Job status
To see all jobs on Charlie (running and queued), use the
qstat -a command (/opt/pbs/bin/qstat). It will generate a table with the job id, name, user, and resource requests (nodes, cpus/tsk, memory, time).Option | Description |
-1 | Add as last argument to keep each job to one line (used with -n and -s) |
-a | Show additional information, including requested resources |
-f <job id> | Show full information about job |
-J | Limit status information to job arrays. |
-n | Show node that job is running on (if used with -a, use -n after -a) |
-p | Show percentage complete instead of time used column (cannot be used with -a) |
-q | Show all queues and the total requested resources for all jobs in each queue (do not use with other options) |
-s | Show job start time, node, cpus, and memory for running jobs. Use with -1 to display output on one line per job |
-t | Show status information for jobs, job arrays, and sub‐ jobs |
-T | Show estimated start time instead of elapsed time |
-u <username> | Show jobs for a specific user |
-w | Use wider fields (up to 8 characters wide instead of 4) |
-x | Show status information for finished and moved jobs |
The status column uses one letter to represent the status of a job. Use the table below to as a reference.
Status | Description |
E | Exiting after having run |
F | Finished |
H | Held |
Q | Queued |
R | Running |
S | Suspended (contact admin to resume) |
W | Waiting for submitter-assigned start time to be reached |
X | Completed execution or has been deleted |
# Show job status and requested resources
qstat -a
# In addition to qstat -a output, shows which node a job is running on
qstat -an1
# Show all queues and the total requested resources for all jobs in each queue
qstat -q
# Show all jobs for user hbigelow
qstat -u hbigelow
# Show all information for job with job id 123456
qstat -f 123456
Last modified 2yr ago