Received: from localhost (unknown [200.46.204.187]) by postgresql.org (Postfix) with ESMTP id 7D42C2E4420 for ; Sun, 13 Jan 2008 17:08:21 -0400 (AST) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.187]) (amavisd-maia, port 10024) with ESMTP id 02767-01-2 for ; Sun, 13 Jan 2008 17:08:12 -0400 (AST) Received: from sss.pgh.pa.us (sss.pgh.pa.us [66.207.139.130]) by postgresql.org (Postfix) with ESMTP id 1C5CC2E38D1 for ; Sun, 13 Jan 2008 17:04:28 -0400 (AST) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss.pgh.pa.us (8.14.1/8.14.1) with ESMTP id m0DL4PJW008812; Sun, 13 Jan 2008 16:04:25 -0500 (EST) To: Kevin Kempter cc: pgsql-admin@postgresql.org Subject: Re: large objects & space usage In-reply-to: <200801131301.04468.kevin@kevinkempterllc.com> References: <200801131301.04468.kevin@kevinkempterllc.com> Comments: In-reply-to Kevin Kempter message dated "Sun, 13 Jan 2008 13:01:04 -0700" Date: Sun, 13 Jan 2008 16:04:25 -0500 Message-ID: <8811.1200258265@sss.pgh.pa.us> From: Tom Lane X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200801/140 X-Sequence-Number: 27914 Kevin Kempter writes: > 3) we did a pg_dump and a reload into a database on the test box and the table > is still 30Gig. You need to find out *exactly* what they did there. The idea that comes to mind is that they dropped and reloaded the table itself, but didn't do anything about cleaning out pg_largeobject, and when they say "the table is still 30Gig" they actually mean the whole database directory is still 30Gig. If there are no other tables storing large objects in the database, dropping and reloading the whole database is probably the simplest solution. Otherwise they're going to need to make sure the unwanted LOs have been cleaned out (see contrib/vacuumlo) and then do a VACUUM FULL or similar on pg_largeobject. regards, tom lane