[slurm-dev] Re: race condition in stepd i/o shutdown

I think that Hydra should kept 0,1,2 open and dup them to /dev/null so that any children’s file descriptor will be greater than 2. This is standard Unix way. Thanks /David/Bigagli da...@schedmd.com === Slurm User Group Meeting, 15-16

[slurm-dev] sacct format %all missing dates, missing aveVMsize

Hi, I am submitting some test jobs via sbatch and when looking at the output, I see 2 items that I don't comprehend fully: 1) sacct -format=%all is not showing any dates for submit, start etc. vs. sacct -format=jobid,. - see example below. 2) I see AveVMSize AveCPU numbers on the 12621.batch

[slurm-dev] Re: race condition in stepd i/o shutdown

Thanks David. That was my gut feeling and what I think I've seen before but I wasn't sure. On Mon, Jul 27, 2015 at 3:19 PM, David Bigagli da...@schedmd.com wrote: I think that Hydra should kept 0,1,2 open and dup them to /dev/null so that any children’s file descriptor will be greater than 2.

[slurm-dev] Slurm 14.11.7 reservation resources wrong

Hello, we have a system with 2 sockets per node and 12 cores per socket. If I create a reservation over one node, Slurm displays CoreCnt=12, but one node has 24 cores. If I take many nodes, Slurm displays CoreCnt=NodeCnt*12 instead of CoreCnt=NodeCnt*24. Could you please tell me what's wrong?