Received: from davinci.ethosmedia.com (davinci.ethosmedia.com [209.10.40.250]) by postgresql.org (Postfix) with ESMTP id 180B04758A7 for ; Sun, 28 Apr 2002 15:39:06 -0400 (EDT) Received: from [209.122.114.82] (account ) by davinci.ethosmedia.com (CommuniGate Pro WebUser 3.5.9) with HTTP id 1387143; Sun, 28 Apr 2002 12:39:06 -0700 From: "Josh Berkus" Subject: Re: Function won't complete To: Tom Lane , "Josh Berkus" Cc: pgsql-sql@postgresql.org X-Mailer: CommuniGate Pro Web Mailer v.3.5.9 Date: Sun, 28 Apr 2002 12:39:06 -0700 Message-ID: In-Reply-To: <7079.1019943371@sss.pgh.pa.us> MIME-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 8bit X-Archive-Number: 200204/346 X-Sequence-Number: 7610 Tom, > This analysis is nonsense, because PG does not rely on WAL for > transaction rollback, and the amount of WAL activity is *not* > proportional to transaction length. (At least not since 7.1.2.) > However, until you drop > your focus on the WAL we'll not find out what's really the > bottleneck... Sorry, Tom. I was used to the problems of 7.1.2, and didn't really "get it" when you told me things had changed. I still say it's the disk I/O, and I think your explanation of dead tuples makes a lot of sense. The debug log is full of this: DEBUG: proc_exit(0) DEBUG: shmem_exit(0) DEBUG: exit(0) DEBUG: reaping dead processes DEBUG: child process (pid 11933) exited with exit code 0 DEBUG: proc_exit(0) DEBUG: shmem_exit(0) DEBUG: exit(0) DEBUG: reaping dead processes DEBUG: child process (pid 11939) exited with exit code 0 And each one of the cycles about takes 5-10 minutes. I'm a little reluctant to dump everything to the list, as we're talking about a lot of data and code. Lemme do some judicious editing and I'll send you a gzip package this week. -Josh Berkus