Hey there, let's clear up the difference between port and targetPort using your Service example—it's a common point of confusion, so great question!

port #

This is the port that the Service itself exposes to the rest of the Kubernetes cluster. Any other pod, deployment, or internal resource that wants to talk to your my-service will use this port. Think of it as the Service's "public-facing" port within the cluster. In your example, if another pod needs to reach your MyApp, it would connect to my-service:80.

targetPort #

This is the port that your backend application pods are actually listening on. The Service acts as a traffic router here: when it gets incoming traffic on its port, it forwards that traffic to the targetPort on the pods matched by the Service's selector (in your case, pods labeled app: MyApp). So in your example, traffic hitting my-service:80 gets sent straight to port 9376 on your MyApp pods.

Quick Traffic Flow Recap #

Internal Cluster Pod → my-service:80 (Service port) → MyApp Pod:9376 (targetPort)

One handy bonus: targetPort doesn't have to be a number! If your pod's container defines a named port (like containerPort: 9376 with name: app-port), you can set targetPort: app-port in the Service. This makes your config more flexible if you ever change the actual port number later—no need to update the Service definition.

内容的提问来源于stack exchange，提问作者Mr.DevEng