Re: Re: [COMMITTERS] pgsql: Fix brain fade in DefineIndex(): it was continuing to access the

Alvaro Herrera <alvherre@commandprompt.com>

From: Alvaro Herrera <alvherre@commandprompt.com>
To: ohp@pyrenet.fr
Cc: Tom Lane <tgl@sss.pgh.pa.us>, Andrew Dunstan <andrew@dunslane.net>, PostgreSQL-development <pgsql-hackers@postgresql.org>
Date: 2007-09-02T22:49:57Z
Lists: pgsql-hackers
ohp@pyrenet.fr wrote:
> Hi Tom and Andrew,
> 
> On Thu, 30 Aug 2007, Tom Lane wrote:
> 
> While waiting for my application for another animal, I made some tests and
> was surprised that cluster test failed with an ordering error.

This is running with CLOBBER_CACHE_ALWAYS set, right?  I think it is
quite possible that an autovacuum came and processed the catalog,
leading to different ordering.

> Maybe an ORDER BY is missing in the test query.

Yeah, an ORDER BY should be enough.

>   SELECT conname FROM pg_constraint WHERE conrelid = 'clstr_tst'::regclass;
>       conname
>   ----------------
> -  clstr_tst_pkey
>    clstr_tst_con
>   (2 rows)
> 
>   SELECT relname, relkind,
> --- 252,259 ----
>   SELECT conname FROM pg_constraint WHERE conrelid = 'clstr_tst'::regclass;
>       conname
>   ----------------
>    clstr_tst_con
> +  clstr_tst_pkey
>   (2 rows)


-- 
Alvaro Herrera                                http://www.CommandPrompt.com/
The PostgreSQL Company - Command Prompt, Inc.