Chapter 4 - Building a K8s cluster with Ansible

[SteveJM]

When running the cluster build playbook (P. 79), I get the following error:

...
TASK [geerlingguy.kubernetes : Ensure kubelet is started and enabled at boot.] ***************************

fatal: [kube1]: FAILED! => {"changed": false, "msg": "Unable to enable service kubelet: Failed to enable unit: Unit file /etc/systemd/system/kubelet.service is masked.\n"}

fatal: [kube2]: FAILED! => {"changed": false, "msg": "Unable to enable service kubelet: Failed to enable unit: Unit file /etc/systemd/system/kubelet.service is masked.\n"}

fatal: [kube3]: FAILED! => {"changed": false, "msg": "Unable to enable service kubelet: Failed to enable unit: Unit file /etc/systemd/system/kubelet.service is masked.\n"}

This is using the base image pas per the Vagrant file on P. 71) .

As the Ansible role is defined in Ansible Galaxy this looks like something out of my control.

[SteveJM]

(Untested) Steps to resolve (these probably apply to ansible-role-kubernetes:
In each host kube1-2:

sudo rm /etc/systemd/system/kubelet.service
sudo systemctl daemon-reload

[geerlingguy]

Just for my own reference, this is in chapter 4, "Building a Kubernetes cluster with Ansible" — I'll test out the playbook in this repo on my laptop to see if I can reproduce the error.

[geerlingguy]

I just re-ran the build fresh on my laptop and didn't seem to run into an error. I'm guessing it was a transient issue with maybe one of the apt caches or something in the image. I would recommend doing a full vagrant destroy then bringing it up again, and make sure you're on the latest versions of Vagrant, all the roles (ansible-galaxy install --force -r requirements.yml), etc.

The playbook completed successfully:

...
TASK [geerlingguy.kubernetes : Join node to Kubernetes master] *****************
changed: [kube2]
changed: [kube3]

RUNNING HANDLER [geerlingguy.kubernetes : restart kubelet] *********************
changed: [kube3]
changed: [kube1]
changed: [kube2]

PLAY RECAP *********************************************************************
kube1                      : ok=56   changed=29   unreachable=0    failed=0    skipped=18   rescued=0    ignored=0   
kube2                      : ok=47   changed=24   unreachable=0    failed=0    skipped=17   rescued=0    ignored=0   
kube3                      : ok=47   changed=24   unreachable=0    failed=0    skipped=17   rescued=0    ignored=0

And I could connect to the cluster:

$ vagrant ssh kube1
vagrant@kube1:~$ sudo su
root@kube1:/home/vagrant# kubectl get pods --all-namespaces
NAMESPACE     NAME                            READY   STATUS    RESTARTS   AGE
kube-system   coredns-6955765f44-72pmw        1/1     Running   0          2m12s
kube-system   coredns-6955765f44-gmrp8        1/1     Running   0          2m12s
kube-system   etcd-kube1                      1/1     Running   0          2m24s
kube-system   kube-apiserver-kube1            1/1     Running   0          2m24s
kube-system   kube-controller-manager-kube1   1/1     Running   0          2m24s
kube-system   kube-flannel-ds-amd64-4dk9t     1/1     Running   0          2m9s
kube-system   kube-flannel-ds-amd64-htq4g     1/1     Running   0          2m13s
kube-system   kube-flannel-ds-amd64-l77g2     1/1     Running   0          2m9s
kube-system   kube-proxy-9nnv7                1/1     Running   0          2m9s
kube-system   kube-proxy-hsc5d                1/1     Running   0          2m9s
kube-system   kube-proxy-jsfx5                1/1     Running   0          2m13s
kube-system   kube-scheduler-kube1            1/1     Running   0          2m24s