Open MPI logo

Open MPI User's Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Open MPI User's mailing list

Subject: Re: [OMPI users] Uninterruptible processes on 2*4 multicore
From: Jeff Squyres (jsquyres) (jsquyres_at_[hidden])
Date: 2010-02-02 06:34:51

Can you send all the information listed on the getting help page on the ompi web site? Also, information about your application would be helpful.

Sent from my PDA. No type good.


From: users-bounces_at_[hidden] <users-bounces_at_[hidden]>
To: users_at_[hidden] <users_at_[hidden]>
Sent: Tue Feb 02 05:34:28 2010
Subject: [OMPI users] Uninterruptible processes on 2*4 multicore



I experience some stange behaviour on multi-core node of our cluster that I presume is linked to openmpi.

When running for a long time, and several pseudo-nodes of a single multicore node are concerned, one process freezes in a uninterrutible mode (D status) and the others seem to wait for a long time (S status).

Concurrent processes over the the other pseudo-nodes are also frozen in D mode.

When forcing the sleeping processes to sleep (kill -STOP), normal activity of other processes is recovered.

When interrupting blocked process at wakeup, it seems to be blocked in poll_dispatch method, and I guess that comment about multithread must be concerned.

Do you know more about this behaviour ?


Thank a lot,




Nb : I'm using openmpi v1.3 on Linux debian etch distribution

nodes are as following (/proc/cpuinfo):

rocessor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 23
model name : Intel(R) Xeon(R) CPU E5440 @ 2.83GHz
stepping : 10
cpu MHz : 2833.422
cache size : 6144 KB
physical id : 0
siblings : 4
core id : 0
cpu cores : 4
apicid : 0
initial apicid : 0
fpu : yes
fpu_exception : yes
cpuid level : 13
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx lm constant_tsc arch_perfm
on pebs bts rep_good nopl pni monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr dca sse4_1 lahf_lm
bogomips : 5666.84
clflush size : 64
cache_alignment : 64
address sizes : 38 bits physical, 48 bits virtual
power management:




Manager Technique Groupe Calcul Scientifique de PERTUIS

lchailan_at_[hidden] //