Received: from sss2.sss.pgh.pa.us (root@sss.pgh.pa.us [209.114.166.2]) by hub.org (8.10.1/8.10.1) with ESMTP id e765jPG01887 for ; Sun, 6 Aug 2000 01:45:25 -0400 (EDT) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss2.sss.pgh.pa.us (8.9.3/8.9.3) with ESMTP id BAA24263; Sun, 6 Aug 2000 01:45:21 -0400 (EDT) To: "Michael Richards" cc: pgsql-sql@postgresql.org Subject: Re: Database in recovery mode In-reply-to: <398B100F.0000AB.63030@frodo.searchcanada.ca> References: <398B100F.0000AB.63030@frodo.searchcanada.ca> Comments: In-reply-to "Michael Richards" message dated "Fri, 04 Aug 2000 14:48:47 -0400" Date: Sun, 06 Aug 2000 01:45:21 -0400 Message-ID: <24260.965540721@sss.pgh.pa.us> From: Tom Lane X-Archive-Number: 200008/45 "Michael Richards" writes: > I've got a postgres system that keeps going into recovery mode. I > can't really find any docs on this. All of the postgres processes > will be in the STOP state and when I try to connect it will say "The > database is in recovery mode". > I suspect there is a query that is causing it to crash in some > specific way that causes it. This is happening about once per day. Sounds like that to me too. Are any core dump files showing up in the database directory (PGDATA/base/DBNAME/core)? If so a stack trace from the corefile would be informative. Also, make sure you are collecting a postmaster log --- do NOT start the postmaster with -S, instead direct its stdout and stderr into a logfile --- and see if anything interesting is recorded therein. You might want to start the postmaster with -d2 so that the actual queries are logged too. regards, tom lane