aronahl/kargo repository overview

⁠Easy Kubernetes Cluster on Ubuntu 16.04 Using Kubespray (Formerly Kargo) v2.4.0.

⁠Requirements

1. Three Ubuntu 16.04 VMs or Bare Metal installs with no swap. I used Parallels VMs with 4G RAM and 4 Cores each.
   • The following packages should be installed:
     • python
     • python-minimal
     • openssh-server
   • All three should have the same user with password-free sudo access. In this guide, my username is ubuntu.
   • Each of the three should have unfiltered network access to the other two.
   • You should have a private ssh key which gives you access to the aforementioned user.
2. A local docker install with the aronahl/kargo image.

⁠Steps

1. Configure the inventory of your three machines. Assuming the IP addresses of your three machines are 10.0.0.1, 10.0.0.2, and 10.0.0.3:

   $ docker run --rm -it -v kube-data:/usr/local/share/kubespray aronahl/kargo python3 ./contrib/inventory_builder/inventory.py 10.0.0.1 10.0.0.2 10.0.0.3
   

   Copy

   This will create a kube-data docker volume populated with Kargo's ansible files and your new inventory config file. Be careful to avoid having an existing config in the volume, or you will end up running the buildout against hosts that you don't intent to change (or that may not exist).

2. Copy your ssh private key to the kube-data volume. In my case, the file is called kube_ecdsa.

   $ docker run --rm -i -v kube-data:/data busybox tee /data/id_ecdsa < kube_ecdsa
   $ docker run --rm -v kube-data:/data busybox chmod 600 /data/id_ecdsa
   

   Copy

3. Configure the cluster.

   $ docker run --rm -it -v kube-data:/usr/local/share/kubespray aronahl/kargo \
     ansible-playbook -i ./inventory.cfg cluster.yml -b -v --private-key=./id_ecdsa \
     -u ubuntu -e kube_version=v1.9.2 -e kube_api_pwd=mysup3rs3cr3tp455w0rd
   

   Copy

4. Download the config for your local kubectl

   $ ssh -i kube_ecdsa [email protected] sudo cat /etc/kubernetes/admin.conf > ~/.kube/config
   

   Copy

5. Use it.

   $ kubectl --namespace=kube-system get pods
   NAME                                    READY     STATUS    RESTARTS   AGE
   calico-node-894bv                       1/1       Running   0          20m
   calico-node-gxws8                       1/1       Running   0          20m
   calico-node-w6sxv                       1/1       Running   0          20m
   kube-apiserver-node1                    1/1       Running   0          19m
   kube-apiserver-node2                    1/1       Running   0          19m
   kube-controller-manager-node1           1/1       Running   0          20m
   kube-controller-manager-node2           1/1       Running   0          20m
   kube-dns-79d99cdcd5-ggh9c               3/3       Running   0          19m
   kube-dns-79d99cdcd5-w6tlb               3/3       Running   0          19m
   kube-proxy-node1                        1/1       Running   0          19m
   kube-proxy-node2                        1/1       Running   0          19m
   kube-proxy-node3                        1/1       Running   0          19m
   kube-scheduler-node1                    1/1       Running   0          20m
   kube-scheduler-node2                    1/1       Running   0          20m
   kubedns-autoscaler-5564b5585f-fjcnt     1/1       Running   0          19m
   kubernetes-dashboard-6bbb86ffc4-5bdr6   1/1       Running   0          19m
   nginx-proxy-node3                       1/1       Running   0          19m
   

   Copy

6. Check out the UI

   $ kubectl proxy
   

   Copy

   Visit http://localhost:8001/api/v1/namespaces/kube-system/services/https:kubernetes-dashboard:/proxy/#!/login⁠ and select skip.