Results 1 to 2 of 2
  1. #1
    Join Date
    May 2013

    Unanswered: Postgres Replication through Pgpool

    Can someone help me in my simple pgpool replication?
    Objective: To implement simple postgres database replication using Pgpool
    Versions used : Postgres 9.2.4 , Pgpool 3.2

    (1) I have created two Centos Virtual machines on my laptop- Centos32, and CentOS32A. Their hostnames are ‘localhost’ and ‘’ (as provided by hostname command and Their IPs are, and ‘ ‘ respectively. (as provided by ifconfig command).

    (2) I installed Postgres 9.2.4 and Pgpool 3.2 on both machines successfully.

    (3) I configured pgpool config files and pcp.conf and pgpool.conf for basic replication. As shown at the end. And restarted the machine.

    (4) As ‘postgres’ user I have created a small database : yash01 on both the hosts. Then created a schema and table ‘sch1.tab1 (aa int)’. Inserted 5 integers (1,2, 3, 4, 5) in the table on both the machines.

    (5) I start the pgpool on host1 – ‘localhost’. I verify that pgpool started correctly by looking at and xxx.log files. And then doing ps –ef and verifying that pgpool and child processes are running.

    (6) On machine one – ‘Localhost’, I insert 5 additional rows in ‘tab1’ in the table (11, 12, 13, 14, 15). Commited it. And then checked it using select.

    (7) I go and verify on machine -2 , ‘’ whether the ‘tab1’ entries are replicated on it.

    I find that it is NOT replicated. I want to know what mistake I am doing?
    Is there something more to be configured - in config files or soemewhere else? I have taken default files pcp.conf and pgpool.conf.sample-replication for replication and made the changes as follows. I did not make any changes to the default database config file.
    Attached Files Attached Files

  2. #2
    Join Date
    Mar 2014
    The glib answer is that using the wrong tool. I've never used pgpool to setup replication, and don't even know if that's possible. pgpool is a proxy interface system for replicated servers allowing read-only queries to be routed to a hot-standby read-only server while write queries only are sent to the master. You really need synchronous replication setup first and the standard postgresql 9.x documentation does a really good job at explaining that setup.

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts