Skip to content

Latest commit

 

History

History
36 lines (24 loc) · 1.52 KB

k3s.md

File metadata and controls

36 lines (24 loc) · 1.52 KB

Notes on k3s

There is a "self-contained" (single-binary, single-node) kubernetes distribution called k3s.

Supercluster can run on k3s, and this is a relatively easy way to test it or run it locally.

Follow these steps:

  • Install k3s: curl -sfL https://get.k3s.io | INSTALL_K3S_VERSION=v1.23.17+k3s1 sh -s - --write-kubeconfig-mode 644 --docker --kubelet-arg=cgroup-driver=systemd --no-deploy traefik

  • Deploy nginx-ingress:

kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/controller-v1.5.1/deploy/static/provider/cloud/deploy.yaml \
&& kubectl wait --namespace ingress-nginx --for=condition=ready pod --selector=app.kubernetes.io/component=controller --timeout=120s
  • Raise the sysctl limits on your ARP cache. This is necessary to run larger supercluster missions. Put something like this in your /etc/sysctl.conf and then run sysctl -p (as root):
net.ipv4.neigh.default.gc_thresh1 = 80000
net.ipv4.neigh.default.gc_thresh2 = 90000
net.ipv4.neigh.default.gc_thresh3 = 100000

net.ipv6.neigh.default.gc_thresh1 = 80000
net.ipv6.neigh.default.gc_thresh2 = 90000
net.ipv6.neigh.default.gc_thresh3 = 100000

  • Set your kubeconfig to the k3s one: export KUBECONFIG=/etc/rancher/k3s/k3s.yaml

  • Build supercluster normally (see getting-started.md)

  • Run supercluster with these additional arguments: --kubeconfig $KUBECONFIG --namespace default --ingress-class nginx --ingress-internal-domain local --ingress-external-host localhost --uneven-sched