Re: pg_autovacuum and REINDEX at the same time (?)

From: Heikki Linnakangas <heikki(at)enterprisedb(dot)com>
To: Janar Kartau <janar(dot)kartau(at)cvkeskus(dot)ee>
Cc: pgsql-bugs(at)postgresql(dot)org
Subject: Re: pg_autovacuum and REINDEX at the same time (?)
Date: 2007-02-28 12:12:18
Message-ID: 45E571A2.2010209@enterprisedb.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

Janar Kartau wrote:
> I know what causes the "sorry, too many clients already" error, but the
> number of connections shouldn't grow so big.. so i guess the table
> (category_tree) got locked.

If that happens again, take a look at pg_stat_activity or ps output to
see what's going on. Something must be using the connections, a locked
table alone won't give you that error message.

> We do REINDEX every 5 minutes because the table gets updated very often
> and the query's took a lot of time. I think since we use autovacuum now,
> it's not needed anymore.

Yeah, a vacuum is normally enough to keep the indexes in good shape. I'd
suggest removing the REINDEX call and either letting autovacuum handle
it, or adding an explicit VACUUM call after the commit.

--
Heikki Linnakangas
EnterpriseDB http://www.enterprisedb.com

In response to

Browse pgsql-bugs by date

  From Date Subject
Next Message Andrew White 2007-02-28 21:59:30 BUG #3089: View/Table Creation/Ownership Bug
Previous Message Janar Kartau 2007-02-28 11:38:50 Re: pg_autovacuum and REINDEX at the same time (?)