On Thu, Jul 19, 2001 at 01:01:26PM +0300, Henri Kaukola wrote:
> Emiliano wrote:
> >
> > Henri Kaukola wrote:
> >
> > > I'm running the latest CVS snapshot (1.4.1-cvs-20010719/SG) on my
> > > system. I'm trying to export a site I've built to another box. However,
> > > Repligard keeps segfaulting on export, both as root and as the mysql
> > > user.
> > >
> > > I've attached my repligard.conf, and can also send the core file
> > > on request.
> > >
> > > $ repligard -c /tmp/repligard.conf -a -e /tmp/test.xmlgz
> >
> > Run this in gdb:
> >
> > $ gdb repligard
> > [bla bla]
> > (gdb) run -c /tmp/repligard.conf -a -e /tmp/test.xmlgz
> > [bla bla]
> > ..segfault...
> > (gdb) backtrace
> >
> > The output of this last command will tell us where it's segfaulting...
> > and may also tell us where other parts of midgard are having problems.
>
>
> Right, did that and here's the whole output:
>
> (gdb) run -c /tmp/repligard.conf -a -e /tmp/text.xml.gz
> Starting program: /usr/local/bin/repligard -c /tmp/repligard.conf -a -e
> /tmp/text.xml.gz
> Message: Reading config file /tmp/repligard.conf
> Message: Exporting to /tmp/text.xml.gz
>
> Program received signal SIGSEGV, Segmentation fault.
> 0x804e9bd in rpd_write_object_tree (value=0x807cb50,
> user_data=0xbffff8d0)
> at export.c:505
> 505 if ((object_type->type == R_TREE_OBJECT) && exp_data->expandtree)
Hm... It means that entry->type does not exist in schema resulting in
object_type equal to NULL. What does 'bt full' say in gdb? Not 'backtrace'
but 'bt full' and also what does 'print entry->type' says?
--
/ Alexander Bokovoy
$ cat /proc/identity >~/.signature
`Lead software developer and analyst for SaM-Solutions Ltd.`
---
"The medium is the message."
-- Marshall McLuhan
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
