This message was deleted.

revision numbers always go up. any write to the datastore increases the revision number. that is how etcd works.

hmm, any other ideas why we're seeing this since a few days: Cluster health check failed: Failed to communicate with API server during namespace check: Get "https://10.43.0.1:443/api/v1/namespaces/kube-system?timeout=45s": context deadline exceeded or Cluster health check failed: Failed to communicate with API server during namespace check: Get "https://10.43.0.1:443/api/v1/namespaces/kube-system?timeout=45s": tunnel disconnect on the downstream clusters Sometimes the agent disconnects as well

the Rancher server communicates with the downstream clusters through a tunnel that is maintained by the agents, so if the agents go down then yeah… you’d get errors on the Rancher side.

i should see that in the rancher-system-agent.service logs? right?

no, the cattle-cluster-agent pod logs in the downstream cluster.

hm, not much besides connection reset by peer. i've done a failover of the loadbalancer in front of rancher, let's see if that helps