Re: pg_multixact not getting truncated

On 11/09/2014 08:00 PM, Josh Berkus wrote:
On 11/08/2014 01:46 PM, Andres Freund wrote:
>> I'm these days suggesting that people should add manual vacuuming for
>> > "older" relations during off peak hours on busy databases. There's too
>> > many sites which service degrades noticeably during a full table vacuum.
> Me too: https://github.com/pgexperts/flexible-freeze

It turns out that not even a program of preventative scheduled vacuuming
helps.  This is because the template0 database anchors the minmxid and
prevents it from being advanced until autovacuum gets around to that
database, at whatever the minmxid threshold is.

-- 
Josh Berkus
PostgreSQL Experts Inc.
http://pgexperts.com