Re: Re: significant performance hit whenever autovacuum runs after upgrading from 9.0 -> 9.1

On Thu, May 24, 2012 at 12:34 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> Lonni J Friedman <netllama(at)gmail(dot)com> writes:
>> No, not lots of subqueries or ORDERing, and most queries only touch a
>> single table.  However, I'm honestly not sure that I'm following where
>> you're going with this.   The problem isn't triggered by explicit
>> queries.  I can disable all external access, and simply wait for
>> autovacuum to kick off, and the box starts to die.
>
> Can you correlate the performance hit with any specific part of
> autovacuum?  In particular, I'm wondering if it matters whether vacuum
> is cleaning tables or indexes --- it alternates between the two, and the
> access patterns are a bit different.  You could probably watch what the
> autovac process is doing with strace to see what it's accessing.

Is there something specific I should be looking for in the strace
output, or is this just a matter of correlating PID and FD to
pg_class.relfilenode ?