There are 2 parameters that control coredumps on Linux.
(1) Core dump support has to be enabled in the kernel.
I'll have to look at the sources, I think there's a simple way
to look in /proc to see if it's enabled.
Of course, after you do number (2) and generate a core file,
it's obvious :-)
'cat /proc/sys/kernel/core_pattern' shows where core files are stored.
(2) The user's 'ulimit' has to be set to allow core files.
If you use 'bash', "ulimit -c" will show you current allowed core size.
("ulimit -a" shows all values, look at your shell's man page for more
info)
I recommend 'ulimit -c unlimited'.
On Fri, 14 Jan 2005 04:31:39 -0800
Guy Harris <gharris@xxxxxxxxx> wrote:
> März, Frank wrote:
>
> > Using the file in tethereal give the following output.
> >
> > /tethereal -r moritz.eth -R "(gtp)" -w test.eth
> > 965444
> > ** ERROR **: file proto.c: line 2072 (alloc_field_info): assertion
> > failed: (*length >= 0)
> > //aborting...
> > Abgebrochen
> >
> > [root@box01 root]#
>
> Can you run Tethereal on the file as yourself, rather than as root, and
> get a core dump?
>
> Does that version of Linux save core dumps somewhere other than in the
> working directory of the process? (I seem to remember some option to
> configure a "core dump name" string in Linux - I forget whether it can
> change the directory into which core dump files are saved or not.)
>
> If not, can you send us the file that causes the crash?
>
> _______________________________________________
> Ethereal-dev mailing list
> Ethereal-dev@xxxxxxxxxxxx
> http://www.ethereal.com/mailman/listinfo/ethereal-dev
>
--
Rich Coe richard.coe@xxxxxxxxxx
General Electric Healthcare Technologies
Global Software Platfroms, Computer Technology Team
