Received: from localhost (maia-1.hub.org [200.46.204.191]) by postgresql.org (Postfix) with ESMTP id 8E4749FA131 for ; Thu, 4 Jan 2007 06:07:31 -0400 (AST) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.191]) (amavisd-new, port 10024) with ESMTP id 80353-06 for ; Thu, 4 Jan 2007 06:07:20 -0400 (AST) X-Greylist: from auto-whitelisted by SQLgrey-1.7.4 Received: from bugaboo.mu (ns1.bugaboo.mu [213.133.111.57]) by postgresql.org (Postfix) with ESMTP id 344C09F931E for ; Thu, 4 Jan 2007 06:07:19 -0400 (AST) Received: from [192.168.0.237] (berlin.programmfabrik.de [::ffff:213.221.101.147]) (AUTH: CRAM-MD5 markus@bluegap.ch) by bugaboo.mu with esmtp; Thu, 04 Jan 2007 11:07:16 +0100 id 0223C240.459CD1D5.00002F00 Message-ID: <459CD1D1.8010002@bluegap.ch> Date: Thu, 04 Jan 2007 11:07:13 +0100 From: Markus Schiltknecht User-Agent: Icedove 1.5.0.9 (X11/20061220) MIME-Version: 1.0 To: anorganic anorganic CC: pgsql-admin@postgresql.org Subject: Re: Replication References: <459CC52A.4030208@bluegap.ch> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200701/19 X-Sequence-Number: 24175 Hi, anorganic anorganic wrote: >> did you check the manual about 'High Availability and Load >> Balancing"? >> http://www.postgresql.org/docs/8.2/static/high-availability.html > > yes but this was really short i think. Interesting... where exactly is it to short for you? What else would you expect? > I need that: s1 and s2 must be multimaster replication asyn. but make > replication in real time. Changes are immediately send to second server. Slony can probably be used for such a setup, although it's primarily designed for single master, async replication. > OR: s1 is master s2 slave, s1 change some columns on s2, and s2 is > master s1 > slave and s2 change another columns in one time. I don't quite follow what you mean here. I assume you mean rows, not columns. Then probably you mean something like what we call data partitioning, where each server is a master for a certain subset of the data. > I read about slony but, it looks that's make changes not immediatly but in > packages = not send one transaction but wait and send package. This > could be a problem. Hm. If that really is a problem, think again about sync vs. async. In async replication, there is a delay by definition. But again, I'm not sure what Slony can do to minimize that delay. > I hope slony online system is right for me. And these question > about slony i have more than one i think it could be rigth way. Somebody > who run it could be great person for me :). Probably ask on the slony mailing list. I'm not exactly a slony expert either. Regards Markus