Amplify "Resource is not in the state stackUpdateComplete" in CI

amplify push waits for each nested CloudFormation stack to reach UPDATE_COMPLETE. One nested stack entered a failed or rollback state instead, so the waiter throws "Resource is not in the state stackUpdateComplete".

What this error means

amplify push fails with "Resource is not in the state stackUpdateComplete". The CloudFormation console shows a nested stack in UPDATE_ROLLBACK_COMPLETE with the real resource error.

amplify

Following resources failed
Resource is not in the state stackUpdateComplete
{ "resourceProvider": "awscloudformation",
  "category": "auth", "resourceName": "myappauth" }

Common causes

A nested stack resource failed to update

A category stack (auth, api, storage) hit an error - a name collision, a quota, or an immutable property change - and rolled back, so the parent never reaches stackUpdateComplete.

A previous push left a stack mid-update

An interrupted earlier deploy left the environment in UPDATE_IN_PROGRESS or a rollback, blocking the next push from succeeding.

How to fix it

Open the failed nested stack for the real error

In CloudFormation, find the nested stack for the named category.
Read its first failed event to get the underlying resource error.
Fix that resource (rename, raise quota, or revert an immutable change) and push again.

Recover a stuck environment

If a prior push left the stack mid-update, continue the rollback in CloudFormation, then re-run the push against a clean state.

Terminal

aws cloudformation continue-update-rollback \
  --stack-name amplify-myapp-dev-123456

How to prevent it

Let each push finish before starting another against the same env.
Test backend category changes in a sandbox env before main.
Avoid renaming resources that force immutable replacements mid-deploy.