About Setting Up the Topology

Setting up the topology consists of configuring Policy Management sites and clusters, including their IP addresses and hierarchy. You can add MPE and MRA clusters to the topology before configuring the individual server profiles. You can set up all the servers in a cluster in the same operation.

The recommended sequence of setting up the Policy Management topology is as follows:
  1. Add a CMP Site 1 cluster as described in Setting Up a CMP Cluster.
    Note: The primary CMP cluster cannot be deleted from the Policy Management topology.
  2. (Optional) Add a Site 2 CMP cluster.

    See Setting Up a CMP Cluster for details.

    Note: A Site 2 CMP cluster can provide georedundancy.
  3. Add non-CMP (MPE and MRA) clusters.

    See Setting Up a Non-CMP Cluster for details.

  4. (Optional) For georedundancy, configure sites and additional MPE and MRA clusters.

    See Setting Up a Georedundant Site and Setting Up a Georedundant Non-CMP Cluster for details.

  5. After setting up the topology, the topology settings are replicated as follows:
    1. The active CMP server replicates the topology configuration, including the cluster settings, to active, standby, and (if present) spare servers over the OAM network. These servers form an MPE or MRA cluster based on the topology configuration. The servers have the following communication restrictions:
      • Active servers communicate with standby servers using LAN connections over the OAM network.
      • Active servers communicate with spare servers using WAN connections over the OAM, SIG-A, SIG-B, or REP network.
      • Active and standby servers share a virtual IP (VIP) address to support automatic failover.
      • If present, the spare server has a unique VIP address.
    2. The HA database runtime process constantly monitors the status of the servers in each cluster and, in the event of a server or servers failure, triggers the following actions:
      • If an active server in a cluster fails, the standby server takes over and becomes the active server.
      • In a georedundant topology, if both the active and standby servers in a cluster fail, HA instructs the spare server to take over and become the active server.
  6. After you set up the topology, view the System tab of each server and determine if there are any configuration mismatches. See About Reapplying a Configuration for more information.
Parent topic: Configuring the Policy Management Topology