Day 3: YAML, Vim, and Saying Goodbye to Nano

Today was all about YAML, the language Kubernetes uses to define everything. (Used a lot in Docker, so a familiar space for me!) Writing YAML manually can be a pain, but Kubernetes has some tricks to make it easier. I also learned a few Vim hacks that made working with YAML files a lot smoother—looks like I’m officially making the switch from Nano.

Editing YAML in the Cluster (But Should You?)

Kubernetes lets you edit resources directly in the cluster using:

kubectl edit pod <pod-name>

This opens the Pod’s YAML file in Vim so I can make quick changes. It’s useful, but in the real world, editing live configs isn’t best practice. Everything should be version-controlled and applied properly. That said, knowing this is a must for the CKA exam, so I made sure to practice.

Generating YAML with Dry Runs

Instead of writing a YAML file from scratch, Kubernetes can generate a template for me:

kubectl run nginx --image=nginx --dry-run=client -o yaml > nginx.yaml

Now I have a nginx.yaml file ready to be modified before deployment. Much easier than starting from zero.

Making Vim Play Nice with YAML

One of the most frustrating things about working with YAML is pasting text into Vim and having the indentation go crazy. I found a simple fix:

:set paste

This prevents Vim from messing with formatting when pasting content. Definitely a lifesaver when copying YAML configs.

YAML Example

Here’s the YAML file I generated and modified:

apiVersion: v1
kind: Pod
metadata:
  name: nginx-yaml
  labels:
    run: nginx-yaml
    method: fromcode
    another: label
spec:
  containers:
  - name: nginx
    image: nginx

This defines a simple Pod running an Nginx container.

Final Thoughts

YAML is everywhere in Kubernetes, so getting comfortable with it is a must. Between learning how to generate, edit, and cleanly paste YAML, today’s session felt like a big step forward. Also, after years of using Nano, it looks like Vim is officially my new editor.

Next up: Interacting with Pods—checking logs, troubleshooting, and getting hands-on with running containers.