Synchronizing Replicated Databases



The replica synchronization process requires pgrd daemon running on each involved host. If you copied pgrd startup script from pgreplica/bin/utilities directory into /etc/rc.d/init.d and you restarted the system, then pgrd is probably already running, otherwise change working directory to pgreplica/bin and run pgrd as daemon:

./pgrd &



To start database synchronization you can run pgr command from any one of involved hosts.

To synchronize the test database which is replicated on Ut and Penguin hosts we can run pgr command on Ut (Ut becomes replica master site in this replica session).

./pgr -sync test



A brief log of replica process is displayed. You can find the entire log file test.log in pgreplica/data/rbase/test/log directory.

If you run pgr without arguments, a brief help message is displayed

./pgr

Usage: pgr [-option] dbname

-help this help

-list list postgreSQL replicated databases

-check check remote hosts status

-sync start replica synchronization process

-resolv stop replica session before updating users tables, then standby

-update complete a -resolv replica session

-standby suspend replica session after updating users tables

-crs commit a standby replica session

-rrs rollback a standby replica session





NEXT SECTION: Data Ownership Models