AWS CodePipeline (CI/CD Pipeline)

CICD Pipeline with CodeCommit CodeBuild and CodeDeploy

CodePipeline Limitations

In this guide, we’ll walk through the key limitations, quotas, and naming rules for AWS CodePipeline. Understanding these restrictions is critical for designing scalable, reliable CI/CD workflows.

Unique Naming Constraints

Every pipeline, stage, and action name must be unique within its scope:

  • Pipeline names must be unique per AWS account and region.
  • Stage names must be unique within a pipeline.
  • Action names must be unique within their stage.

You can reuse common stage names (for example, Source or Deploy) across different pipelines, but you cannot duplicate them inside the same pipeline.

The image outlines naming restrictions for pipelines, stages, and actions, alongside a CI/CD infinity loop diagram with stages labeled "Source," "Deploy," and "Test."

Note

Choose descriptive stage and action names to improve pipeline readability and troubleshooting.

Regional Quotas and Requesting Increases

Most CodePipeline quotas are applied per AWS Region. For example, the default limit of 1,000 pipelines refers to pipelines per region.

The image shows a world map with location markers and text indicating that some quotas apply per region, with instructions to request a quota increase through the Support Center console.

Warning

If you hit a regional quota, you can request an increase via the AWS Support Center. Approval may take up to two weeks.

Core Service Limits

Below are the default CodePipeline quotas you should plan for in each region:

ResourceDefault LimitScope
Pipelines1,000Per account per region
Stages per pipeline2–50Per pipeline
Actions per stage1–50Per stage

Actions within a stage can execute in parallel or sequentially. The only overall cap is the per-stage limit.

The image displays AWS pipeline limits, including the maximum number of pipelines, stages, and actions per account, stage, and pipeline.

Additional Considerations

  • Manual approval actions will time out after 7 days without a decision.
  • Artifacts are typically stored in Amazon S3. While S3 supports virtually unlimited objects, general S3 limits still apply.

For the full list of quotas and best practices, see the AWS CodePipeline Limits documentation.

Watch Video

Watch video content

Previous
CodeDeploy Demo