Re: [HTCondor-users] Worker node should only accept jobs with rhel9

[Vikrant Aggarwal <ervikrant06@xxxxxxxxx>]

Thanks Todd for your reply.

Yes, I checked the output of the reverse machine. It's showing no issues in slot match making. it's not a negotiator issue, something related to the _expression_ of the worker node because If I change it to "owner" instead ofÂOsversion version, job matchmaking works. Workernode is running 24.0.3 and submitter node running 9.0.17 (again version should also not cause any issue)Â

Job 77.0 has the following attributes:

  TARGET.Osversion = "rhel9"
  TARGET.RequestCpus = 1
  TARGET.RequestDisk = 3
  TARGET.RequestMemory = 1024

The Requirements _expression_ for this slot reduces to these conditions:

   ÂClusters
Step  ÂMatched ÂCondition
----- Â-------- Â---------
[12] Â Â Â Â Â1 ÂisUndefined(IcingaAlerts)
[14] Â Â Â Â Â1 ÂisUndefined(DontStartJobs)
[16] Â Â Â Â Â1 ÂisUndefined(Benchmarking)
[34] Â Â Â Â Â1 ÂTARGET.Osversion == "rhel9"
[36] Â Â Â Â Â1 ÂWithinResourceLimits

Negotiator log shows "No match found". It's strange. Do I need to addÂOsversion anywhere? osrel is already present in STARTD_ATTRS.Â

Thanks & Regards,

Vikrant Aggarwal

On Tue, Mar 11, 2025 at 1:33âPM Todd L Miller via HTCondor-users <htcondor-users@xxxxxxxxxxx> wrote:

> requirements = target.osrel == "rhel9"
> +Osversion = "rhel9"

    I haven't been able to reproduce your problem in my testing,
although I had to use `OpSysAndVer` for the attribute name because my
startds don't advertise `osrel`. I suggest using `condor_q -bet` (and the
reverse and machine flags) to figure out why jobs and/or machines aren't
matching machines/and or jobs.

-- ToddM
_______________________________________________
HTCondor-users mailing list
To unsubscribe, send a message to htcondor-users-request@xxxxxxxxxxx with a
subject: Unsubscribe

Join us in June at Throughput Computing 25: https://osg-htc.org/htc25

The archives can be found at: https://www-auth.cs.wisc.edu/lists/htcondor-users/