Ticket 7046

Summary: After upgrade in slurm18, epoch 0 start_time has not effect in sacct -j jobid request
Product: Slurm Reporter: Regine Gaudin <regine.gaudin>
Component: AccountingAssignee: Albert Gil <albert.gil>
Status: RESOLVED DUPLICATE QA Contact:
Severity: 3 - Medium Impact    
Priority: --- CC: matthieu.hautreux, regine.gaudin
Version: 18.08.6   
Hardware: Linux   
OS: Linux   
Site: CEA 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:
CLE Version: Version Fixed:
Target Release: --- DevPrio: ---
Emory-Cloud Sites: ---

Description Regine Gaudin 2019-05-17 07:04:19 MDT 
After upgrade of slurmdbd and controller in slurm-18.08.06, 
sacct -j jobid does not allow to have jobid accounting info despite epoch 0 seems to be used while sacct -j jobid -S date is ok

1) On controller or client in slurm-18

[root@vm1 ~]# which sacct
/usr/bin/sacct
[root@vm1 ~]#  rpm -qf /bin/sacct
slurm-18.08.6-2.x86_64

root@vm1 ~]# sacct -j 2
       JobID    JobName  Partition    Account  AllocCPUS      State ExitCode 
------------ ---------- ---------- ---------- ---------- ---------- -------- 


[root@vm1 ~]# sacct -j 2 -S 2019-05-16 (yesterday)
       JobID    JobName  Partition    Account  AllocCPUS      State ExitCode 
------------ ---------- ---------- ---------- ---------- ---------- -------- 
2                 valid         vm       root         48     FAILED      2:0 
2.0               valid                  root          3     FAILED      2:0 


[root@vm1 ~]#  sacct -j 2 -S 2019-05-16 -v
sacct: Jobs Eligible in the time window from Thu May 16 00:00:00 2019 to Fri May 17 12:56:23 2019
sacct: Accounting storage SLURMDBD plugin loaded
       JobID    JobName  Partition    Account  AllocCPUS      State ExitCode 
------------ ---------- ---------- ---------- ---------- ---------- -------- 
2                 valid         vm       root         48     FAILED      2:0 
2.0               valid                  root          3     FAILED      2:0 

[root@vm1 ~]#  sacct -j 2  -v
sacct: Jobs Eligible in the time window from Epoch 0 to Fri May 17 12:56:28 2019
sacct: Accounting storage SLURMDBD plugin loaded
       JobID    JobName  Partition    Account  AllocCPUS      State ExitCode 
------------ ---------- ---------- ---------- ---------- ---------- -------- 
[root@vm1 ~]#


2)While on client in slurm 17 (slurmdbd 18)

[root@vm3 ~]# which sacct
/usr/bin/sacct

[root@vm3 ~]# rpm -qf /usr/bin/sacct
slurm-17.11.6-1.x86_64

       JobID    JobName  Partition    Account  AllocCPUS      State ExitCode 
------------ ---------- ---------- ---------- ---------- ---------- -------- 
2                 valid         vm       root         48     FAILED      2:0 
2.0               valid                  root          3     FAILED      2:0 

[root@vm3 ~]# sacct -j 2 -S 2019-05-16
       JobID    JobName  Partition    Account  AllocCPUS      State ExitCode 
------------ ---------- ---------- ---------- ---------- ---------- -------- 
2                 valid         vm       root         48     FAILED      2:0 
2.0               valid                  root          3     FAILED      2:0 

[root@vm3 ~]# sacct -j 2 -v
sacct: Jobs eligible from Thu Jan 01 00:00:00 1970 - Now
sacct: Accounting storage SLURMDBD plugin loaded with AuthInfo=(null)
       JobID    JobName  Partition    Account  AllocCPUS      State ExitCode 
------------ ---------- ---------- ---------- ---------- ---------- -------- 
2                 valid         vm       root         48     FAILED      2:0 
2.0               valid                  root          3     FAILED      2:0 

[root@vm3 ~]# sacct -j 2 -S 2019-05-16 -v
sacct: Jobs eligible from Thu May 16 00:00:00 2019 - Now
sacct: Accounting storage SLURMDBD plugin loaded with AuthInfo=(null)
       JobID    JobName  Partition    Account  AllocCPUS      State ExitCode 
------------ ---------- ---------- ---------- ---------- ---------- -------- 
2                 valid         vm       root         48     FAILED      2:0 
2.0               valid                  root          3     FAILED      2:0 

Many CEA epilogs script are using sacct -j so problem is relatively important

Thanks
 
Regine
Comment 1 Albert Gil 2019-05-17 09:38:27 MDT 
Hi Regine,

You are totally right and this is a known problem in 18.08.6 already fixed in 18.08.7. See bug 6697 for more details.

Sorry for the inconveniences,
Albert

*** This ticket has been marked as a duplicate of ticket 6697 ***