RancherClientUnreachable

This alert pops up when client clusters are inaccessible through Rancher.

Solutions

on the cloud zone

Ensure other client zones managed by the parent cloud zone are unaffected. Otherwise, there is probably a larger problem which requires analysis (rancher server in cloud zone, ingress-controller, network,...)

on the client zone

The logs on the affected client zone should be analyzed first in order to try to identify a root cause (cluster agent logs, events). Finally, the rancher cluster agent pods on the affected client zones should be restarted

On the affected client zone:

kubectl -n cattle-system get pods
kubectl -n cattle-system get events

For each of the pods shown:

kubectl -n cattle-system describe pod cattle-cluster-agent-XXXXX
kubectl -n cattle-system delete pod cattle-cluster-agent-XXXXX