nobody got same problem?
2018-04-09 13:05 GMT+02:00 Nicolò Parmiggiani
:
> Hi,
>
> i have a problem with Preemption.
>
> When high priority job suspend lower priority job all its ok. But when
> high priority job ends and slurm resume lower priority one, sometimes
Hi Matt,
You might want to take a look at my Slurm Wiki, which focuses on
CentOS/RHEL 7: https://wiki.fysik.dtu.dk/niflheim/SLURM. Complete
instructions for Slurm installation, configuration, etc. is in the Wiki.
/Ole
On 04/11/2018 02:26 PM, Matt Hohmeister wrote:
I’m brand-new to Slurm,
Hello,
I'm writing my own "job_submit.lua" for controlling in what
partition a user can run a "srun" and how many CPUs and nodes are
allowed. I want to allow only "srun" in partition "interactive" with
only one core and one node. I have wrote this script but I'm
I'm brand-new to Slurm, and setting it up on a single RHEL 7.4 VM as a proof of
concept before I deploy it. After following the instructions on
https://www.slothparadise.com/how-to-install-slurm-on-centos-7-cluster/ (sorry,
site not working now), I can get slurmd to start perfectly, but
It looks like your slurm.conf is specifying /var/spool as your Save state
directory, and `fatal: Incorrect permissions on state save loc: /var/spool`
indicates that SlurmUser (another configuration in slurm.conf) does not
have access to write to it. It might be a good to make a directory
"sysadmin.caos" writes:
> srun: error: slurm_job_submit: parameter error 65534 4294967294 1
4294967294 is the special value slurm.NO_VAL, meaning the parameter was
not specified. It is for 32 bit parametres. I guess (but haven't
double checked) that 65534 is
Hi,
I would like to know why the sacct command which I am usinig that to
get some reports, shows two lines for each job. For example
[mahmood@rocks7 g]$ sacct --format=start,end,elapsed,ncpus,cputime,user
Start EndElapsed NCPUS
CPUTime User
Hey, folks. I have a relatively simple queueing setup on Slurm 17.02 with a
1000 CPU-day AssocGrpCPURunMinutesLimit set. When the cluster is less busy
than typical, I may still have users run up against the 1000 CPU-day limit,
even though some nodes are idle.
What’s the easiest way to force a job
Actually, I’ve gotten it all working; I just had overlooked some things in
slurm.conf.
I had previously trying to get Son of Grid Engine working, but after ripping
out half my hair, I went to Slurm, since it’s what our research computing
center uses.
Matt Hohmeister
Systems and Network
On 12/04/18 04:00, Mahmood Naderan wrote:
Hi,
Hi Mahmood,
I would like to know why the sacct command which I am usinig that to
get some reports, shows two lines for each job.
sacct reports one line per job step by default, not per job.
If you add the 'JobName' field to your sacct
May I ask what is the purpose to view intermediate job steps?
Also, today when I run sacct command, I only see the last two jobs
[mahmood@rocks7 g]$ sacct -a
JobIDJobName PartitionAccount AllocCPUS State ExitCode
-- -- -- --
11 matches
Mail list logo