Re: Possible work-around for 9.1 partial vacuum bug?

From: Michael Paquier <michael(dot)paquier(at)gmail(dot)com>
To: Jim Nasby <jim(at)nasby(dot)net>
Cc: Pg Hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Possible work-around for 9.1 partial vacuum bug?
Date: 2013-12-09 01:46:29
Message-ID: CAB7nPqShq7eXo-tU123tnKMOrLf69RUhi07RfsoNUKmiEk0Asw@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Mon, Dec 9, 2013 at 6:33 AM, Jim Nasby <jim(at)nasby(dot)net> wrote:
> If I'm understanding the vacuum truncate bug correctly, it can be avoided if
> every 2^31 transactions[1] you:
>
> SET vacuum_freeze_table_age = 0;
> VACUUM FREEZE;
>
> table_age = 0 disables partial vacuum and then everything[1] gets frozen,
> eliminating the risk. Or am I missing something?
Yes, this will fix any latent error by forcing a freeze on all the
rows of all tables, but not the ones that already happened if your
system has already done more than 2^31 transactions. In this case take
care of any constraint violation.

Regards,
--
Michael

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Andrew Gierth 2013-12-09 03:13:46 Re: WITHIN GROUP patch
Previous Message Mark Kirkwood 2013-12-08 22:19:48 Re: ANALYZE sampling is too good