Re: High availibility with PostgreSQL

From: Andrew Sullivan <andrew(at)libertyrms(dot)info>
To: pgsql-general(at)postgresql(dot)org
Subject: Re: High availibility with PostgreSQL
Date: 2002-07-23 14:32:08
Message-ID: 20020723103208.B23294@mail.libertyrms.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Tue, Jul 23, 2002 at 06:29:14AM +0000, IH8MSFT wrote:
> I'm trying to find out what the current best solutions for failover are
> currently available for PostgreSQL on Linux. A commercial soultion would
> be OK with us.
>
> I've scoured google groups but only came up with the eserver stuff which
> seems to have not been updated since 2000.

We use it. It works. The commercial version from PostgreSQL, Inc.
is better than the code in contrib/ by a long shot.

It is _not_ a hot failover system, however, and I'm not even sure it
could be made so. It's a store-and-forward sort of replication
system. If the master database fails, you don't want one of the
slaves to take over automatically, because you'll have "stranded"
data -- stuff that made it in to your master before the failure, but
which did not get replicated.

If you are running high-quality servers, the chances of total failure
are not that big. What is more likely is a sort of "emergency
shutdown" case, where you have to sync to the slave and then cut over
to it while you fix the (now former) master. The rserv code will
allow you to do that, although it's a bit of a pain to resync the
systems.

I believe you can get more information from <pgsql(at)pgsql(dot)com>. I am
not associated with PostgreSQL, Inc, by the way, except as a
customer.

A

--
----
Andrew Sullivan 87 Mowat Avenue
Liberty RMS Toronto, Ontario Canada
<andrew(at)libertyrms(dot)info> M6K 3E3
+1 416 646 3304 x110

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Carmen Wai 2002-07-23 14:35:40 Case in-sensitive
Previous Message Jan Wieck 2002-07-23 14:23:34 Re: Linux max on shared buffers?