Re: [Netatalk-devel] segfault (was: still wrong mandir)

[Carsten N. <de...@sn...>]

On Mon, 31 Dec 2001, Joe Clarke wrote:
> [snip]
>=20
> > >
> > > > I've still got problems to start atalkd (1.5pre8 and later) on a =
SuSE 6.3, kernel 2.2.13, via-rhine
> > > > NIC driver.
> > >
> > > What kind of problems?
> >
> > When I start atalkd I get the following error:
> >
> > =09AppleTalk not up! Check your syslog for the reason. Child died.
> >
> > /var/log/messages stays untouched. :-(
> >
> > strace -f shows that a child died because of a segment violation.
> > I could sent you strace's output.
> >
>=20
> Can you produce a core dump to run gdb on?  What about the getiface.c f=
ile
> I sent you?  Did that make a difference?

Sorry, didn't tell you.
Your getiface.c makes no difference.

Is there a way to enable core-dumps via the /proc fs or otherwise?

>=20
> >
> > Another thing:
> >
> > My latest CVS update gives a compilation error in line 159 of etc/ata=
lkd/config.c.
> > This was introduced by your latest patch.
> >
> >
> >     /* We store our buffer pointer in argv, too, so we can free it la=
ter.
> >      * (But don't tell anyone.)
> >      */
> >     argv[ argv ] =3D buffer;                            <---- LINE 15=
9
> >
> > You probably mean `argv[argc]'? ;-))
>=20
> This should be fixed, but I also added some other fixes that might caus=
e
> core dumps.  I haven't tested it, I just made sure it killed the compil=
e
> errors and warnings.

Yep, saw Joe's latest posting.
I already did a cvs update.

>=20
> Joe
>=20
> >
> > >
> > >  - Sebastian
> >
> > Thank you,
> > kind regards,
> > happy new year
> >
> > =09Carsten

Thank you

=09Carsten