Lists: | pgsql-jdbc |
---|
From: | henk53602(at)hotmail(dot)com (henk53) |
---|---|
To: | pgsql-jdbc(at)postgresql(dot)org |
Subject: | Replicating JDBC proxy |
Date: | 2004-05-14 12:38:07 |
Message-ID: | 135a85f7.0405140438.7837b73c@posting.google.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Lists: | pgsql-jdbc |
I would like to have a constantly updated copy of my DB. In order to
realize this I have though of the following scheme:
A server is equiped with 3 NICs and is running a special 'proxy' JDBC
implementation.
The only thing this JDBC proxy does is:
1) Forwarding every update request on both ports. On each port a
server is running a JDBC compliant DB (in my case, postgresql)
2) Forwarding every select request on only 1 port.
Theoretically, this would give me a total synchronized DB, wouldn't
it? However, this sounds relative simple to implement. I've seen a
couple of replication efforts undertaken for postgresql that seems to
take forever to finish.
Am I overlooking something?
From: | Dave Cramer <pg(at)fastcrypt(dot)com> |
---|---|
To: | henk53 <henk53602(at)hotmail(dot)com> |
Cc: | "pgsql-jdbc(at)postgresql(dot)org" <pgsql-jdbc(at)postgresql(dot)org> |
Subject: | Re: Replicating JDBC proxy |
Date: | 2004-05-21 11:13:34 |
Message-ID: | 1085138014.1591.23.camel@localhost.localdomain |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Lists: | pgsql-jdbc |
check out c-jdbc
It does all you want and more..
Dave
On Fri, 2004-05-14 at 08:38, henk53 wrote:
> I would like to have a constantly updated copy of my DB. In order to
> realize this I have though of the following scheme:
>
> A server is equiped with 3 NICs and is running a special 'proxy' JDBC
> implementation.
> The only thing this JDBC proxy does is:
>
> 1) Forwarding every update request on both ports. On each port a
> server is running a JDBC compliant DB (in my case, postgresql)
>
> 2) Forwarding every select request on only 1 port.
>
> Theoretically, this would give me a total synchronized DB, wouldn't
> it? However, this sounds relative simple to implement. I've seen a
> couple of replication efforts undertaken for postgresql that seems to
> take forever to finish.
>
> Am I overlooking something?
>
> ---------------------------(end of broadcast)---------------------------
> TIP 9: the planner will ignore your desire to choose an index scan if your
> joining column's datatypes do not match
>
>
>
> !DSPAM:40ad9632179968186840688!
>
>
--
Dave Cramer
519 939 0336
ICQ # 14675561