Received: from NIH2WAAE (smtp5.site1.csi.com [149.174.183.74]) by hub.org (8.8.8/8.8.8) with ESMTP id GAA14049 for ; Sun, 27 Sep 1998 06:29:43 -0400 (EDT) (envelope-from ndarlow@csi.com) Received: from mail pickup service by csi.com with Microsoft SMTPSVC; Sun, 27 Sep 1998 06:29:10 -0400 Sender: ndarlow@csi.com Received: from csi.com (lon-dnbsi-109.gbr.compuserve.net [195.232.123.109]) by hil-img-ims-1.compuserve.com (8.8.6/8.8.6/IMS-1.6) with ESMTP id GAA13297 for ; Sun, 27 Sep 1998 06:28:49 -0400 (EDT) Sender: neil@hub.org Message-ID: <360E149B.245ECE2E@csi.com> Date: Sun, 27 Sep 1998 11:34:03 +0100 From: Neil Darlow X-Mailer: Mozilla 4.06 [en] (X11; I; Linux 2.0.35 i586) MIME-Version: 1.0 To: pgsql-interfaces@postgresql.org Subject: libpgtcl core dump with socket access. Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, My environment is Red Hat 5.1 Intel, kernel 2.0.35-2, glibc-2.0.7 running PostgreSQL-6.3.2 and Tcl/Tk-8.0.2. Colleagues have reported that the problem does not occur with the original *stock* Red Hat 5.1 distribution. If you connect to a database locally with: set dbc [pg_connect "dbname"] then there is no problem. Executing: set result [pg_exec $dbc "select * from non_existent_table"] results in an error message being reported by tclsh or wish. If you connect to a database with: set dbc [pg_connect -conninfo "dbname=database host=hostname port=portname user=username password=password"] then execute: set result [pg_exec $dbc "select * from non_existent_table"] always results in a core dump. I have checked other libpgtcl interface commands and they appear to work ok. It just seems to be pg_exec over socket connections. Performing the same operations with psql always results in error messages being returned. Has anyone else noticed this problem? Is there a solution? Neil Darlow (UK).