checkjob  jobID

Check the status of a job, identified by its Maui Job ID. This command produces output somewhat overcrowded with information; in many cases, it is better to use diagnose -j.
Example:

[140] bourbaki:/home/ai2/gregory>checkjob gregory.127


checking job gregory.127

State: Running
Creds:  user:gregory  group:user  class:Batch  qos:med
WallTime: 00:00:22 of 00:05:00
SubmitTime: Sun Apr 11 15:25:08
  (Time Queued  Total: 00:00:21  Eligible: 00:00:21)

StartTime: Sun Apr 11 15:25:29
Total Tasks: 1

Req[0]  TaskCount: 1  Partition: DEFAULT
Network: ETHER  Memory >= 0  Disk >= 0  Swap >= 0
Opsys: Linux  Arch: i386  Features: [NONE]
Allocated Nodes:
[.8:1]


IWD: /cluster-home/gregory/testing/grab  Executable:  /var/spool/clubmask/gregory.127
Bypass: 0  StartCount: 1
PartitionMask: [ALL]
Flags:       ADVRES:gregoryLIST16640.0 BYNAME PREEMPTOR

Reservation 'gregory.127' (-00:00:22 -> 00:04:38  Duration: 00:05:00)
PE:  1.00  StartPriority:  50

Explanation: This job is currently running in QOS "med". It is using a single CPU on node 8. Its declared wallclock time limit is 5 minutes, of which it has used so far 22 seconds. Its working directory is /cluster-home/gregory/testing/grab.
One more thing to note is the ADVRES flag; if shown, it tells you the reservation with which the job is associated.

See also: diagnose -j,showq,cmwait

Top