[SCore-users-jp] Re: [SCore-users] SCore's strange error message...

kameyama @ pccluster.org kameyama @ pccluster.org
2004年 9月 10日 (金) 20:06:25 JST


In article <E1C5iyV-0006a4-00 @ mailhub1.shef.ac.uk> "Mr. H. Xia" <h.xia @ sheffield.ac.uk> wrotes:
> I've met this kind of error message for many times. When I specify less nodes
> , it seems no problems at all but when I use more nodes, it comes out like th
> e following.
> 
> SCOUT: Spawning done.
> <0:0> SCORE: 100 nodes (50x2) ready.
> <37:1>SCore: *** SIGNAL EXCEPTION eip=0x081953f7, cr2=0x 98f7b70 ***
> <37:1>SCore: gs=0x0000, fs=0x0000, es=0x002b, ds=0x002b
> <37:1>SCore: edi=0x0860b020, esi=0x00000000, ebp=0x00000014, esp=0xbfffea14
> <37:1>SCore: ebx=0xbfffeab0, edx=0x098f7b84, ecx=0xba930e95, eax=0x08b84588
> <37:1>SCore: trapno=0x0000000e, err=0x00000006, eip=0x081953f7, cs=0x0023
> <37:1>SCore: esp_at_signal=0xbfffea14, ss=0x002b, oldmask=0x80000000, cr2=0x0
> 98f7b70
> <18:1> Trying to attach GDB (DISPLAY=snowdon.leeds.ac.uk:25.0): Exception sig
> nal (SIGSEGV)
> SCORE: Program aborted.

This means the program is segmentation fault at 0x081953f7.
If you run scrun with debug option
    % scrun -nodes=100,debug ...
and your X server can access snowdon.leeds.ac.uk:25.0 from the compute hosts,
gdb is excuted at the time.

If X server dose not access from the compute host,
please unexport DISPLAY environment variable, and scrun with debug optiion.
So you can get backtrace where the program is segmentation fault.


                       from Kameyama Toyohisa
_______________________________________________
SCore-users mailing list
SCore-users @ pccluster.org
http://www.pccluster.org/mailman/listinfo/score-users



SCore-users-jp メーリングリストの案内