JP Rosevear (jpr@cr168790-a.nmkt1.on.wave.home.com)
Thu, 29 Apr 1999 01:13:00 +0000
Dan Burcaw wrote:
> We're actually thinking it's a glibc issue...
> Looking into it futhur.
I'm curious as to why you think this and where the code is that you think maybe
causing the problem (its not that I don't believe you, its just that I want to
understand).
-JP
>
> On Thu, 29 Apr 1999, JP Rosevear wrote:
>
> > Tracing back further I believe that the error is caused by a tuple pointer
> > being null when returning from the system cache - could this be some sort
> > of PPC related compile error (with gcc that is). Is there possibly a
> > gcc/egcs option that could correct this?
> >
> > -JP
> >
> > P Rosevear wrote:
> >
> > > Dan Burcaw wrote:
> > >
> > > > Good news,
> > > >
> > > > New PostgreSQL rpms on the FTP site in devel/. If they work well I'll
> > > > place them in the updates directory.
> > > >
> > > > They should fixed all the problems people were having.
> > > >
> > > > Watch the FTP sites.. they'll show up later this evening.
> > >
> > > A bit of improvement. Now when I create and destroy the db I get the
> > > following:
> > >
> > > createdb test
> > > destroydb test
> > > ERROR: typeidTypeRelid: Invalid type - oid = 0
> > >
> > > Only one error instead of two. \d <table name> still has the same
> > > error. BTW its a 7600/120 if that makes any difference. I've also
> > > tried kernel 2.2.6 on as well just in case that made a diff.
> > >
> > > I believe the error is actually sent by
> > > postgresql-6.4.2/src/backend/paser/parse_type.c on line 188.
This archive was generated by hypermail 2.0b3 on Fri Apr 30 1999 - 11:30:18 MDT