UPDATE_ROLLBACK_FAILED
state, continues rolling it back to the UPDATE_ROLLBACK_COMPLETE
state. Depending on the cause of the failure, you can manually fix the error and continue the rollback. By continuing the rollback, you can return your stack to a working state (the UPDATE_ROLLBACK_COMPLETE
state), and then try to update the stack again. A stack goes into the UPDATE_ROLLBACK_FAILED
state when AWS CloudFormation cannot roll back all changes after a failed stack update. For example, you might have a stack that is rolling back to an old database instance that was deleted outside of AWS CloudFormation. Because AWS CloudFormation doesn't know the database was deleted, it assumes that the database instance still exists and attempts to roll back to it, causing the update rollback to fail.