Skip to content

Amplify CloudFormation Stack in UPDATE_ROLLBACK_IN_PROGRESS — Unable to Recover or Push Changes

0

Hi AWS Team,

I'm encountering a critical issue with my Amplify project. The root CloudFormation stack is stuck in the UPDATE_ROLLBACK_IN_PROGRESS state and I'm unable to recover or push updates using the Amplify CLI.

Details: Stack Name: amplify-myvillageprojectadmi-dev-04552

Region: us-east-1

Environment: dev

CLI Command Used: amplify push, amplify status, amplify env checkout dev

Error:

🛑 This AWS::CloudFormation::Stack resource is in a UPDATE_ROLLBACK_IN_PROGRESS state.

What I’ve Tried:

  • amplify pull --appId <appId> --envName dev
  • amplify env checkout dev
  • CloudFormation "Continue rollback" with skipping resources
  • Force push
  • Validated schema file (GraphQL schema compiles fine)
  • Checked AWS credentials and region configuration
  • Internet/VPN/network is stable
  • Used aws cloudformation describe-stacks to confirm the stack status

Background: There is changes made to the schema recently, and after running amplify push, it failed midway. We now cannot complete the rollback, and CLI commands related to this environment are all stuck.

❓What I Need Help With:

  • How can I recover the environment without deleting the stack or data?
  • How can I safely re-enable Amplify CLI functionality?
  • Is there a way to forcefully sync the local project with current CloudFormation resources

I would highly appreciate step-by-step guidance, as I am a beginner with CloudFormation internals.

1 Answer
0

When a CloudFormation stack is in the UPDATE_ROLLBACK_IN_PROGRESS state, it means CloudFormation is actively trying to roll back changes from a failed update. For your Amplify project's stack, here's how you can address this issue:

First, you need to wait for the rollback to complete naturally. CloudFormation should automatically transition from UPDATE_ROLLBACK_IN_PROGRESS to UPDATE_ROLLBACK_COMPLETE once the rollback finishes. This can take some time depending on the resources involved.

If the stack gets stuck in UPDATE_ROLLBACK_FAILED state instead of completing the rollback, you can use the continue-update-rollback command:

aws cloudformation continue-update-rollback --stack-name amplify-myvillageprojectadmi-dev-04552

This command helps when CloudFormation can't roll back all changes after a failed stack update, often because it's trying to roll back to resources that no longer exist.

If specific resources are causing the rollback to fail, you can skip them using the --resources-to-skip parameter with the continue-update-rollback command. However, be aware that skipping resources can result in inconsistencies between your stack and actual resources, which may cause future updates to fail.

Once your stack reaches the UPDATE_ROLLBACK_COMPLETE state, you should be able to use Amplify CLI commands again. At this point, you can:

  1. Run amplify pull to sync your local environment with the cloud resources
  2. Make your schema changes incrementally rather than all at once
  3. Test each change with amplify push

If you continue to face issues after the stack reaches UPDATE_ROLLBACK_COMPLETE, you might need to:

  1. Check CloudWatch logs for specific error details
  2. Verify IAM permissions for the roles used by Amplify
  3. Consider reaching out to AWS Support if the problem persists

Remember that making smaller, incremental changes to your GraphQL schema can help avoid these types of failures in the future.
Sources
ContinueUpdateRollback - AWS CloudFormation
continue-update-rollback — AWS CLI 2.27.41 Command Reference
Troubleshooting CloudFormation - AWS CloudFormation

answered a year ago

  • The above mentioned information is correct.

  • I tried already above but not worked any of above solution

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.