Re: autovacuum not prioritising for-wraparound tables

On Sun, Jan 27, 2013 at 4:17 AM, Simon Riggs <simon(at)2ndquadrant(dot)com> wrote:
> On 25 January 2013 17:19, Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
>> We
>> could easily run across a system where pg_class order happens to be
>> better than anything else we come up with.
>
> I think you should read that back to yourself and see if you still
> feel the word "easily" applies here.

I absolutely do. You will not convince me that whacking around the
behavior of autovacuum in a maintenance release is a remotely sane
thing to do. There are plenty of things wrong with the way autovacuum
works today, and I am all in favor of fixing them - but not in the
back-branches. Every time we whack behavior around in the back
branches, no matter how innocuous it looks, somebody's environment
gets broken, and then they won't apply patch releases, and it causes
all sorts of headaches. At least, that's my experience at
EnterpriseDB. YMMV.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company