This message was deleted.

Hi on the

local

cluster check if the Secret in

fleet-default

called

custom-<id>-machine-plan

has

0

data

kubectl -n fleet-default get secrets | grep <YOUR_MACHINE_ID>

Hi @prehistoric-gpu-28362 - sorry for the late reply. The secret actually has data

Hi, it's ok, not the same issue I'm seeing then.

I'll have a look, thanks

That sounds like it could ranchers certs though I've not encountered that before so don't know how to fix it, sorry.

I could fix my issue by manually copying the machine plan bash script to the node and run it by hand. After adding a few nodes, the cluster was somewhat stable and I could just run the registration command again on the outstanding nodes. Rancher detected the nodes as "known" and added it to the cluster with no issues.

I ran into this issue after migrating Rancher to a new cluster. Even after restoring the secrets things were still wonky, but manually running the script from the appropriate machine-bootstrap secret did the trick. Thanks for sharing your solution, @agreeable-sugar-99934 !

Even if I'm not @agreeable-sugar-99934, you're welcome @wooden-easter-8724 😉

Wow, how embarrassing 😳 sorry about that!