On 5/29/26 09:40, Kevin Buckley (Pawsey) via slurm-users wrote:
On Wednesday, May 27th, 2026 at 17:53, Ole Holm Nielsen via slurm-users <slurm-users@lists.schedmd.com> wrote:
The safe way to enable the new purge parameters is to introduce them very gradually, for example:
PurgeJobScriptAfter=2000days PurgeJobEnvAfter=2000days
and lower the values little by little over time (restarting slurmdbd each time).
Do you mean to start off with a value of 2000 days?
Yes, 2000 days purging interval would do no harm and ensure that you don't overload the slurmdbd server with database operations, so it's good for an initial test where you check the slurmdbd performance over a couple of days where daily purging takes place. I have an extended discussion of Slurm database purging in this Wiki page: https://wiki.fysik.dtu.dk/Niflheim_system/Slurm_database/#setting-database-p...
Some sites (although perhaps not mine!) could have completely refreshed the hardware resourses that Slurm is controlling, within a 5 year period.
The compute node hardware is a separate resource from the Slurm database server. The age of the Slurm database contents is what matters when it comes to purging of records. Would you agree on these considerations? Best regards, Ole