Ole Holm Nielsen
> wrote:
>
> Hi Hoot,
>
> I'm glad that you have figured out that GrpTRESMins is working as documented
> and kills running jobs when the limit is exceeded. This would only occur if
> you lower the GrpTRESMins limit after a job has started.
>
> /
canceled.
Thanks for your help.
> On Apr 24, 2023, at 1:55 PM, Ole Holm Nielsen
> wrote:
>
> On 24-04-2023 18:33, Hoot Thompson wrote:
>> In my reading of the Slurm documentation, it seems that exceeding the limits
>> set in GrpTRESMins should result in terminating a
Is there a mechanism for terminating an active job based on a predetermined
threshold - CPU usage, budget, etc? GrpTRESMins reads like it should work but
in practice it doesn’t seem to.
Thanks in advance!
So Ole, any thoughts on the config info I sent?
I’m still not certain if terminating a running job based on GrpTRESMins is even
possible or supposed to work.
Hoot
> On Apr 24, 2023, at 3:21 PM, Hoot Thompson wrote:
>
> See below…...
>
>> On Apr 24, 2023, at 1:55 PM
See below…...
> On Apr 24, 2023, at 1:55 PM, Ole Holm Nielsen
> wrote:
>
> On 24-04-2023 18:33, Hoot Thompson wrote:
>> In my reading of the Slurm documentation, it seems that exceeding the limits
>> set in GrpTRESMins should result in terminating a running job. Howe
In my reading of the Slurm documentation, it seems that exceeding the limits
set in GrpTRESMins should result in terminating a running job. However, in
testing this, The ‘current value’ of the GrpTRESMins only updates upon job
completion and is not updated as the job progresses. Therefore jobs a
installation solution.
>
> Since requirements can be so different, and because Slurm is a fantastic
> software that can be configured for many different scenarios, IMHO a support
> contract with SchedMD is the best way to get consulting services, get general
> help, and report bugs.
So an update, GrpTRES registers a value while a job is running but GRpTRESMins
does not. So I still have something wrong. GrpTRESMins reads in the docs like
it is in fact an aggregate number.
> On Apr 20, 2023, at 1:01 PM, Ole Holm Nielsen
> wrote:
>
> On 20-04-2023 18:23, H
And it indeed does show current value for a running job!! Do I feel stupid :-)
> On Apr 20, 2023, at 1:01 PM, Ole Holm Nielsen
> wrote:
>
> On 20-04-2023 18:23, Hoot Thompson wrote:
>> Ole,
>> Earlier I found your Slurm_tools posting and found it very useful. Thi
Ah, I thought that was the aggregate of past and current jobs.
> On Apr 20, 2023, at 1:01 PM, Ole Holm Nielsen
> wrote:
>
> On 20-04-2023 18:23, Hoot Thompson wrote:
>> Ole,
>> Earlier I found your Slurm_tools posting and found it very useful. This
>> rema
ubuntu 1 0
0.00 0.00
Clearly I’m still missing something or I don’t understand how it’s supposed to
work.
Hoot
> On Apr 20, 2023, at 2:10 AM, Ole Holm Nielsen
> wrote:
>
> Hi Hoot,
>
> On 4/20/23 00:15, Hoot Thompson wrote:
Thank you for this. I’ll give it a read but no promises that I won’t be back
with more questions!
Hoot
> On Apr 20, 2023, at 2:10 AM, Ole Holm Nielsen
> wrote:
>
> Hi Hoot,
>
> On 4/20/23 00:15, Hoot Thompson wrote:
>> Is there a ‘how to’ or recipe document for
Is there a ‘how to’ or recipe document for setting up and enforcing resource
limits? I can establish accounts, users, and set limits but 'current value' is
not incrementing after running jobs.
Thanks in advance
I'm running slurm on an AWS cluster and there seems to be a race
condition whereby the PrologSlurmctld script runs on occasion when
compute nodes try to transition to CG from P but fall back to P when
nodes are not available. Have you seen this behavior and is there a way
to prevent it? I'm run
Can the prolog script be configured to only run on a cluster head node
as opposed to compute nodes?
Thank you for the support. I will be back with any additional questions.
BTW, if it changes or adds to your thoughts, I'm working in AWS on a
parallelcluster.
Hoot
On 1/21/22 4:12 AM, Ole Holm Nielsen wrote:
On 1/21/22 10:05, Diego Zuccato wrote:
Il 21/01/2022 07:51, Ole Holm Nielsen ha scr
How do you change the default memory per node from the current 1MB to
something much higher?
Thanks in advance.
*ubuntu@node*:*/shared*$ sinfo -o "%20N%10c%10m%25f%10G "
NODELISTCPUSMEMORYAVAIL_FEATURES GRES
hpc-demand-dy-c5n18x361 dynamic,c5n.18xlarge,c5n1(null)
Ok, a fresh start after installing the two recommended packages and things
appear to be working. Thank for the help!
On 9/23/21, 3:04 PM, "slurm-users on behalf of Hoot Thompson"
wrote:
Do I need to specify the json path in the configure process?
On 9/23/21, 2:45 PM, &q
Do I need to specify the json path in the configure process?
On 9/23/21, 2:45 PM, "slurm-users on behalf of Hoot Thompson"
wrote:
If this useful, note that there's no attempt to build anything in the
serializer/json directory.
Making all in serializer
ma
7;
make[5]: Nothing to be done for 'all-am'.
make[5]: Leaving directory '/home/ubuntu/slurm-21.08.1/src/plugins/serializer'
make[4]: Leaving directory '/home/ubuntu/slurm-21.08.1/src/plugins/serializer'
Making all in site_factor
make[4]: Entering directory '/home/u
What's getting built is
serializer_url_encoded.a
serializer_url_encoded.la
serializer_url_encoded.so
if this helps.
On 9/23/21, 2:10 PM, "slurm-users on behalf of Hoot Thompson"
wrote:
On Ubuntu 20.04 I installed ...
libjson-c-dev
Libhttp-parser-dev
Th
On Ubuntu 20.04 I installed ...
libjson-c-dev
Libhttp-parser-dev
That work? No joy if so.
On 9/23/21, 1:30 PM, "slurm-users on behalf of Ole Holm Nielsen"
wrote:
On 23-09-2021 16:01, Hoot Thompson wrote:
> In upgrading to 21.08.1, slurmctld status reports:
>
In upgrading to 21.08.1, slurmctld status reports:
Sep 23 13:49:52 ip-10-10-7-17 systemd[1]: Started Slurm controller daemon.
Sep 23 13:49:52 ip-10-10-7-17 slurmctld[1323]: fatal: Unable to find plugin:
serializer/json
Sep 23 13:49:52 ip-10-10-7-17 systemd[1]: slurmctld.service: Main process
I have the REST API basically working but I am having a problem with job
submission syntax. The error I receive is ‘Unable to parse query”. I have
followed the guides found on-line to no avail. Is there somewhere to look for
what the issue may be?
24 matches
Mail list logo