Hi,
we are having problems with the configuration of our queuing system.
We use OpenPBS with maui.
Jobs are rejected because of insufficient Swap:
output of checkjob -v :
======================================
checking job 215
State: Idle (User: mk Group: sfbusr Account: [NONE])
WallTime: 0:00:00 (Limit: 0:10:00)
QueueTime: Wed Apr 12 17:05:42
Total Tasks: 1
Req[0] TaskCount: 1 Partition: ALL
Network: [NONE] Memory >= 0 Disk >= 0 Features: [NONE]
Opsys: [NONE] Arch: [NONE] Class: [small 1]
ExecSize: 0 ImageSize: 0
Dedicated Resources Per Task: Procs: 1
TasksPerNode: 1 NodeCount: 0
IWD: [NONE] Executable: [NONE]
QOS: DEFAULT Bypass: 0 StartCount: 0
Partition Mask: [ALL]
Flags: RESTARTABLE
PE: 1.00 StartPriority: 7121
job cannot run in partition DEFAULT (idle procs do not meet
requirements)
idle procs: 14 feasible procs: 0
Rejection Reasons: [Swap : 7][State : 1]
Detailed Rejection Information:
sfb663-1 rejected : Swap
sfb663-2 rejected : Swap
sfb663-3 rejected : Swap
sfb663-4 rejected : Swap
sfb663-5 rejected : Swap
sfb663-6 rejected : Swap
sfb663-7 rejected : Swap
master rejected : State
==========================================
(master rejecting is OK, it has no pbs_mom runnning)
output of checknode:
==========================================
checking node sfb663-1
State: Idle Opsys: DEFAULT Arch: linux
Configured Resources: Procs: 2 Mem: 2 Swap: 5 Disk: 1
Utilized Resources: [NONE]
Dedicated Resources: [NONE]
Speed: 1.00 Load: 0.000
Partition: DEFAULT
Network: [DEFAULT]
Features: [NONE]
Classes: [verylong 2:2][medium 2:2][default 2:2][small 2:2][long 2:2]
node has been in current state for 0:36:01
Reservations:
Total Time: 35:09:32:56 Up: 33:09:19:01 (94.32%) Busy: 0:00:00 (0.00%)
===========================================
what I don't understand is:
Why does it report "Mem: 2 Swap: 5 Disk: 1" ?!
it has 4 GB of physical memory, and 4 GB of swap:
output of free:
==========================================
total used free shared buffers cached
Mem: 4024520 1186708 2837812 0 160996 892004
-/+ buffers/cache: 133708 3890812
Swap: 4080488 0 4080488
==========================================
so the question is: where does the "5" for swap come from? where does
the "2" of Mem come from?
...martin
_______________________________________________
mauiusers mailing list
[email protected]
http://www.supercluster.org/mailman/listinfo/mauiusers
