Day1EKS Training #SelfReflection:

Duration: session 1(4th July,2020).

Things I learnt:

1. Theoretical Information about :

- Concept of Node (Docker Host or Container Host).

- How Kubernetes manage the pods on distributed nodes and prevent the fault tolerance.

- How different programs like APIServer, Controller, Scheduler etc. works in Kubernetes to serve the clients desire.

- EKS launch the Slaves nodes on different availability zone that's why they provide disaster recovery support.

- To do lots of customization on AWS EKS we use one independent program called "eksctl". Concept of Node Group.

- How eksctl internally using "cloud formation" to build the whole Kubernetes architecture on AWS cloud. Concept of Namespace in Kubernetes.

- How Elastic Load Balancer works to connect the pods with a public IP, so that clients can access the contents, this public IP connecting concept is known as "Public Facing".

- Difference between LoadBalancer service and NodePort service in K8s.

- In EKS if you expose pods using LoadBalancer service type then EKS automatically configure ELB.

- How EKS makes our sensitive data inside pods permanent by adding EBS volumes using PVC.

2. Practical's done:

- Creating AWS IAM with administrator power. 

- Next Installing and setting up eksctl program. 

- Next creating one yml code and mentioning our desired node groups and how much nodes we need inside these groups. 

apiVersion: eksctl.io/v1alpha5
kind: ClusterConfig


metadata:
  name: scluster
  region: ap-south-1


nodeGroups:
   - name: ng1
     desiredCapacity: 2
     instanceType: t2.micro
   - name: ng2
     desiredCapacity: 1

     instanceType: t2.small

- Next running this yml file and creating the EKS cluster. 

- Next we Configured the kubectl program to run Kubernetes commands from our local machine to the master node of EKS. 

- Next creating one namespace and making it as default. 

- Next launching webserver pods and exposing it. 

- Copying our local machine php codes inside one pod. 

- Describing gp2 storage class and getting informations like provisioner is EBS, Format Type is ext4, volumeBindingMode etc. 

- Next creating one yml code for pvc and running that code. 

apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: lwpvc1
spec:
  storageClassName: lwsc1
  accessModes:
    - ReadWriteOnce
  resources:
    requests:

      storage: 10Gi

- Editing the deployment on the fly and adding volume to the pods. 

- Editing the gp2 and disabling the default sc option.

apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
   name: lwsc1
provisioner: kubernetes.io/aws-ebs
parameters: 
  type: io1

reclaimPolicy: Retain

-Deleting all the pods,service,deployment,replicaset etc.

-Deleting the cluster.

Thank you watching :)

Please let me know if this was useful !