From: "Ron Mayer" <ron@intervideo.com>
To: <josh@agliodbs.com>,
	"Achilleus Mantzios" <achill@matrix.gatewaynet.com>
Cc: <pgsql-performance@postgresql.org>
Subject: Re: [SQL] 7.3 analyze & vacuum analyze problem
Date: Wed, 30 Apr 2003 15:16:58 -0700
Message-ID: <POEDIPIPKGJJLDNIEMBEMEOACKAA.ron@intervideo.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Importance: Normal
In-Reply-To: <200304301148.18322.josh@agliodbs.com>


Josh wrote...
> Achilleus,
> 
> > I am afraid it is not so simple.
> > What i (unsuccessfully) implied is that 
> > dynacom=# VACUUM ANALYZE status ;
> > VACUUM
> > dynacom=# ANALYZE status ;
> > ANALYZE
> > dynacom=#
> >
> > [is enuf to damage the performance.]
> 
> You're right, that is mysterious.   If you don't get a response from one of 
> the major developers on this forum, I suggest that you post those EXPLAIN 
> results to PGSQL-BUGS.

I had the same problem a while back.

http://archives.postgresql.org/pgsql-bugs/2002-08/msg00015.php
http://archives.postgresql.org/pgsql-bugs/2002-08/msg00018.php
http://archives.postgresql.org/pgsql-bugs/2002-08/msg00018.php

Short summary: Later in the thread Tom explained my problem as free 
space not being evenly distributed across the table so ANALYZE's 
sampling gave skewed results.  In my case, "pgstatuple" was a 
good tool for diagnosing the problem, "vacuum full" fixed my table
and a much larger fsm_* would have probably prevented it.