X-Original-To: pgsql-sql@postgresql.org Received: from spampd.localdomain (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with ESMTP id 233564764DA for ; Tue, 6 May 2003 10:32:55 -0400 (EDT) Received: from megazone.bigpanda.com (megazone.bigpanda.com [63.150.15.178]) by postgresql.org (Postfix) with ESMTP id C71F3476345 for ; Tue, 6 May 2003 10:32:44 -0400 (EDT) Received: by megazone.bigpanda.com (Postfix, from userid 1001) id A9564D62E; Tue, 6 May 2003 07:32:45 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by megazone.bigpanda.com (Postfix) with ESMTP id 9F0BA5C0C; Tue, 6 May 2003 07:32:45 -0700 (PDT) Date: Tue, 6 May 2003 07:32:45 -0700 (PDT) From: Stephan Szabo To: SZUCS Gabor Cc: Subject: Re: VACUUM FULL and ALTER TABLE do *nothing* in 7.3.2 In-Reply-To: <008801c313d6$ad89e390$0a03a8c0@fejleszt2> Message-ID: <20030506073053.G7678-100000@megazone23.bigpanda.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Spam-Status: No, hits=-16.3 required=5.0 tests=BAYES_01,EMAIL_ATTRIBUTION,IN_REP_TO autolearn=ham version=2.50 X-Spam-Level: X-Spam-Checker-Version: SpamAssassin 2.50 (1.173-2003-02-20-exp) X-Archive-Number: 200305/88 X-Sequence-Number: 13253 On Tue, 6 May 2003, SZUCS Gabor wrote: > Dear Achilleus, > > Dunno, it's a multiuser environment. However, I'll try asking everyone to > stop working for a minute... > > ... oops, I just see that it finished about half an hour before (taking an > amazing 76 minutes for ALTER TABLE to complete). About that time, I started > and exited a client. Seems it solved the problem. Now "vacuum full analyze" > finished in a couple of seconds. > > Do you have any idea about an explanation? Does vacuum wait for the end of > each and every transaction? Both alter table and vacuum full want to take exclusive locks on the table IIRC. If say that transaction that was idle had done *anything* on the table and was sitting open after that, both of those would wait until the transaction released the lock.