Watchdog fails, or the communication link between the watchdog and one or both of the servers fails
Scenario
The watchdog fails, the connection between the watchdog and the primary server is broken, or the connection between the watchdog and the secondary server is broken.
Symptoms
Either the watchdog process has failed or, if the watchdog is still running but has lost contact with one of the servers, one of the following symptoms occurs:
▪ The watchdog poll of the primary server fails but the secondary server state is reported to be SECONDARY ACTIVE
▪ The watchdog poll of the secondary server fails but the primary server state is reported to be PRIMARY ACTIVE.
In either case, the servers are likely to be functioning as normal.
Remedy
The following table describes the steps that should be taken by the watchdog and the administrator in order to return the service to normal operation.
Description
Illustration
If the watchdog fails, or if one or both of the polls fail, the primary and secondary server continue normal operation.
or
The administrator fixes the issue.
The watchdog checks the state of each server with the command:
ADMIN COMMAND 'hotstandby state';
If the states of the servers have changed and one server is no longer in service, refer to the applicable scenario for instructions:
