DevOps Classroomnotes 13/Feb/2022

Exercise 1: Create a Pod Manifest to run alpine with sleep 1d

  • Refer Here for the changes made
  • Now lets look at describing out kubectl describe <resource-name> <object-name>

Using Port forwarding

  • Lets launch the pod from activity1

Getting More Information With Logs


Activity 3: Create a Pod with container which has CMD to get container to exited state

  • Refer Here for the changes
  • As the container gets into the exited state pod tries to restart the container and this happens continuously
  • To deal with these kind of continuous restarts we will be learning probes.

Activity 4: Start activity 1

  • create a pod with spec activity1.yaml
  • Now shutdown the node where activity1 is running. (Simulating some error on the node)
  • Node is shutdown, Pod is not accesible, but k8s is not scheduling this pod to run on other node

Workload Resources

  • Kubernetes Provides different workload resources
  • Replicaset:
    • This is used to create multiple replications of your Pod which is running your application
  • Deployment
  • Statefulset
  • DaemonSet
  • Job and CronJob
  • If we try to use the above workloads then the desired state is maintained
  • Since Pod cannot maintain its own state i.e. if the Pod is Down, schedule the Pod on a new node
  • Running our application in the Workload resources which maintain some kind of desired state is what we want.
  • Before we understand workload resources, we need to understand the concept of labels and annotations in kubernetes


  • Labels are key/value pairs that can be attached to k8s objects such as Pods, replicasets etc.
  • They are arbitrary and are useful for identifying information about k8s objects
  • They provide the foundation for grouping objects
  • Labels have very simple syntax.
  • Keys can be broken down into two parts an optional prefix and a name seperated by slahs (dots, underscores, dashes(hypens)) can be used in names
  • Values are strings with maximum of 63 characters
  • Examples:

| key | value |
| — | ——|
| | 1.0.0 |
| appVersion | 1.0.0 |
| app.version | 1.0.0 |
| project | qtidentitiy |
* Lets try to apply labels
* Now lets get labels
* Label Selectors: These are used to filter k8s objects based on set of labels
* Lets create labels in the declarative approach. Refer Here for the Pod manifest
* Selectors in the Yaml will be in the format of

  app: web
  env: prod


  • Annotations are also key value pairs like labels which provide a place to store additional metadata for k8s objects with the sole purpose of assisting tools and libraries
  annotations: ""


  • If we want multiple replicas of the application in Pod running and maintaing the state of number of replicas we use replica sets

Activity 5: Lets create a replicaset to create 2 replicas of jenkins (Pod)

  • Refer Here for api reference
  • Refer Here for Replicaset Spec
  • Refer Here for the ReplicaSet Manifest created
  • Now delete one pod manually to see what replica set does

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

About continuous learner

devops & cloud enthusiastic learner