Deploying and Validating Kata Containers

This blog provides a step-by-step guide to installing and validating Kata Containers on IBM Cloud, covering essential configurations and hands-on deployment.

Mar. 06, 25 · Tutorial

Likes (1)

Comment

Save

5.4K Views

In Part 1, we explored how Kata Containers bring together the agility of Kubernetes pods and the robust security of virtual machines, offering a powerful solution for cloud-native workloads.

In this part, we’ll provide a hands-on guide to deploying and validating Kata Containers in IBM Cloud, a platform well-suited for secure, scalable, and enterprise-grade cloud solutions. From configuring the runtime to testing its performance and security, we’ll walk you through the entire process. By the end of this guide, you’ll have the practical knowledge to integrate Kata Containers into your IBM Cloud environment and enhance the security of your containerized applications.

Let’s turn theory into practice and unlock the potential of Kata Containers in IBM Cloud!

Prerequisites

Before starting, ensure you have the following:

An IBM Cloud Kubernetes Service (IKS) cluster running in VPC
Kubernetes CLI (kubectl) configured for your cluster
Basic familiarity with Kubernetes YAML configurations

For this exercise, we’ve considered a simple cluster configuration:

Number of nodes: 3
Node flavor: bx2.4x16

Step 1: Install Kata Containers on the Cluster

To deploy Kata Containers, you’ll use the Kata Deploy daemonset, which automates the installation process. Run the following command to deploy it:

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

This daemonset will install Kata runtime configurations across all worker nodes in your cluster.

Step 2: Create the KataConfig Custom Resource

Define a KataConfig Custom Resource (CR) to manage the deployment of Kata runtimes on the cluster.

Use the following YAML file:

    YAML
   
 

   kind: KataConfig
metadata:
   name: kataconfig
spec:
   runtimeClass: kata-qemu
  

Apply this configuration with:

    Shell
   
   kubectl apply -f kataconfig.yaml

This CR automates the setup of the Kata runtime on all worker nodes.

Step 3: Deploy a Sample Workload

To test the integration of Kata Containers, deploy a sample workload. Use the following example deployment YAML:

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

Step 4: Validate the Deployment

Check the status of the running pods to ensure the workload is deployed successfully:

    Shell
   
   kubectl get pods

Example output:

    Plain Text
   
   NAME                                    READY   STATUS    RESTARTS   AGE
php-apache-kata-qemu-78ccf8958f-nbn6d   1/1     Running   0          46h

Inspect the pod details to confirm it is using the kata-qemu runtime:

    Shell
   
   kubectl describe pod php-apache-kata-qemu-78ccf8958f-nbn6d

Key details to verify:

Runtime class name: kata-qemu
Node selector: katacontainers.io/kata-runtime=true

Step 5: Verify the Underlying VM

To ensure the container is running inside a lightweight VM, log into one of your cluster nodes and check the running processes:

    Shell
   
   ps -eaf | grep "kvm"

You should see a process corresponding to the Kata Container VM, such as the following:

Command:

    Shell
   
   ps -eaf | grep "kvm"

Output:

    Plain Text
   
 

   root      49189  49179  0 Dec03 ?        00:02:57 /var/opt/kata/bin/qemu-system-x86_64 \
  -name sandbox-6afa9c35ac34dd2e7364ddae072358233b57efa390b5c9f2b6b30986be48e686 \
  -uuid 7e89b75a-6dab-4d21-80da-f7d86ee3d4f7 \
  -machine q35,accel=kvm,nvdimm=on \
  -cpu host,pmu=off \
  -qmp unix:fd=3,server=on,wait=off \
  -m 2048M,slots=10,maxmem=17040M \
  -device pci-bridge,bus=pcie.0,id=pci-bridge-0,chassis_nr=1,shpc=off,addr=2,io-reserve=4k,mem-reserve=1m,pref64-reserve=1m \
  -device virtio-serial-pci,disable-modern=true,id=serial0 \
  -device virtconsole,chardev=charconsole0,id=console0 \
  -chardev socket,id=charconsole0,path=/run/vc/vm/6afa9c35ac34dd2e7364ddae072358233b57efa390b5c9f2b6b30986be48e686/console.sock,server=on,wait=off \
  -device nvdimm,id=nv0,memdev=mem0,unarmed=on \
  -object memory-backend-file,id=mem0,mem-path=/var/opt/kata/share/kata-containers/kata-ubuntu-latest.image,size=268435456,readonly=on \
  -device virtio-scsi-pci,id=scsi0,disable-modern=true \
  -object rng-random,id=rng0,filename=/dev/urandom \
  -device virtio-rng-pci,rng=rng0 \
  -device vhost-vsock-pci,disable-modern=true,vhostfd=4,id=vsock-2212957142,guest-cid=2212957142 \
  -chardev socket,id=char-c824ec5354a6e25f,path=/run/vc/vm/6afa9c35ac34dd2e7364ddae072358233b57efa390b5c9f2b6b30986be48e686/vhost-fs.sock \
  -device vhost-user-fs-pci,chardev=char-c824ec5354a6e25f,tag=kataShared,queue-size=1024 \
  -netdev tap,id=network-0,vhost=on,vhostfds=5,fds=6 \
  -device driver=virtio-net-pci,netdev=network-0,mac=d6:4a:fd:51:f7:06,disable-modern=true,mq=on,vectors=4 \
  -rtc base=utc,driftfix=slew,clock=host \
  -global kvm-pit.lost_tick_policy=discard \
  -vga none \
  -no-user-config \
  -nodefaults \
  -nographic \
  --no-reboot \
  -object memory-backend-file,id=dimm1,size=2048M,mem-path=/dev/shm,share=on \
  -numa node,memdev=dimm1 \
  -kernel /var/opt/kata/share/kata-containers/vmlinux-6.1.62-140 \
  -append tsc=reliable no_timer_check rcupdate.rcu_expedited=1 i8042.direct=1 i8042.dumbkbd=1 i8042.nopnp=1 i8042.noaux=1 noreplace-smp reboot=k cryptomgr.notests net.ifnames=0 pci=lastbus=0 root=/dev/pmem0p1 rootflags=dax,data=ordered,errors=remount-ro ro rootfstype=ext4 console=hvc0 console=hvc1 quiet systemd.show_status=false panic=1 nr_cpus=4 selinux=0 systemd.unit=kata-containers.target systemd.mask=systemd-networkd.service systemd.mask=systemd-networkd.socket scsi_mod.scan=none \
  -pidfile /run/vc/vm/6afa9c35ac34dd2e7364ddae072358233b57efa390b5c9f2b6b30986be48e686/pid \
  -smp 1,cores=1,threads=1,sockets=4,maxcpus=4

root      49197      2  0 Dec03 ?        00:00:00 [kvm-nx-lpage-re]
root      49201      2  0 Dec03 ?        00:00:00 [kvm-pit/49189]
root     116536 116078  0 04:22 ?        00:00:00 grep --color=auto kvm
  

Conclusion

In this article, we explored the setup and deployment of Kata Containers on an IBM Cloud Kubernetes Service (IKS) cluster within a VPC. With its strong isolation and security capabilities, Kata Containers offer a reliable runtime solution for workloads that demand enhanced security or multi-tenancy.

IBM Cloud Kubernetes Kata (programming)

Opinions expressed by DZone contributors are their own.

Related

Trending