What Three Components Are Needed for SAP NetWeaver ABAP High Availability on AWS?

Quick Answer

When a question asks which components need protection in a multi-AZ active/passive cluster for the SAP NetWeaver ABAP stack, it's really asking which parts of the architecture hold state that would be lost or corrupted on failover if not specifically replicated or made cluster-aware. The database instance is one of those components because it holds all persistent business data - if it isn't part of the highly available cluster design, a failure there takes down the entire system regardless of how resilient the application layer is. The Enqueue Replication Server (ERS) is another, because it continuously replicates the lock table from the primary ASCS instance to the standby node; without that replication, a failover would force all currently locked resources to be released, which can produce data inconsistencies or broken transactions for users who were mid-process at the time of failure. The pattern to notice is that the NetWeaver ABAP stack has more than one place where state lives outside the application servers themselves - the database and the enqueue lock table are both examples - and each needs its own explicit high-availability treatment in a cluster, rather than assuming that clustering the application servers alone is sufficient. Whenever you see a select-THREE question about NetWeaver high availability, expect the answer to include the database, the ASCS and ERS pairing, and possibly shared storage, since those are the classic single points of failure in this architecture.