Skip site navigation (1) Skip section navigation (2)

Re: out-of-order caution

From: "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov>
To: "Simon Riggs" <simon(at)2ndQuadrant(dot)com>
Cc: "pgsql-hackers" <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: out-of-order caution
Date: 2011-10-27 17:51:50
Message-ID: 4EA953E60200002500042788@gw.wicourts.gov (view raw or flat)
Thread:
Lists: pgsql-hackers
Simon Riggs <simon(at)2ndQuadrant(dot)com> wrote:
> On Thu, Oct 27, 2011 at 4:41 PM, Kevin Grittner
> <Kevin(dot)Grittner(at)wicourts(dot)gov> wrote:
>> On the docs page for the SELECT statement, there is a caution
>> which starts with:
>>
>> | It is possible for a SELECT command using ORDER BY and FOR
>> | UPDATE/SHARE to return rows out of order. This is because ORDER
>> | BY is applied first.
>>
>> Is this risk limited to queries running in READ COMMITTED
>> transactions?  If so, I think that should be mentioned in the
>> caution.
> 
> I think it should say that if this occurs with SERIALIZED
> transactions it will result in a serialisation error.
> 
> Just to say there is no effect in serializable mode wouldn't be
> helpful.
 
Hmm.  At first reading I thought this was related to the
mixed-snapshot issue in READ COMMITTED, but now I'm not so sure. 
Does anyone know which isolation levels are affected?  Barring that,
can anyone point to an existing test which demonstrates the problem?
 
If this can happen in snapshot isolation with just one reader and
one writer, I doubt that SSI helps with it.  :-(
 
-Kevin

In response to

Responses

pgsql-hackers by date

Next:From: Robert HaasDate: 2011-10-27 18:19:31
Subject: Re: out-of-order caution
Previous:From: Bruce MomjianDate: 2011-10-27 17:43:56
Subject: Re: Bug in walsender when calling out to do_pg_stop_backup (and others?)

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group