I added that line and restarted the service via
# systemctl restart slurmctld
However, still I get the same error.
Moreover, when I salloc, I don't see slurm/ in cgroup path
[shams@hpc ~]$ salloc
salloc: Granted job allocation 293
[shams@hpc ~]$ bin/show_my_cgroup --debug
bash: bin/show_my_cgrou
>depends on whether "ConstrainSwapSpace=yes" appears in cgroup.conf.
Thanks for the detail.
On the head node, mine is
# cat cgroup.conf
CgroupAutomount=yes
CgroupReleaseAgentDir="/etc/slurm/cgroup"
ConstrainCores=no
ConstrainRAMSpace=no
Is that the root of the problem?
Regards,
Mahmood
apologies for a long response; didn't have time for a shorter one ;)
>you have it backwards. slurm creates a cgroup for the job (step)
and uses the cgroup control to tell the kernel how much memory to
permit the job-step to use.
I would like to know how can I increase the threshold in slur
That's a different project. I'm asking if there is a REST client library
for making REST calls in the slurm source code
https://github.com/SchedMD/slurm.
On Fri, Jan 24, 2020 at 12:35 PM Renfro, Michael wrote:
> The slurm-web project [1] has a REST API [2]. Never used it myself, just
> used the
The slurm-web project [1] has a REST API [2]. Never used it myself, just used
the regular web frontend for viewing queue and node state.
[1] https://edf-hpc.github.io/slurm-web/index.html
[2] https://edf-hpc.github.io/slurm-web/api.html
> On Jan 24, 2020, at 1:22 PM, Dean Schulze wrote:
>
> Ex
Since there isn't a list for slurm development I'll ask here. Does the
slurm code include a library for making REST calls? I'm writing a plugin
that will make REST calls and if slurm already has one I'll use that,
otherwise I'll find one with an appropriate open source license for my
plugin.
Tha
>you have it backwards. slurm creates a cgroup for the job (step)
>and uses the cgroup control to tell the kernel how much memory to
>permit the job-step to use.
I would like to know how can I increase the threshold in slurm config
files. I can not find it.
According to [1], " No value is provi
>how much memory are you requesting from Slurm in your job?
#SBATCH --mem=38GB
also,
# sacctmgr list association format=user,grptres%30 | grep shams
shams cpu=10,mem=40G
Regards,
Mahmood
Excuse me, I have confused with that.
While the cgroup value is 68GB, I run on terminal and see the VSZ is about
80GB and the program runs normally.
However, with slurm on that node, I can not run.
how much memory are you requesting from Slurm in your job?
Why on terminal I can run, but I can
Excuse me, I have confused with that.
While the cgroup value is 68GB, I run on terminal and see the VSZ is about
80GB and the program runs normally.
However, with slurm on that node, I can not run.
Why on terminal I can run, but I can not run via slurm?
I wonder if slurm gets the right value from
There are differences for X11 between Slurm versions so it may help to know
which version you have.
I tried some of your commands on our slurm 19.05.3-2 cluster, and
interestingly on the session on the compute node I don't see the cookie for
the login node: This was with MobaXterm:
[user@prdubrv
I see this
# cat /sys/fs/cgroup/memory/memory.memsw.usage_in_bytes
71496372224
which is about 68GB.
As I said, running from terminal has no problem.
Is is just fine to set a larger value (130GB) as below?
echo 139586437120 > /sys/fs/cgroup/memory/memory.memsw.usage_in_bytes
of course not. "u
I see this
# cat /sys/fs/cgroup/memory/memory.memsw.usage_in_bytes
71496372224
which is about 68GB.
As I said, running from terminal has no problem.
Is is just fine to set a larger value (130GB) as below?
echo 139586437120 > /sys/fs/cgroup/memory/memory.memsw.usage_in_bytes
Regards,
Mahmood
Yes, it uses a large value for virtual size.
Since I can run it via terminal (outside of slurm), I think kernel
parameters are OK.
In other words, I have to configure slurm for that purpose.
Which slurm configuration parameter is in charge of that?
Regards,
Mahmood
On Fri, Jan 24, 2020 at 5:22
I have a weird problem which I can't get to the bottom of.
We have a cluster which allows users to start interactive sessions which
forward any X11 sessions they generated on the head node. This generally works
fine, but on the account of one user it doesn't work. The X11 connection to
the he
On 24/1/20 3:46 am, Mahmood Naderan wrote:
Has anyone run blast on multiple nodes via slurm?
I don't think blast is something that can run across nodes (or at least
it didn't used to be). There is/was something called "mpiblast" that
could do that.
If you'll excuse the plug this sounds li
Answering my own question after much help from Josko Plazonic-
local mem_to_use = 0
-- This is the logic -
-- either min_mem_per_node or min_mem_per_cpu will be set
-- Both can't be set, so only act in those two cases
if job_desc.min_mem_per_node ~= nil then
mem_to_use = job_desc.min_me
Resurrecting an older thread where I need to obtain the value for memory
in a submitted job. Turns out this is not an easy case with the method
I'm trying to use so hope that there is just some variable I am overlooking.
The trivial case was simply to look at job_desc.pn_min_memory. And this
Does your Slurm cgroup or node OS cgroup configuration limit the virtual
address space of processes? The "Error memory mapping" is thrown by blast when
trying to create a virtual address space that exposes the contents of a file on
disk (see "man mmap") so the file can be accessed via pointers
Hi,
Although I can run the blastx command on terminal on all nodes, I can not
use slurm for that due to a so called "memory map error".
Please see below that I pressed ^C after some seconds when running via
terminal.
Fri Jan 24 15:29:57 +0330 2020
[shams@hpc ~]$ blastx -db ~/ncbi-blast-2.9.0+/bin/
Hi,
Has anyone run blast on multiple nodes via slurm? The question should be
asked from blast guys but I didn't find their discussion mailing list.
I see the example on [1] which uses "-N 1" and "--ntasks-per-node".
So that limits to one node run only.
Thanks for any comment.
[1] http://hpc.medi
21 matches
Mail list logo
