[slurm-users] Re: Bug? sbatch not respecting MaxMemPerNode setting

2024-09-05 Thread Angel de Vicente via slurm-users
Hello, Brian Andrus via slurm-users writes: > Unless you are using cgroups and constraints, there is no limit > imposed. [...] > So your request did not exceed what slurm sees as available (1 cpu > using 4GB), so it is happy to let your script run. I suspect if you > look at the usage, you wil

[slurm-users] Re: Bug? sbatch not respecting MaxMemPerNode setting

2024-09-05 Thread Angel de Vicente via slurm-users
Hello again, Angel de Vicente via slurm-users writes: > [...] I don't understand is why the first three submissions > below do get stopped by sbatch while the last one happily goes through? > >>> , >>> | $ sbatch -N 1 -n 1 -c 76 -p short --mem-per-cpu=4000M test.batch >>> | sbatch: error: Ba

[slurm-users] Make a job pending in the plugin

2024-09-05 Thread Benjamin Jin via slurm-users
Hello all, I am tyring to build a custom plugin to force some jobs to be pended. In the official document, `ESLURM*` errors are only valid for `job_submit_lua`. I tried to send `ESLURM_JOB_PENDING`, but it only rejects the job submission. Does anyone know how to pend a job in job_submit plugin?

[slurm-users] Re: Make a job pending in the plugin

2024-09-05 Thread Ole Holm Nielsen via slurm-users
On 9/5/24 11:13, Benjamin Jin via slurm-users wrote: I am tyring to build a custom plugin to force some jobs to be pended. In the official document, `ESLURM*` errors are only valid for `job_submit_lua`. I tried to send `ESLURM_JOB_PENDING`, but it only rejects the job submission. Does anyone k

[slurm-users] Configuration for nodes with different TmpFs locations and TmpDisk sizes

2024-09-05 Thread Jake Longo via slurm-users
Hi all, We have a number of machines in our compute cluster that have larger disks available for local data. I would like to add them to the same partition as the rest of the nodes but assign them a larger TmpDisk value which would allow users to request a larger tmp and land on those machines. T

[slurm-users] Re: Configuration for nodes with different TmpFs locations and TmpDisk sizes

2024-09-05 Thread Cutts, Tim via slurm-users
I’ve always had local storage mounted in the same place, in /tmp. In LSF clusters, I just let LSF’s lim get on with autodetecting how big /tmp was and setting the tmp resource automatically. I presume SLURM can do the same thing, but I’ve never checked. Tim -- Tim Cutts Scientific Computing

[slurm-users] salloc not starting shell despite LaunchParameters=use_interactive_step

2024-09-05 Thread Loris Bennett via slurm-users
Hi, With $ salloc --version slurm 23.11.10 and $ grep LaunchParameters /etc/slurm/slurm.conf LaunchParameters=use_interactive_step the following $ salloc --partition=interactive --ntasks=1 --time=00:03:00 --mem=1000 --qos=standard salloc: Granted job allocation 18928869 sal

[slurm-users] Re: salloc not starting shell despite LaunchParameters=use_interactive_step

2024-09-05 Thread Jason Simms via slurm-users
I know this doesn't particularly help you, but for me on 23.11.6 it works as expected and immediately drops me onto the allocated node. In answer to your question, yes, as I understand it the default/expected behavior is to return the shell directly. Jason On Thu, Sep 5, 2024 at 8:18 AM Loris Ben

[slurm-users] Re: salloc not starting shell despite LaunchParameters=use_interactive_step

2024-09-05 Thread Carsten Beyer via slurm-users
Hi Loris, we use SLURM 23.02.7 (Production) and 23.11.1 (Testsystem). Our config contains a second parameter InteractiveStepOptions in slurm.conf: InteractiveStepOptions="--interactive --preserve-env --pty $SHELL -l" LaunchParameters=enable_nss_slurm,use_interactive_step That works fine for u

[slurm-users] Re: salloc not starting shell despite LaunchParameters=use_interactive_step

2024-09-05 Thread Jason Simms via slurm-users
Ours works fine, however, without the InteractiveStepOptions parameter. JLS On Thu, Sep 5, 2024 at 9:53 AM Carsten Beyer via slurm-users < slurm-users@lists.schedmd.com> wrote: > Hi Loris, > > we use SLURM 23.02.7 (Production) and 23.11.1 (Testsystem). Our config > contains a second parameter In

[slurm-users] Re: salloc not starting shell despite LaunchParameters=use_interactive_step

2024-09-05 Thread Carsten Beyer via slurm-users
Thanks Jason for the hint. Looks like, the parameter was kept in slurm.conf from previous SLURM versions at our site.  Works also without setting InteractiveStepOptions in slurm.conf. Best Regards, Carsten Am 05.09.24 um 15:55 schrieb Jason Simms via slurm-users: Ours works fine, however, wi

[slurm-users] Re: salloc not starting shell despite LaunchParameters=use_interactive_step

2024-09-05 Thread Loris Bennett via slurm-users
Jason Simms via slurm-users writes: > Ours works fine, however, without the InteractiveStepOptions parameter. My assumption is also that default value should be OK. It would be nice if some one could confirm that 23.11.10 was working for them. However, we'll probably be upgrading to 24.5 fairl

[slurm-users] Re: salloc not starting shell despite LaunchParameters=use_interactive_step

2024-09-05 Thread Paul Edmon via slurm-users
Its definitely working for 23.11.8, which is what we are using. -Paul Edmon- On 9/5/24 10:22 AM, Loris Bennett via slurm-users wrote: Jason Simms via slurm-users writes: Ours works fine, however, without the InteractiveStepOptions parameter. My assumption is also that default value should b

[slurm-users] Nodelist syntax and semantics

2024-09-05 Thread Jackson, Gary L. via slurm-users
Is there a description of the “nodelist” syntax and semantics somewhere other than the source code? By “nodelist” I mean expressions like “name[000,099-100]” and how this one, for example, expands to “name000, name099, name100”. -- Gary smime.p7s Description: S/MIME cryptographic signatu

[slurm-users] Re: Nodelist syntax and semantics

2024-09-05 Thread Paul Edmon via slurm-users
I think this might be the closest to one: https://slurm.schedmd.com/slurm.conf.html#SECTION_NODE-CONFIGURATION From the third paragraph: "Multiple node names may be comma separated (e.g. "alpha,beta,gamma") and/or a simple node range expression may optionally be used to specify numeric ranges