Quoting Davis Ford <davisf...@gmail.com>:
Hi Moe, I have tried this..for example the node ORL-APP5 is one of the
nodes that is down.
If your node has Sockets=1 CoresPerSocket=1 ThreadsPerCore=1, then how
can Procs=4?
[root@ORL-APP5 ~]# slurmd -C
NodeName=ORL-APP5 Procs=4 Sockets=1 CoresPerSocket=1 ThreadsPerCore=1
RealMemory=629 TmpDisk=3852
[root@ORL-APP5 ~]# scontrol show node ORL-APP5
NodeName=ORL-APP5 Arch=i686 CoresPerSocket=1
CPUAlloc=0 CPUErr=0 CPUTot=4 Features=(null)
Gres=(null)
NodeAddr=192.168.206.47 NodeHostName=ORL-APP5
OS=Linux RealMemory=629 Sockets=4
State=DOWN* ThreadsPerCore=1 TmpDisk=3852 Weight=1
BootTime=2011-10-31T20:23:06 SlurmdStartTime=2012-01-17T17:42:55
Reason=Low socket*core*thread count [slurm@2012-01-17T17:38:27]
ORL-APP5 is defined in this section of the slurm.conf NodeName:
NodeName=ORL-APP[3,5-6] NodeAddr=192.168.206.[45,47-48] Procs=4
CoresPerSocket=1
ThreadsPerCore=1 RealMemory=300 TmpDisk=200 State=UNKNOWN
Unless I'm missing something, the hardware is a match with the config? In
the slurmctld logfile, I see this is stated about ORL-APP5:
[2012-01-17T17:39:12] error: Node ORL-APP5 has low socket*core*thread count
(1 < 4)
Maybe I'm missing something here? The node has 4 sockets, 4 cpus, and 1
core per socket. The only thing I didn't spec. on the NodeName line was
the 4 sockets. Is this misconfigured?
Thanks,
Davis
