Kubernetes Classroom Series – 18/Oct/2019, Namespaces & Jobs

Namespaces

• Problem-1: Multiple Environments created in same cluster, creates confusion when we execute kubectl commands

kubectl get pods
# all the pods of different enviroments

kubect get deployments
# all the deployments of different  environments

• Problem-2: You are ok with multiple environments in a cluster. lets assume you are creating springpetclinic

apiVersion: v1
kind: Pod
metadata:
  name: spclinic
  ...
  ...
  ...

  

• Problem-2 :

  • Here the problem is we might be using same spec/manifest for different environment. At least same names for pods, rc, deployments etc.
  • K8S will not allow to create two objects of same kind with same name

• To avoid problems like above, We create a virtual cluster and Namespace is virtual cluster

• The default Namespace of Kubernetes is default

Exploring namespaces

• Execute the following commands

kubectl get namespaces

• Create two YAML files for test & dev environments

apiVersion: v1
kind: Namespace
metadata:
  name: dev-env

apiVersion: v1
kind: Namespace
metadata:
  name: dev-env

• Execute kubectl apply commands and then kubectl get namespaces
• create a Pod with the following info

---
apiVersion: v1
kind: Pod
metadata:
  name: nginx
  namespace: dev-env
spec:
  containers:
    - image: nginx
      name: nginx-demo
      ports:
        - containerPort: 80
          protocol: TCP

• Execute kubecl apply -f command
• Then execute the following

kubectl get pods
kubectl get pods --all-namespaces
kubectl get pods -n dev-env

• Create one more nginx pod in test-env

apiVersion: v1
kind: Pod
metadata:
  name: nginx
  namespace: dev-env
spec:
  containers:
    - image: nginx
      name: nginx-demo
      ports:
        - containerPort: 80
          protocol: TCP

• Now execute the following commands

kubectl apply -f nginx-test-pod.yml
kubectl get pods -n test-env
kubectl get pods --all-namespaces

Jobs & Cron Jobs

• Jobs will create containers which run for finite time.
• Cron Jobs repeatedly will create containers which run for finite time.
• Will appear under api group batch
• Sample Job

apiVersion: batch/v1
kind: Job
metadata:
  # Unique key of the Job instance
  name: example-job
spec:
  template:
    metadata:
      name: example-job
    spec:
      containers:
      - name: pi
        image: perl
        command: ["perl"]
        args: ["-Mbignum=bpi", "-wle", "print bpi(2000)"]
      # Do not restart containers after they exit
      restartPolicy: Never