[slurm-dev] Re: srun and openmpi

Hi Michael, I'm not sure if this is right or not, I don't have much experience with OpenMPI. On most slurm installs I believe the MPI type defaults to none, have you tried adding --mpi=pmi2 or --mpi=openmpi to your srun command? Ian On Wed, Dec 16, 2015 at 9:50 AM, Michael Di Domen

[slurm-dev] Re: Problem in loading modules in slurm Batch script.

Hello Hezi, Have you tried making the shell for the batch script a login shell? #!/bin/bash -l I've had to do that in the past to make the module command behave on our Centos 7.1 systems. Ian On Thu, Dec 3, 2015 at 1:30 AM, Hezi Ismah-Moshe wrote: > > Hi, > > When I try to l

[slurm-dev] Re: scripted use of sacctmgr

Hi Thomas, We use sacctmgr -i in our scripts to remove the prompt, works great. Thanks, Ian On Tue, Oct 13, 2015 at 10:53 AM, Thomas Orgis wrote: > > Hi, > > is it possible to avoid sacctmgr insisting on confirmation on database > changes? Specifically, with 14.11.8, this does n

[slurm-dev] Question about Node State

? Do I need to issue some command across the slaves (in the past I've found that manually relaunching the slurmd application has helped "$ bpsh -a slurmd"). Thank you, -- ~ Ian Lee Lawrence Livermore National Laboratory (W) 925-423-4941

[slurm-dev] Re: slurm.conf consistent across all nodes

t across all >> nodes. >> >> How do we handle parameters that are different on a per node basis? >> >> For example, RealMemory may be X on node 1, but Y on node 2. >> >> Am I missing something? >> >> >> > -- Ian Logan Virtualization and Unix Systems Administrator Information and Communication Techologies - New Mexico State University Phone: 575-646-3054 Email: i...@nmsu.edu

[slurm-dev] Re: Trouble with Prolog scripts

Thanks Kilian, This looks like exactly what I was looking for. ~ Ian Lee Lawrence Livermore National Laboratory (W) 925-423-4941 -Original Message- From: Kilian Cavalotti [mailto:kilian.cavalotti.w...@gmail.com] Sent: Tuesday, October 07, 2014 9:33 PM To: slurm-dev Subject: [slurm-dev

[slurm-dev] Trouble with Prolog scripts

natively, how would I configure SLURM so that I can do what I really want to do which is to drain a node if the diskspace of a particular disk is insufficient? Thank you, ~ Ian Lee Lawrence Livermore National Laboratory (W) 925-423-4941=

[slurm-dev] Re: No standard output

first compute node are two different host. That's why I can't see any output in the control node. Again, thank you for your reply! Best regards, Ian Malcolm

[slurm-dev] No standard output

59:15] debug2: Tree head got back 1 [2014-01-04T16:59:15] debug2: Tree head got them all [2014-01-04T16:59:16] debug2: node_did_resp SLURM1 miao@SLURM0:~$ srun -n2 -l hostname 1: SLURM2 0: SLURM1 miao@SLURM0:~$ Thank you very much! Best regards Ian Malcolm

[slurm-dev] slurm-dev strigger does not work

vents from "strigger --get" and slurmctld.log. Although the SLURM help page says that a check for trigger events is performed every 15 seconds, nothing returns after that time ><" The event handling program I used is simply to echo a message. My SLURM version is 2.1.0. Thanks Ian

[slurm-dev] strigger does not work

vents from "strigger --get" and slurmctld.log. Although the SLURM help page says that a check for trigger events is performed every 15 seconds, nothing returns after that time ><" The event handling program I used is simply to echo a message. Thanks Ian