Ticket 15744

Summary: Archiving DB data triggers: error: invalid units from purge '-21736'
Product: Slurm Reporter: Marc Caubet Serrabou <marc.caubet>
Component: slurmdbdAssignee: Director of Support <support>
Status: RESOLVED FIXED QA Contact:
Severity: 4 - Minor Issue    
Priority: ---    
Version: 22.05.6   
Hardware: Linux   
OS: Linux   
Site: Paul Scherrer 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 Marc Caubet Serrabou 2023-01-06 01:27:20 MST 
Hi,

since hour DB is become very big, I wanted to Archive records with the dump tool provided with slurmdbd.

These are the slurmdbd.conf settings:
# ARCHIVE
ArchiveDir=/archivedir
#ArchiveScript=
ArchiveEvents=yes
PurgeEventAfter=1825days
ArchiveJobs=yes
PurgeJobAfter=1825days
ArchiveResvs=yes
PurgeResvAfter=1825days
ArchiveSteps=yes
PurgeStepAfter=1825days
ArchiveSuspend=yes
PurgeSuspendAfter=1825days
ArchiveTXN=yes
PurgeTXNAfter=1825days
ArchiveUsage=yes
PurgeUsageAfter=1825days


# LOGGING
DebugLevelSyslog=info
DebugLevel=verbose
DebugFlags=DB_ARCHIVE
LogFile=/var/log/slurm/slurmdbd.log
LogTimeFormat=iso8601_ms

After forcing a "sacctmgr archive dump" command, I get the following errors:

[2023-01-06T09:19:42.798] error: invalid units from purge '-21736'
[2023-01-06T09:19:42.798] error: Parsing purge meg_event_table
[2023-01-06T09:19:42.798] error: Cluster meg rollup failed
[2023-01-06T09:19:42.798] error: invalid units from purge '-21736'
[2023-01-06T09:19:42.798] error: Parsing purge merlin5_event_table
[2023-01-06T09:19:42.798] error: Cluster merlin5 rollup failed
[2023-01-06T09:19:42.799] error: invalid units from purge '-21736'
[2023-01-06T09:19:42.799] error: Parsing purge gmerlin6_event_table
[2023-01-06T09:19:42.799] error: Cluster gmerlin6 rollup failed
[2023-01-06T09:19:42.802] error: invalid units from purge '-21736'
[2023-01-06T09:19:42.802] error: Parsing purge merlin6_event_table
[2023-01-06T09:19:42.802] error: Cluster merlin6 rollup failed
[2023-01-06T09:22:09.691] error: invalid units from purge '-21736'
[2023-01-06T09:22:09.691] error: Parsing purge gmerlin6_event_table

On a smaller test instance, I had no problems testing this and archiving worked properly. How can we fix this problem?

Thanks a lot,
Marc
Comment 1 Marc Caubet Serrabou 2023-01-06 02:07:31 MST 
Hi,

Closing ticket. 1825 days was a very big number which triggered an overflow. Setting this to the equivalent of 3+ years worked. Now I am progressively archiving in chunks of 1 year.

Sorry for the noise.
Marc