Adopt existing clusters on vSphere

[andyzheung]

As I say in the following:
#7572

I think this is very common, if I can enhance my exist k8s cluster to get autoscaler capibility.

for more details , I want to enhance my exist kubernetes use capi and capv.

[24sama]

Also need, looking forward to this feature.

[fabriziopandini]

@andyzheung could you kindly expand the issue comment providing a better explanation of what is your ask, so people can quickly understand the problem?

If the question is to adopt existing clusters into Cluster API, if fear the answer is the same as you got in the discussion; this is not supported, and given the complexity of the topic, IMO it requires a group of folks committed to writing a proposal and implementing it

[andyzheung]

@andyzheung could you kindly expand the issue comment providing a better explanation of what is your ask, so people can quickly understand the problem?

If the question is to adopt existing clusters into Cluster API, if fear the answer is the same as you got in the discussion; this is not supported, and given the complexity of the topic, IMO it requires a group of folks committed to writing a proposal and implementing it
=======================

In fact， I have a kubernetes cluster that created by kubedam on vmware vsphere already, but not by cluster-api.
Now I want to enhance this cluster to get the cluster autoscaler capibility.
So I find that cluster-api+cluster api vsphere can solve my problem ,so I try it...

what i want :

what I use the cr and controller:
cluster-api.zip
refer to:
https://github.com/kubernetes-sigs/cluster-api-provider-vsphere/blob/v0.8.0/templates/cluster-template.yaml
https://github.com/kubernetes-sigs/cluster-api-provider-vsphere/releases?page=2
https://github.com/kubernetes-sigs/cluster-api/releases?page=6

I also find that I deploy the cr to the system the capi controller crashed:

my cluster version:
1、kubernetes: 1.22.0
2、cert-manager: 1.4.0
3、cluster-api-component: 0.3.16
4、infrastructure-componemntsv0.7.12

my problem is that:
1、if I have the way use cluster api to enhance my exist k8s cluster to get CA?
2、why I deploy the cr and controller will crash? Is version not match? which version I need to try.
really need your help, thank you so much.
@fabriziopandini

[fabriziopandini]

Thanks for adding more details (please move them in the comment on top of the issue)
This confirms my initial guess, and thus the answer as of today is

this is not supported, and given the complexity of the topic, IMO it requires a group of folks committed to writing a proposal and implementing it

Based on the fact the community never agreed/committed on implementing something similar, IMO it will be easier and faster to move workloads to a new cluster created by Cluster API, but let's see if we can reach critical mass this time.

/triage accepted
/title Adopt existing clusters on vSphere
/kind proposal

[andyzheung]

thx for reply. I don't understand IMO means, have any details for it?
the second question is that why I deploy the cr and controller will crash? Is version mismatch?
cluster-api.zip

[chiukapoor]

/retitle Adopt existing clusters on vSphere

[k8s-ci-robot]

@chiukapoor: Re-titling can only be requested by trusted users, like repository collaborators.

In response to this:

/retitle Adopt existing clusters on vSphere

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[fabriziopandini]

IMO stands for "In my opinion", sorry for using abbreviations.

the second question is that why I deploy the cr and controller will crash? Is version mismatch?

given my understanding of the issue, I don't think it is worth investing time in avoiding the error above now because you are using the controller for a scenario that is not supported - adopting an existing cluster not created with Cluster API -. Fixing this error will eventually be part of the work that will start after a proposal for addressing this scenario is merged.

[andyzheung]

given my understanding of the issue, I don't think it is worth investing time in avoiding the error above now because you are using the controller for a scenario that is not supported - adopting an existing cluster not created with Cluster API -. Fixing this error will eventually be part of the work that will start after a proposal for addressing this scenario is merged.

---

oh, thx.
But the error I mentioned is not that I used the exist cluster, I just follow this:
https://github.com/kubernetes-sigs/cluster-api-provider-vsphere/blob/v0.8.0/templates/cluster-template.yaml
and my version is:
my cluster version:
1、kubernetes: 1.22.0
2、cert-manager: 1.4.0
3、cluster-api-component: 0.3.16
4、infrastructure-componemntsv0.7.12
@fabriziopandini

[fabriziopandini]

v0.3.16 is out of support, please use a more recent Cluster API version, also, if a problem is different from the scenario described in this thread, move the discussion to another issue

[andyzheung]

thx, i will talk to other issue.@fabriziopandini

[k8s-triage-robot]

This issue has not been updated in over 1 year, and should be re-triaged.

You can:

• Confirm that this issue is still relevant with /triage accepted (org members only)
• Close this issue with /close

For more details on the triage process, see https://www.kubernetes.dev/docs/guide/issue-triage/

/remove-triage accepted

[sbueringer]

No activity for > 1 year

/close

[k8s-ci-robot]

@sbueringer: Closing this issue.

In response to this:

No activity for > 1 year

/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.