• Now on to automating the flow using GitHub Action CI(How to replicate this at your end):
• Step1: Setup the following secrets for secret injection into pipeline in a secure way.
• 1. DOCKER HUB USERNAME - to push image to docker hub
• 2. DOCKER HUB PASSWORD- to push image to docker hub
• 3. AWS_ACCESS_KEY_ID- for terraform to provision EKS CLUSTER
• 4. AWS_SECRET_ACCESS_KEY- for terraform to provision EKS CLUSTER
• 5. STATE_BUCKET_NAME - where TF will store its state file(S3 bucket was setup from portal)
• 6. STATE_PATH - where TF will store its state file(S3 bucket was setup from portal)

• Step2: Setup bucket with proper permission to store state file

• Step3: All Github action pipeline YML are in

• Step4: Docker Image build pipeline

• Trigger whenever Dockerfile in the repo is update.Comment explain what each step does.

• Step5: Terraform INIT PLAN APPLY:

• Run the TF Workfflow. Runs whenever the folder hosting the tf code i.e. "terraform" folder in the repo is updated.

• Step6: If you want to destroy the complete infra run the TF Destroy(if needed). Run on manual trigger.

• Capture of Pipeline runs:

• Using new sample application, Pods are in ready state:

• Live and Ready probes are working

• Deployed with TF:

• Unable to start the webserver to the the server bind. Will use better sample code to fix this.

• FIX: Used code from this repo: https://github.com/jakshaym1234/rubysampleadjust

  • Binded the application to port 8000 as the non root user cant bind to port 80
  • Added liveliness and readness port also to watch for /index

• SUCCESS AT LAST. APPLICATION IS AVAILABLE USING THE ALB PROVISIONED BY THE HELM CHART:

• 

• 

• Write to me at kmjakshay@gmail.com if something is not working for you. Thanks