k3s: Join a new worker node to an existing cluster

3 min read

On a previous post we saw how ridiculously easy is to bootstrap a k3s cluster on a Raspberry Pi but what do we need to do to join new worker nodes to the cluster?

The installation of a new node to make it join to an existing cluster is as follows:

# curl -sfL https://get.k3s.io | sh -

To install a workerd node we will have to set some environment variables for telling the installer for join an existing cluster. We will need:

  • Token
  • Master's IP

The token we can get it from the master node on the following location /var/lib/rancher/k3s/server/node-token:

# cat /var/lib/rancher/k3s/server/node-token
K173832c8dd5a175bf2123d840ad491850199cbd19309ab3b37827047cdd6319b04::server:faddf0a734d338cd66d4ab19fb4bed73

Having the master node IP we just need to specify that it is https protocol on port 6443 for having the URL. To tell the installer we will have to set the following environment variables K3S_URL and K3S_TOKEN as follows:

# curl -sfL https://get.k3s.io | K3S_URL=https://10.12.1.40:6443 K3S_TOKEN="K173832c8dd5a175bf2123d840ad491850199cbd19309ab3b37827047cdd6319b04::server:faddf0a734d338cd66d4ab19fb4bed73" sh -

The last step is to enable and start the k3s-agent (instead of the k3s-server that we would enable on the master):

# systemctl enable --now k3s-agent

Running kubectl get nodes you'll be able to see that the new node has joined the cluster:

# kubectl get nodes
NAME                    STATUS   ROLES                  AGE   VERSION
nauvoo.pet2cattle.com   Ready    control-plane,master   19d   v1.20.4+k3s1
tycho.pet2cattle.com    Ready    <none>                 26s   v1.20.4+k3s1

Out of curiosity, we can check how the k3s-agent unit file uses an environment file (/etc/systemd/system/k3s-agent.service.env) to store the variables variables K3S_URL and K3S_TOKEN:

# cat /etc/systemd/system/k3s-agent.service
[Unit]
Description=Lightweight Kubernetes
Documentation=https://k3s.io
Wants=network-online.target
After=network-online.target

[Install]
WantedBy=multi-user.target

[Service]
Type=exec
EnvironmentFile=/etc/systemd/system/k3s-agent.service.env
KillMode=process
Delegate=yes
# Having non-zero Limit*s causes performance problems due to accounting overhead
# in the kernel. We recommend using cgroups to do container-local accounting.
LimitNOFILE=1048576
LimitNPROC=infinity
LimitCORE=infinity
TasksMax=infinity
TimeoutStartSec=0
Restart=always
RestartSec=5s
ExecStartPre=-/sbin/modprobe br_netfilter
ExecStartPre=-/sbin/modprobe overlay
ExecStart=/usr/local/bin/k3s \
    agent \

The contents of the k3s-agent.service.env will look like:

# cat /etc/systemd/system/k3s-agent.service.env
K3S_TOKEN=K173832c8dd5a175bf2123d840ad491850199cbd19309ab3b37827047cdd6319b04::server:faddf0a734d338cd66d4ab19fb4bed73
K3S_URL=https://10.12.1.40:6443

To manually start the k3s agent we can also use the options --server and --token inetad of the environment variables:

# k3s agent --server https://10.12.1.40:6443 --token "K108a732b7cfb59036f2362848d61823733359bbdf152192f7ebc6ad4b3078fd659::server:68e6f6ceae26d8c276e4a1a714f0b1ed"

Posted on 05/04/2021

Categories