Planning for SAP NetWeaver

SAP NetWeaver High Availability feature is supported for the following configurations:
  • Two-node cluster deployment
  • PowerHA® SystemMirror® 7.2.3 or later: Four-node cluster deployment

Review the following information before you configure Smart Assist for SAP NetWeaver:

  • The following SAP Central Services high availability configurations are supported through the wizard:
    • Advanced Business Application Programming (ABAP) System Central Services (ASCS) High Availability
    • JAVA System Central Service instance (SCS) High Availability
    • SAP Application Server only High Availability
    • SAP router failover
    • SAP Web Dispatcher failover
    • Start after dependency for SAP HANA database
  • You must install SAP NetWeaver components before you configure resources by using Smart Assist.
  • User-defined resources such as Persistent IP are deleted when a Smart Assist policy is activated. You must configure the user-defined resources again.
  • You must configure the Network File System (NFS) tie breaker by using the clmgr command if you configure even number of nodes in the cluster.

Standalone Enqueue Server 2 support

The Standalone Enqueue Server 2 (ENSA2) in SAP NetWeaver Application Server for an SAP Advanced Business Application Programming (ABAP) system provides improved scalability and advanced SMENQ transaction. You can use the advanced SMENQ transaction instead of the previous SM12 transaction to monitor and administer ENSA2. The standalone ENSA2 provides a simplified high availability architecture and the Enqueue Replicator 2 (ERS2) provides an improved failover process.

In PowerHA SystemMirror Version 7.2.3, or later, Smart Assist for SAP NetWeaver discovers and supports Enqueue Server (ENSA) by using Enqueue Replicator Server (ERS) and ENSA2 with ERS2. The ENSA2 can be configured either in a two-node cluster configuration or a multi-node cluster configuration. In a two-node cluster configuration, ABAP SAP Central Services (ASCS) fails over to the node where ERS is running. In a multi-node cluster configuration, ASCS fails over to any online node in the cluster where ERS is not running.