Received: from localhost (unknown [200.46.204.183]) by postgresql.org (Postfix) with ESMTP id A966737BFDB for ; Mon, 29 Sep 2008 20:10:43 -0300 (ADT) Received: from postgresql.org ([200.46.204.86]) by localhost (mx1.hub.org [200.46.204.183]) (amavisd-maia, port 10024) with ESMTP id 71786-03 for ; Mon, 29 Sep 2008 20:10:40 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey-1.7.6 Received: from polaris.pinpointresearch.com (66-7-238-130.static-ip.telepacific.net [66.7.238.130]) by postgresql.org (Postfix) with ESMTP id 8C84237BFD1 for ; Mon, 29 Sep 2008 20:10:40 -0300 (ADT) Received: from [192.168.1.179] (betelgeuse.pinpointresearch.com [192.168.1.179]) by polaris.pinpointresearch.com (Postfix) with ESMTP id 3FB0BE00FC11; Mon, 29 Sep 2008 16:10:38 -0700 (PDT) Message-ID: <48E1606E.7070702@pinpointresearch.com> Date: Mon, 29 Sep 2008 16:10:38 -0700 From: Steve Crawford User-Agent: Thunderbird 2.0.0.17 (X11/20080925) MIME-Version: 1.0 To: Tom Lane CC: Scott Marlowe , Peter Kovacs , pgsql-admin@postgresql.org Subject: Re: Do we need vacuuming when tables are regularly dropped? References: <9815.1222690583@sss.pgh.pa.us> <48E10C95.2050207@pinpointresearch.com> <25195.1222728353@sss.pgh.pa.us> In-Reply-To: <25195.1222728353@sss.pgh.pa.us> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Virus-Scanned: Maia Mailguard 1.0.1 X-Spam-Status: No, hits=0 tagged_above=0 required=5 tests=none X-Spam-Level: X-Archive-Number: 200809/346 X-Sequence-Number: 30439 Tom Lane wrote: > "Scott Marlowe" writes: > >> Also, there was a time when you couldn't do vacuum full on system >> tables do to locking issues, and had to take the db down to single >> user mode to do so. >> > > There was a short period when *concurrent* vacuum fulls on just the > wrong combinations of system catalogs could deadlock (because they both > needed to look up stuff in the other one). AFAIK we fixed that. It's > never been the case that it didn't work at all. > > regards, tom lane > Never personally had trouble with vacuum full or reindex on system tables. CLUSTER, however, is another story. While I've never run across anything explicitly documenting that clustering system tables is forbidden, I've also never used a version of PostgreSQL that allows it (though I've never tried in single-user mode): postgres@[local]=> CLUSTER pg_class USING pg_class_oid_index ; ERROR: "pg_class" is a system catalog Should the docs (http://www.postgresql.org/docs/8.3/interactive/sql-cluster.html) be updated to note this restriction? Cheers, Steve