DevOps Classroomnotes 16/Sep/2023

Managed Kubernetes or Kubernetes as a Service

Managed Kuberenetes (Kubernetes as a Service) is the offerings from various cloud providers.
In these Control-Plane is managed by Cloud Service Providers and they charge you hourly for that
For worker nodes, storage and other resources charges are as usual.
Popular Kuberentes offerings
- AKS (Azure Kubernetes Services)
- EKS (Elastic Kuberentes Services)
- GKE (Google Kubernetes Engine)

AKS

Install kubectl in your system

choco install kubernetes-cli
# mac
brew install kubectl

Creating AKS Cluster
- CLI Refer Here
- Portal Refer Here

Activity: Create a replica set

Apply the replicaset

---
apiVersion: apps/v1
kind: ReplicaSet
metadata:
  name: nginx-rs
  labels:
    app: nginx
    purpose: svcdemo
spec:
  replicas: 3
  minReadySeconds: 1
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
        - name: nginxc
          image: nginx:1.25
          ports:
            - containerPort: 80
              protocol: TCP
          env:
            - name: purpose
              value: learning
            - name: why
              value: dontknow

* Refer Here for the changes that include adding environmental variables in the container

Lets create the service and expose it using the LoadBalancer. Azure internally takes request from cloud controller manager and creates a Load Balancer and all the necessary rules to expose the nginx service on Azure Load Balancer

Apply the service
We get the external ip which represents the ipaddress of the loadbalancer. Now access by using http://<lb-ip>:<port>
Refer Here for the changeset
Exercise:
- Write a spec to expose jenkins/jenkins which runs on 8080 port using nodePort on Port 30001
- DNS:
  - A record
  - CName Record
Service with external name

---
apiVersion: v1
kind: Service
metadata:
  name: nginx-svc-external
  labels:
    app: nginx
    purpose: svcdemo
spec:
  type: ExternalName
  externalName: nginx.qt.com
  selector:
    app: nginx

Deployments

Refer Here for the official docs
Overview
Deployment is a workload which creates
- Replica Sets: These create
  - Pods: Which inturn creates
    - Containers: This is where the applcation runs
Kubernetes Deployments creates Replicaset which can be persisted which helps us in performing rolling updates and rollbacks
Lets create a basic template for deployment of nopCommerce. The image will be shaikkhajaibrahim/nopcommerceaug23

---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: nop-deploy
spec:
  minReadySeconds: 5
  replicas: 4
  selector:
    matchLabels:
      app: nop
  strategy:
    type: RollingUpdate
    rollingUpdate:
      maxSurge: 25%
      maxUnavailable: 25%
  template:
    metadata:
      labels:
        app: nop
        version: v1.1
    spec:
      containers:
        - name: nop
          image: shaikkhajaibrahim/nopcommerceaug23
          ports:
            - containerPort: 80
              protocol: TCP

Apply the deployment and check the objects

Performing Rolling updates and Rollback using Deployments

Lets create a k8s manifest to deploy an application shaikkhajaibrahim/deploy-sample
Lets create with 10 replicas and a service exposed via loadbalancer
Refer Here for the changes done
Apply the deployment and set the current namespace to prod

 kubectl config set-context --current --namespace=prod

Now create service
Access the application
Now we have some extra commands
Annotation: Refer Here
We will be adding an annotation to add change cause kubernetes.io/change-cause
apply the deployment Refer Here for changes
Lets rollback to revision 1