Re: Re: new patch of MERGE (merge_204) & a question about duplicated ctid

Robert Haas <robertmhaas@gmail.com>

From: Robert Haas <robertmhaas@gmail.com>
To: Florian Pflug <fgp@phlo.org>
Cc: Heikki Linnakangas <heikki.linnakangas@enterprisedb.com>, Stephen Frost <sfrost@snowman.net>, Simon Riggs <simon@2ndquadrant.com>, Greg Smith <greg@2ndquadrant.com>, Marko Tiikkaja <marko.tiikkaja@cs.helsinki.fi>, Boxuan Zhai <bxzhai2010@gmail.com>, pgsql-hackers@postgresql.org
Date: 2011-01-03T17:09:32Z
Lists: pgsql-hackers
On Mon, Jan 3, 2011 at 11:36 AM, Florian Pflug <fgp@phlo.org> wrote:
> On Jan3, 2011, at 17:21 , Robert Haas wrote:
>> On Mon, Jan 3, 2011 at 11:08 AM, Heikki Linnakangas
>>> In serializable mode you get a serialization error.
>>
>> I don't think this part is true.  You can certainly do this:
>>
>> CREATE TABLE test (a int);
>> BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE;
>> SELECT * FROM test;
>> <in another session, insert (1) into test>
>> LOCK TABLE test IN SHARE MODE; -- or just LOCK TABLE test, if you prefer
>> SELECT * FROM test;  -- still ain't there
>> INSERT INTO test VALUES (1);
>
> In SERIALIZABLE mode, you need to take any table-level locks before obtaining
> a snapshot. There's even a warning about this in the docs somewhere IIRC...

That should be safe, if people do it that way.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company