Hi, I’m seeing a Railway deploy issue that looks platform-side rather than app-specific.

Project:

• Swift backend using a Dockerfile
• region us-west1
• runtime V2

What happens:

• railway up --verbose uploads successfully
• builder starts normally
• for my real service, the build often gets all the way through swift build -c release --product bitecam-server
• for a separate minimal smoke-test Swift service, Railway also fails the same way
• deploy then ends with only Deploy failed
• railway status --json shows the latest deployment as FAILED or stuck as BUILDING with deploymentStopped: true
• railway logs often only shows stale logs from the older successful deployment, not the failing one

Important isolation step:

• I created a brand-new minimal Swift smoke service with no app logic, no database, no Cloudflare/R2 integration, just a tiny HTTP server in Swift
• it builds and runs locally
• Railway still fails it with the same generic deploy failure

So this does not appear to be specific to my app code.

Current evidence:

• smoke service deployment id: 8cdc4cd4-5ec2-49c3-aaf1-aee389815f51
• service name: swift-smoke
• main app service also showed similar behavior, with deployments getting stuck or failing after build

Questions:

1. Is there a known issue right now with Swift Docker deployments on Railway?
2. Is there a way to get the real failing deployment logs instead of stale logs from the prior successful deployment?
3. Is the deploy failing during build, image import, or runtime promotion/healthcheck?
4. Do you recommend any specific workaround for Swift container deploys right now?

If useful, I can also share the minimal Dockerfile and Package.swift for the smoke test.