HA Media Interface Keepalive

In an HA node, it is possible for the two systems in the node to lose communication via the management (rear, wancom) interfaces. For example, wancom 1 and wancom 2 might become disconnected, and cause the heartbeat synchronization to fail. This type of failure causes communication errors because both systems try to assume the active role and thereby access resources reserved for the active system.

To avoid these types of conditions, you can enable an option instructing the standby system to take additional time before going to the active state. This check occurs through the system’s media interfaces. Using it, the standby can determine whether or not there has been a true active failure.

In cases when the standby determines the active system has not truly failed, it will go out of service because it will have determined it no longer has up-to-date data from its active counterpart. You can restore functionality by re-establishing management (rear) interface communication between the system in the node, and then re-synchronizes the standby by rebooting it.

When you enable the media interface keepalive, the standby system in the HA node sends ARP requests to determine if the media interfaces’ virtual IP address are active. There are two possible outcomes:

• If it receives responses to its ARP requests, the standby takes itself out of service—to prevent a conflict with the active.
• If it does not receive responses to its ARP requests within a timeout value you set, then standby assumes the active role in the HA node.