Received: from sss2.sss.pgh.pa.us (sss.pgh.pa.us [209.114.166.2]) by hub.org (8.9.3/8.9.3) with ESMTP id LAA44978; Sun, 23 Jan 2000 11:18:39 -0500 (EST) (envelope-from tgl@sss.pgh.pa.us) 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 LAA06055; Sun, 23 Jan 2000 11:18:13 -0500 (EST) To: gav cc: pgsql-general@postgreSQL.org, pgsql-sql@postgreSQL.org Subject: Re: [SQL] pqReadData... In-reply-to: <13697.000123@nlr.ru> References: <13697.000123@nlr.ru> Comments: In-reply-to gav message dated "Sun, 23 Jan 2000 16:44:12 +0300" Date: Sun, 23 Jan 2000 11:18:13 -0500 Message-ID: <6052.948644293@sss.pgh.pa.us> From: Tom Lane gav writes: > I have the database created with the postgresql 6.5.3. I make many > inserts (10-30 thousands) in one of tables in one transaction. > Sometimes I receive the error: > pqReadData() -- backend closed the channel unexpectedly. > This probably means the backend terminated abnormally > before or while processing the request > What is it? Is this a bug? It's a bug, but we'll need a lot more info than that to fix it. For example: what were the last few commands you issued before the crash? Is there anything in the postmaster log file? (6.5.* libpq has a bad habit of dropping backend error messages if the channel is closed immediately after the message arrives. So if the backend reports a FATAL error and quits, the above is all you see. But the FATAL message ought to be logged in the postmaster log file.) Is there a core dump file from the failed backend? (It'd be in .../data/base/yourdatabasename/core) If so, can you use gdb to get a backtrace from the core file? regards, tom lane