[slurm-users] Create users
pedmon at cfa.harvard.edu
Thu Sep 13 09:09:40 MDT 2018
This is true. I've seen this happen before in our set up if sacctmgr
happens to wedge. It doesn't happen very frequently though and doesn't
impact us that much. We've been running in this fashion for years now
with only very infrequent degradation. We do about 80,000 jobs per day.
Plus we only add them if they don't already exist so the impact is only
when new users appear.
On 09/13/2018 10:48 AM, Douglas Jacobsen wrote:
> At one point in time we would also use the job_submit.lua to add
> users, however, I cannot recommend it in general since job_submit runs
> while locks are held within slurmcltd, which could have dramatic
> performance or even functionality impacts if there are delays in
> adding the user.
> Doug Jacobsen, Ph.D.
> NERSC Computer Systems Engineer
> Acting Group Lead, Computational Systems Group
> National Energy Research Scientific Computing Center
> dmjacobsen at lbl.gov <mailto:dmjacobsen at lbl.gov>
> ------------- __o
> ---------- _ '\<,_
> ----------(_)/ (_)__________________________
> On Thu, Sep 13, 2018 at 7:40 AM Eric F. Alemany <ealemany at stanford.edu
> <mailto:ealemany at stanford.edu>> wrote:
> Hi Paul
> You said
> “Another way would be to make all your Linux users and then map
> that in to Slurm using sacctmgr.”
> I am curious to know how you do that.
> Thank you
> *Eric F. Alemany*
> /System Administrator for Research/
> Division of Radiation & Cancer Biology
> Department of Radiation Oncology
> Stanford University School of Medicine
> Stanford, California 94305
> Tel:1-650-498-7969 <tel:1-650-498-7969> No Texting
> Fax:1-650-723-7382 <tel:1-650-723-7382>
> On Sep 13, 2018, at 01:09, Loris Bennett
> <loris.bennett at fu-berlin.de <mailto:loris.bennett at fu-berlin.de>>
>> Hi Paul,
>> I'd be interested in seeing your Lua submit script, if you're
>> willing to
>> Until now I had thought that the most elegant way of setting up Slurm
>> users would be via a PAM module analogous to pam_mkhomedir, the
>> option being to use pam_script.
>> However, given that we do have users who somehow never get round to
>> submitting a job before their HPC access expires, setting up the
>> account when the first job is submitted seems quite appealing.
>> Paul Edmon <pedmon at cfa.harvard.edu
>> <mailto:pedmon at cfa.harvard.edu>> writes:
>>> So useradd is adding a Linux user, which sacctmgr creates a
>>> Slurm user.
>>> What we do is that we run AD for our Linux user managment. We
>>> then in our job submit lua script look to see if the user has an
>>> account in slurm and if they don't we create it.
>>> Another way would be to make all your Linux users and then map
>>> that in to Slurm using sacctmgr.
>>> It really depends on if your Slurm users are a subset of your
>>> regular users or not.
>>> -Paul Edmon-
>>> On 9/12/2018 12:21 PM, Andre Torres wrote:
>>> Hi all,
>>> I’m new to slurm and I’m confused regarding user creation. I
>>> have an installation with 1 login node and 5 compute nodes. If I
>>> create a user across all the nodes with the same uid and gid I
>>> can execute jobs but
>>> I can’t understand the difference between user creation with
>>> “useradd” command and the “sacctmgr” command
>>> sacctmgr create account name=test
>>> sacctmgr create user jdoe account=test
>>> Also, is there anyway of creating a user at login node and
>>> replicate to the compute nodes ? What is the best practice for
>>> user creation ?
>>> Thanks in advance
>> Dr. Loris Bennett (Mr.)
>> ZEDAT, Freie Universität Berlin Email
>> loris.bennett at fu-berlin.de <mailto:loris.bennett at fu-berlin.de>
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the slurm-users