[SOLVED] HA failover not working

Rancher
1

hello all,
first of all thank you for providing such a good system.
i’ve a question about HA. Right now, in our staging system, I’ve deployed my HA Cluster in the following configuration:
1x rancher/rancher server
2x all roles server
2x worker server.

2018-05-14%2016_20_08-Rancher
2018-05-14%2016_20_08-Rancher1184×271 18.7 KB

for testing purpose, I’ve shutted down 1 of the 2 “all roles server” (kube3) and the cluster stopped working with the following error message:

This cluster is currently Unavailable; areas that interact directly with it will not be available until the API is ready.

Failed to communicate with API server: Get https://10.10.1.77:6443/api/v1/componentstatuses: net/http: request canceled (Client.Timeout exceeded while awaiting headers)

what’s wrong in my configuration?
thank you very much
best regards
Nick

2

I think an odd number of servers is required for etcd and controlplane roles (specifically 1, 3, or 5). Try adding another “All roles” server to see if it works then.

3

Hi Pib,
you were right. Adding a 3rd server with all roles worked correctly, once a server is unavailable.
thank you very much.
best regards

Nick