Apply Pod Security Standards at the Namespace Level

Note

This tutorial applies only for new clusters.

Pod Security Admission is an admission controller that applies Pod Security Standards when pods are created. It is a feature GA’ed in v1.25. In this tutorial, you will enforce the baseline Pod Security Standard, one namespace at a time.

You can also apply Pod Security Standards to multiple namespaces at once at the cluster level. For instructions, refer to Apply Pod Security Standards at the cluster level.

Before you begin

Install the following on your workstation:

Create cluster

  1. Create a kind cluster as follows:

    1. kind create cluster --name psa-ns-level

    The output is similar to this:

    1. Creating cluster "psa-ns-level" ...
    2.  ✓ Ensuring node image (kindest/node:v1.29.0) 🖼 
    3.  ✓ Preparing nodes 📦  
    4.  ✓ Writing configuration 📜 
    5.  ✓ Starting control-plane 🕹️ 
    6.  ✓ Installing CNI 🔌 
    7.  ✓ Installing StorageClass 💾 
    8. Set kubectl context to "kind-psa-ns-level"
    9. You can now use your cluster with:
    11. kubectl cluster-info --context kind-psa-ns-level
    13. Not sure what to do next? 😅  Check out https://kind.sigs.k8s.io/docs/user/quick-start/

  2. Set the kubectl context to the new cluster:

    1. kubectl cluster-info --context kind-psa-ns-level

    The output is similar to this:

    1. Kubernetes control plane is running at https://127.0.0.1:50996
    2. CoreDNS is running at https://127.0.0.1:50996/api/v1/namespaces/kube-system/services/kube-dns:dns/proxy
    4. To further debug and diagnose cluster problems, use 'kubectl cluster-info dump'.

Create a namespace

Create a new namespace called example:

  1. kubectl create ns example

The output is similar to this:

  1. namespace/example created

Enable Pod Security Standards checking for that namespace

  1. Enable Pod Security Standards on this namespace using labels supported by built-in Pod Security Admission. In this step you will configure a check to warn on Pods that don’t meet the latest version of the baseline pod security standard.

    1. kubectl label --overwrite ns example \
    2.    pod-security.kubernetes.io/warn=baseline \
    3.    pod-security.kubernetes.io/warn-version=latest

  2. You can configure multiple pod security standard checks on any namespace, using labels. The following command will enforce the baseline Pod Security Standard, but warn and audit for restricted Pod Security Standards as per the latest version (default value)

    1. kubectl label --overwrite ns example \
    2.   pod-security.kubernetes.io/enforce=baseline \
    3.   pod-security.kubernetes.io/enforce-version=latest \
    4.   pod-security.kubernetes.io/warn=restricted \
    5.   pod-security.kubernetes.io/warn-version=latest \
    6.   pod-security.kubernetes.io/audit=restricted \
    7.   pod-security.kubernetes.io/audit-version=latest

Verify the Pod Security Standard enforcement

  1. Create a baseline Pod in the example namespace:

    1. kubectl apply -n example -f https://k8s.io/examples/security/example-baseline-pod.yaml

    The Pod does start OK; the output includes a warning. For example:

    1. Warning: would violate PodSecurity "restricted:latest": allowPrivilegeEscalation != false (container "nginx" must set securityContext.allowPrivilegeEscalation=false), unrestricted capabilities (container "nginx" must set securityContext.capabilities.drop=["ALL"]), runAsNonRoot != true (pod or container "nginx" must set securityContext.runAsNonRoot=true), seccompProfile (pod or container "nginx" must set securityContext.seccompProfile.type to "RuntimeDefault" or "Localhost")
    2. pod/nginx created

  2. Create a baseline Pod in the default namespace:

    1. kubectl apply -n default -f https://k8s.io/examples/security/example-baseline-pod.yaml

    Output is similar to this:

    1. pod/nginx created

The Pod Security Standards enforcement and warning settings were applied only to the example namespace. You could create the same Pod in the default namespace with no warnings.

Clean up

Now delete the cluster which you created above by running the following command:

  1. kind delete cluster --name psa-ns-level

What’s next