Re: some strange bugs related to upgrade from 8.1 to 8.3

"Pavel Stehule" <pavel(dot)stehule(at)gmail(dot)com> writes:
> 2008/11/4 Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>:
>> "Pavel Stehule" <pavel(dot)stehule(at)gmail(dot)com> writes:
>>> a) server crash after creating tsearch2 function (I use tsearch2
>>> contrib from 8.3)
>>
>> I couldn't reproduce that with the script you gave.

> I tested it on fe8, 32 bit without problem, so it's maybe related to 64bit.

Can't reproduce it on 64-bit either. Looking closer, I don't believe
that you were running this script at all --- the crash backtrace
includes

#14 0x00000000005e0a0c in exec_simple_query (
query_string=0x1f69ae8 "CREATE FUNCTION plpgsql_call_handler()
RETURNS language_handler\n AS '/usr/lib64/pgsql/plpgsql',
'plpgsql_call_handler'\n LANGUAGE c;") at postgres.c:986

and there is no such command in this script.

Something else weird I just noticed: in this script, you've got

--
-- Name: position; Type: TABLE; Schema: public; Owner: lmc; Tablespace:
--

CREATE TABLE "position" (
id integer NOT NULL,
title character varying(255) NOT NULL,
description text,
fulltext_index tsvector
);

ALTER TABLE public."position" OWNER TO lmc;

--
-- Name: view_position_uniq; Type: VIEW; Schema: public; Owner: lmc
--

CREATE VIEW view_position_uniq AS
SELECT p.uniq_key FROM "position" p;

The CREATE VIEW fails because there's no uniq_key column in "position".
Either you edited this script before sending it in, or there's something
a bit broken about pg_dump or the database you dumped from.

regards, tom lane