This message was deleted.

I'm sometimes seeing similar issues (but I have no solution yet). Are you also using VMware vSphere as the infrastructure provider for the clusters?

@salmon-morning-84088 In one cluster the nodes are physical and in the other cluster they're on azure. I've seen issues about cluster state stuck in Unavailable of Configuring, but nothing that exactly matches an intermittent "upgrading" state.

Are you using RKE2? And if so: using Rancher Cluster Templates?

Hi, did you create an issue on GitHub for that ?

I have not.

> Are you using RKE2? And if so: using Rancher Cluster Templates? Yes to RKE2. I'm not sure if we're using Rancher Cluster Templates, I'm somewhat new to the env.

If you do not know, you probably don't. I assume you don't have Rancher Prime and can request support directly from them?

We're working on that. We're just moving from POC stages to some PRD, @salmon-morning-84088 What are you doing for support? do you have Rancher Prime? I wonder if there other support options. Thanks for the replies. I'll post here if I find a solution.

I think they're are two tiers of Rancher Prime: standard and priority. You can read about them here: https://www.suse.com/support/rancher-prime/

> Hi, did you create an issue on GitHub for that ? @clever-mouse-42560 I did end up entering a bug

Sounds every similar to a bug I reported a while back: https://github.com/rancher/rancher/issues/44530

what is the source of truth for these clusters? are they imported? (that part is missing from the video)

@broad-ambulance-17822 Yes, they're all imported

the docs mention Imported (aka Registered) clusters should be upgraded with Rancher — https://ranchermanager.docs.rancher.com/how-to-guides/new-user-guides/kubernetes-clusters-in-rancher-setup/register-existing-clusters#additional-features-for-registered-rke2-and-k3s-clusters … if you’d like you’re welcome to scale down the

system-upgrade-controller

to zero but then Rancher has no responsibility for upgrades any longer, they are happening outside of Rancher at that point.

@broad-ambulance-17822. Thanks. I understand imported clusters should be upgraded by Rancher and believe me I agree with that approach :-) The company I recently joined no longer wants to use `system-upgrade-controller`(SUC), so the issue is how do we clean up Rancher after SUC was installed on imported cluster and failed? We've tried scaling down SUC, let SUC finish the upgrade, removing all SUC objects, but Rancher UI still shows clusters in

upgrading

or blinking from

Active

to

Upgrading

I noticed - and somebody else also reported - that Rancher creates upgrade Plans automatically Somebody else is unable to completely remove SUC as well. Our goal is to ensure Rancher and its clusters are in a healthy state so we have the confidence to transition applications from Non-prod to Prod environments.