High Availability failover not working after reboot during maintenance window (i.e. Windows patch update)

After rebooting the 'Active' server in a High-Availability pool, this should trigger a failover event to the 'Standby' server. The 'Standby' server takes over as the 'Active' server while the other pool members reboot. When the rebooted server returns online within the Registry and the Database, it remains the 'Active' server, preventing further failover in the pool. 

This issue occurs due to how the ownership logic is determined in SolarWinds High Availability. The previous logic relied on the local cache and registry setting on each poller to determine who was the active server.

After SolarWinds Platform 2024.2, the logic has been simplified, and the standby poller will remain on standby unless its heartbeat is older than 30 seconds. If the heartbeat is older than 30 seconds, then the standby will update the HA_PoolMembers tables and become the active node.

(Screenshots property of © 2018 Microsoft Corporation)

SolarWinds recommends upgrading to SolarWinds Platform 2024.2.1 or later for the improved HA logic to resolve this issue.

If you are unable to upgrade, complete the following steps.

1. RDP to the machine that shows as the 'Standby' in the web console
2. Open up SolarWinds Platform Service Manager and click 'Shutdown Everything' at the top to ensure all services are stopped
3. Open up regedit.exe
4. Navigate to Computer\HKEY_LOCAL_MACHINE\SOFTWARE\WOW6432Node\SolarWinds\HighAvailability
5. Edit the 'RunType' value and add 'Standby' to the end of the string as shown below. (Capitalization Matters)

(Screenshots property of © 2018 Microsoft Corporation)

6. Afterwards, the database entry will update itself automatically, and the HA pool will be back in working order. These same steps apply to 'AdditionalPoller' pool members as well.

Note: These steps may not work for other SolarWinds High Availability scenarios where the pool is not functional, and a more thorough cleanup may be needed.