Hello, Quite recently I upgraded to Slurm 17.02.2. For some reason every day I've noticed that a number of compute nodes go into draining or drained state after running jobs. For example, this morning I see...
[root@blue30 openmpi-1.10.1]# sinfo -Nl | grep drain red0165 1 batch* draining 12 2:6:1 1 0 1 (null) batch job complete f red0691 1 batch* drained 12 2:6:1 1 0 1 (null) batch job complete f red0692 1 batch* draining 12 2:6:1 1 0 1 (null) batch job complete f red0882 1 batch* draining 12 2:6:1 1 0 1 (null) batch job complete f red0895 1 batch* drained 12 2:6:1 1 0 1 (null) batch job complete f red0897 1 batch* drained 12 2:6:1 1 0 1 (null) batch job complete f red0901 1 batch* draining 12 2:6:1 1 0 1 (null) batch job complete f I never saw this behaviour with v16.05.2. Does anyone understand why this is happening, please? Is this a bug introduced in 17.02.2? Best regards, David
