Think of it this way: scontrol is to control the behavior of the cluster
now(ish). Once jobs are done, they can't really be controlled any more,
so you use sacct to look back at their history. That's a logical way of
thinking about it.
Architecturally, I assume it's because scontrol gets it's data from the
state information, which only concerns the current state of the cluster
and sacct, gets its data from the accounting DB, which is for historical
information.
That's just a guess - I haven't read the source code, but the last I
checked, state data was in /var/lib/slurm, or something like that, and
accounting information was in a mysql DB, so two different data formats.
Prentice
On 6/17/25 7:42 AM, Davide DelVento via slurm-users wrote:
Yeah, that's an annoying thing which I have not understood why has
been designed that way. The information is there and it can be
queried, just with a different command and spitting it out in a
different format. The syntax is
sacct -j XXX
which gives you only some fields, or
sacct -o fields,you,want -j XXX
the (super long) list of possible fields (case insensitive) can be
queried with
sacct -e
HTH
On Tue, Jun 17, 2025 at 4:45 AM Gestió Servidors via slurm-users
<slurm-users@lists.schedmd.com> wrote:
Hello,
Is there any way to get all information (like submit script or
submit node) from a job that is completed? Something like
“scontrol show jobid=XXX” when job is “running” or “pending”. I
need to inspect the submit script of a job but I only know job_id.
Thanks.
--
slurm-users mailing list -- slurm-users@lists.schedmd.com
To unsubscribe send an email to slurm-users-le...@lists.schedmd.com
--
Prentice Bisbal
HPC Systems Engineer III
Computational & Information Systems Laboratory (CISL)
NSF National Center for Atmospheric Research (NSF NCAR)
https://www.cisl.ucar.edu
https://ncar.ucar.edu
--
slurm-users mailing list -- slurm-users@lists.schedmd.com
To unsubscribe send an email to slurm-users-le...@lists.schedmd.com
