On Mon, Mar 3, 2014 at 7:52 AM, Robert Haas wrote:
> But all that having been said, a deadline is a deadline, so if anyone
> wishes to declare this untimely please speak up.
Hearing only crickets, committed.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Comp
On Thu, Feb 27, 2014 at 1:06 PM, Andres Freund wrote:
> As Robert previously complained a database wide VACUUM FULL now (as of
> 3cff1879f8d03) reliably increases the relfrozenxid for all tables but
> pg_class itself. That's a bit sad because it means doing a VACUUM FULL
> won't help in a anti-wra
Hi,
As Robert previously complained a database wide VACUUM FULL now (as of
3cff1879f8d03) reliably increases the relfrozenxid for all tables but
pg_class itself. That's a bit sad because it means doing a VACUUM FULL
won't help in a anti-wraparound scenario.
The reason for that is explained by the