Re: pg_upgrade on high number tables database issues

From: Bruce Momjian <bruce(at)momjian(dot)us>
To: Pavel Stehule <pavel(dot)stehule(at)gmail(dot)com>
Cc: Jeff Janes <jeff(dot)janes(at)gmail(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: pg_upgrade on high number tables database issues
Date: 2014-03-10 19:11:24
Message-ID: 20140310191124.GB16713@momjian.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Mon, Mar 10, 2014 at 07:40:42PM +0100, Pavel Stehule wrote:
>
> >
> > There were several bottlenecks in this area removed in 9.2 and 9.3.
> > Unfortunately the worst of those bottlenecks were in the server, so they
> depend
> > on what database you are upgrading from, and so won't help you much
> upgrading
> > from 9.1.
>
> Yes, I assume 9.3 will be much better, though Jeff is right that if it
> is pg_dump locking that is hurting you, you  might not see a win even in
> 9.3.
>
>
> I'll see it next year when we plan to migrate to 9.4
>
> I though so some form of "superlock" can be interesting, because nobody can
> work with database when it is upgraded.

Remember pg_upgrade is using pg_dump, which then connecting to a
backend, so passing that super-lock mode there is not ideal. The fixes
in 9.3 improve locking in all user cases, not just upgrades.

--
Bruce Momjian <bruce(at)momjian(dot)us> http://momjian.us
EnterpriseDB http://enterprisedb.com

+ Everyone has their own god. +

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Pavel Stehule 2014-03-10 19:12:20 Re: pg_upgrade on high number tables database issues
Previous Message Robert Haas 2014-03-10 19:09:41 Re: extension_control_path