High Availability

High Availability (HA) is a network configuration used to ensure that planned and unplanned outages do not disrupt service. In an HA configuration, Oracle Communications Session Border Controllers (SBC) are deployed in a pair to deliver continuous high availability for interactive communication services. Two SBCs operating in this way are called an HA node. The HA node design ensures that no stable call is dropped in the event of an outage.

In an HA node, one SBC operates in the active mode and the other SBC operates in the standby mode.

• Active. The active member of the HA node is the system actively processing signal and media traffic. The active member continuously monitors itself for internal processes and IP connectivity health. If the active member detects a condition that can interrupt or degrade service, it hands over its role as the active member of the HA node to the standby member.
• Standby. The standby member of the HA node is the backup system. The standby member is fully synchronized with the active member's session status, but it does not actively process signal and media traffic. The standby member monitors the status of the active member and it can assume the active role without the active system having to instruct it to do so. When the standby system assumes the active role, it notifies network management using an SNMP trap.

The SBC establishes active and standby roles in the following ways.

• If an SBC boots up and is alone in the network, it is automatically the active system. If you pair a second SBC with the first one to form an HA node, the second system automatically establishes itself as the standby.
• If both SBCs in the HA node boot up at the same time, they negotiate with each other for the active role. If both systems have perfect health, then the SBC with the lowest HA rear interface IPv4 address becomes the active SBC. The SBC with the higher HA rear interface IPv4 address becomes the standby SBC.

If the rear physical link between the two SBCs is unresponsive during boot up or operation, both will attempt to become the active SBC. In this circumstance, processing does not work properly.

The standby SBC assumes the active role when:

• it does not receive a checkpoint message from the active SBC for a certain period of time.
• it determines that the active SBC health score declined to an unacceptable level.
• the active SBC relinquishes the active role.

To produce a seamless switch over from one SBC to the other, the HA node members share their virtual MAC and virtual IP addresses for the media interfaces in a way that is similar to Virtual Router Redundancy Protocol (VRRP). Sharing these addresses eliminates the possibility that the MAC address and the IPv4 address set on one SBC in an HA node will be a single point of failure. Within the HA node, the SBCs advertise their current state and health to one another in checkpointing messages to apprise each one of the other one's status. Using the Oracle HA protocol, the SBCs communicate with UDP messages sent out and received on the rear interfaces. During a switch over, the standby SBC sends out an ARP request using the virtual MAC address to establish that MAC address on another physical port within the Ethernet switch. To the upstream router, the MAC address and IP address are still alive. Existing sessions continue uninterrupted.