Hi everyone,
I would like to verify my understanding of PAM360 High Availability options with the embedded PostgreSQL database:
Primary + Secondary (Shared DB):
The Secondary server does not maintain an independent database—it connects over the network to the PostgreSQL service running locally on the Primary server. If the Primary host suffers a power outage or hardware failure, the Secondary server stops working entirely because its database is hosted locally on the dead Primary. There is no database redundancy.
Primary + Read-Only (Independent DB):
The Read-Only node runs its own independent replicated PostgreSQL instance, providing genuine database redundancy. However, bringing full read/write capabilities back during a disaster is manual: an administrator must detach replication, run promotion scripts, and manually rebuild the replication pairing once the original primary recovers.
Primary + Secondary + Read-Only (3-Node Setup):
If the Primary host goes down, the Secondary server is unusable because the shared DB on the Primary is dead. Resuming operations requires the exact same manual promotion process on the Read-Only node as the 2-node model. The Secondary node only protects against an application-level web service crash while the Primary host and database remain healthy.