Idea ID 2697884
In HAC configuration (2 DPS), we had a crash of the HyperV cluster hosting DPS 1 which was the active one (meaning all HAC processes running on it).
We expected that all those HAC processes switch to the other DPS but only some did leading by a longer outage.
The RnD confirm there is no internal mechanism to ensure all processes are at least active in one DPS and need to improve this mechanism (ex: like for the bus, store regularly on DB the status of each services and switch to the other if one is not running).
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.