The Primary database (in the PRIMARY ACTIVE state) on node 1 and the Secondary database (in the SECONDARY ACTIVE state) on node 2 both encounter connection failure to each other.
Recovery
In the recovery from the HotStandby link failure the HAC instances ping the External Reference Entity (ERE) to find out if it is the network or the opposite server that has failed. The recovery proceeds automatically as follows:
1 Upon the connection failure, both databases move automatically to the PRIMARY UNCERTAIN (node 1) and SECONDARY ALONE state (node 2), respectively.
2 The direct connections of both HAC instances to the remote server fail.
3 Both HAC instances ping the ERE by using the operating system’s ping utility.
4 If the ping fails, the local server is retained or set to the SECONDARY ALONE state.
5 If the ping succeeds, the successful HAC tries to connect to the remote database server.
6 If the connect effort to the remote database server fails, the HAC concludes that its part of the network connection is operational and sets the local server to the PRIMARY ALONE state.
7 The HAC instance on the Primary database attempts to re-establish the connection to the Secondary database.
8 Once the network becomes operational and the connect succeeds, the Primary database and the Secondary database move automatically to the PRIMARY ACTIVE and SECONDARY ACTIVE state, respectively.