[torqueusers] New Nodes

Brett Ellis ellis at cs.utk.edu
Wed Sep 21 09:27:44 MDT 2005

I'm sending this to torqueusers first because -W attributes specified
in maui.cfg work.

As short a description as I can make it.  Let's for this example
say I have a 2 node cluster x86 running debian sarge, specified as
follows in the PBS/server_priv/nodefile

foo01 np=2 x86 foo_cluster
foo02 np=2 x86 foo_cluster

Now I've added another 2 node cluster which are
amd64's running ubuntu, so I've made my

foo01 np=2 x86 foo_cluster
foo02 np=2 x86 foo_cluster
bar01 np=2 amd64 bar_cluster
bar02 np=2 amd64 bar_cluster

Restarted torque and maui daemons appropriately.

Now here's what happens

qsub -I -l nodes=bar01:ppn=2+bar02:ppn=2

   Works as advertised

qsub -I -l nodes=2:ppn=2:x86

   Works as advertised

qsub -I -l nodes=2:ppn=2:amd64

   Hangs and goes into Maui's deferred pile.

Maui reports

checking job 314

State: Idle  EState: Deferred
Creds:  user:joe  group:joe  class:batch  qos:DEFAULT
WallTime: 00:00:00 of 1:00:00
SubmitTime: Wed Sep 21 11:23:36
   (Time Queued  Total: 00:00:01  Eligible: 00:00:01)

StartDate: 00:00:01  Wed Sep 21 11:23:38
Total Tasks: 2

Req[0]  TaskCount: 2  Partition: ALL
Network: [NONE]  Memory >= 0  Disk >= 0  Swap >= 0
Opsys: [NONE]  Arch: [NONE]  Features: [amd64]

IWD: [NONE]  Executable:  [NONE]
Bypass: 0  StartCount: 0
PartitionMask: [ALL]
Flags:       RESTARTABLE

job is deferred.  Reason:  NoResources  (cannot create reservation for 
job '314' (intital reservation attempt)

A checknode shows the following feature line

Features:   [amd64][bar_cluster]

Can anyone point me in the right direction to troubleshoot this?


More information about the torqueusers mailing list