Primary instance stops updating the heartbeat point

There are multiple situations that cause IF-Node1 to stop updating the heartbeat point. Examples include:

• The interface crashes.

• The interface is unable to write data for any reason.

• The interface hangs in a function call that does not return.

When IF-Node1 stops updating the heartbeat point, IF-Node2 determines if it should assume primary role using the following operations:

1. At each failover update interval, IF-Node2 reads and stores the value of the heartbeat point for IF-Node1.

2. If the heartbeat point value for IF-Node1 does not update between two readings, IF-Node2 stops deleting older queued data and continues to queue data until it knows the state of IF-Node1.

3. If the heartbeat point value for IF-Node1 does not update for four failover update intervals, IF-Node2 transitions to the AssumingControl state.

4. In AssumingControl state, IF-Node2 immediately writes its failover ID to the active ID point.

5. IF-Node2 remains in AssumingControl state for two update intervals.

6. If the active ID value is still the same as the failover ID for IF-Node2, IF-Node2 transitions to PrimaryReady state and writes all queued data to Data Archive.

Total failover time for this scenario is between three and five failover update intervals. The exact amount of time depends on when IF-Node1 failed and when IF-Node2 read the control points. When using hot failover, overlapping data is dependent on this timing.