Confidential Containers & Secure Enclaves (Kata/Sandboxed + HW Support)

Duration: ~25 minutes

Overview

Learn how to deploy confidential containers using Kata containers and secure enclaves with hardware support. You’ll understand how confidential computing protects workload data and code from host-level access, meeting requirements for sensitive workloads and regulatory compliance.

Explore confidential container and enclave options on OpenShift and identify which workloads justify the stronger isolation model.

Why it matters

Confidential containers provide hardware-backed isolation that protects workload data and code even from privileged host access. This is critical for sensitive workloads, multi-tenant environments, and regulatory compliance where data must remain protected throughout its lifecycle.

What does it solve

  • Host-level data access

  • Multi-tenant data protection

  • Regulatory compliance

  • Sensitive workload isolation

  • Trust boundary extension

Your Mission

Complete the examples below to close the attack paths this lab covers—treat each step as defending the cluster, not just clicking through commands.

Click each step only if you need a hint.

Check Kata container support
oc get nodes -o jsonpath='{.items[*].status.nodeInfo.containerRuntimeVersion}'
Create RuntimeClass for Kata
oc apply -f - <<'EOF'
apiVersion: node.k8s.io/v1
kind: RuntimeClass
metadata:
  name: kata
handler: kata
EOF
Deploy workload with Kata runtime
oc apply -f - <<'EOF'
apiVersion: v1
kind: Pod
metadata:
  name: confidential-app
  namespace: 301-11-demo
spec:
  runtimeClassName: kata
  containers:
  - name: app
    image: registry.access.redhat.com/ubi9/ubi:latest
    command: ["sleep", "infinity"]
EOF
Verify runtime
oc get pod confidential-app -n 301-11-demo -o jsonpath='{.spec.runtimeClassName}'
Cleanup
oc delete project 301-11-demo --wait=false

Debrief

Confidential containers push trust toward hardware-backed isolation so a privileged host or noisy neighbor is a weaker attacker.

What breaks without this:

  • Shared-host visibility into memory/process for sensitive workloads

  • Regulatory need for stronger isolation than cgroups alone

Controls that matter: Kata/confidential runtimes, hardware support (TDX/SEV/etc.), encrypted memory patterns, and still stacking SCC/NetworkPolicy.

Quick facts: expect some overhead; select for sensitive or regulated workloads on less-trusted hosts.

giphy

Cleanup

Before moving to the next module, run the lab cleanup script to reset transient resources from this module.

cd ~/openshift-security-roadshow
bash setup/lab-cleanup.sh --module 301-11