Inconsistent deploy success even with successful builds

[AlexanderChiclana]

Environment information

I can provide other details if needed, but having issues running command 

Using custom image: public.ecr.aws/docker/library/node:18.17.0

Description

I have 30+ production branches on Amplify that have identical codebases. They are Gatsby JS sites which are fairly large and can take up to an hour each to build. They are connected to various webhooks to trigger rebuilds.

Since the Gen 2 dashboard has gone live we have been experiencing inconsistent deployments across these branches. Typically 4 or 5 out of the 30 branches will fail to deploy after the webhook is triggered, and manually rebuilding them off the same codebase will work inconsistently as well.

When I review the logs of the failed branches, there is nothing out of the ordinary there and no fatal error messages. The builds which fail to deploy end like this in their logs, which is identical to how the successful build logs finish

Completed phase: postBuild Completed Frontend Build Build completed successfully Starting caching... Creating cache artifact... Created cache artifact Uploading cache artifact... Uploaded cache artifact Caching completed Starting build artifact upload process... Uploading build artifact '__artifacts.zip'... Build artifact upload completed Starting environment caching... Uploading environment cache artifact... Uploaded environment cache artifact Environment caching completed