[Mauiusers] maui crashes with segv (version 3.2.6.p16)

[Steve Traylen]

We've increased the following so far with out disaster:

MAX_MCLASS from 16 to 64 allowing increasing the number of queues.

MMAX_JOB from 4096 to 32768 and MAX_MJOB_TRACE from 4096 to 32768.  
This allows the number of jobs in the system to be increased.


MAX_MRES from 1024 to 8192. This limit was reached when we had more  
than 1024 job slots.


    Steve

On Oct 31, 2006, at 10:13 AM, Bas van der Vlies wrote:

> maui version: latest snapshot
>
> increased MAX_JOB in msched.h to:
>  #define MMAX_JOB          32768
>  #define MAX_MJOB       MMAX_JOB
>
>
>
> We have a large number of jobs in the queue for this moment 20,000.  
> When  i start maui it crashes with a segv.
>
> Program received signal SIGSEGV, Segmentation fault.
> 0x080cd7b5 in MJobGetStartPriority (J=0x0, PIndex=0,  
> Priority=0xbfc61220, Mode=0, BPtr=0x0, BSpace=0x0) at MPriority.c:781
> (gdb) bt
> #0  0x080cd7b5 in MJobGetStartPriority (J=0x0, PIndex=0,  
> Priority=0xbfc61220, Mode=0, BPtr=0x0, BSpace=0x0) at MPriority.c:781
> #1  0x080af317 in MQueuePrioritizeJobs (Q=0xb4fc680,  
> JobIndex=0xbfc61270) at MQueue.c:224
> #2  0x080d1e4f in MQueueSelectAllJobs (Q=0xb4fc680, PLevel=4,  
> P=0x8735d80, JIList=0xbfcc27b0, DoPrioritize=1, UpdateStats=1,  
> TrackIdle=1, Msg=0x0) at MPolicy.c:699
> #3  0x08078e97 in MSchedProcessJobs (OldDay=0x0,  
> GlobalSQ=0xbfce27b0, GlobalHQ=0xbfcc27b0) at MSched.c:6909
> #4  0x0804aa41 in main (ArgC=2, ArgV=0xbfd027b0) at Server.c:192
>
> My question is are there other sides that run torque/maui with a  
> large number of jobs in the queue?
>
> Regards
> -- 
> --
> ********************************************************************
> *                                                                  *
> *  Bas van der Vlies                     e-mail: basv at sara.nl      *
> *  SARA - Academic Computing Services    phone:  +31 20 592 8012   *
> *  Kruislaan 415                         fax:    +31 20 6683167    *
> *  1098 SJ Amsterdam                                               *
> *                                                                  *
> ********************************************************************
> _______________________________________________
> mauiusers mailing list
> mauiusers at supercluster.org
> http://www.supercluster.org/mailman/listinfo/mauiusers

-- 
Steve Traylen
steve.traylen at cern.ch
CERN, IT-GD-OPS.



-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 1569 bytes
Desc: not available
Url : http://www.supercluster.org/pipermail/mauiusers/attachments/20061031/b560bf6d/smime.bin