squeue - list jobs in the queuing system
sstat - display statistics of a running job
sacct - display information about completed jobs
Information about the status of tasks can be obtained by command squeue .
Calling the command without providing additional parameters will display information about all tasks.
For information on the status of your tasks, please specify the option --user=nazwa_użytkownika or -u nazwa_użytkownika:
abcd@ui: ~ $ squeue -u abcd
JOBID PARTITION NAME USER ST TIME NODES NODELIST(REASON)
142 test bash abcd R 0:18 1 wn1701
or longer version:
abcd@ui: ~ $ squeue -u abcd --long
Mon Jul 05 10:44:43 2021
JOBID PARTITION NAME USER STATE TIME TIME_LIMI NODES NODELIST(REASON)
142 test bash abcd RUNNING 8:23 UNLIMITED 1 wn1701
Where:
| JOBID | Task number |
| PARTITION | Partition name |
| %1 | Job Name |
| USER | Owner of the task |
| STATE | Task Status (see below) |
| TIME | Time of execution |
| (PHP 3, PHP 4) | Maximum task execution time (INFINITY, means unlimited time) |
| NODES | Number of node divided task |
| NODELIST | Name of nodes on which the task is run |
The status may be:
| Status | Code | Description |
|---|---|---|
| COMPLETED | CD | The task has been successfully completed |
| COMPLETING | CG | The job is at the end of the job. |
| FAILED | F | The job ended with a mistake. |
| PENDING | PD | Expects resources to be allocated |
| PREEMPTED | PR | The task has been exonerated by another task |
| RUNNING | R | Started (on-the-job) |
| SUSPENDED | S | Detention of resources |
The command scancel is used to terminate a job and remove it from the queue. To delete a job, you need to know its JOBID, which can be read using the squeue command.
abcd@ui: ~ $ squeue -u abcd
JOBID PARTITION NAME USER ST TIME NODES NODELIST(REASON)
613 test bash abcd R 0:05 1 wn1701
614 test bash abcd R 1:05 1 wn1702
615 test bash abcd R 1:05 1 wn1703
abcd@ui ~>scancel 613
abcd@ui ~>squeue -u abcd
JOBID PARTITION NAME USER ST TIME NODES NODELIST(REASON)
614 test bash abcd R 1:05 1 wn1702
615 test bash abcd R 1:05 1 wn1703
We can also delete several tasks with one command
abcd@ui: ~ $ squeue -u abcd
JOBID PARTITION NAME USER ST TIME NODES NODELIST(REASON)
613 test bash abcd R 0:05 1 wn1701
614 test bash abcd R 1:05 1 wn1702
615 test bash abcd R 1:05 1 wn1703
abcd@ui ~>scancel 613 615
abcd@ui ~>squeue -u abcd
JOBID PARTITION NAME USER ST TIME NODES NODELIST(REASON)
614 test bash abcd R 1:05 1 wn1702
Or all tasks
abcd@ui: ~ $ squeue -u abcd
JOBID PARTITION NAME USER ST TIME NODES NODELIST(REASON)
613 test bash abcd R 0:05 1 wn1701
614 test bash abcd R 1:05 1 wn1702
615 test bash abcd R 1:05 1 wn1703
abcd@ui: ~ $ scancel -u abcd
abcd@ui: ~ $ squeue -u abcd
JOBID PARTITION NAME USER ST TIME NODES NODELIST(REASON)
More detailed job status can be obtained using the sstat command. The command requires the JOBID parameter obtained using the squeue command. Due to the large number of displayed columns, it is recommended to use the --fields= option.
Example sstat command:
abcd@ui: ~ $ sstat 581969
JobID MaxVMSize MaxVMSizeNode MaxVMSizeTask AveVMSize MaxRSS MaxRSSNode MaxRSSTask AveRSS MaxPages MaxPagesNode MaxPagesTask AvePages MinCPU MinCPUNode MinCPUTask AveCPU NTasks AveCPUFreq ReqCPUFreqMin ReqCPUFreqMax ReqCPUFreqGov ConsumedEnergy MaxDiskRead MaxDiskReadNode MaxDiskReadTask AveDiskRead MaxDiskWrite MaxDiskWriteNode MaxDiskWriteTask AveDiskWrite TRESUsageInAve TRESUsageInMax TRESUsageInMaxNode TRESUsageInMaxTask TRESUsageInMin TRESUsageInMinNode TRESUsageInMinTask TRESUsageInTot TRESUsageOutAve TRESUsageOutMax TRESUsageOutMaxNode TRESUsageOutMaxTask TRESUsageOutMin TRESUsageOutMinNode TRESUsageOutMinTask TRESUsageOutTot
------------ ---------- -------------- -------------- ---------- ---------- ---------- ---------- ---------- -------- ------------ -------------- ---------- ---------- ---------- ---------- ---------- -------- ---------- ------------- ------------- ------------- -------------- ------------ --------------- --------------- ------------ ------------ ---------------- ---------------- ------------ -------------- -------------- ------------------ ------------------ -------------- ------------------ ------------------ -------------- --------------- --------------- ------------------- ------------------- --------------- ------------------- ------------------- ---------------
581969.0 864K r21c01b06 0 864K 840K r21c01b06 0 840K 0 r21c01b06 0 0 00:00:00 r21c01b06 0 00:00:00 1 3.41M Unknown Unknown Unknown 0 752302 r21c01b06 0 752302 18100 r21c01b06 0 18100 cpu=00:00:00,+ cpu=00:00:00,+ cpu=r21c01b06,ene+ cpu=00:00:00,fs/d+ cpu=00:00:00,+ cpu=r21c01b06,ene+ cpu=00:00:00,fs/d+ cpu=00:00:00,+ energy=0,fs/di+ energy=0,fs/di+ energy=r21c01b06,f+ fs/disk=0 energy=0,fs/di+ energy=r21c01b06,f+ fs/disk=0 energy=0,fs/di+
And the same example with the use of the --fields
abcd@ui ~ $ sstat --fields=JobID,MaxVMSize,MaxVMSizeNode,MaxVMSizeTask,AveCPU 581969
JobID MaxVMSize MaxVMSizeNode MaxVMSizeTask AveCPU
------------ ---------- -------------- -------------- ----------
581969.0 872K r21c01b06 0 00:00:00