Hi all,
when I get an email back from slurm concerning an sbatch script, I notice
the name of the job in the email subject header has been truncated, e.g.:
SLURM Job_id=506585 Name=PicardToolsCollectMultip Failed, Run time 00:30:50
The name should read PicardToolsCollectMultiple
Is there a way
Hi,
Anyone who has a successful build?
I got following errors after "configure":
checking for support of printf("%s", NULL)... configure: error: printf("%s",
NULL) results in abort, upgrade to OpenSolaris release 119
Thanks.
Fred
Paul,
I had such ulimit statements in a wide variety of files, including that one ...
and nothing made any difference until I used Christopher's suggestion of
placing the LimitXXX lines directly in the service files. At any rate, we are
working now.
Paul.
> On Jun 23, 2015, at 16:00, Paul v
(Apologies for this slightly off-topic question.)
We are currently using Gold
(http://www.adaptivecomputing.com/products/open-source/gold/) to manage
allocations and accounting, but are looking for alternative solutions.
It would be very interesting to know what people on this list use for
accou
Why would you use gold rather than Slurm's internal mechanism, which
are better integrated with Slurm and vastly more scalable? See:
http://slurm.schedmd.com/accounting.html
Quoting Bjørn-Helge Mevik :
(Apologies for this slightly off-topic question.)
We are currently using Gold
(http://www
On 24/06/15 16:04, Bjørn-Helge Mevik wrote:
(Apologies for this slightly off-topic question.)
We are currently using Gold
(http://www.adaptivecomputing.com/products/open-source/gold/) to manage
allocations and accounting, but are looking for alternative solutions.
It would be very interesting
There is no truncation in current code (src/slurmctld/agent.c):
if (job_ptr->array_task_id != NO_VAL) {
mi->message = xstrdup_printf("SLURM Job_id=%u_%u (%u) Name=%s "
"%s%s%s",
job_p
Can someone explain why these node lists might overlap? This is causing a
lot of pain for the user trying to use the PECAN-1km reservation. I've
tried recreating the other reservation, and it is grabbing nodes that
should not be available to it:
[root@master.stampede]# scontrol show res=PECAN-1km
>From looking at your list of nodes I dont see any node overlapping. The
names are really long and it is somewhat confusing but in seeing these
nodes they are not overlapping on the reservations.
i.e. c401 has listed in the first reservation - node 001-004 and 101-102,
201-204, 301-304, 401-404...
Sorry, this one wasn't the overlapping one. See below:
ReservationName=Claremont-Training-2015-06-25
StartTime=2015-06-24T12:00:00 EndTime=2015-06-25T12:00:00
Duration=1-00:00:00
Nodes=c401-[403-404,501-504,601-604,701-704,801-804,901-904],c402-[001-004,
101-104,201-204,301-302] NodeCnt=36 Co
Thanks, Jackie. I sent the wrong Claremont reservation with the overlap
(se my update). I'm aware of the OVERLAP flag but we never set that. Thus
my confusion. See my update message when it hits the list.
We did recently upgrade slurm from 2.6.3 to 14.11.3, and these
reservations existed before t
Thank you all for your help and for your suggestions, that's really
appreciated! I will give a try to the cgroups plugins and to the
--exclusive option and experiment with slurm to find out what's more
appropriate for my server.
On 06/23/2015 02:59 PM, Morris Jette wrote:
See the task/cgroups
Hi,
hm, weird. We're running slurm 2.6.0.
Was this ever the case in a previous version? Or is there any reason why a
the Name value would be truncated?
How would I go about debugging such a thing?
When I look at squeue's output, it is truncated in the same way, but not
when I issue `scontrol show
It's open source. Help yourself to it.
Quoting Kenny Billiau :
Hi,
hm, weird. We're running slurm 2.6.0.
Was this ever the case in a previous version? Or is there any reason why a
the Name value would be truncated?
How would I go about debugging such a thing?
When I look at squeue's output,
2015-06-24 22:27 GMT+02:00 Moe Jette :
>
> It's open source. Help yourself to it.
Like it! :)
Hey!
I've got one user I trust and know that he isn't going to do anything
malicious, he needs a direct acces to file in dev (/dev/cpu/*/msr in
particular).
Have anybody checked how to do such a thing in slurm? We are thinking abuot
doing it in prologue and changing back in epilogue, checking if
2015-06-24 16:43 GMT+02:00 Veronique Legrand :
>
> On 24/06/15 16:04, Bjørn-Helge Mevik wrote:
>
>> (Apologies for this slightly off-topic question.)
>>
>> We are currently using Gold
>> (http://www.adaptivecomputing.com/products/open-source/gold/) to manage
>> allocations and accounting, but are
2015-06-24 23:12 GMT+02:00 Marcin Stolarek :
>
>
> 2015-06-24 16:43 GMT+02:00 Veronique Legrand :
>
>>
>> On 24/06/15 16:04, Bjørn-Helge Mevik wrote:
>>
>>> (Apologies for this slightly off-topic question.)
>>>
>>> We are currently using Gold
>>> (http://www.adaptivecomputing.com/products/open-sou
Be sure to test it first before trying anything else:
https://stackoverflow.com/questions/18661976/reading-dev-cpu-msr-from-userspace-operation-not-permitted.
We ran into this issue once when we had a "trusted" person and we
couldn't easily grant him access to the MSRs. We couldn't find a goo
Sory for previous mails.. (keyboard problem)
We are using slurm accounting with xdmod (http://xdmod.sourceforge.net/)
for graphical presentation. It's nice and I hope with group of people using
and developing this tool will make it even better :)
cheers,
marcin
+1 for SLURM accounting with XDMoD.
=
Trey Dockendorf
Systems Analyst I
Texas A&M University
Academy for Advanced Telecommunications and Learning Technologies
Phone: (979)458-2396
Email: treyd...@tamu.edu
Jabber: treyd...@tamu.edu
On Wed, Jun 24, 2015 at 4:19 PM, Marc
I would try to make the dev a reserveable device in a gres conf. Not entirely
sure that it will work, but it would be where I would start.
Buddy
Sent via my Android phone.
Original message
From: Marcin Stolarek
Date:06/24/2015 4:08 PM (GMT-06:00)
To: slurm-dev
Subject: [slur
On Wed, Jun 24, 2015 at 2:37 PM, Trey Dockendorf wrote:
> +1 for SLURM accounting with XDMoD.
Ditto.
Cheers,
--
Kilian
On 24/06/15 21:11, Fred Liu wrote:
> Anyone who has a successful build?
>
> I got following errors after "configure":
>
> checking for support of printf("%s", NULL)... configure: error: printf("%s",
> NULL) results in abort, upgrade to OpenSolaris release 119
That's checking for support for
Dear Morris,
Thank you for the suggestion.
Cheers,
Tyanko
On 23 June 2015 at 20:53, Morris Jette wrote:
> See --ntasks-per-node option.
>
>
> On June 23, 2015 2:23:32 AM PDT, Tyanko Aleksiev
> wrote:
>>
>> Dear all,
>>
>> I'm trying to run the quantum chemistry package "GAMESS" using SLURM.
>
On 25/06/15 00:28, Moe Jette wrote:
> Why would you use gold rather than Slurm's internal mechanism, which are
> better integrated with Slurm and vastly more scalable? See:
> http://slurm.schedmd.com/accounting.html
This is what we use for our quarterly quotas on projects.
We then use Karaage t
26 matches
Mail list logo