0
votes

I have problem with my rancher installation:

  1. I had to upgrade Amazon EKS kubernetes version through Rancher UI inerface from 1.12 to 1.13
  2. Then I followed this: https://docs.aws.amazon.com/eks/latest/userguide/update-stack.html because all worker nodes in the cluster had the old kubernetes version
  3. The upgrade according to Cloudformation interface went well
  4. All nodes from Rancher UI cluster view disappeared and all my services became unavailable

I'm not Rancher or kubernetes master obviously. And I have no idea why it went wrong. What should I check?

2

2 Answers

0
votes

How you have created your worker node, is it via launch configuration template or with EKS managed node group.

If it is via launch config template then first create a node group and make pod scheduled over there & after that you can upgrade your node group from the management console or IAC.

0
votes

Please try to terminate the instances from the AWS EC2 console one by one. The Auto Scaling Groups (ASGs) will recreate the replacement instances which will be visible to the cluster, and then the pods/services will come back.