Because it fails at "scheduling build" before Railpack prints anything, this is almost certainly not an issue with your Python Flask code, requirements.txt, or Procfile. That stage happens before Railway actually initiates the image build.

Since multiple deployments are dying there with zero build output, your service is likely stuck on a bad/unavailable builder node, or experiencing a temporary infrastructure glitch with the Metal build environment.

A few things you can try right now to bypass this:

1. Trigger a manual redeploy from the exact same commit. If Railway has cordoned off the bad builder node, a retry may land on a healthy one.
2. Go into your Service Settings and temporarily turn off the "Metal Build Environment" toggle if it's available, then redeploy.
3. If you use a railway.json file, explicitly force the Railpack provider so the build configuration isn't ambiguous.

Ultimately, a trying a fresh redeploy or toggling off Metal usually acts as a solid immediate workaround!