Tom Lane wrote:
> Kai Hessing <kai(dot)hessing(at)hobsons(dot)de> writes:
>> No one any idea? *sigh*
>
> What makes you think it's a deadlock and not a very slow query? I'd be
> checking if the tables were all ANALYZEd and comparing EXPLAIN output
> to the old database ...
*hmm* the difference of 5 seconds on 8.0.8 compared to killing the
process after 40 hours?
All databases are ANALYZEd and EXPLAIN output cannot be compared because
it hangs also up on 8.0.8.