X-Original-To: pgsql-general-postgresql.org@postgresql.org Received: from localhost (wm.hub.org [200.46.204.128]) by postgresql.org (Postfix) with ESMTP id 79AC29FB345 for ; Wed, 16 Aug 2006 17:21:33 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.128]) (amavisd-new, port 10024) with ESMTP id 44896-01 for ; Wed, 16 Aug 2006 20:21:25 +0000 (UTC) X-Greylist: from auto-whitelisted by SQLgrey- Received: from sss.pgh.pa.us (sss.pgh.pa.us [66.207.139.130]) by postgresql.org (Postfix) with ESMTP id 338BD9FB30D for ; Wed, 16 Aug 2006 17:21:25 -0300 (ADT) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss.pgh.pa.us (8.13.6/8.13.6) with ESMTP id k7GKLJFI028315; Wed, 16 Aug 2006 16:21:19 -0400 (EDT) To: =?ISO-8859-1?Q?Poul_M=F8ller_Hansen?= cc: pgsql-general@postgresql.org Subject: Re: Segmentation Fault In-reply-to: <44E337BB.2030708@pbnet.dk> References: <44E308B2.8030008@pbnet.dk> <19838.1155738694@sss.pgh.pa.us> <44E337BB.2030708@pbnet.dk> Comments: In-reply-to =?ISO-8859-1?Q?Poul_M=F8ller_Hansen?= message dated "Wed, 16 Aug 2006 17:20:27 +0200" Date: Wed, 16 Aug 2006 16:21:19 -0400 Message-ID: <28314.1155759679@sss.pgh.pa.us> From: Tom Lane X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200608/750 X-Sequence-Number: 99521 =?ISO-8859-1?Q?Poul_M=F8ller_Hansen?= writes: >> Given that you're using duration logging and JDBC, I wonder whether you >> didn't trip over this recently-identified bug: >> http://archives.postgresql.org/pgsql-hackers/2006-08/msg00815.php >> Patch is here: >> http://archives.postgresql.org/pgsql-committers/2006-08/msg00278.php > Sorry but didn't, but I noticed this: > Also I must notice that the segfault only occur if > log_min_duration_statement is set to 0 I don't believe that statement actually ... it might have chanced to act that way in one or two trials for Sergey, but since the bug essentially consists in access to already-freed-and-perhaps-reused memory, it's not very predictable whether it will fail visibly or not. In any case the problem could occur for any duration-logging attempt. regards, tom lane