Commands and Arguments in Kubernetes

Welcome to this lesson on configuring commands and arguments in Kubernetes pods. In this session, we'll learn how to adjust container behaviors by overriding default settings defined in the Dockerfile via the pod definition.

Previously, we built a simple Docker image—named "ubuntu-sleeper"—that executes a sleep command for a specified number of seconds. By default, running a container with this image makes it sleep for five seconds. However, you can easily change this behavior by passing a command-line argument.

Overriding Default Behavior with Arguments

Suppose we want to create a pod using the "ubuntu-sleeper" image. We begin with a basic pod definition where the pod's name and image are specified. When the pod is created, it starts a container that runs the default sleep command (sleeping for five seconds) and then exits. To modify the sleep duration to 10 seconds, simply append an additional argument in the pod specification. Any argument provided in the Docker run command correlates with the args property in the pod definition file (formatted as an array).

Consider the following examples:

Docker commands:

docker run --name ubuntu-sleeper ubuntu-sleeper
docker run --name ubuntu-sleeper ubuntu-sleeper 10

Pod definition YAML:

apiVersion: v1
kind: Pod
metadata:
  name: ubuntu-sleeper-pod
spec:
  containers:
  - name: ubuntu-sleeper
    image: ubuntu-sleeper
    args: ["10"]

Dockerfile Instructions and Their Mappings

The Dockerfile for the "ubuntu-sleeper" image is defined with both an ENTRYPOINT and a CMD:

FROM Ubuntu
ENTRYPOINT ["sleep"]
CMD ["5"]

• The ENTRYPOINT instruction specifies the command to run when the container starts.
• The CMD instruction provides default parameters for that command.

By specifying the args field in the pod definition file, the CMD instruction is overridden, which effectively changes the sleep duration from 5 to 10 seconds.

Overriding the ENTRYPOINT

Now, consider a scenario where you want to override the ENTRYPOINT itself (for example, switching from "sleep" to an alternative command like "sleep2.0"). In Docker, you would use the --entrypoint option in the docker run command. In Kubernetes, this is achieved by providing the command field in the pod definition. Here, the command field corresponds to the Dockerfile’s ENTRYPOINT, while the args field continues to override the CMD instruction.

Docker command example with overridden ENTRYPOINT:

docker run --name ubuntu-sleeper \
  --entrypoint sleep2.0 \
  ubuntu-sleeper 10

Pod definition YAML with overridden ENTRYPOINT:

apiVersion: v1
kind: Pod
metadata:
  name: ubuntu-sleeper-pod
spec:
  containers:
    - name: ubuntu-sleeper
      image: ubuntu-sleeper
      command: ["sleep2.0"]
      args: ["10"]

Deployment of the Pod

Once your pod definition is ready, deploy the pod using the following command:

kubectl create -f pod-definition.yml

Key Fields Summary

Below is a table summarizing the Kubernetes pod definition fields that map to Dockerfile instructions:

This concludes the lesson on managing commands and arguments within your Kubernetes pods. Next, proceed to the coding exercises section to practice these concepts and refine your troubleshooting skills with real-world scenarios.

Happy coding!