This tutorial will walk you through the creation
of kind
cluster with CRI-O Container Runtime by creating a custom node image
.
It assumes you've already installed golang>=1.22.2
,kind>=v0.22.0
,docker>=26.1.0
.
We need kind
sources to build the
base image
:
$ git clone [email protected]:kubernetes-sigs/kind.git
$ cd kind/images/base
$ make quick
./../../hack/build/init-buildx.sh
docker buildx build --load --progress=auto -t gcr.io/k8s-staging-kind/base:v20240508-19df3db3 --pull --build-arg GO_VERSION=1.21.6 .
### ... some output here
The image gcr.io/k8s-staging-kind/base:v20240508-19df3db3
is our
base image
.
We'll use it for
node image
building.
Before start building node image
, we need kubernetes sources at $GOPATH
.
$ mkdir -p "$GOPATH"/src/k8s.io/kubernetes
$ K8S_VERSION=v1.30.0
$ git clone --depth 1 --branch ${K8S_VERSION} https://github.com/kubernetes/kubernetes.git "$GOPATH"/src/k8s.io/kubernetes
# some output
Now let's build the node image
:
$ kind build node-image --base-image gcr.io/k8s-staging-kind/base:v20240508-19df3db3
Starting to build Kubernetes
+++ [0508 15:41:04] Verifying Prerequisites....
+++ [0508 15:41:04] Building Docker image kube-build:build-14d7110ae1-5-v1.30.0-go1.22.2-bullseye.0
+++ [0508 15:42:49] Creating data container kube-build-data-14d7110ae1-5-v1.30.0-go1.22.2-bullseye.0
+++ [0508 15:42:50] Syncing sources to container
+++ [0508 15:42:54] Running build command...
+++ [0508 15:42:46] Building go targets for linux/arm64
k8s.io/kubernetes/cmd/kube-apiserver (static)
k8s.io/kubernetes/cmd/kube-controller-manager (static)
k8s.io/kubernetes/cmd/kube-proxy (static)
k8s.io/kubernetes/cmd/kube-scheduler (static)
k8s.io/kubernetes/cmd/kubeadm (static)
k8s.io/kubernetes/cmd/kubectl (static)
k8s.io/kubernetes/cmd/kubelet (non-static)
+++ [0508 15:45:16] Syncing out of container
+++ [0508 15:45:22] Building images: linux-arm64
+++ [0508 15:45:22] Starting docker build for image: kube-apiserver-arm64
+++ [0508 15:45:22] Starting docker build for image: kube-controller-manager-arm64
+++ [0508 15:45:22] Starting docker build for image: kube-scheduler-arm64
+++ [0508 15:45:22] Starting docker build for image: kube-proxy-arm64
+++ [0508 15:45:22] Starting docker build for image: kubectl-arm64
+++ [0508 15:45:32] Deleting docker image registry.k8s.io/kubectl-arm64:v1.30.0
+++ [0508 15:45:32] Deleting docker image registry.k8s.io/kube-scheduler-arm64:v1.30.0
+++ [0508 15:45:35] Deleting docker image registry.k8s.io/kube-proxy-arm64:v1.30.0
+++ [0508 15:45:35] Deleting docker image registry.k8s.io/kube-controller-manager-arm64:v1.30.0
+++ [0508 15:45:40] Deleting docker image registry.k8s.io/kube-apiserver-arm64:v1.30.0
+++ [0508 15:45:40] Docker builds done
Finished building Kubernetes
Building node image ...
Building in container: kind-build-1715175957-1495463435
Image "kindest/node:latest" build completed.
Now let's build our image with CRI-O on top of kindest/node:latest
:
FROM kindest/node:latest
ARG CRIO_VERSION
ARG PROJECT_PATH=prerelease:/$CRIO_VERSION
RUN echo "Installing Packages ..." \
&& apt-get clean \
&& apt-get update -y \
&& DEBIAN_FRONTEND=noninteractive apt-get install -y \
software-properties-common vim gnupg \
&& echo "Installing cri-o ..." \
&& curl -fsSL https://pkgs.k8s.io/addons:/cri-o:/$PROJECT_PATH/deb/Release.key | gpg --dearmor -o /etc/apt/keyrings/cri-o-apt-keyring.gpg \
&& echo "deb [signed-by=/etc/apt/keyrings/cri-o-apt-keyring.gpg] https://pkgs.k8s.io/addons:/cri-o:/$PROJECT_PATH/deb/ /" | tee /etc/apt/sources.list.d/cri-o.list \
&& apt-get update \
&& DEBIAN_FRONTEND=noninteractive apt-get --option=Dpkg::Options::=--force-confdef install -y cri-o \
&& sed -i 's/containerd/crio/g' /etc/crictl.yaml \
&& systemctl disable containerd \
&& systemctl enable crio
Next let's build image with
prerelease:v1.30
CRI-O version:
$ CRIO_VERSION=v1.30
$ docker build --build-arg CRIO_VERSION=$CRIO_VERSION -t kindnode/crio:$CRIO_VERSION .
# some output
Note
In case you're using buildx
the error like below might happen:
$ docker build --build-arg CRIO_VERSION=$CRIO_VERSION -t kindnode/crio:$CRIO_VERSION .
[+] Building 1.4s (2/2) FINISHED docker-container:kind-builder
=> [internal] load build definition from Dockerfile 0.0s
=> => transferring dockerfile: 977B 0.0s
=> ERROR [internal] load metadata for docker.io/kindest/node:latest 1.2s
------
> [internal] load metadata for docker.io/kindest/node:latest:
------
WARNING: No output specified with docker-container driver. Build result will only remain in the build cache. To push result image into registry use --push or to load image into docker use --load
Dockerfile:1
--------------------
1 | >>> FROM kindest/node:latest
2 |
3 | ARG CRIO_VERSION
--------------------
ERROR: failed to solve: kindest/node:latest: failed to resolve source metadata for docker.io/kindest/node:latest: docker.io/kindest/node:latest: not found
Check if you're using buildx
:
$ cat ~/.docker/config.json
{
"auths": {},
"aliases": {
"builder": "buildx"
}
}
$ mv ~/.docker/config.json ~/.docker/config.jsonBACKUP
# disable buildx for a while
With the built node image,
we can create the kind cluster:
kind: Cluster
apiVersion: kind.x-k8s.io/v1alpha4
nodes:
- role: control-plane
kubeadmConfigPatches:
- |
kind: InitConfiguration
nodeRegistration:
criSocket: unix:///var/run/crio/crio.sock
- |
kind: JoinConfiguration
nodeRegistration:
criSocket: unix:///var/run/crio/crio.sock
- role: worker
kubeadmConfigPatches:
- |
kind: JoinConfiguration
nodeRegistration:
criSocket: unix:///var/run/crio/crio.sock
Let's create a cluster:
$ kind create cluster --image kindnode/crio:$CRIO_VERSION --config ./kind-crio.yaml
Creating cluster "kind" ...
✓ Ensuring node image (kindnode/crio:v1.30) 🖼
✓ Preparing nodes 📦 📦
✓ Writing configuration 📜
✓ Starting control-plane 🕹️
✓ Installing CNI 🔌
✓ Installing StorageClass 💾
✓ Joining worker nodes 🚜
Set kubectl context to "kind-kind"
You can now use your cluster with:
kubectl cluster-info --context kind-kind
Have a question, bug, or feature request? Let us know! https://kind.sigs.k8s.io/#community 🙂
Let's try a simple deployment kubectl apply -f httpbin.yaml
:
apiVersion: v1
kind: ServiceAccount
metadata:
name: httpbin
---
apiVersion: v1
kind: Service
metadata:
name: httpbin
labels:
app: httpbin
service: httpbin
spec:
ports:
- name: http
port: 8000
targetPort: 8080
selector:
app: httpbin
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: httpbin
spec:
replicas: 1
selector:
matchLabels:
app: httpbin
version: v1
template:
metadata:
labels:
app: httpbin
version: v1
spec:
serviceAccountName: httpbin
containers:
- image: docker.io/kong/httpbin
imagePullPolicy: IfNotPresent
name: httpbin
# Same as found in Dockerfile's CMD but using an unprivileged port
command:
- gunicorn
- -b
- 0.0.0.0:8080
- httpbin:app
- -k
- gevent
env:
# Tells pipenv to use a writable directory instead of $HOME
- name: WORKON_HOME
value: /tmp
ports:
- containerPort: 8080
Use port-forward
:
$ kubectl port-forward svc/httpbin 8000:8000 -n default
Forwarding from 127.0.0.1:8000 -> 8080
Forwarding from [::1]:8000 -> 8080
In another terminal:
curl -X GET localhost:8000/get
{
"args": {},
"headers": {
"Accept": "*/*",
"Host": "localhost:8000",
"User-Agent": "curl/8.4.0"
},
"origin": "127.0.0.1",
"url": "http://localhost:8000/get"
}