Re: Rare corruption of pg_class index

Tom Lane wrote:
> Alvaro Herrera <alvherre@commandprompt.com> writes:
> > Greg Sabino Mullane wrote:

> > It would help if you could get a stack trace at the moment of the
> > problem, but I'm not sure how to do that.
> 
> Perhaps insert an abort() call right before the elog(ERROR)
> that's reporting this.

Yeah, but doing this in a production environment is not likely to go
very far ...

> (I think there are three possibilities,
> but they're all in heapam.c so you might as well just hack them all.)

Maybe it would be good to run under log_error_verbosity='verbose' for a
while to discover whether it's always the same one.

-- 
Alvaro Herrera                                http://www.CommandPrompt.com/
PostgreSQL Replication, Consulting, Custom Development, 24x7 support