Received: from beta.bkc.lv (beta.bkc.lv [195.244.128.53]) by hub.org (8.9.3/8.9.3) with SMTP id JAA75347 for ; Tue, 30 Nov 1999 09:34:56 -0500 (EST) (envelope-from emils@mail.usis.bkc.lv) Received: from mail.usis.bkc.lv (mail.usis.bkc.lv [195.244.132.2]) by beta.bkc.lv (NTMail 3.02.13) with ESMTP id fa962369 for ; Tue, 30 Nov 1999 16:34:46 +0200 Received: from USIS-RIGA/SpoolDir by mail.usis.bkc.lv (Mercury 1.44); 30 Nov 99 16:44:09 EET Received: from SpoolDir by USIS-RIGA (Mercury 1.44); 30 Nov 99 16:43:56 EET From: "Emils Klotins" Organization: USIS Riga To: pgsql-sql@postgresql.org Date: Tue, 30 Nov 1999 16:43:50 EET MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Subject: Addendum: PG6.5.3: CASE w. diff THEN types -- prob with Linux(?) Priority: normal X-mailer: Pegasus Mail for Win32 (v3.12b) Message-Id: <14344673472060@bkc.lv> FYI. I posted a note couple of days ago about CASE WHEN clause causing backend termination. I was able to get my hands on FreeBSD machine and the statement does not break the backend: [ PostgreSQL 6.5.3 on i386-unknown-freebsd3.3, compiled by gcc 2.7.2.3 ] => select case when 3=1 then 'rrr'::text else 5 end; ERROR: CASE/WHEN types 'int4' and 'text' not matched => The same statement on RedHat 5.2 Linux machine breaks the backend: [ PostgreSQL 6.5.3 on i586-pc-linux-gnu, compiled by gcc 2.7.2.3 ] => select case when 3=1 then 'rrr'::text when 3=2 else 5 end; NOTICE: Message from PostgreSQL backend: The Postmaster has informed me that some other backend died abnormally and possibly corrupted shared memory. I have rolled back the current transaction and am going to terminate your database system connection and exit. Please reconnect to the database system and repeat your query. pqFlush() -- backend closed the channel unexpectedly. This probably means the backend terminated abnormally before or while processing the request. => \q Emils