I'm getting the error [failed to create cluster: failed to join node with kubeadm: command "docker exec --privileged kind-cassandra-worker3 kubeadm join --config /kind/kubeadm.conf --ignore-preflight-errors=all --v=6" failed with error: exit status 1].
It's not always the same worker number. I'm running on centOS in a Hyper-V VM on Windows 10. All is fine when using the cloud instance provided.
Cannot create kind cluster in Step 6 of Week 5 exercise
this is most likely due to lack of resources on the Hyper-V VM. You will need a good amount of RAM (around 16GB) to run the week5 exercises especially, since the kind cluster uses 6 kubernetes nodes.
You can experiment with the kind config by reducing the number of worker nodes.
For example, the following config yaml would only set up one worker node (you will need a master node (i.e. the control-plane) in any setup):
kind: Cluster apiVersion: kind.sigs.k8s.io/v1alpha3 networking: apiServerPort: 45451 nodes: - role: control-plane - role: worker
For the full exercise however, with 3-node cassandra cluster running in kind and prometheus and grafana, you will need the RAM resources to support the exercise.
I hope this explains the issue you are seeing.
If we understand well your settings:
Windows 10 => hosting VirtualBox => running a CentOS VM => running Kind
When you want to create the kind cluster
kind create cluster --name kind-cassandra --config ./setup-your-cluster/01-kind-config.yaml
You got error
failed to create cluster: failed to join node with kubeadm: command "docker exec --privileged kind-cassandra-worker3 kubeadm join --config /kind/kubeadm.conf --ignore-preflight-errors=all --v=6" failed with error: exit status 1
Exit status 1 is not enough to propose any diagnostic. Can extract here some logs ?
This look like a limitation of the KIND tool, or ressource issue. KIND is a simple tool to work on local environnement.
6 People are following this question.
