https://rancher.com/ logo
Title
a

acceptable-leather-15942

03/01/2023, 9:54 PM
Hi All 👋 , Using the Ranger UI I’m trying to create a cluster on Digitalocean using RKE2. I want to install Digitaloceans own cloud provider to enable loadbalancer creation and such, but when I switch the cloud provider setting to
external
the cluster fails to install. Do I need to add a
Cloud Provider Config
in the field that comes up?
c

creamy-pencil-82913

03/01/2023, 10:38 PM
you need to put a manifest to deploy the digital ocean cloud provider in the additional manifests section. the “cloud provider config” section is only for if you’re using a built-in cloud provider.
if you don’t want to deploy it as an additional manifest, you’ll need to log in to one of the server nodes and deploy the cloud provider manually
a

acceptable-leather-15942

03/01/2023, 10:41 PM
Thanks for your reply! I almost gave up 😬 .. I just found a post here on slack of someone else posting a similar solution, but it doesn’t seem to have an effect. I would like to try and apply it manually to see if that works, but I don’t know how to apply that. There is no
kubectl
c

creamy-pencil-82913

03/01/2023, 10:41 PM
there is, you just need to put it in your path
💯 1
export
PATH=$PATH:/var/lib/rancher/rke2/bin/ KUBECONFIG=/etc/rancher/rke2/rke2.yaml
If you have a manifest that you can apply with kubectl, its probably easier to just paste that in to the rancher UI though, and let the installer handle it for you
a

acceptable-leather-15942

03/01/2023, 10:43 PM
That’s what I did. No success 😕
c

creamy-pencil-82913

03/01/2023, 10:44 PM
where in the UI did you put it
a

acceptable-leather-15942

03/01/2023, 10:45 PM
In the
Add-on Config
section
c

creamy-pencil-82913

03/01/2023, 10:45 PM
Down at the bottom, under Additional Manifest?
a

acceptable-leather-15942

03/01/2023, 10:45 PM
Yes
c

creamy-pencil-82913

03/01/2023, 10:46 PM
That should be the correct place for it.
Did you put it there when you originally created the cluster, or did you add it after the fact?
a

acceptable-leather-15942

03/01/2023, 10:47 PM
Before. Fresh cluster
c

creamy-pencil-82913

03/01/2023, 10:47 PM
Hmm. You might check the logs on the server nodes to see if there were any errors deploying the manifest
or if it did get deployed, see if there are errors in the pod logs
a

acceptable-leather-15942

03/01/2023, 10:47 PM
I just tried
kubectl get nodes
on one of the nodes but getting
The connection to the server localhost:8080 was refused - did you specify the right host or port?
c

creamy-pencil-82913

03/01/2023, 10:48 PM
sounds like you didn’t export the kubeconfig path as I suggested up above
a

acceptable-leather-15942

03/01/2023, 10:48 PM
you’re right 😅
image.png
Ah it seems to have applied the config
Thanks for the help @creamy-pencil-82913. This will help me on my way. #LifeSaver
c

creamy-pencil-82913

03/01/2023, 10:53 PM
CreateContainerConfigError looks wrong though
suggests something in your manifest is invalid
a

acceptable-leather-15942

03/01/2023, 10:53 PM
Yeah I checked the logs. I’m missing a secret 🤓
c

creamy-pencil-82913

03/01/2023, 10:54 PM
👍