Re: support for distributed transactions - Mailing list pgsql-novice

From Richard Broersma Jr
Subject Re: support for distributed transactions
Date
Msg-id 403639.76475.qm@web31802.mail.mud.yahoo.com
Whole thread Raw
In response to support for distributed transactions  ("Brian Oki (boki)" <boki@cisco.com>)
List pgsql-novice
--- "Brian Oki (boki)" <boki@cisco.com> wrote:

> My question is this: How does PostgreSQL 8.2.5 execute DML statements
> (insert, update, delete, select) on remote nodes as part of the same
> transaction? Where is the syntax specified? Or, is there a different
> model supported?

I am not to sure.  If you don't get many good answers, I would repost this question on the
PG-general mailing list since it is more active and will have users that can give you a well
informed comment.

There are two ideas that I have that may or may not be what you are looking for.
My first idea is basically replication.  You write to the "master" database and the changes are
cascaded to various nodes.  The two projects that come to mind are slony and mammoth replicator.
http://slony.info/
http://www.commandprompt.com/products/mammothreplicator/

The other idea is to write to tables on external server db-clusters.
http://pgfoundry.org/projects/dbi-link/

I assume that sql statements issued from dbi-link can be acid compliant, however, I doubt that it
supports referential integrity across the various nodes.

Regards,
Richard Broersma Jr.

pgsql-novice by date:

Previous
From: Tom Lane
Date:
Subject: Re: support for distributed transactions
Next
From: "Brett Maton"
Date:
Subject: Problem with PG_GETARG_CSTRING