Re: pg_autovacuum integration attempt #2 - Mailing list pgsql-patches

From Rod Taylor
Subject Re: pg_autovacuum integration attempt #2
Date
Msg-id 1090080321.91769.8.camel@jester
Whole thread Raw
In response to Re: pg_autovacuum integration attempt #2  ("Matthew T. O'Connor" <matthew@zeut.net>)
List pgsql-patches
> Exactly.  However, I can work around the GUC ordering issue just by
> having the postmaster check all the required variables before launching
> autovac.  What I wanted to do was have GUC prevent autovacuum = true
> when the stats collector is not enabled, reguardless of what is in
> postgresql.conf.

If the user has requested autovacuum, then why not just enable it? The
user shouldn't need to know that autovacuum is implemented via widget X
-- enable widget X for the user.


Are you also going to complain about a default statement_timeout that
doesn't allow vacuum to finish, or will you work around it by setting
statement_timeout = 0 on your connection?


pgsql-patches by date:

Previous
From: Tom Lane
Date:
Subject: Re: pg_autovacuum integration attempt #2
Next
From: Tom Lane
Date:
Subject: Re: pg_autovacuum integration attempt #2