X-Original-To: pgsql-admin-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [200.46.204.2]) by svr1.postgresql.org (Postfix) with ESMTP id DED71D1D065 for ; Fri, 9 Apr 2004 13:33:08 -0300 (ADT) Received: from svr1.postgresql.org ([200.46.204.71]) by localhost (neptune.hub.org [200.46.204.2]) (amavisd-new, port 10024) with ESMTP id 54867-05 for ; Fri, 9 Apr 2004 13:33:07 -0300 (ADT) Received: from sss.pgh.pa.us (sss.pgh.pa.us [66.207.139.130]) by svr1.postgresql.org (Postfix) with ESMTP id C0AADD1D06D for ; Fri, 9 Apr 2004 13:33:06 -0300 (ADT) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss.pgh.pa.us (8.12.11/8.12.11) with ESMTP id i39GX28i026549; Fri, 9 Apr 2004 12:33:02 -0400 (EDT) To: "scott.marlowe" Cc: hal , pgsql-admin@postgresql.org Subject: Re: Multiple disks In-reply-to: References: Comments: In-reply-to "scott.marlowe" message dated "Fri, 09 Apr 2004 09:10:17 -0600" Date: Fri, 09 Apr 2004 12:33:02 -0400 Message-ID: <26548.1081528382@sss.pgh.pa.us> From: Tom Lane X-Virus-Scanned: by amavisd-new at postgresql.org X-Archive-Number: 200404/183 X-Sequence-Number: 13156 "scott.marlowe" writes: > The next way is to individually link thinks like indexes onto other > volumes. The procedure is basically, create the index, figure out which > file in $PGDATA/base/oidofyourdbhere is the index, shut down postgresql, > copy to file elsewhere, softlink it, restart apache. This setup will not > survive reindexing or dropping / recreating the index. It should also be noted that this hack doesn't cope well with tables or indexes exceeding 1Gb. You'd have to separately symlink each segment, and deal with creating new links every time the table/index grows past another 1Gb boundary. Tablespaces will be a lot nicer when we have them. For now, a RAID array is definitely the best bet. regards, tom lane