Hi All,
Apologies for cross posting, but I wasn't sure where it would fit
best. I have a problem that a job is stuck in the queued state in
torque, i.e. qstat shows the state as queued, but maui says the job is
active, i.e. showq lists it as active and running. In showq the job
runtime is not counting down and the job is definitely not running on
any of the nodes it is supposed to. Diagnose -j says:
Name State Par Proc QOS WCLimit R Min User
Group Account QueuedTime Network Opsys Arch Mem Disk
Procs Class Features
21 Running DEF 144 DEF 00:02:00 1 144 mcdiypp2
nmrc - 00:28:33 [NONE] [NONE] [NONE] >=0 >=0
NC0 [short_2h:1] [NONE]
And qstat -f says
Job Id: 21.steel.mib.man.ac.uk
Job_Name = qsubtest.com
Job_Owner = [EMAIL PROTECTED]
job_state = Q
queue = short_2h
server = steel.mib.man.ac.uk
Checkpoint = u
ctime = Wed Sep 24 14:08:27 2008
Error_Path = steel.mib.man.ac.uk:/home/mcdiypp2/qsubtest.com.e21
Hold_Types = n
Join_Path = n
Keep_Files = n
Mail_Points = a
mtime = Wed Sep 24 14:37:54 2008
Output_Path = steel.mib.man.ac.uk:/home/mcdiypp2/qsubtest.com.o21
Priority = 0
qtime = Wed Sep 24 14:08:27 2008
Rerunable = True
Resource_List.nodect = 18
Resource_List.nodes = 18:ppn=8
Resource_List.walltime = 00:02:00
Variable_List = PBS_O_HOME=/home/mcdiypp2,PBS_O_LANG=en_GB.UTF-8,
PBS_O_LOGNAME=mcdiypp2,
PBS_O_PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/b
in:/usr/games:/opt/torque/2.3.3/bin:/opt/torque/2.3.3/sbin:/opt/maui/3
.2.6p19/bin:/opt/maui/3.2.6p19/sbin:/opt/openmpi/1.2.6/bin,
PBS_O_MAIL=/var/mail/mcdiypp2,PBS_O_SHELL=/bin/bash,
PBS_SERVER=steel.mib.man.ac.uk,PBS_O_HOST=steel.mib.man.ac.uk,
PBS_O_WORKDIR=/home/mcdiypp2,PBS_O_QUEUE=route
etime = Wed Sep 24 14:08:27 2008
exit_status = -3
submit_args = qsubtest.com
start_time = Wed Sep 24 14:08:28 2008
start_count = 1756
I don't understand the priority being zero, as maui lists the
startpriority as 60. Something appears to be not communicating
somewhere. Could someone shed some light on it?
Philip Peartree
_______________________________________________
mauiusers mailing list
[email protected]
http://www.supercluster.org/mailman/listinfo/mauiusers
