[pgpool-general: 1125] Re: Using PgPool-II on multiple hosts

[Wolf Schwurack]

Sorry but I not sure what you are trying to do. If you're asking if you can connect directly to the postgres database without pgpool - yes. I have 4 servers 2 for pgpool, 1 postgres master and 1 postgres replicate. I can connect to the database using pgpool or directly to the master or replicate. I'm only using pgpool for the application connections. If I need to do DBA work on the database I connect to the master.

Wolf

From: Csaba Simon [mailto:simocsab at gmail.com]
Sent: Wednesday, October 24, 2012 9:25 AM
To: Wolf Schwurack
Cc: pgpool-general at pgpool.net
Subject: Re: [pgpool-general: 1107] Using PgPool-II on multiple hosts

As far as I know Slony is also a master/slave replication, moreover it is asynchronously, so if the master site server fails the newest pieces of data can be lost.

I can describe the problem in an other way: can a PostgresSQL database behind PgPool-II be modified by other applications directly? If it can, then tehh other application can be another PgPool-II. In this case I suppose the developers have to take care about distinct sets of identifiers for the sites and special recovery scripts/actions for the case of online recovery...

Csaba
On 24 October 2012 17:11, Wolf Schwurack <wolf at uen.org<mailto:wolf at uen.org>> wrote:
If you are looking for pgpool on multiple host this is the way to go. If one of the pgpool server fails then the other will take over. This in the only method if you want multiple pgpool servers. You don't have to use master/slave. You could set it up using slony if you like

Wolf

From: Csaba Simon [mailto:simocsab at gmail.com<mailto:simocsab at gmail.com>]
Sent: Wednesday, October 24, 2012 9:01 AM
To: Wolf Schwurack
Cc: pgpool-general at pgpool.net<mailto:pgpool-general at pgpool.net>
Subject: Re: [pgpool-general: 1107] Using PgPool-II on multiple hosts

Hi,

Thanks for your quick answer, but this part of the documentation is about a master-slave server pair with a standby server. In contrast with it our architecture would have two PgPool-II nodes on different company sites and they would serve 2-2 databases nodes (2 PostgresSQL db servers per site), all of the nodes should be active and synchronized while working normally.

Is not it possible just use two individual PgPool-II servers for the 4 databases?

Csaba
On 24 October 2012 15:29, Wolf Schwurack <wolf at uen.org<mailto:wolf at uen.org>> wrote:

http://www.pgpool.net/pgpool-web/contrib_docs/watchdog_master_slave/en.html
Take a look at this doc

Wolf

From: pgpool-general-bounces at pgpool.net<mailto:pgpool-general-bounces at pgpool.net> [mailto:pgpool-general-bounces at pgpool.net<mailto:pgpool-general-bounces at pgpool.net>] On Behalf Of Csaba Simon
Sent: Wednesday, October 24, 2012 6:24 AM
To: pgpool-general at pgpool.net<mailto:pgpool-general at pgpool.net>
Subject: [pgpool-general: 1107] Using PgPool-II on multiple hosts

Hi,

Can I use PgPool-II on multiple hosts avoiding my single PgPool-II instance to be the single point of failure in the architecture?

For example I would like to use four active PostgresSql database nodes with two active PgPool-II nodes in replication mode. Is this a usual and usable architecture?

Csaba


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.sraoss.jp/pipermail/pgpool-general/attachments/20121024/8a0c8a73/attachment-0001.html>