search

hetznercloud / hcloud-cloud-controller-manager

Kubernetes cloud-controller-manager for Hetzner Cloud
Apache License 2.0
733 stars 118 forks source link

Failure in EnsureLoadBalancer: missing prefix hcloud:// #70

Closed alexeyzimarev closed 4 years ago

alexeyzimarev commented 4 years ago

I have a running cluster and I have previously provisioned a load balancer using the instructions provided for the ingress controller. It all works fine.

However, now I am trying to do a similar thing for another service and I keep getting this:

error processing service gitlabsync/production-auto-deploy-aspnetcore (will retry): failed to ensure load balancer: hcloud/loadBalancers.EnsureLoadBalancer: hcops/LoadBalancerOps.ReconcileHCLBTargets: hcops/providerIDToServerID: missing prefix hcloud://:

I experience this issue with any new service of type LoadBalancer with hcloud annotations.

It seems some requirements have changed but it's not documented what do I suppose to do. It looks the same as #50 but I don't really want to re-provision the cluster since it works just fine and I have quite a lot of things running there.

alexeyzimarev commented 4 years ago

I checked #52 as well and I am sure I followed the instructions from another tutorial and I have this:

root@master-1:/etc/systemd/system/kubelet.service.d# cat 20-hetzner-cloud.conf
[Service]
Environment="KUBELET_EXTRA_ARGS=--cloud-provider=external"
LKaemmerling commented 4 years ago

Please keep in mind that you need to create the cluster with this Argument, just a restart is not enough. Actually the only way that this message can be triggered is a missing cloud provider flag on cluster creation.

alexeyzimarev commented 4 years ago

As I mentioned before, I created the cluster according to the guidelines and I cat the file just to be sure it's there.

In addition, I also mentioned that I managed to provision one load balancer as described in the documentation about four weeks ago but suddenly it doesn't work anymore.

LKaemmerling commented 4 years ago

Maybe there was a k8s update that removed/unloaded the setting? Actually there was no release on our side (or at least no requirement change) .

LKaemmerling commented 4 years ago

Duplicate of #80

morphace commented 3 years ago

I've got the same issue as @alexeyzimarev - although I followed the steps in the readme.md (except that I use cilium instead flannel) I get the same error message.

I use Ubuntu 18.04 and k8s V 1.18.5. Is there anything I can check, to verify that my config is OK?

morphace commented 3 years ago

Just got it to work 👍

1.) I moved to Ubuntu 20.04 and k8s 1.20 2.) I added

[Service] Environment="KUBELET_EXTRA_ARGS=--cloud-provider=external"

to the worker nodes as well (before I did it only on the master node).

I don't know which one made it work, but it works.

MuhanedYahya commented 4 months ago

https://github.com/kubernetes/kubeadm/issues/1021#issuecomment-408090344 for me this solution worked