Re: pg_upgrade improvements - Mailing list pgsql-hackers

From Harold Giménez
Subject Re: pg_upgrade improvements
Date
Msg-id CABQCq-QHes7Bs+ke40JB0X7Zk2CMEHNKzU9J-CepOwBeC7XzOw@mail.gmail.com
Whole thread Raw
In response to Re: pg_upgrade improvements  (Bruce Momjian <bruce@momjian.us>)
List pgsql-hackers


On Wed, Apr 11, 2012 at 5:40 PM, Bruce Momjian <bruce@momjian.us> wrote:
On Wed, Apr 04, 2012 at 07:26:58PM -0700, Harold Giménez wrote:
> There could be incoming connections for a number of
> reasons: either the user or the user's applications are reestablishing
> connections, or something like collectd on the localhost attempts to
> connect during that small window.

Well, we did address this in PG 9.2 by having pg_upgrade use a
non-default port number when starting servers, 50432.  You can do that
too in PG 9.1 by just specifying non-default port numbers when you run
pg_upgrade.  We do start the server with a special --binary-upgrade
mode, and we could do all sorts of connection limits in that mode, but
having the port number be different seemed the clearest solution.

The non-default port number is a good solution, better than modifying pg_hba. 
Thanks for pointing that out!

-Harold

pgsql-hackers by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: pg_upgrade improvements
Next
From: Robert Haas
Date:
Subject: Re: Last gasp