Skip to content
This repository has been archived by the owner on Dec 5, 2023. It is now read-only.

ProcessPayload.__call__ doesn't fail if S3 upload fails #36

Open
jkeifer opened this issue Jan 11, 2022 · 1 comment
Open

ProcessPayload.__call__ doesn't fail if S3 upload fails #36

jkeifer opened this issue Jan 11, 2022 · 1 comment
Labels
bug Something isn't working

Comments

@jkeifer
Copy link
Member

jkeifer commented Jan 11, 2022

I saw an issue where the cirrus-geo process lambda was not given appropriate S3 permissions to allow uploading the payload to the S3 payloads bucket. However, the call to ProcessPayload.__call__ did not fail despite the S3 error returned, and the payload was ultimately dispatched to the specified workflow.

That __call__ needs to be audited to ensure it and all the functions it calls are not swallowing exceptions, and that upon hitting an exception it has the expected behavior of not processing that payload.

@jkeifer jkeifer added the bug Something isn't working label Jan 11, 2022
@jkeifer
Copy link
Member Author

jkeifer commented Jan 11, 2022

Here is the error I saw in the process logs: An error occurred (AccessDenied) when calling the PutObject operation: Access Denied.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
bug Something isn't working
Projects
None yet
Development

No branches or pull requests

1 participant