I was wondering if anybody experienced and/or resolved the following issue:
Windows Failover cluster Setup:
- Two Windows 2016 or 2012-R2 server nodes: A and B with current Windows patches.
- Generic Application DLL resource: implements IsAlive(), LookAlive(), Online() and Offline()
- Virtual IP address resource: as a dependency of the Generic Application
-
Policy: configured to failover at the first failure
- Period for Restarts=15:00
- Maximum restarts in the specified period=0
- Delay between restarts=0
Issue:
When IsAlive() fails on A primary server, the cluster manager:
- Does not call Offline() on A (leaving A online)
- Moves VIP address from A to B
- Calls Online() on B
As a result, both A and B Application resources are online.