Re: postgres deadlocks

On Wed, Jul 25, 2001 at 06:32:48PM -0400, Joseph Shraibman wrote:
> But *something* got idle in trasaction.
>
> Even more fustrating it seemed that my whole database was locked, not
> just one table, and what could have locked the whole thing except
> vacuum?

My guess is that if vacuum was processing say pg_shadow when it locked, that
could prevent other people from logging in. Strange though, because vacuum
checks for all locks *before* actually doing anything, so vacuum would be
waiting, not actually locked.

Kill the transaction with kill -INT is the way out.

You should find out what exactly was being updated that stop
mid-transaction.

HTH,
--
Martijn van Oosterhout <kleptog(at)svana(dot)org>
http://svana.org/kleptog/
> It would be nice if someone came up with a certification system that
> actually separated those who can barely regurgitate what they crammed over
> the last few weeks from those who command secret ninja networking powers.