Where is the latest documentation for this?

[SaltedEggIndomee]

I've been trying for weeks to set up a vanilla Kubernetes cluster on AWS and still unable to pinpoint an official and latest tutorial to use this plugin.

Every guide on the internet uses --cloud-provider=aws flag, which we know will be deprecated very soon.

Also, on the half-baked instructions in this GitHub README, it says :

The flag --cloud-provider=external needs to be passed to kubelet, kube-apiserver, and kube-controller-manager.

But when you look at Kubernetes official documentation at https://kubernetes.io/docs/tasks/administer-cluster/running-cloud-controller/ :

kube-apiserver and kube-controller-manager MUST NOT specify the --cloud-provider flag.

So which is it??

[mikejoh]

Not that it's related to cloud-provider-aws documentation but to your observation regarding the official docs on how and when the --cloud-provider flag should be not used/used: kubernetes/website#20137

[andrewsykim]

/assign

[zimmertr]

Hi, I would also be interested in this. I'm trying to put together a Kubeadm-based cluster on AWS at the moment and struggling to get the Cloud Provider integration to advance beyond my worker nodes being tainted with:

node.cloudprovider.kubernetes.io/uninitialized

Their hostnames are the EC2 DNS name:

$> hostname -f
ip-10-0-10-91.us-gov-west-1.compute.internal

My Kubeadm config files set the cloud provider to external in four places

# Present in both KubeletConfiguration and InitConfiguration
nodeRegistration:
  kubeletExtraArgs:
    cloud-provider: external

# Present in ClusterConfiguration
apiServer:
  extraArgs:
    cloud-provider: external

# Present in ClusterConfiguration
controllerManager:
  extraArgs:
    cloud-provider: external

The EC2 instances as well as my route tables, subnets, etc are tagged with:

"kubernetes.io/cluster/${var.K8S_CLUSTER_NAME}" = "kubernetes.io/cluster/${var.K8S_CLUSTER_NAME}"

And my EC2 instances were launched with an instance profile with the IAM policies outlined in the README:

$> aws ec2 describe-instances --instance-ids INSTANCE.ID | jq '.Reservations[].Instances[].IamInstanceProfile[]'
"arn:aws-us-gov:iam::ACCOUNT.ID:instance-profile/PROFILE-NAME"

[zimmertr]

Got this squared away thanks to the help of @andrewsykim.

I had to make a few changes. First of all, if you look in the manifests directory on the root of the repo there is a DaemonSet and an RBAC schema. These both need to be deployed and operational on your cluster for your nodes to be initialized.

You'll notice that the DaemonSet doesn't have an image reference. This is because the AWS Cloud Controller Manager isn't officially publicly hosted yet. You'll need to build it yourself or use one on DockerHub. Here's the one I made from the newest commit from three days ago.

You might also notice that there is an --cloud-provider=aws argument passed to the container despite being the external provisioner. This is intentional.

Lastly, you'll need to ensure that all of your EC2 instances have this tag: KubernetesCluster = var.K8S_CLUSTER_NAME where var.K8S_CLUSTER_NAME is the name of your cluster.

[andrewsykim]

Looking for feedback on what the documentation site should look like #102

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[fejta-bot]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten

[fejta-bot]

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close

[k8s-ci-robot]

@fejta-bot: Closing this issue.

In response to this:

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/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.

[andrewsykim]

/reopen
/remove-lifecycle rotten

[k8s-ci-robot]

@andrewsykim: Reopened this issue.

In response to this:

/reopen
/remove-lifecycle rotten

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.

[andrewsykim]

We have a docs site here https://cloud-provider-aws.sigs.k8s.io/ but we need help with content

[andrewsykim]

/help wanted

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[nckturner]

/remove-lifecycle stale

[nckturner]

We are in the process of updating the docs and adding a new docs site. I'll keep this issue updated.

[nckturner]

Follow progress here:
https://kubernetes.github.io/cloud-provider-aws/

Also the README has been updated: https://github.com/kubernetes/cloud-provider-aws/blob/master/README.md but ideally that information should be included in the home page of the docs site too.

[nckturner]

/assign

[nckturner]

Documentation will be available here going forward: https://cloud-provider-aws.sigs.k8s.io/

We will continue working on the docs, but I think we've made enough initial progress that I can close this issue. Please feel free to open specific issues if you feel certain areas are lacking documentation.