From 661d5bd7b341dd7932604e8dc2547e77f412f934 Mon Sep 17 00:00:00 2001 From: Enrico Nasca Date: Thu, 18 Jan 2024 15:06:12 +0100 Subject: [PATCH] Increase the kill wait time To allow large jobs to be cancelled gracefully, both the KillWait parameter and UnkillableStepTimeout have been increased to 5 minutes. --- ansible/roles/slurm_install/templates/slurm.conf.j2 | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/ansible/roles/slurm_install/templates/slurm.conf.j2 b/ansible/roles/slurm_install/templates/slurm.conf.j2 index 77e74096..e8a0dc70 100644 --- a/ansible/roles/slurm_install/templates/slurm.conf.j2 +++ b/ansible/roles/slurm_install/templates/slurm.conf.j2 @@ -26,11 +26,11 @@ TaskPlugin=task/affinity,task/cgroup # # # TIMERS -KillWait=120 +KillWait=300 #MinJobAge=300 #SlurmctldTimeout=120 #SlurmdTimeout=300 -UnkillableStepTimeout=120 +UnkillableStepTimeout=300 # # # SCHEDULING