After rke upgrade to HA rancher cluster, rancher intermmitantly reports "cluster is currently unavailable"

Rancher
1

Greetings,
I just followed these instructions:

https://rancher.com/docs/rke/latest/en/upgrades/

To upgrade the cluster that I have rancher HA installed on. Looks like it went from 1.14.5 to 1.17.2

The rke command reported the upgrade went successful, but now in the Rancher UI, when viewing the upgraded ‘local’ cluster, it reports:

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

For 15 seconds, the message then disappears, then the message comes back - follows this 15 seconds on, 15 seconds off pattern indefinitely.

Perhaps also worth noting that under nodes, it only shows one of the nodes having the upgraded version; the other two nodes still show 1.14.5

I had upgraded the cluster in preparation for upgrading rancher itself, but am hesitant to move forward until I get this sorted.

Is there something I need to do to have rancher properly interface with the upgraded cluster that it resides on?

2

So what happened was the nodes of the ‘local’ cluster that rancher was on, after the rke update, were in a ‘NotReady’ state and needed a reboot to finish updating and get in a ‘ready’ state; once I did that, rancher was, not surprisingly, unable to communicate with the cluster at all, I presume since 1.17.2 is an unsupported version for rancher 2.2.8. I then upgraded rancher to 2.3.5, and all is well.