On 19/09/16 22:58, Christopher Samuel wrote: > Thanks so much Ulf, you've just answered a puzzle I've been seeing on an > x86 cluster I'm helping out with!
...and stopping slurmctld & slurmd's (slurmdbd was left going), moving /var/spool/slurm/jobs/node_state* out of the way and starting everything back up fixed it. We saved the state of drained nodes beforehand and reapplied them after of course. Thanks so much Ulf! -- Christopher Samuel Senior Systems Administrator VLSCI - Victorian Life Sciences Computation Initiative Email: sam...@unimelb.edu.au Phone: +61 (0)3 903 55545 http://www.vlsci.org.au/ http://twitter.com/vlsci