[torqueusers] Maui dies immediately on job submission

Lippert, Kenneth B. Kenneth.Lippert at alcoa.com
Tue Feb 27 13:01:09 MST 2007


Thanks for the pointer, here is the stack trace...


gdb maui
Detaching after fork from child process 31683.

Program received signal SIGSEGV, Segmentation fault.
MStatUpdateActiveJobUsage (J=0x2231ec0) at MStats.c:1033
1033      if ((RQ->NAccessPolicy == mnacSingleJob) &&
(gdb) where
#0  MStatUpdateActiveJobUsage (J=0x2231ec0) at MStats.c:1033
#1  0x0000000000466d41 in MQueueAddAJob (J=0x2231ec0) at MQueue.c:1605
#2  0x0000000000476643 in MJobStart (J=0x2231ec0) at MJob.c:6485
#3  0x0000000000465e34 in MQueueScheduleIJobs (Q=0x7fbfff1450,
P=0xcb37a8) at MQueue.c:966
#4  0x000000000042f0a5 in MSchedProcessJobs (OldDay=Variable "OldDay" is
not available.
) at MSched.c:7001
#5  0x0000000000405b4d in main (ArgC=1, ArgV=Variable "ArgV" is not
available.
) at Server.c:192
(gdb)



 

-----Original Message-----
From: Josh Butikofer [mailto:josh at clusterresources.com] 
Sent: Tuesday, February 27, 2007 2:49 PM
To: Lippert, Kenneth B.
Cc: torqueusers at supercluster.org
Subject: Re: [torqueusers] Maui dies immediately on job submission

Look at the following URL (section 14.6.2.1) for instructions on how to
properly start Maui under a
debugger:

http://www.clusterresources.com/products/maui/docs/14.6troubleshootingsy
stemerrors.shtml

-- 
Joshua Butikofer
Cluster Resources, Inc.

josh at clusterresources.com
Voice: (801) 717-3707
Fax:   (801) 717-3738
--------------------------


Lippert, Kenneth B. wrote:
>  
> 
> 
> You are overcomplicating things :)
> 
> The probably is that maui is segfaulting and never gets to start a
job.
> 
> We just need a backtrace of maui segfaulting in gdb.
> 
> 
> 
> 
> 
> okay,  but I'm not sure how to do that.  If I start Maui from gdb, it
> just forks and gdb says "completely normally".  I can see the process
> running with "ps".  When i submit a job from another window, the Maui
> process dies, but I don't see any messages in the gdb session.
> 
> Thanks for the quick response!
> 
> -k
> _______________________________________________
> torqueusers mailing list
> torqueusers at supercluster.org
> http://www.supercluster.org/mailman/listinfo/torqueusers


More information about the torqueusers mailing list