CLUSTER STATUS

Updated: April 1, 2018

Queue Status:

To view queue status:

Use qstat -f

more information on qstat here: http://docs.adaptivecomputing.com/torque/4-1-3/Content/topics/commands/qstat.htm

State and meaning:

d: disabled

r: running

E: error

a: alarm

u: unreachable

more information on queue states here: http://www.softpanorama.org/HPC/Grid_engine/Queues/queue_states.shtml

Fixes:

'dr' state: qdel -f <jobid> - force kill a job (good for jobs in 'dr' state)

if the job still remains in 'dr' state have the administrator force kill the job

'E' state: qmod -c <queue name> - clears the error state ('E') for a queue

'au' state: Attempt a soft PXE reboot of node, if that is not successful the node will have to be hard booted.

qmod options:

qmod [options] [job or queue]

'-d' disables the specified queue

'-e' enables the specified queue (changes state from 'd')

more information on qmod here: https://linux.die.net/man/1/qmod