AW: Big projet, please help

From: Zeugswetter Andreas SB <ZeugswetterA(at)wien(dot)spardat(dot)at>
To: "'ohp(at)pyrenet(dot)fr'" <ohp(at)pyrenet(dot)fr>, "'pgsql-hackers(at)postgresql(dot)org'" <pgsql-hackers(at)postgresql(dot)org>
Subject: AW: Big projet, please help
Date: 2000-06-08 09:02:21
Message-ID: 219F68D65015D011A8E000006F8590C604AF7DC9@sdexcsrv1.f000.d0188.sd.spardat.at
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers


> Hi all,
>
> I'm faced to a big problem!!
>
> I have to do this for a customer:
>
> create a database on my server; this database will be queried
> and updated
> through the web --- this is easy
>
> H!ave the same databse on my customer server. This databse
> will be queried
> and updated by the customer . -esay too
>
> The 2 databases have to be synchronized both ways ! Huh How can I do
> that???
>
> Can you give me some pointers... I'd love to do it with postgresql
> instead of going to Oracle just because they can replicate bases...

In an environment with moderate to low update activity at least on one side
the simplest and most reliable replication mechanism is usually done with
triggers
that work on the primary key.
They provide a synchronous replication on the basis of all or nothing,
and thus solve the problem of concurrent update to the same row
from both sides.
The tricky part is usually how to break the trigger chain. In Informix you
can
set global stored procedure variables to "local" and "remote" and only
trigger if that
variable is set to "local". (create trigger .... when myconn()="local" ...)

Andreas

Browse pgsql-hackers by date

  From Date Subject
Next Message Zeugswetter Andreas SB 2000-06-08 09:23:43 AW: Sigh, LIKE indexing is *still* broken in foreign lo cales
Previous Message Zeugswetter Andreas SB 2000-06-08 08:48:34 AW: Look at heap_beginscan()