AWS CodeDeploy

CICD are essential practices in software development. In continuous integration the system builds the code with every commit to the repository and stores the artifacts into a storage space, further making them available for testing and deployment processes. On the other hand continuous deployment automates the deployment process enabling swift and consistent releases. This entire process is important as it decreases the manual error and makes product deployment more efficient and reliable.

AWS CodePipeline simplifies this entire continuous integration and continuous deployment process by automating the build, test, and deployment phases of the software development process. Here in this article, I will walk through the steps of how to integrate CodeBuild AWS CodeBuild, and AWS CodeDeploy with AWS CodePipeline to automate the entire code commit, code build, and code deploy stages in a continuous integration and continuous deployment process.

 

Steps To Use AWS CodePipeline For Continuous Integration And Deployment

Step 1: First create a CodeCommit repository.

Step 2 : On the top right corner of page , click Clone URL to copy the HTTPS URL 

Step 3 : Now go to a local editor terminal (for e.g. visual studio) and write the git clone using the copied URL . But here you will see a new screen will pop out and it will be asking GIT credential .

git clone <copied URL>

Step 4 : Now to generate GIT credential , go to the IAM service . Here go to the IAM user account that you are using using now . There you will find Security Credential, select that and scroll down to HTTPS GIT credentials for AWS CodeCommit . Here click Generate credentials.

Step 5 : Now we have the GIT credentials . So we can now git clone using the HTTPS URL of CodeCommit repository (the same command used in step 3) . You can now successfully clone the repository after entering the GIT credentials .

Step 6 : Now go inside the repository and create a index.html . Basically here all the changes are in local environment , so to make sync the changes with the CodeCommit repository you need to push all changes to the master branch of CodeCommit repository.

git add .
git commit -m "first commit in codecommit"  (you can give any message inside the inverted commas)
git push origin master

Step 7 : Now we have all these code present inside the CodeCommit repository . We need to build these code now . Before using AWS CodeBuild , we need a buildspec.yml file inside the repository . This buildspec.yml defines the set of build commands and settings which is necessary for providing a structured way to specify how CodeBuild should build and package your application . Create a buildspec.yml file and push it CodeCommit repository .

  version : 0.2

  phases:
    install:
      commands:
        - echo Start with NGINX install
        - sudo apt-get update
        - sudo apt-get install nginx -y
    build:
      commands:
        - echo Building...
        - cp index.html /var/www/html/
  artifacts:
    files:
      - /var/www/html/index.html

Step 8 : Now go to AWS CodeBuild and create a build . Here provide the project name . Add CodeCommit as code source and here also mention the branch you are using . Select Ubuntu as operating system , standard runtime and use the latest image . Here automatically a new service role will be created and attached to the build . Mention the build file ,but if you have already selected buildspec.yml then then there is no need for mentioning in the Buildspec name . Select no artifacts . Here you can untick the Cludwatch logs and create the build .

Video Player

00:00

00:19

Step 9 : After creating code build you can start the build to see whether your build is successful or not . This build is creating some artifacts which needs to be stored . So select edit and then select artifacts .

Step 10 : Here I have used Amazon S3 bucket to store the artifacts . Mention the s3 bucket previously created here and in names write artifacts.zip (<any-name>.zip)

Step 11 : Then start the build . You can check the phase details status .

Step 12 : After build status is successful , now we have to deploy our website . Create an application on AWS CodeDeploy . Here mention the application name and select EC2/On-premises as compute platform.

Step 13 : In the previous step we have taken EC2 as compute platform . So now we have to create an EC2 instance. This EC2 instance needs is used as deployment platform which is interacting with AWS S3 bucket , AWS CodeDeploy service . So after creating EC2 instance you have to create a IAM role and attach the role to the EC2 instance with some policies AmazonEC2FullAccess , AmazonS3FullAccess and AWSCodeDeployFullAccess .

Step 14 : Create another IAM role for deployment group with the policies AmazonEC2FullAccess , AmazonS3FullAccess , AWSCodeDeployFullAccess , AmazonEC2RoleforAWSCodeDeploy , AWSCodeDeployRole and AmazonEC2RoleforAWSCodeDeployLimited . This role will attach to the deployment group that we will create in the next step .

Step 15 : Now go to AWS CodeDeploy and create a deployment group . Here mention deployment group name , attach the IAM role you have created in the previous step , select EC2 for environment configuration and attach the EC2 instance created in step 13 . And here we will install codedeploy agent manually and so select never in install codedeploy agent .

Video Player

00:00

00:16

Step 16 : Now we have to install the codedeploy agent on EC2 instance . You can install the codedeploy by following the official documentation of AWS or you can directly install codedeploy agent using the following script (make sure that you are using us-east-1 or else you have change the region name in the script) :

sudo apt-get update 
sudo apt-get install ruby-full ruby-webrick wget -y 
cd /tmp 
wget https://aws-codedeploy-us-east-1.s3.us-east-1.amazonaws.com/releases/codedeploy-agent_1.3.2-1902_all.deb 
mkdir codedeploy-agent_1.3.2-1902_ubuntu22 
dpkg-deb -R codedeploy-agent_1.3.2-1902_all.deb codedeploy-agent_1.3.2-1902_ubuntu22 
sed 's/Depends:.*/Depends:ruby3.0/' -i ./codedeploy-agent_1.3.2-1902_ubuntu22/DEBIAN/control 
dpkg-deb -b codedeploy-agent_1.3.2-1902_ubuntu22/ 
sudo dpkg -i codedeploy-agent_1.3.2-1902_ubuntu22.deb 
systemctl list-units --type=service | grep codedeploy 
sudo service codedeploy-agent status

Step 17 : Like CodeBuild stage , CodeDeploy stage also needs a configuration file called appspec.yml which provides instructions to CodeDeploy on how to deploy your application . Create appspec.yml in the local environment .

version: 0.0
os: linux
files:
  - source: /
    destination: /var/www/html
hooks:
  AfterInstall:
    - location: scripts/install_nginx_service.sh
      timeout: 300
      runas: root
  ApplicationStart:
    - location: scripts/start_nginx_service.sh
      timeout: 300
      runas: root

Here there are two scripts used , one script install the nginx service and another script to start the nginx service .

#!/bin/bash

sudo apt-get update
sudo apt-get install -y nginx

start_nginx_service.sh

#!/bin/bash

sudo service nginx start

After creating all these files you have to push all the changes to AWS CodeCommit repository .

Step 18 : Now you need to update the artifacts files in the buildspec.yml and then push the changes to AWS CodeCommit repository .

Step 19 : Now here we have to create a deployment using revise location as S3 Bucket's URL . Previously we have used artifacts.zip to store the artifacts in S3 bucket , so mention the revision file type as .zip .

Step 20 : After this check the CodeCommit repository if all the files are present or not . Now again start the codebuild and then create the deployment .

Step 21 : After the successful deployment then access the Public IP of EC2 instance to access the webpage .

Step 22: Till now all the above steps are done manually . To automate the CodeCommit , CodeBuild and CodeDeploy stage we need to use AWS CodePipeline . So create an AWS CodePipeline .

Step 23 : Select code source as AWS CodeCommit and mention the repository name and branch also . Here choose AWS CodePipeline in Change detection options (After selecting this it will automatically detect changes in the repository and start the code build, code deploy stage).

Step 24 : Select AWS CodeBuild as build provider and enter the project name .

Step 25 : Here also mention the AWS CodeDeploy as deploy provider and mention the application name and deployment group name .

Step 26 : After this the pipeline will be executed successfully .

Step 27 : Now access the webpage using the Public IP of EC2 instance .

Step 28 : To check whether the AWS CodePipeline working properly or not , make some changes in the index.html and push the changes to the AWS CodeCommit repository .

Step 29 : After changes are pushed to AWS CodeCommit repository , you will notice that pipeline will automatically start and can see the recent commit message also .

Step 30 : Finally access the updated webpage using the Public IP of EC2 instance .

Conclusion

So you have now learned why continuous integration and continuous deployment is important . Also we have gone through the various steps to automate the different stages like code commit , code build and code deploy using AWS CodePipeline . Finally we successfully accessed the website using the Public IP of EC2 instance .

 

FAQS On How to use AWS CodePipeline for continuous integration and deployment

 

1. What are different source code repositories that can integrate with AWS CodePipeline ?

CodePipeline integrates with different source code repositories such as AWS CodeCommit , GitHub, Bitbucket ,S3 Bucket , etc .

 

2. How execution of pipeline triggered in AWS CodePipeline ?

Pipeline executions can be triggered manually , based on a schedule or automatically in response to any change in the source code repository .

 

3. What are artifacts ?

Artifacts are deployable packages or build outputs that are produced during the pipeline action .

 

4. How can you troubleshoot any type of issues in AWS CodePipeline ?

Cloud monitoring tools such as AWS cloudWatch logs and CloudTrail can be used to find any type of issues in the AWS CodePipeline .

 

5. Can AWS CodePipeline integrate with third party tools ?

The answer is yes . For e.g AWS CodePipeline can integrate with Jenkins to build the code or it can also integrate with different source code repositories like Github , Bitbucket , etc .