Message-ID: <4EFB2597.1080005@beccati.com>
Date: Wed, 28 Dec 2011 15:20:07 +0100
From: Matteo Beccati <php@beccati.com>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64;
	rv:8.0) Gecko/20111105 Thunderbird/8.0
MIME-Version: 1.0
To: Tom Lane <tgl@sss.pgh.pa.us>
CC: pgsql-performance@postgresql.org
Subject: Re: PostgreSQL 9.0.4 blocking in lseek?
References: <4EA82B7C.1040008@google.com> <9398.1319761978@sss.pgh.pa.us>
	<4EFAE7F6.30603@beccati.com>
In-Reply-To: <4EFAE7F6.30603@beccati.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit

Hi,

> I'm also trying an EXPLAIN ANALYZE for the SELECT part, but it seems to
> take a while too and is seemingly calling only gettimeofday.

An update on this. Apart from gettimeofday calls, which I filtered out
when logging, I've seen about 80 lseeks recorded every 1h10m (each 1us
apart):

2011-12-28 11:06:25.546661500 lseek(10, 0, SEEK_END) = 27623424
2011-12-28 11:06:25.546662500 lseek(10, 0, SEEK_END) = 27623424
2011-12-28 11:06:25.546663500 lseek(10, 0, SEEK_END) = 27623424
...
2011-12-28 12:16:56.144663500 lseek(10, 0, SEEK_END) = 27623424
...
2011-12-28 13:28:20.436549500 lseek(10, 0, SEEK_END) = 27623424
...

I've then decided to interrupt the EXPLAIN ANALYZE. But I suppose that
the database as it is now will likely allow to further debug what's
happening.


Cheers
-- 
Matteo Beccati

Development & Consulting - http://www.beccati.com/