[Pgpool-general] configure replication with pgpool installed on separate host

[Tatsuo Ishii]

> I was using the following docs to get things working and both of these
> docs install pgpool on the same host as the primary server.  But it
> did not make sense for a production environment: if the primary server
> crashes, so does pgpool...

Right. Following docs are not perfect in this regard. They intended to
minimize the computer resoucees needed for starter. I would like to
add appropreate comments to them. Suggestions are welcome.

> http://pgpool.projects.postgresql.org/pgpool-II/doc/tutorial-en.html
> http://pgpool.projects.postgresql.org/contrib_docs/pgpool-II_for_beginners.pdf
> 
> So my web server application needs to send database requests to the
> pgpool head node, then it replicates to the primary and secondary
> database servers.

Another idea is installing pgpool to each web server. The web app
accesses to local pgpool. This way, if you have multiple web servers(I
think this is quite common), you could avoid the SPOF(single point of
failure) problem.
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese: http://www.sraoss.co.jp