kata-deploy

kata-deploy

kata-deploy provides a Dockerfile, which contains all of the binaries and artifacts required to run Kata Containers, as well as reference DaemonSets, which can be utilized to install Kata Containers on a running Kubernetes cluster.

Note: installation through DaemonSets successfully installs katacontainers.io/kata-runtime on a node only if it uses either containerd or CRI-O CRI-shims.

Kubernetes quick start

Install Kata on a running Kubernetes cluster

k3s cluster

For your k3s cluster, run:

$ git clone https://github.com/kata-containers/kata-containers.git

Check and switch to the stable branch of your choice, if wanted, and then run:

$ cd kata-containers/tools/packaging/kata-deploy
$ kubectl apply -f kata-rbac/base/kata-rbac.yaml
$ kubectl apply -k kata-deploy/overlays/k3s

RKE2 cluster

For your RKE2 cluster, run:

$ git clone https://github.com/kata-containers/kata-containers.git

Check and switch to the stable branch of your choice, if wanted, and then run:

$ cd kata-containers/tools/packaging/kata-deploy
$ kubectl apply -f kata-rbac/base/kata-rbac.yaml
$ kubectl apply -k kata-deploy/overlays/rke2

[k0s] cluster

For your k0s cluster, run:

$ git clone https://github.com/kata-containers/kata-containers.git

Check and switch to "main", and then run:

$ cd kata-containers/tools/packaging/kata-deploy
$ kubectl apply -f kata-rbac/base/kata-rbac.yaml
$ kubectl apply -k kata-deploy/overlays/k0s

Note

The supported version of k0s is v1.27.1+k0s and above, since the k0s support leverages a special dynamic containerd configuration mode:

From 1.27.1 onwards k0s enables dynamic configuration on containerd CRI runtimes. This works by k0s creating a special directory in /etc/k0s/containerd.d/ where user can drop-in partial containerd configuration snippets.

k0s will automatically pick up these files and adds these in containerd configuration imports list. If k0s sees the configuration drop-ins are CRI related configurations k0s will automatically collect all these into a single file and adds that as a single import file. This is to overcome some hard limitation on containerd 1.X versions. Read more at containerd#8056

However, this would also require a magic string set in the beginning of the line for /etc/k0s/containerd.toml:

# k0s_managed=true

Vanilla Kubernetes cluster

Installing the latest image

The latest image refers to pre-release and release candidate content. For stable releases, please, use the "stable" instructions.

$ kubectl apply -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/kata-rbac/base/kata-rbac.yaml
$ kubectl apply -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/kata-deploy/base/kata-deploy.yaml

Installing the stable image

The stable image refers to the last stable releases content.

Note: if you use a tagged version of the repo, the stable image does match that version. For instance, if you use the 2.2.1 tagged version of the kata-deploy.yaml file, then the version 2.2.1 of the kata runtime will be deployed.

$ kubectl apply -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/kata-rbac/base/kata-rbac.yaml
$ kubectl apply -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/kata-deploy/base/kata-deploy-stable.yaml

Ensure kata-deploy is ready

$ kubectl -n kube-system wait --timeout=10m --for=condition=Ready -l name=kata-deploy pod

Run a sample workload

Workloads specify the runtime they'd like to utilize by setting the appropriate runtimeClass object within the Pod specification. The runtimeClass examples provided define a node selector to match node label katacontainers.io/kata-runtime:"true", which will ensure the workload is only scheduled on a node that has Kata Containers installed

runtimeClass is a built-in type in Kubernetes. To apply each Kata Containers runtimeClass:

  $ kubectl apply -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/runtimeclasses/kata-runtimeClasses.yaml

The following YAML snippet shows how to specify a workload should use Kata with Dragonball:

spec:
  template:
    spec:
      runtimeClassName: kata-dragonball

The following YAML snippet shows how to specify a workload should use Kata with Cloud Hypervisor:

spec:
  template:
    spec:
      runtimeClassName: kata-clh

The following YAML snippet shows how to specify a workload should use Kata with Firecracker:

spec:
  template:
    spec:
      runtimeClassName: kata-fc

The following YAML snippet shows how to specify a workload should use Kata with StratoVirt:

spec:
  template:
    spec:
      runtimeClassName: kata-stratovirt

The following YAML snippet shows how to specify a workload should use Kata with QEMU:

spec:
  template:
    spec:
      runtimeClassName: kata-qemu

To run an example with kata-dragonball:

$ kubectl apply -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/examples/test-deploy-kata-dragonball.yaml

To run an example with kata-clh:

$ kubectl apply -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/examples/test-deploy-kata-clh.yaml

To run an example with kata-fc:

$ kubectl apply -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/examples/test-deploy-kata-fc.yaml

To run an example with kata-stratovirt:

$ kubectl apply -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/examples/test-deploy-kata-stratovirt.yaml

To run an example with kata-qemu:

$ kubectl apply -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/examples/test-deploy-kata-qemu.yaml

The following removes the test pods:

$ kubectl delete -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/examples/test-deploy-kata-dragonball.yaml
$ kubectl delete -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/examples/test-deploy-kata-clh.yaml
$ kubectl delete -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/examples/test-deploy-kata-fc.yaml
$ kubectl delete -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/examples/test-deploy-kata-stratovirt.yaml
$ kubectl delete -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/examples/test-deploy-kata-qemu.yaml

Remove Kata from the Kubernetes cluster

Removing the latest image

$ kubectl delete -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/kata-deploy/base/kata-deploy.yaml
$ kubectl -n kube-system wait --timeout=10m --for=delete -l name=kata-deploy pod

After ensuring kata-deploy has been deleted, cleanup the cluster:

$ kubectl apply -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/kata-cleanup/base/kata-cleanup.yaml

The cleanup daemon-set will run a single time, cleaning up the node-label, which makes it difficult to check in an automated fashion. This process should take, at most, 5 minutes.

After that, let's delete the cleanup daemon-set, the added RBAC and runtime classes:

$ kubectl delete -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/kata-cleanup/base/kata-cleanup.yaml
$ kubectl delete -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/kata-rbac/base/kata-rbac.yaml
$ kubectl delete -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/runtimeclasses/kata-runtimeClasses.yaml

Removing the stable image

$ kubectl delete -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/kata-deploy/base/kata-deploy-stable.yaml
$ kubectl -n kube-system wait --timeout=10m --for=delete -l name=kata-deploy pod

After ensuring kata-deploy has been deleted, cleanup the cluster:

$ kubectl apply -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/kata-cleanup/base/kata-cleanup-stable.yaml

The cleanup daemon-set will run a single time, cleaning up the node-label, which makes it difficult to check in an automated fashion. This process should take, at most, 5 minutes.

After that, let's delete the cleanup daemon-set, the added RBAC and runtime classes:

$ kubectl delete -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/kata-cleanup/base/kata-cleanup-stable.yaml
$ kubectl delete -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/kata-rbac/base/kata-rbac.yaml
$ kubectl delete -f https://raw.githubusercontent.com/kata-containers/kata-containers/main/tools/packaging/kata-deploy/runtimeclasses/kata-runtimeClasses.yaml

kata-deploy details

Dockerfile

The Dockerfile used to create the container image deployed in the DaemonSet is provided here. This image contains all the necessary artifacts for running Kata Containers, all of which are pulled from the Kata Containers release page.

Host artifacts:

• cloud-hypervisor, firecracker, qemu, stratovirt and supporting binaries
• containerd-shim-kata-v2 (go runtime and rust runtime)
• kata-collect-data.sh
• kata-runtime

Virtual Machine artifacts:

• kata-containers.img and kata-containers-initrd.img: pulled from Kata GitHub releases page
• vmlinuz.container and vmlinuz-virtiofs.container: pulled from Kata GitHub releases page

DaemonSets and RBAC

Two DaemonSets are introduced for kata-deploy, as well as an RBAC to facilitate applying labels to the nodes.

Kata deploy

This DaemonSet installs the necessary Kata binaries, configuration files, and virtual machine artifacts on the node. Once installed, the DaemonSet adds a node label katacontainers.io/kata-runtime=true and reconfigures either CRI-O or containerd to register three runtimeClasses: kata-clh (for Cloud Hypervisor isolation), kata-qemu (for QEMU isolation), kata-fc (for Firecracker isolation) and kata-stratovirt (for StratoVirt isolation). As a final step the DaemonSet restarts either CRI-O or containerd. Upon deletion, the DaemonSet removes the Kata binaries and VM artifacts and updates the node label to katacontainers.io/kata-runtime=cleanup.

Kata cleanup

This DaemonSet runs of the node has the label katacontainers.io/kata-runtime=cleanup. These DaemonSets removes the katacontainers.io/kata-runtime label as well as restarts either CRI-O or containerd systemctl daemon. You cannot execute these resets during the preStopHook of the Kata installer DaemonSet, which necessitated this final cleanup DaemonSet.