[SCore-users-jp] [SCore-users] PM errors

[Shinji Sumimoto]

Hi, Nick.

From: Nick Birkett <nrcb at streamline-computing.com>
Subject: [SCore-users-jp] [SCore-users] PM errors
Date: Fri, 6 Sep 2002 07:26:27 +0100
Message-ID: <200209060626.g866Qij02645 at zeralda.streamline.com>

nrcb> Hi, we having been running SCore for 18 months now using PBS batch scheduling
nrcb> and running over Myrinet 2000, but are now getting some errors with a code 
nrcb> that was working ok (DLPOLY chemistry code):
nrcb> 
nrcb> > SCore-D 4.0 connected.
nrcb> > <3> ULT:SYSCALLPANIC(../recv.c:85) PM Error (pmReceive) (32:Broken
nrcb> > pipe)
nrcb> > <5> SCore-D:WARNING Some job(s) will not stop (4 more retry)
nrcb> > <5> SCore-D:WARNING Force to stop JOB 1
nrcb> > <5> SCore-D:WARNING Some job(s) will not stop (3 more retry)
nrcb> > <5> SCore-D:WARNING Force to stop JOB 1
nrcb> > <5> SCore-D:WARNING Some job(s) will not stop (2 more retry)
nrcb> > <5> SCore-D:WARNING Force to stop JOB 1
nrcb> > <5> SCore-D:WARNING Some job(s) will not stop (1 more retry)
nrcb> > <5> SCore-D:WARNING Force to stop JOB 1
nrcb> > <5> SCore-D:WARNING Failed to stop job(s).
nrcb> > <5> SCore-D:WARNING Force to kill JOB 1
nrcb> 
nrcb> Anyone know what is the problem and how i can fix it ?

This error means the Myrinet NIC on node 3 has reset by timeout on
packet receiving. If the error is not occurred again, you do not have
to care about the error. If the error is occurred again, the error may
come from hardware problems.

The description of the timeout is described LANai 9 data sheet
provided by Myricom web pages.

Shinji.

nrcb>  I know it is an old version of SCore, but this is a heavily loaded system and
nrcb> we cannot upgrade just yet.
nrcb> 
nrcb> 
nrcb> Cheers,
nrcb> 
nrcb> Nick
nrcb> 
nrcb> _______________________________________________
nrcb> SCore-users mailing list
nrcb> SCore-users at pccluster.org
nrcb> http://www.pccluster.org/mailman/listinfo/score-users
nrcb> _______________________________________________
nrcb> SCore-users-jp mailing list
nrcb> SCore-users-jp at pccluster.org
nrcb> http://www.pccluster.org/mailman/listinfo/score-users-jp
nrcb> 
------
Shinji Sumimoto, Fujitsu Labs