Re: Coredump question

From: Trey T. Morita (tmorita@texas.net)
Date: 04/19/96


On Fri, 19 Apr 1996, Daniel W. Burke wrote:

> At 05:30 PM 4/19/96 EDT, you wrote:
> >
> >	We just upgraded our system to Linux 1.2.13 and now whenever the 
> >mud dumps a corefile, if coredumpsize is unlimited we get a 330+ MB core, 
> >if we set the size to 6000 kbytes we get a file of EXACTLY 6000 kbytes, 
> >and on top of that gdb says:
> >
> >Couldn't fetch registers from core file: File in wrong format.
> >Couldn't fetch registers from core file: File in wrong format.
> >#0 0x0 in ?? ()
> >(gdb)
> >
> >	Any ideas on whats going wrong?  So far we've only installed the 
> >a, d, n, and the disk with the Linux source on it.  Did we forget 
> >something _VERY_ important?
> >
> >	Many thanks in advance,
> >
> >	- Sean
> 
> I've been trying to figure out the EXACT same error......
> well, I'm not getting 300meg cores, but I'm getting that error from gdb :(
> 
> SO far anyone I've asked hasn't had a clue :(
> 
I understand this is a bug in gdb in the Slackware 3.0 distribution. Try 
finding a newer version of gdb. I'm running an old distribution (pre 2.3) 
with the kernel patched up from 1.1.59 to 1.2.13, and my gdb has always 
worked. As a temporary work-around, run the MUD from within gdb, and try 
to catch those crash bugs as they happen.

Trey



This archive was generated by hypermail 2b30 : 12/18/00 PST