replication only

On 05/08/2008 08:21, Jef Peeraer wrote:

would be the best option to go with. Slony i presume, although schema
chanages are not propagated.

Schema changes *are* propagated in Slony, using the EXECUTE SCRIPT
statement:

http://www.slony.info/documentation/stmtddlscript.html

Ray.

------------------------------------------------------------------
Raymond O'Donnell, Director of Music, Galway Cathedral, Ireland
rod@iol.ie
Galway Cathedral Recitals: http://www.galwaycathedral.org/recitals
------------------------------------------------------------------

In response to Jef Peeraer <jef.peeraer@telenet.be>:

i read about the replication possibilities with postgresql. If i just need
some replication ( without failover stuff ) to 1 standby server, what
would be the best option to go with.

Your description of you requirements is very lacking, so much so that
any attempt at suggesting a "best" option would be pointless.

Provide some more information on your requirements and people will be
able to answer intelligently.

--
Bill Moran
Collaborative Fusion Inc.
http://people.collaborativefusion.com/~wmoran/

wmoran@collaborativefusion.com
Phone: 412-422-3463x4023

On Tue, 5 Aug 2008, Bill Moran wrote:

In response to Jef Peeraer <jef.peeraer@telenet.be>:

i read about the replication possibilities with postgresql. If i just need
some replication ( without failover stuff ) to 1 standby server, what
would be the best option to go with.

Your description of you requirements is very lacking, so much so that
any attempt at suggesting a "best" option would be pointless.

Provide some more information on your requirements and people will be
able to answer intelligently.

my apology, i tried to be brief.
i've got a main database, where all transactions end up. Daily we make a
backup (pg_dumpall) and restore the first database in a sort of backup
database. This backup database is mainly used to do reporting.
To automate this process for the backup database and to minimize the
out-of-sync state, i should like to use replication.
The second backup database is only used as read-only,
except for some third party software that writes to 1 table.

jef peeraer

In response to Jef Peeraer <jef.peeraer@telenet.be>:

On Tue, 5 Aug 2008, Bill Moran wrote:

In response to Jef Peeraer <jef.peeraer@telenet.be>:

i read about the replication possibilities with postgresql. If i just need
some replication ( without failover stuff ) to 1 standby server, what
would be the best option to go with.

Your description of you requirements is very lacking, so much so that
any attempt at suggesting a "best" option would be pointless.

Provide some more information on your requirements and people will be
able to answer intelligently.

my apology, i tried to be brief.
i've got a main database, where all transactions end up. Daily we make a
backup (pg_dumpall) and restore the first database in a sort of backup
database. This backup database is mainly used to do reporting.
To automate this process for the backup database and to minimize the
out-of-sync state, i should like to use replication.
The second backup database is only used as read-only,
except for some third party software that writes to 1 table.

You'd probably be best with Slony, just don't replicate that 1 table
that you'll need to write to.

--
Bill Moran
Collaborative Fusion Inc.
http://people.collaborativefusion.com/~wmoran/

wmoran@collaborativefusion.com
Phone: 412-422-3463x4023

On Tue, Aug 5, 2008 at 7:34 AM, Bill Moran
<wmoran@collaborativefusion.com> wrote:

In response to Jef Peeraer <jef.peeraer@telenet.be>:

You'd probably be best with Slony, just don't replicate that 1 table
that you'll need to write to.

That's one of my favorite things about slony.