Received: from localhost (unknown [200.46.204.183]) by developer.postgresql.org (Postfix) with ESMTP id 88D702E005E for ; Tue, 29 Apr 2008 18:12:54 -0300 (ADT) Received: from developer.postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.183]) (amavisd-maia, port 10024) with ESMTP id 63747-02 for ; Tue, 29 Apr 2008 18:12:51 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey-1.7.5 Received: from lists.commandprompt.com (host-159.commandprompt.net [207.173.203.159]) by developer.postgresql.org (Postfix) with ESMTP id A9BA32E005A for ; Tue, 29 Apr 2008 18:12:51 -0300 (ADT) Received: from perhan.alvh.no-ip.org (200-126-123-196.bk8-dsl.surnet.cl [200.126.123.196]) (authenticated bits=0) by lists.commandprompt.com (8.13.8/8.13.8) with ESMTP id m3TLDQr0005115 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Tue, 29 Apr 2008 14:13:29 -0700 Received: by perhan.alvh.no-ip.org (Postfix, from userid 1000) id F1FD247C5B; Tue, 29 Apr 2008 17:12:44 -0400 (CLT) Date: Tue, 29 Apr 2008 17:12:44 -0400 From: Alvaro Herrera To: Ian Westmacott Cc: pgsql-admin@postgresql.org Subject: Re: 8.1.8 autovacuum missing databases Message-ID: <20080429211244.GM5652@alvh.no-ip.org> References: <1209499373.18622.47.camel@vega.intellivid.com> <20080429203335.GK5652@alvh.no-ip.org> <1209503103.18622.68.camel@vega.intellivid.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1209503103.18622.68.camel@vega.intellivid.com> User-Agent: Mutt/1.5.17+20080114 (2008-01-14) X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.0 (lists.commandprompt.com [207.173.203.159]); Tue, 29 Apr 2008 14:13:29 -0700 (PDT) X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200804/292 X-Sequence-Number: 29056 Ian Westmacott wrote: > On Tue, 2008-04-29 at 16:33 -0400, Alvaro Herrera wrote: > > Is autovacuum dying before being able to finish the vacuuming of > > template1 or the other database? > > Not as far as I can tell. There are no indications of any crash or > error in the log file (I just bumped the log level up to debug1). Just > autovacuum processing template1 or the one other database every minute. > I ran VACUUM ANALYZE on both template1 and the one database manually > just to make sure there wasn't a lock or something preventing > completion, but they both completed without error. Okay. On 8.2 you could have the problem that leftover temp tables are messing up the frozenxid calculations, but I don't think that can happen on 8.1 at all. Can you verify whether age(pg_database.datfrozenxid) is shrinking after vacuuming? If the age() of a database is higher than the applicable max_freeze_age, then it will always be chosen. -- Alvaro Herrera http://www.CommandPrompt.com/ The PostgreSQL Company - Command Prompt, Inc.