19 days ago
Builds via CLI have failed repeatedly since yesterday before any Dockerfile step starts. Logs only show “scheduling build on Metal builder builder-wmnzmk”. Redeploying produces the same infrastructure error. Project: bed0e3b3-fa4b-4bc2-a7fb-dcabca09cd9b, service: f5569fee-b396-4df8-bdba-044377cc269b, environment: f2f1c163-1dee-4bb0-9a3b-5dce942fa3ba.
1 Replies
Status changed to Awaiting Railway Response Railway • 19 days ago
19 days ago
Your builds are working again. The last one went through at 16:38 UTC and is live now.
The failures came down to where railway up was run from. Your service has Root Directory set to mdoctor-backend, so the builder unpacks your upload and looks for a mdoctor-backend folder inside it. Running railway up from inside mdoctor-backend means that folder isn't in the upload, so the build stopped before it ever read your Dockerfile. That's why you only saw the scheduling line and nothing after it. Redeploying reused the same upload, so it failed identically every time.
Run railway up from the repo root and you'll be fine. If you'd rather deploy from inside mdoctor-backend, clear the Root Directory setting on the service instead.
Status changed to Awaiting User Response Railway • 19 days ago
Status changed to Solved mykal • 19 days ago