Docker: container platform for running and managing containers.
Kubernetees: container orchestration platform.

Why Kubernetees?

Prb 1 : (Single host nature of your container -limitation) - if there are 100 container under Docker (one VM). If 1st cont1 consuming lot of resources. the 99th or other container is dead or not even started. This is decided by kernel.
Prb 2: (Auto healing) in kubernetess. If container is dieing, docker itself will not bring back. Here comes kubernetess
Prb 3: (Auto scaling) - if sudden newtork traffic increases, we need load balancers whether manually or automatically increase conatainers( Not available in DOcker.)
Prb 4: (enterprise level) - Docker doesn't support enterprise level of standards.

Kubernetess:
- Sol 1: Kubernetess is a cluster, Group of nodes (master/ node architecture).
- Sol 2: kubernetess controls and fix the damage by replacing failing pods before they go down. when One pod/container goes down. Using autohealing , even before pod goes down, whenever apiserver receives a signal container is going down it will rollout a new pod is created.
- Sol 3: Kubernetess solves network traffic by increasing replica set in yaml (1 to 10) or HPA(Horizontal Pod Auto scaling) control when load reached threshold.
- Sol 4: load balancing support, firewall support by CNCF . Lots of enhancement , developing integration support for different tools, load balancer, Prometheus monitoring tools

k8s Architecture

in Docker - container runtime -> Docker shim responsible for container running.
In kubernetess

Worker node: (Data plane) - actually executing the actions

Container runtime actually run container- supports multi container runtime -> Docker shim, container-d, crio-o.
Kubelet - responsible for creation of pods. Ensures POD is always in the running state. If anything goes wrong will inform one of the components called API server in control panel of master.

kube proxy: responsible for generating IP adresses, networing (uses IP table for networking configuration) and default load balanicing capability..

Master node : (Control plane) - actually controlling the actions

API server - responsible for exposing the kubernetess to the outer world. core component (Basically takes requests from external world). (Who decides the information)
scheduler - responsible for scheduling the pods or kubernetess resources. (who acts on the information)..
etcd - key value store , contains backup of cluster related info's
controller manager - ensures controllers like replica set are running cloud
controller manager - like terraform

Difference between Minikube and Kubectl

kubectl: command line tool for kubernetess, that we can directly interact with kubernetess clusters.
minikube: command line tool used for creating kubernetess cluster

Below are Development environments

Minikube
k3s
kind
k3d
Micro k8s

have you used kubenetess distribution in production. If you have used, any kubernetess distribution. did you manage installation and upgrade?

kubernetess: opensource container orchestration platform. But there are certain distribution provides support and build for user experience to its customer. such as Openshift, rancher, tanzu, EKS, AKS, GKE. These are used in production.

Difference between Kubernetess and EKS

If On top of EC2 install Kubernetess and use it. if there is any misconfiguration or we will not get any support.
On the other hand if we use distributed supports providing like EKS, management and scalability handled by platform like AWS.
Distribution provides additional wrapper, plugin and command line tools. But at the end of day solution is same.

KOPS (Kubernetess Operations)

KOPS is one of the most widely used tools for installing kubernetess.
lifecycle of kubernetess is managed by KOPS.
lifecycle: Installation, upgradation, configuration and deletion are part of lifecycle of Kubernetes cluster.

KOPS installation for production use

Dependencies for installing kubernetess Python3 AWS CLI kubectl
Install KOPS..

  chmod +x kops-linux-amd64
  sudo mv kops-linux-amd64 /usr/local/bin/kops

Set up AWS CLI configuration on your EC2 Instance or Laptop. Run aws configure
Kubernetes Cluster Installation
1. Create S3 bucket for storing the KOPS objects.
  aws s3api create-bucket --bucket kops-thi-storage --region us-east-1
2. create the cluster kops create cluster --name=demok8scluster.k8s.local --state=s3://kops-thi-storage --zones=us-east-1a --node-count=1 --node-size=t2.micro --master-size=t2.micro --master-volume-size=8 --node-volume-size=8
Build the cluster kops update cluster demok8scluster.k8s.local --yes --state=s3://kops-thi-storage

Difference between KOPS and Minikube

Both KOPS and minikube is a tool that used for kubernetess cluster management. But used in different cases.
KOPS: is a tool designed for deploying across multiple nodes and managing production gradde kubernetess clusters.
Minikube is a tool designed for single node kubernetess cluster on your local machine.

kubernetess Pods

pod is a wrapper of typically containing one or more containers that share resources like networking and storage.
In kubernetess instead of container we deploy pod Advantages of pod:
- if there is 2 or multiple containers, pod allows you to shared network, shared storage access the container using cluster IP address generated by kube proxy

Deploy your first APP in kubernetess

prerequisites

install kubectl
install minikube

kubectl installation

curl.exe -k -LO "https://dl.k8s.io/release/v1.32.0/bin/windows/amd64/kubectl.exe"
kubectl version

minikube installation

open powershell as a admin

New-Item -Path 'c:\' -Name 'minikube' -ItemType Directory -Force
Invoke-WebRequest -OutFile 'c:\minikube\minikube.exe' -Uri 'https://github.com/kubernetes/minikube/releases/latest/download/minikube-windows-amd64.exe' -UseBasicParsing
minikube

$oldPath = [Environment]::GetEnvironmentVariable('Path', [EnvironmentVariableTarget]::Machine)
if ($oldPath.Split(';') -inotcontains 'C:\minikube'){
  [Environment]::SetEnvironmentVariable('Path', $('{0};C:\minikube' -f $oldPath), [EnvironmentVariableTarget]::Machine)
}
(start cluster)
minikube start --memory=4096 --driver=hyperkit

kubectl get nodes

creating pods

Check available nodes: (single-host minikube is already present)
kubectl get nodes
creat pod.yml file

i
:wq

apiVersion: v1
kind: Pod
metadata:
  name: nginx
spec:
  containers:
  - name: nginx
    image: nginx:1.14.2
    ports:
    - containerPort: 80

Deploy the pod: kubectl create -f pod.yml
Check the pod status:

  kubectl get pods -o wide
(copy IP address where application is exposed)

Access the Application login to kubernetess cluster(minikube)
```
 minikube ssh
 curl <ip address>
 #application is running
```

For reference: https://kubernetes.io/pt-br/docs/reference/kubectl/cheatsheet/

Kubernetees - Deploy the first Application