Custom resource deployment fails in v6.0.0

[bchunduri]

Hi Team,

We are deploying the new version of AWS QnABot v6.0.0 using the Cloudformation stacks and it failed to create all the custom resources with error statement as shown in below example:

CloudFormation did not receive a response from your Custom Resource. Please check your logs for requestId [xxxxxxx-xxxxxx-xxxxxx-xxxx]. If you are using the Python cfn-response module, you may need to update your Lambda function code so that CloudFormation can attach the updated version.

There were no cloudwatch events recorded against any of the requestId's from the failed resources and the triggered lambda logs throwing this error:

"errorType": "Runtime.ImportModuleError",
"errorMessage": "Error: Cannot find module 'index'\nRequire stack:\n- /var/runtime/index.mjs",

Please investigate, review and assist us to unblock this problem. Appreciate your swift action.

[abhirpat]

Hi @bchunduri, thank you for reporting this issue.

Please try following:

• Could you please try clicking on stack that has failed > Events > Detect RootCause and see if it points you to the actual error? Also, please try selecting lambda log groups and run cloudwatch query mentioned in the section Use Log Insights to query logs from CloudWatch groups.
• Is your stack name longer than 26 character ? If so, please try using using latest version which has the fix (see CHANGELOG).

Additionally:

• Are you using main or VPC template or extended template (deployed using Github repo and npm commands)?
• Which parameters are you customizing in your Cloudformation stack? Could you please provide steps to reproduce ?

Please let us know if any questions.

Thanks,
Abhishek

[bchunduri]

Hi @abhirpat, thanks for the quick suggestions and let me review your points:

1. Unfortunately log insights query could not display any events of the failed resource. And by debugging and backtracking, it is assume to throw errors onto the cfn-proxy log group.
2. The root stack is exactly 26 characters and other nested stacks go on a higher count and if that is the case, I need to either adapt that little change fixing this bug in v6.0.1
3. I am using the vpc template and passed all relevant custom parameters to the main and all nested stacks created by the root stack

Thanks,

Babu

[abhirpat]

@bchunduri Thank you. Could you please try using latest template v6.0.2 and update us if you have this issue?

[michaelin-96]

@bchunduri, I deployed v6.0.0 VPC template with 26 characters in the stack name and wasn't able to replicate the issue. I was able to deploy successfully.

[bchunduri]

Hi @michaelin-96 , thanks and have you tried stack names with more than 26 characters ?

[bchunduri]

@abhirpat , wonder to see another minor version released in no time ! This leave us in dilemma to hold over to a stable release :(

[michaelin-96]

@bchunduri, if you create a stack name with over 26 characters for v6.0.0 then it will fail. See #741. This is addressed v6.0.1 and onwards.

[DeccanAjay]

@bchunduri - thank you for bringing this to our attention. Please use the latest version to update your stack. We will look into this more and be in touch shortly.

[fhoueto-amz]

Hi @bchunduri
I am closing this. Please feel free to reopen or open a new issue, if you have more details to help reproduce.