2022-07-29 14:03:07 +02:00

10 KiB

Raw Blame History

stage	group	info	type
Release	Release	To determine the technical writer assigned to the Stage/Group associated with this page, see https://about.gitlab.com/handbook/engineering/ux/technical-writing/#assignments	howto

Deploy to AWS from GitLab CI/CD (FREE)

GitLab provides Docker images with the libraries and tools you need to deploy to AWS. You can reference these images in your CI/CD pipeline.

If you're using GitLab.com and deploying to the Amazon Elastic Container Service (ECS), read about deploying to ECS.

Authenticate GitLab with AWS

To use GitLab CI/CD to connect to AWS, you must authenticate. After you set up authentication, you can configure CI/CD to deploy.

Sign on to your AWS account.
Create an IAM user.
Select your user to access its details. Go to Security credentials > Create a new access key.
Note the Access key ID and Secret access key.

In your GitLab project, go to Settings > CI/CD. Set the following CI/CD variables:

Environment variable name	Value
`AWS_ACCESS_KEY_ID`	Your Access key ID.
`AWS_SECRET_ACCESS_KEY`	Your secret access key.
`AWS_DEFAULT_REGION`	Your region code. You might want to confirm that the AWS service you intend to use is available in the chosen region.

Variables are protected by default. To use GitLab CI/CD with branches or tags that are not protected, clear the Protect variable checkbox.

Use an image to run AWS commands

If an image contains the AWS Command Line Interface, you can reference the image in your project's .gitlab-ci.yml file. Then you can run aws commands in your CI/CD jobs.

For example:

deploy:
  stage: deploy
  image: registry.gitlab.com/gitlab-org/cloud-deploy/aws-base:latest
  script:
   - aws s3 ...
   - aws create-deployment ...

GitLab provides a Docker image that includes the AWS CLI:

Images are hosted in the GitLab Container Registry. The latest image is registry.gitlab.com/gitlab-org/cloud-deploy/aws-base:latest.
Images are stored in a GitLab repository.

Alternately, you can use an Amazon Elastic Container Registry (ECR) image. Learn how to push an image to your ECR repository.

You can also use an image from any third-party registry.

Deploy your application to ECS

Introduced in GitLab 12.9.

The Deploy-ECS.gitlab-ci.yml template was moved to AWS/Deploy-ECS.gitlab-ci.yml in GitLab 13.2.

You can automate deployments of your application to your Amazon ECS cluster.

Prerequisites:

Authenticate AWS with GitLab.
Create a cluster on Amazon ECS.
Create related components, like an ECS service, a database on Amazon RDS, and so on.
Create an ECS task definition, where the value for the containerDefinitions[].name attribute is the same as the Container name defined in your targeted ECS service. The task definition can be:
- An existing task definition in ECS.
- In GitLab 13.3 and later, a JSON file in your GitLab project. Use the template in the AWS documentation and save the file in your project. For example <project-root>/ci/aws/task-definition.json.

To deploy to your ECS cluster:

In your GitLab project, go to Settings > CI/CD. Set the following CI/CD variables. You can find these names by selecting the targeted cluster on your Amazon ECS dashboard.

Environment variable name	Value
`CI_AWS_ECS_CLUSTER`	The name of the AWS ECS cluster that you're targeting for your deployments.
`CI_AWS_ECS_SERVICE`	The name of the targeted service tied to your AWS ECS cluster.
`CI_AWS_ECS_TASK_DEFINITION`	If the task definition is in ECS, the name of the task definition tied to the service.
`CI_AWS_ECS_TASK_DEFINITION_FILE`	If the task definition is a JSON file in GitLab, the filename, including the path. For example, `ci/aws/my_task_definition.json`. If the name of the task definition in your JSON file is the same name as an existing task definition in ECS, then a new revision is created when CI/CD runs. Otherwise, a brand new task definition is created, starting at revision 1.

WARNING: If you define both CI_AWS_ECS_TASK_DEFINITION_FILE and CI_AWS_ECS_TASK_DEFINITION, CI_AWS_ECS_TASK_DEFINITION_FILE takes precedence.

Include this template in .gitlab-ci.yml:
```
include:
  - template: AWS/Deploy-ECS.gitlab-ci.yml
```
The AWS/Deploy-ECS template ships with GitLab and is available on GitLab.com.
Commit and push your updated .gitlab-ci.yml to your project's repository.

Your application Docker image is rebuilt and pushed to the GitLab Container Registry. If your image is located in a private registry, make sure your task definition is configured with a repositoryCredentials attribute.

The targeted task definition is updated with the location of the new Docker image, and a new revision is created in ECS as result.

Finally, your AWS ECS service is updated with the new revision of the task definition, making the cluster pull the newest version of your application.

NOTE: ECS deploy jobs wait for the rollout to complete before exiting. To disable this behavior, set CI_AWS_ECS_WAIT_FOR_ROLLOUT_COMPLETE_DISABLED to a non-empty value.

WARNING: The AWS/Deploy-ECS.gitlab-ci.yml template includes two templates: Jobs/Build.gitlab-ci.yml and Jobs/Deploy/ECS.gitlab-ci.yml. Do not include these templates on their own. Only include the AWS/Deploy-ECS.gitlab-ci.yml template. These other templates are designed to be used only with the main template. They may move or change unexpectedly. Also, the job names within these templates may change. Do not override these job names in your own pipeline, because the override stops working when the name changes.

Deploy your application to EC2

Introduced in GitLab 13.5.

GitLab provides a template, called AWS/CF-Provision-and-Deploy-EC2, to assist you in deploying to Amazon EC2.

When you configure related JSON objects and use the template, the pipeline:

Creates the stack: Your infrastructure is provisioned by using the AWS CloudFormation API.
Pushes to an S3 bucket: When your build runs, it creates an artifact. The artifact is pushed to an AWS S3 bucket.
Deploys to EC2: The content is deployed on an AWS EC2 instance.

Configure the template and JSON

To deploy to EC2, complete the following steps.

Create JSON for your stack. Use the AWS template.
Create JSON to push to S3. Include the following details.
```
{
  "applicationName": "string",
  "source": "string",
  "s3Location": "s3://your/bucket/project_built_file...]"
}
```
The source is the location where a build job built your application. The build is saved to artifacts:paths.
Create JSON to deploy to EC2. Use the AWS template.
Make the JSON objects accessible to your pipeline:
- If you want these JSON objects saved in your repository, save the objects as three separate files.
  
  In your .gitlab-ci.yml file, add CI/CD variables that point to the file paths relative to the project root. For example, if your JSON files are in a <project_root>/aws folder:
```
variables:
  CI_AWS_CF_CREATE_STACK_FILE: 'aws/cf_create_stack.json'
  CI_AWS_S3_PUSH_FILE: 'aws/s3_push.json'
  CI_AWS_EC2_DEPLOYMENT_FILE: 'aws/create_deployment.json'
```
- If you do not want these JSON objects saved in your repository, add each object as a separate file type CI/CD variable in the project settings. Use the same variable names as above.
In your .gitlab-ci.yml file, create a CI/CD variable for the name of the stack. For example:
```
variables:
  CI_AWS_CF_STACK_NAME: 'YourStackName'
```

In your .gitlab-ci.yml file, add the CI template:

include:
  - template: AWS/CF-Provision-and-Deploy-EC2.gitlab-ci.yml

Run the pipeline.
- Your AWS CloudFormation stack is created based on the content of your CI_AWS_CF_CREATE_STACK_FILE variable. If your stack already exists, this step is skipped, but the provision job it belongs to still runs.
- Your built application is pushed to your S3 bucket then and deployed to your EC2 instance, based on the related JSON object's content. The deployment job finishes when the deployment to EC2 is done or has failed.

10 KiB Raw Blame History