JSON PATH Use case Kubernetes

Hello and welcome! I’m Mumshal Manambath. In this guide, you’ll learn how to leverage JSONPath with the kubectl command to query, filter, and format Kubernetes resources. We’ll cover:

• Viewing kubectl output in raw JSON
• Building JSONPath expressions step by step
• Real-world JSONPath examples
• Loop constructs (range) for iteration
• Custom columns and table sorting
• Practice exercises to reinforce your skills

Prerequisites

You should already have a basic grasp of JSONPath syntax and be comfortable running kubectl commands against a Kubernetes cluster.

Why Use JSONPath with kubectl?

Managing production Kubernetes clusters often means inspecting hundreds of nodes and thousands of objects—deployments, pods, ReplicaSets, services, secrets, and more. JSONPath lets you:

• Extract specific fields across many resources
• Generate concise summaries (e.g., node names with CPU counts)
• Filter output dynamically based on custom criteria

Manual parsing is error-prone and time-consuming. JSONPath queries with kubectl provide a programmable way to slice and dice cluster data.

Understanding kubectl Output

By default, kubectl formats API responses into human-readable tables. To see the raw JSON payload:

kubectl get nodes -o json

Example output (truncated):

{
  "apiVersion": "v1",
  "items": [
    {
      "kind": "Node",
      "metadata": {
        "name": "master",
        "labels": { … }
      },
      "status": {
        "capacity": {
          "cpu": "4"
        },
        "nodeInfo": {
          "architecture": "amd64"
        }
      }
    }
    // more nodes…
  ]
}

If you prefer more columns in the table view, use:

kubectl get nodes -o wide

NAME     STATUS   ROLES    AGE   VERSION     INTERNAL-IP    OS-IMAGE              KERNEL-VERSION
master   Ready    master   40m   v1.11.3     172.17.0.44    Ubuntu 16.04.2 LTS    4.4.0-18148-generic
node01   Ready    <none>   40m   v1.11.3     172.17.0.63    Ubuntu 16.04.2 LTS    4.4.0-18148-generic

Neither the default table nor kubectl describe lets you build fully custom reports (for example, listing node CPU, taints, and container images in one view). This is where JSONPath shines.

Getting Started with JSONPath

1. Identify the base resource:

   kubectl get pods
   

2. Append -o json to inspect the structure:

   kubectl get pods -o json
   

3. Draft a JSONPath expression against the JSON. Example:

   .items[0].spec.containers[0].image
   

4. Run with -o jsonpath and wrap your expression in single quotes:

   kubectl get pods -o jsonpath='{.items[0].spec.containers[0].image}'
   

Basic JSONPath Queries

Formatting with Newlines and Tabs

Add "\n" and "\t" for readability:

kubectl get nodes -o jsonpath='{.items[*].metadata.name}{"\n"}{.items[*].status.capacity.cpu}'

Produces:

master node01
4 4

Iteration Using range

Use range to iterate through items:

kubectl get nodes -o jsonpath='
{range .items[*]}{.metadata.name}{"\t"}{.status.capacity.cpu}{"\n"}{end}
'

Output:

master    4
node01    4

Custom Columns with kubectl

Define headers and expressions with -o custom-columns (no need for .items[*]):

kubectl get nodes -o custom-columns=NODE:.metadata.name,CPU:.status.capacity.cpu

NODE     CPU
master   4
node01   4

Sorting Resources

Order your output by any JSON field:

kubectl get nodes --sort-by=.metadata.name
kubectl get nodes --sort-by=.status.capacity.cpu

Conclusion

You now know how to extract, format, and sort Kubernetes data using JSONPath with kubectl. Practice these examples and try the exercises to master JSONPath queries in your clusters.

Links and References

• Kubernetes API Concepts
• kubectl Cheat Sheet
• Custom Columns Documentation
• jsonpath.com for testing expressions

Happy querying!