PostgreSQL 12 High Availability Cookbook - Third Edition

How to do it...

For the purposes of this recipe, pgha1 will remain our primary node, the standby will be pgha2, and our witness is pgha3. The VIP is 10.0.30.50. Follow these steps to test the failover process:

1. Issue the following command on pgha1 to stop PostgreSQL on a Debian system:

sudo systemctl stop postgresql@12-main

Red Hat derivatives should use this command to stop PostgreSQL instead:

sudo systemctl stop postgresql-12

2. Watch the repmgr logs on pgha2 as the postgres user:

tail -f /var/log/postgresql/repmgr.log

3. Watch the state of the cluster on pgha3 with this command as the postgres user:

watch -n 1 repmgr cluster show

4. Once pgha2 assumes the primary role, execute this command to check the state of the VIP on pgha2:

ip addr ...