the answer just answer posting the question.
The resolution for my issue was to add FAIR_TREE to PriorityFlags.
Fairshare factor for the user who submitted the job was low because of his
high utilization of resources and sshare -u login was showing 0.00000 for
him.

cheers,
Marcin

2018-03-15 11:00 GMT+01:00 Marcin Stolarek <stolarek.marcin at gmail.com>:

> I'm working on a priority multifactor plugin configuration and I'm not
> sure if I'm missing something or the behaviour I see is the result of bug.
> Basically
>
> # sshare | grep XX
>  XX                                    1    0.071429        4367
> 0.031536   0.736368
>
> which I read as fairshare factor = 0.73, quite hight
>
> # sprio -w
>           JOBID   PRIORITY        AGE  FAIRSHARE
>         Weights                  1000      10000
>
> and than the jobs is account YY and XX respectively:
> # sprio
>           JOBID   PRIORITY        AGE  FAIRSHARE
>         9150547          1          0          0
> <--YY
>         9265691          8          8          0
> <--XX
>
> I'm using TRESBilling, however, I expected fairshare priority component to
> be simply fairshare factor (from sshare) * fairshare weight..
>
> cheers,
> Marcin
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.schedmd.com/pipermail/slurm-users/attachments/20180315/e1712487/attachment.html>