Date: Wed, 23 Apr 2014 23:34:27 +0200
From: Andres Freund <andres@2ndquadrant.com>
To: Alvaro Herrera <alvherre@2ndquadrant.com>
Cc: Bruce Momjian <bruce@momjian.us>, Stephen Frost <sfrost@snowman.net>,
 pgsql-hackers@postgresql.org
Subject: Re: MultiXactId error after upgrade to 9.3.4
Message-ID: <20140423213427.GA12442@awork2.anarazel.de>
References: <20140330185349.GD4582@tamriel.snowman.net>
 <20140331115452.GT9567@eldon.alvh.no-ip.org>
 <20140331115713.GA18358@alap3.anarazel.de>
 <20140423180102.GP25695@eldon.alvh.no-ip.org>
 <20140423182804.GA21458@momjian.us>
 <20140423184214.GQ25695@eldon.alvh.no-ip.org>
 <20140423184716.GA22722@momjian.us>
 <20140423185120.GR25695@eldon.alvh.no-ip.org>
 <60097392-07aa-4310-8632-43b7b09c1822@email.android.com>
 <20140423193005.GS25695@eldon.alvh.no-ip.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20140423193005.GS25695@eldon.alvh.no-ip.org>
Precedence: bulk
Sender: pgsql-hackers-owner@postgresql.org

On 2014-04-23 16:30:05 -0300, Alvaro Herrera wrote:
> Andres Freund wrote:
> 
> > I think this patch is a seriously bad idea. For one, it's not actually
> > doing anything about the problem - the tuple can be accessed without
> > freezing getting involved.
> 
> Normal access other than freeze is not a problem, because other code
> paths do check for HEAP_XMAX_INVALID and avoid access to Xmax if that's
> set.
> 
> > For another, it will increase wall traffic for freezing of short lived
> > tuples considerably, without any benefit.
> 
> True.  I didn't actually try to run this; it was just for demonstration
> purposes.  It'd need some cooperation from heap_tuple_needs_freeze in
> order to work at all, for one thing.

I think if you want to add something like this it should be added
*inside* the if (MultiXactIdPrecedes(multi, cutoff_multi)) block in
FreezeMultiXactId(). There it seems to make quite a bit of sense.

Greetings,

Andres Freund

-- 
 Andres Freund	                   http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, Training & Services


-- 
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers