This message was deleted.

This post solved my problem (https://www.suse.com/support/kb/doc/?id=000021274). Specifically running this on the rancher container:

kubectl annotate <http://clusters.management.cattle.io|clusters.management.cattle.io> <REPLACE_WITH_CLUSTERID> io.cattle.agent.force.deploy=true

I'm not sure what else that does differently from updating

CATTLE_SERVER

and

CATTLE_CA_CHECKSUM

in the

cattle-cluster-agent

deployment like I did via

kubectl

directly against the cluster but it worked.