Details
-
Bug
-
Resolution: Won't Fix
-
Major
-
0.7.0
-
None
Description
Once an auto-failover has occurred the internal reset counter isn't automatically reset to prevent subsequent auto-failovers in order to allow manual operators to manually inspect the cluster:
This counter prevents the server from automatically failing over additional nodes until you identify the issue that caused the failover and resolve it. If the internal counter indicates a node has failed over, the server will no longer automatically failover additional nodes in the cluster
|
So in order to allow unsupervised auto-failover we'll need to call the resetCount.
I think the way to offer this as an option in the spec. Enabled by default, but can be disabled.