Hi Slurm Folks,
I've run into a problem with how Slurm schedules jobs submitted to
multiple partitions.
I'm running Slurm 20.02.3. Our cluster is divided into two partitions
by node funding group. All users have rights to and submit to both
partitions (i.e., jobs specify -p part1,part2).
But you can change the time limit at runtime with "scontrol update job
...". Therefore an environment variable can have an outdated value.
> There's an outstanding feature request for that:
>
> https://bugs.schedmd.com/show_bug.cgi?id=8383
>
> While waiting on that, we've taken to injecting it
Troy,
I have added my vote for that feature. Meanwhile, I am looking at doing
it through the job_submit.lua, but I am unable to access
job_desc.environment (it is NULL). Looks like others have had that issue
as well, but I see no resolution.
How are you setting environment variables from
That I don't know. I would imagine not as it doesn't touch the job
records. It just touches the totals that sshare uses to compute
fairshare. But that's just a guess on my part. We don't really use
sreport for reporting, we instead use grafana and XDMod for historic data.
-Paul Edmon-
On
Does setting RawUsage=0 for a user affect historical usage information,
used by, e.g., sreport?
Jason
On Mon, Jul 27, 2020 at 2:22 PM Paul Edmon wrote:
> You want this handy feature in sacctmgr:
> *RawUsage*= This allows an administrator to reset the raw usage
> accrued to an account. The only
You want this handy feature in sacctmgr:
/RawUsage/=
This allows an administrator to reset the raw usage accrued to an
account. The only value currently supported is 0 (zero). This is a
settable specification only - it cannot be used as a filter to list
accounts.
See:
Dear all,
Apologies for the basic question. I've looked around online for an answer
to this, and I haven't found anything that has helped accomplish exactly
what I want. That said, it is also probable that what I am asking isn't a
best practice, or isn't actually necessary, etc. I'd welcome any
There's an outstanding feature request for that:
https://bugs.schedmd.com/show_bug.cgi?id=8383
While waiting on that, we've taken to injecting it into the job's environment
ourselves in the Lua submit filter.
--Troy
On 7/27/20, 12:45 PM, "slurm-users on behalf of Brian Andrus"
All,
Trying to see the best way to identify the time limit of a job from
within the job.
I had hoped it was one of the SLURM variables, but apparently not.
Is there an easy way outside doing an 'squeue' or 'scontrol' call from
the job to find out the max time of itself?
I hate doing RPC
