LoadBalancer service stuck at pending even though the LB has been created and has public IP about hcloud-cloud-controller-manager HOT 7 CLOSED

You need to recreate the whole cluster with --cloud-provider=external. See Step 1 in https://github.com/hetznercloud/hcloud-cloud-controller-manager

from hcloud-cloud-controller-manager.

The provider ID will be set from the cloud controller :)

from hcloud-cloud-controller-manager.

The problem was that I updated the kubelet config after creating the cluster. I created a new test cluster with the --cloud-provider=external setting from the beginning and everything worked fine.

For those who have a similar issue with an existing cluster with Rancher, this is what I did:

1. updated the kubelet config
2. installed the cloud controller
3. created a new node pool, making sure the controller adds labels/annotations as expected (meaning that it's working)
4. drained and removed the old node pool
5. created service of type LoadBalancer.

The load balancer was correctly provisioned and I was able to use the ingress resources with Nginx just fine.

Thanks for the new release and for the load balancers!

from hcloud-cloud-controller-manager.

I am using RKE on hetzner and now i am stuck with the same issue , load balancer is creating but cnt get the IP address .

Events:
Type Reason Age From Message

Normal EnsuringLoadBalancer 25s (x5 over 102s) service-controller Ensuring load balancer
Warning SyncLoadBalancerFailed 25s (x5 over 101s) service-controller Error syncing load balancer: failed to ensure load balancer: hcloud/loadBalancers.EnsureLoadBalancer: hcops/LoadBalancerOps.ReconcileHCLBTargets: hcops/providerIDToServerID: missing prefix hcloud://:

from hcloud-cloud-controller-manager.

yeah i did but i also have a question now below :
spec:
externalID: master
podCIDR: 10.244.0.0/24
providerID: hcloud://123456 # <-- Server ID

it is asking for provideID and i am using multi master so do i can entry any master ip or am i still missing something ?

from hcloud-cloud-controller-manager.

Hi @vitobotta ,
I updated kubelet config of existed cluster in rancher like this:

kubelet:
      extra_env:
        - KUBELET_EXTRA_ARGS=--cloud-provider=external

Cluster and worker nodes updated, but error doesn't disappear. What I missed?

from hcloud-cloud-controller-manager.

@vitobotta , I notice that --cloud-provider=external has not been set in kubelet command:
/opt/rke-tools/entrypoint.sh kubelet --client-ca-file=/etc/kubernetes/ssl/kube-ca.pem --cloud-provider= --resolv-conf=/etc/resolv.conf
How do I set kubelet config properly in rancher ui?

from hcloud-cloud-controller-manager.