Hi HPC team,
Any reason for almost all the nodes on baobab being drain/drng ?
Cheers,
Malte
Dear @Malte.Algren
yes there is a reason: we need to change a parameter on Slurm and this may kill a running job. This is why we set all the nodes to drain. As soon as they are in drain, we apply the modification and we resume the node which return in production.
Thanks for the notification