Ticket 3926 - Bug in srun when server switches from pn_min_memory to mem_per_cpu
Summary: Bug in srun when server switches from pn_min_memory to mem_per_cpu
Status: RESOLVED FIXED
Alias: None
Product: Slurm
Classification: Unclassified
Component: Contributions (show other tickets)
Version: 17.11.x
Hardware: Linux Linux
: 4 - Minor Issue
Assignee: Danny Auble
QA Contact:
URL:
Depends on:
Blocks: 3879
  Show dependency tree / graph
 
Reported: 2017-06-24 17:58 MDT by Thomas Opfer
Modified: 2017-07-03 11:38 MDT (History)
3 users (show)

See Also:
Site: -Other-
Slinky Site: ---
Alineos Sites: ---
Atos/Eviden Sites: ---
Confidential Site: ---
Coreweave sites: ---
Cray Sites: ---
DS9 clusters: ---
Google sites: ---
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: ---
Machine Name: Lichtenberg High Performance Computer
CLE Version:
Version Fixed: 17.02.6 17.11.0-pre2
Target Release: ---
DevPrio: ---
Emory-Cloud Sites: ---

Attachments
Fix that makes srun work correctly. (918 bytes, patch)
2017-06-24 17:58 MDT, Thomas Opfer 		Details | Diff
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 Thomas Opfer 2017-06-24 17:58:19 MDT 
Created attachment 4826 [details]
Fix that makes srun work correctly.

When srun creates a job allocation and the server decides to use memory per node instead of memory per core or vice versa, it is not detected correctly in the calling srun instance. I attached a fix.

This fix also makes the proposed fix for bug 3879 working.
Comment 2 Danny Auble 2017-06-26 14:14:29 MDT 
Thomas, I believe you have found a good one here, thanks!

Added in commit 3af8a44ded5e
Comment 3 Danny Auble 2017-06-26 14:15:01 MDT 
Please reopen if anything else is needed on this.
Comment 6 Thomas Opfer 2017-07-03 10:49:19 MDT 
(In reply to Danny Auble from comment #3)
> Please reopen if anything else is needed on this.

I just saw that the fix was not merged to the master branch (while the change in the NEWS file was).
Comment 8 Thomas Opfer 2017-07-03 11:38:21 MDT 
Oops, you are right. The github browser is driving me crazy, it opens wrong branches regularly...

Sorry for that.