Closed cnw004 closed 4 years ago
@cnw004 , we've observed this behavior when kops itself is upgraded. Even without any changes to Kube version, it'll try to upgrade various internal components like protokube etc.
@pradeepnnv thank you so much, this is exactly what the issue was. Do you have any experience doing this? Is updating the kops version dangerous to the cluster itself? It seems to be wanting to make a ton of changes and that makes me nervous. I know that the matrix here says that it shouldn't be a problem
@cnw004 , we didn't run into any issues when did similar updates. We had kops at a higher version than Kube itself. But i think it's best not to use a very latest kops for managing an older version of kube cluster. Best to upgrade both of them in tandem.
@pradeepnnv sounds good, thanks for the tips. I really appreciate you helping me out with this, I was totally lost. Closing out the issue because this is resolved on my end!
1. What
kops
version are you running? The commandkops version
, will display this information.2. What Kubernetes version are you running?
kubectl version
will print the version if a cluster is running or provide the Kubernetes version specified as akops
flag.3. What cloud provider are you using? AWS
4. What commands did you run? What is the simplest way to reproduce this issue?
kops update cluster
nothing to update, everything looks goodkops edit cluster
changed kubernetesVersion from 1.11.6 to 1.12.0kops update cluster
saw a bunch of things were going to change, didn't like thatkops edit cluster
changed kubernetesVersion from 1.12.0 back to 1.11.6kops update cluster
still seeing a bunch of things that are going to change, most of them having to do with going from 1.11.6 to 1.12.0 (which is no longer in the cluster config)5. What happened after the commands executed? The change from 1.11.6 to 1.12.0 seems to have staged changes that are sticking around even after changing the version back to 1.11.6. Note that I never passed the --yes flag. I just ran update to validate what was going to change and then reverted back after I saw how much it was going to do.
6. What did you expect to happen? After I changed from 1.12.0 back to 1.11.6 I expected the changes seen by
kops update cluster
to go away, because no changes should have been expected7. Please provide your cluster manifest. Execute
kops get --name my.example.com -o yaml
to display your cluster manifest. You may want to remove your cluster name and other sensitive information.8. Please run the commands with most verbose logging by adding the
-v 10
flag. Paste the logs into this report, or in a gist and provide the gist link here.9. Anything else do we need to know? No