X-Original-To: pgsql-admin-postgresql.org@postgresql.org Received: from localhost (wm.hub.org [200.46.204.128]) by postgresql.org (Postfix) with ESMTP id 567CA9FB3EF for ; Thu, 31 Aug 2006 13:56:42 -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 95052-02 for ; Thu, 31 Aug 2006 16:56:39 +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 2A3479FB303 for ; Thu, 31 Aug 2006 13:56:39 -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 k7VGubsX009659; Thu, 31 Aug 2006 12:56:38 -0400 (EDT) To: RW cc: pgsql-admin@postgresql.org Subject: Re: pg_dump: schema with OID 16396 does not exist In-reply-to: <44F69D08.80308@tauceti.net> References: <44F5753F.9080409@tauceti.net> <26697.1156945365@sss.pgh.pa.us> <44F69D08.80308@tauceti.net> Comments: In-reply-to RW message dated "Thu, 31 Aug 2006 10:25:44 +0200" Date: Thu, 31 Aug 2006 12:56:37 -0400 Message-ID: <9658.1157043397@sss.pgh.pa.us> From: Tom Lane X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200608/352 X-Sequence-Number: 22895 RW writes: >> Look through the system catalogs to find the object(s) that claim to be >> in that namespace, and then tell us what they are and their history. > Found it in pg_conversion: > mps_dev=# select * from pg_conversion where connamespace = 16396; > conname | connamespace | conowner | conforencoding | > contoencoding | conproc | condefault > ---------------------+--------------+----------+----------------+---------------+-------------------+------------ > iso_8859_1_to_utf_8 | 16396 | 16389 | 8 > | 6 | iso8859_1_to_utf8 | t > (1 row) Ah. This is a known bug: CREATE CONVERSION neglected to enter a dependency from the new conversion to its namespace, hence a DROP SCHEMA on the namespace would not remove the pg_conversion entry. This is fixed in HEAD but hasn't been back-patched for some reason --- I'll go deal with that now. In the meantime I'd suggest a manual DROP CONVERSION to clean up ... hmm ... except you probably can't name that conversion anymore in SQL, what with it not having a schema. I think it would work well enough to just DELETE the specific row in pg_conversion. regards, tom lane