While testing new Linux kernels on our cluster, we got messages like this in /var/log/messges:
Feb 27 11:21:55 compute-15-1 kernel: cgroup: slurmstepd (10089) created nested cgroup for controller "memory" which has incomplete hierarchy support. Nested cgroups may change behavior in the future. Feb 27 11:21:55 compute-15-1 kernel: cgroup: "memory" requires setting use_hierarchy to 1 on the root. Is this something we should worry about? We run Slurm 14.03.7, and we got the messages with kernel versions 3.8.13-55.1.5.el6uek.x86_64 and 3.19.1-1.el6.elrepo.x86_64. We use cgroups to limit the memory jobs can use. Our cgroup.conf contains: CgroupMountpoint=/dev/cgroup CgroupAutomount=yes ConstrainSwapSpace=yes and our slurm.conf contains: TaskPlugin=task/cgroup ProctrackType=proctrack/cgroup SelectType=select/cons_res SelectTypeParameters=CR_CPU_Memory -- Regards, Bjørn-Helge Mevik, dr. scient, Department for Research Computing, University of Oslo
