Ticket 16523 - Limiting the number of jobs and cores per user per partition
Summary: Limiting the number of jobs and cores per user per partition
Status: OPEN
Alias: None
Product: Slurm
Classification: Unclassified
Component: Limits (show other tickets)
Version: 21.08.8
Hardware: Linux Linux
: 6 - No support contract
Assignee: Jacob Jenson
QA Contact:
URL:
Depends on:
Blocks:
 
Reported: 2023-04-13 15:05 MDT by Navneet Khetrapal
Modified: 2023-04-13 16:02 MDT (History)
0 users

See Also:
Site: -Other-
Alineos Sites: ---
Atos/Eviden Sites: ---
Confidential Site: ---
Coreweave sites: ---
Cray Sites: ---
DS9 clusters: ---
HPCnow Sites: ---
HPE Sites: ---
IBM Sites: ---
NOAA SIte: ---
NoveTech Sites: ---
Nvidia HWinf-CS Sites: ---
OCF Sites: ---
Recursion Pharma Sites: ---
SFW Sites: ---
SNIC sites: ---
Tzag Elita Sites: ---
Linux Distro: Rocky Linux
Machine Name: Cruntch4
CLE Version:
Version Fixed:
Target Release: ---
DevPrio: ---
Emory-Cloud Sites: ---

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this ticket.
Description Navneet Khetrapal 2023-04-13 15:05:44 MDT 
Hi Everyone,

I am having trouble in enforcing limits on maximum number of jobs and cores per user per partition. I tired using QOS for this purpose on one partition:

1) Created a test account: 
sacctmgr add account testac

2) Added a user to the account:
sacctmgr add user nsk0051 account=testac

3) Added a qos:
sacctmgr add qos share.64

4) Limited the number of jobs and CPUs for the qos:
sacctmgr modify qos share.64 set GrpJobs=1 GrpTRES=cpu=64

5)sacctmgr show qos share.64 format=name,GrpJobs,GrpTRES
      Name GrpJobs       GrpTRES
---------- ------- -------------
  share.64       1        cpu=64

6) Added QOS=share.64 and AccountingStorageEnforce=qos to partition line in slurm.conf 

Ideally the maximum running jobs should be 1 and cpus should be 64 in this case. But users able to exceed this. The QOS limits don't seem to imposed.  Can anyone help me to please resolve this issue ?