I've been following this post to create user access to my kubernetes cluster (running on Amazon EKS). I did create key, csr, approved the request and downloaded the certificate for the user. Then I did create a kubeconfig file with the key and crt. When I run kubectl with this kubeconfig, I'm recognized as system:anonymous
.
$ kubectl --kubeconfig test-user-2.kube.yaml get pods
Error from server (Forbidden): pods is forbidden: User "system:anonymous" cannot list pods in the namespace "default"
I expected the user to be recognized but get denied access.
$ kubectl --kubeconfig test-user-2.kube.yaml version
Client Version: version.Info{Major:"1", Minor:"11", GitVersion:"v1.11.1", GitCommit:"b1b29978270dc22fecc592ac55d903350454310a", GitTreeState:"clean", BuildDate:"2018-07-18T11:37:06Z", GoVersion:"go1.10.3", Compiler:"gc", Platform:"darwin/amd64"}
Server Version: version.Info{Major:"1", Minor:"10", GitVersion:"v1.10.3", GitCommit:"2bba0127d85d5a46ab4b778548be28623b32d0b0", GitTreeState:"clean", BuildDate:"2018-05-28T20:13:43Z", GoVersion:"go1.9.3", Compiler:"gc", Platform:"linux/amd64"}
$ kubectl --kubeconfig test-user-2.kube.yaml config view
apiVersion: v1
clusters:
- cluster:
insecure-skip-tls-verify: true
server: REDACTED
name: kubernetes
contexts:
- context:
cluster: kubernetes
user: test-user-2
name: kubernetes
current-context: kubernetes
kind: Config
preferences: {}
users:
- name: test-user-2
user:
client-certificate-data: REDACTED
client-key-data: REDACTED
# running with my other account (which uses heptio-authenticator-aws)
$ kubectl describe certificatesigningrequest.certificates.k8s.io/user-request-test-user-2
Name: user-request-test-user-2
Labels: <none>
Annotations: <none>
CreationTimestamp: Wed, 01 Aug 2018 15:20:15 +0200
Requesting User:
Status: Approved,Issued
Subject:
Common Name: test-user-2
Serial Number:
Events: <none>
I did create a ClusterRoleBinding with admin
(also tried cluster-admin
) roles for this user but that should not matter for this step. I'm not sure how I can further debug 1) if the user is created or not or 2) if I missed some configuration.
Any help is appreciated!
This seems to be a limitation of EKS. Even though the CSR is approved, user can not authenticate. I used the same procedure on another kubernetes cluster and it worked fine.
I got this back from AWS support today.
Thanks for your patience. I have just heard back from the EKS team. They have confirmed that the aws-iam-authenticator has to be used with EKS and, because of that, it is not possible to authenticate using certificates.
I haven't heard whether this is expected to be supported in the future, but it is definitely broken at the moment.
As mentioned in this article:
When you create an Amazon EKS cluster, the IAM entity user or role (for example, for federated users) that creates the cluster is automatically granted system:master permissions in the cluster's RBAC configuration. To grant additional AWS users or roles the ability to interact with your cluster, you must edit the aws-auth ConfigMap within Kubernetes.
Check if you have aws-auth ConfigMap applied to your cluster:
kubectl describe configmap -n kube-system aws-auth
If ConfigMap is present, skip this step and proceed to step 3. If ConfigMap is not applied yet, you should do the following:
Download the stock ConfigMap:
curl -O https://amazon-eks.s3-us-west-2.amazonaws.com/1.10.3/2018-07-26/aws-auth-cm.yaml
Adjust it using your NodeInstanceRole ARN in the rolearn:
. To get NodeInstanceRole value check out this manual and you will find it at steps 3.8 - 3.10.
data:
mapRoles: |
- rolearn: <ARN of instance role (not instance profile)>
Apply this config map to the cluster:
kubectl apply -f aws-auth-cm.yaml
Wait for cluster nodes becoming Ready:
kubectl get nodes --watch
Edit aws-auth
ConfigMap and add users to it according to the example below:
kubectl edit -n kube-system configmap/aws-auth
# Please edit the object below. Lines beginning with a '#' will be ignored,
# and an empty file will abort the edit. If an error occurs while saving this file will be
# reopened with the relevant failures.
#
apiVersion: v1
data:
mapRoles: |
- rolearn: arn:aws:iam::555555555555:role/devel-worker-nodes-NodeInstanceRole-74RF4UBDUKL6
username: system:node:{{EC2PrivateDNSName}}
groups:
- system:bootstrappers
- system:nodes
mapUsers: |
- userarn: arn:aws:iam::555555555555:user/admin
username: admin
groups:
- system:masters
- userarn: arn:aws:iam::111122223333:user/ops-user
username: ops-user
groups:
- system:masters
Save and exit the editor.