Re: ALTER TABLE ... NOREWRITE option

From: Robert Haas <robertmhaas(at)gmail(dot)com>
To: Josh Berkus <josh(at)agliodbs(dot)com>
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: ALTER TABLE ... NOREWRITE option
Date: 2012-12-05 19:15:46
Message-ID: CA+Tgmoac0tBfyy81tMmCQoBuL3OCo=HJsEDxZ6geO9X42Bmo0g@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Wed, Dec 5, 2012 at 1:41 PM, Josh Berkus <josh(at)agliodbs(dot)com> wrote:
> That's why we need a mechanism which either logs, or aborts on specific
> actions. From the perspective of the DevOps staff, abort is possibly
> the better option, but there may be issues with it on our end. That was
> the attraction of the original NOREWRITE patch, although as I said that
> suffers from new keywords and a total lack of extensibility.

You know, event triggers seem like an awfully good solution to this
problem. All we'd need is a new event called table_rewrite:

CREATE EVENT TRIGGER my_event_trigger
ON table_rewrite
EXECUTE PROCEDURE consider_whether_to_throw_an_error();

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

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2012-12-05 19:17:04 Re: PITR potentially broken in 9.2
Previous Message Andres Freund 2012-12-05 19:12:31 Re: PITR potentially broken in 9.2