Details
-
Improvement
-
Resolution: Fixed
-
Major
-
5.0.0
-
None
Description
This is an issue when creating clusters using cluster_run.
Clustering currently only looks to see if the any of the n1ql protocols is found in the nodServices endpoint, and if yes, and finds the local node, it assigns the ip address to itself.
The long and the short of it is that multiple nodes end up with the same name.
This in turn affects monitoring - which can't differentiate in between nodes, and does not know to which service to send requests.
It is unclear if with the current architecture things can be improved upon...