- Newest
- Most votes
- Most comments
Don't know why, but after updating the local CDK runtime to 2.31.0, without changing our library reference version (2.28.1), the cdk deploy now succeeds for the notification rule and the other resources.
UPDATE (WORK AROUND)
Following further comment as per below, to work around this problem, either have to define the CodePipeline constructs in another stack or comment it out first, so that dependency resources can initially all succeed. Then let the CodePipeline deployment fail small and quick on first attempt, and deploy again which will be more likely to succeed.
This seems to happen too often when deploying new stacks
App-stack | 159/212 | 4:52:22 pm | CREATE_FAILED | AWS::CodeStarNotifications::NotificationRule | App-stack/App-stack-pipeline/App-stack-pl/NotifyOnExecutionStateChange (AppstackpipelineNotifyOnExecutionStateChange5933C726) Resource handler returned message: "Invalid request provided: AWS::CodeStarNotifications::NotificationRule" (RequestToken: GUID, HandlerErrorCode: InvalidRequest)
It has a very nasty side effect of breaking and stalling the deployment-destruction of resources, just because of this random error.
Relevant content
- Accepted Answerasked 2 years ago
- AWS OFFICIALUpdated a year ago
- AWS OFFICIALUpdated a year ago
- AWS OFFICIALUpdated a month ago