New secondary node remains in a "down" status after a switchover. This occurs when different certificate names are used on the primary and secondary nodes.

[gagannarang226]

We've encountered an issue when using different certificates or certificate paths on primary and secondary nodes with postgresql-auto-failover.conf.

When the postgresql-auto-failover.conf files are generated, they include the specific certificate path and name. During a switchover, pg_auto_failover uses pg_rewind to synchronize data between the primary and secondary nodes. In this process, pg_rewind updates the postgresql-auto-failover.conf file of the former primary node to match that of the new primary.

This creates a problem because the former primary node then expects a different certificate or certificate path than what it actually has. As a result, it remains in a down state because it cannot locate its certificate.