Automatically Build A Multi-Nodes K3OS Cluster In 3 Minutes - from 0 This is mainly for development and test.
K3OS shipped k3s, which is full-featured k8s distribution that can even be used in production.
You should have followings installed:
- bash ( for windows, use git-bash or msys2)
- packer
- vagrant
- virtual box
> cd k3os-vbox-up
# setup env
> . env.sh
# generate vagrant boxes
> ./pack.sh
# boot-up vms in vbox
> ./up.shThat's it! By default you have 3 vm running with k3s: 1 master, 2 nodes:
> k3s-master [~]$ kubectl get pod --all-namespaces -o wide
NAMESPACE NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE
READINESS GATES
k3os-system system-upgrade-controller-66d5c7dd6b-c4dkz 1/1 Running 0 51m 10.42.0.4 k3s-master <none>
<none>
kube-system metrics-server-7566d596c8-prppp 1/1 Running 0 51m 10.42.0.2 k3s-master <none>
<none>
kube-system local-path-provisioner-6d59f47c7-xvh79 1/1 Running 0 51m 10.42.0.3 k3s-master <none>
<none>
kube-system coredns-7944c66d8d-hxb82 1/1 Running 0 51m 10.42.1.3 k3s-node-1 <none>
<none>
kube-system helm-install-traefik-wbrmc 0/1 Completed 0 51m 10.42.1.2 k3s-node-1 <none>
<none>
kube-system svclb-traefik-pfcbk 2/2 Running 0 50m 10.42.0.5 k3s-master <none>
<none>
kube-system svclb-traefik-6nsz2 2/2 Running 0 50m 10.42.1.4 k3s-node-1 <none>
<none>
kube-system svclb-traefik-wq2z8 2/2 Running 0 50m 10.42.2.3 k3s-node-2 <none>
<none>
kube-system traefik-758cd5fc85-4thxp 1/1 Running 0 50m 10.42.2.2 k3s-node-2 <none>
<none>Edit env.sh, change some environment vars.
For example, to increase nodes to 5 in env.sh:
export NODES_NUM=5 # k3s nodes number
Then run:
> . env.sh
> ./up.shJust a few minutes then you get:
> k3s-master [~]$ kubectl get node
NAME STATUS ROLES AGE VERSION
k3s-master Ready master 61m v1.18.9+k3s1
k3s-node-1 Ready <none> 61m v1.18.9+k3s1
k3s-node-2 Ready <none> 61m v1.18.9+k3s1
k3s-node-3 Ready <none> 111s v1.18.9+k3s1
k3s-node-5 Ready <none> 87s v1.18.9+k3s1
k3s-node-4 Ready <none> 12s v1.18.9+k3s1