`squeue`¶

The squeue command allows users to view job and job step information for jobs managed by Slurm in the scheduling queue.

To report jobs for a specific user:

squeue -u user01

JOBID   PARTITION       NAME    USER    ST      TIME    NODES   NODELIST(REASON)
48765   gpu-a100        bash    user01  R       0:19    1       ptolemy-gpu-01

To report jobs for a specific partition:

squeue --partition gpu-a100

JOBID   PARTITION       NAME    USER    ST      TIME    NODES   NODELIST(REASON)
48765   gpu-a100        bash    name    R       0:19    1       ptolemy-gpu-01
48762   gpu-a100        lammps  joey    R       32:01   2       ptolemy-gpu-[01-02]
48768   gpu-a100        vasp    sanders PD      0:00    4       (Resources)

To report the start time of pending jobs:

squeue --partition gpu-a100 --start

JOBID   PARTITION       NAME    USER    ST      START_TIME              NODES   NODELIST(REASON)
48725   gpu-a100        vasp    name    R       2026-01-07T16:53:21     4      (Resources)
48726   gpu-a100        gromacs joey    R       2026-01-08T02:28:47     4      (Priority)
48727   gpu-a100        lammps  sanders PD      2026-01-08T08:00:00     8      (Reservation)

The NODELIST(REASON) field records why the job is not running. Common reasons may include:

Priority – Resources are reserved for higher priority jobs.
Resources – Required resources are in use.
Dependency – Job dependencies have not yet been satisfied.
Reservation – Waiting for the reserved timeslot.
AssociationJobLimit – User or Account job limit reached.
AssociationResourceLimit – User or Account resource limit reached.
AssociationTimeLimit – User or Account time limit reached.
QOSJobLimit – QOS job limit reached.
QOSResourceLimit – QOS resource limit reached.
QOSTimeLimit – QOS time limit reached.

squeue¶

`squeue`¶