Problems rebooting K3S server that was imported into Rancher v2.3.3

I recently put together a 6 node cluster based on K3S and PI4’s running on Ubuntu 19.10. The cluster seems to be running ok and I’ve successfully imported it into Rancher V2.3.3

After a while one of the Pi’s would go offline, so far I haven’t been able to find out what caused the problem. So I went to reboot each one of them with a monitor attached so I can dig into it further - I believe it is because the WiFi went offline… but can’t verify without the console.

I rebooted each of the worker nodes- each one came back online to ranch quickly.

However when I rebooted the K3S server node it shows up as ‘unavailable’ with ‘kubelet stopped posting node status’. I’ve rebooted a few times and still have that problem.

BTW, when the server node is rebooting, I see the cluster go ‘offline’ in rancher - which I expected… Just don’t know how to resolve the ‘kublet stopped working’ problem…

Second question, since I’m here, is there any intent is updating the Apps Catalog with an architecture indicator - to know what will and won’t run on arm64 architectures?