Alternate Formatting of squeue

This KB Article References: High Performance Computing
This Information is Intended for: Instructors, Researchers, Staff, Students
Created: 11/21/2023 Last Updated: 06/19/2024

We now have a new command, fqueue, that allows you to run squeue with different format featuring a larger space for Partition and Name, allowing easier identification of the queue and the job. However, as you will see below, the fields on time has been cut so that the seconds are no longer seen after a job has been running for more than a day.

The current output of squeue:

             JOBID PARTITION     NAME     USER ST       TIME  NODES NODELIST(REASON)
           1359646      v100 Test_PP1 hpc_supp PD       0:00      1 (PartitionTimeLimit)
           1360070 large-28c    test2      hpc PD       0:00     40 (QOSMaxJobsPerUserLimit)
           1360150      p100 Hit9-Sgl      hpc PD       0:00      1 (Resources)
           1360183 short-28c NaCN-fre hpc_supp PD       0:00      1 (Priority)
           1358752  gpu-long apo_SglA hpc_supp PD       0:00      1 (Dependency)
           1358739 medium-28 parallel      hpc PD       0:00     28 (BadConstraints)
           1359361 extended- CC753210 hpc_supp  R 4-23:15:52      2 sn[126-127]
           1360167 long-28co PH3O-fre hpc_supp  R      30:11      1 sn004
       1359405_226 extended- mcmc_emu hpc_supp  R 4-19:19:30      1 sn117

 


Output of fqueue:

JOBID          PARTITION           NAME           USER     ST TIME    NODES NODELIST(REASON)
1359646        v100                Test_PP1_Ojas  hpc_supp PD 0:00    1     (PartitionTimeLimit)
1360070        large-28core        test2          hpc_supp PD 0:00    40    (QOSMaxJobsPerUserLimit)
1360150        p100                Hit9-Sgl1      hpc      PD 0:00    1     (Resources)
1360183        short-28core        NaCN-freq-hf-l hpc_supp PD 0:00    1     (Priority)
1358752        gpu-long            apo_SglA_6     hpc_supp PD 0:00    1     (Dependency)
1358739        medium-28core       parallel_job   hpc      PD 0:00    28    (BadConstraints)
1360166        large-28core        testmpi_large_ hpc      R  35:18   57    sn[060-066,086-102,118-125,128-152]
1359361        extended-28core     CC753210       hpc_supp R  4-23:15 2     sn[126-127]
1359835        v100                HitB1-SglA_5   hpc_supp R  4:54:02 1     sn-nvda1
1360167        long-28core         PH3O-freq-hf-l hpc_supp R  29:52   1     sn004
1359405_226    extended-28core     mcmc_emu       hpc_supp R  4-19:19 1     sn117

 

To run fqueue, first confirm that you loaded the slurm module:

module load slurm

 

Then run,

fqueue

with any other flags you would normally use with squeue, such as -u or --job.

 

If none of these formats are to your liking, you can always adjust the output of squeue by using --format flag.
For example, the fqueue command currently formats the output by specifying:

--format="%14i %19P %14j %8u %2t %7M %5D %R" "$@"


More information can be found on the official documentation or man page for squeue as always.

More general information on how to check a Slurm job's status can be found here.

For More Information Contact


IACS Support System

Still Need Help? The best way to report your issue or make a request is by submitting a ticket.

Request Access or Report an Issue