a month ago
Since ~08:30 UTC today (2026-07-16), every build in project "creative-nurturing" (fc243087-3f17-4f47-bcd4-502930b83944), staging environment, fails immediately after 'scheduling build on Metal builder "builder-bdnvyr"' with no further build log output at all - the build never starts. 10+ attempts over ~2 hours across three services (api, worker, frontend), every one pinned to builder-bdnvyr, all failing within seconds.
Example failed deployment IDs: a18bcd76-2ae2-40cf-a289-3229340da887, 6e990cf7-09f6-4032-93fc-4f6e83402d28, 064fef35-2ddd-4831-bf50-3e0aeb4f11b0.
These are Dockerfile-based deploys via "railway up" from each service's subdirectory - the identical flow built fine yesterday (2026-07-15) on builder-zdgkoi (deployment 33499a2d-2878-43f9-a47a-91af8562b662). Yesterday one build also died at "scheduling build" and succeeded on retry 8 minutes later; today the failure is persistent. The same commits build fine in GitHub CI, so it isn't the Dockerfiles.
Could you check/recycle builder-bdnvyr or move our project's builds to a healthy builder?
Note: the project belongs to workspace "AI DRO's Projects" - that's why the Service selector shows N/A from this account. The project/deployment IDs above are exact.
2 Replies
a month ago
This thread has been opened as a bounty so the community can help solve it.
Status changed to Open Railway • about 1 month ago
a month ago
Can you try redeploy latest commit or disable the build cache?
To redeploy latest commit, open your service > open the command palette (Ctrl-K/Cmd-K) > select Redeploy source image
To disable build cache, add NO_CACHE=1 to your environment variable
a month ago
Thanks — tried both suggestions today (2026-07-17): set NO_CACHE=1 on all three services and re-ran "railway up" fresh (these are CLI deploys of local source, so there's no source image to redeploy). Same instant failure, still pinned to builder-bdnvyr: deployment e6ebf7f1-4c34-4757-8424-c2b493324ac5 (20:03 +04) died seconds after 'scheduling build on Metal builder "builder-bdnvyr"' with zero further build output. That's now >30 hours of every staging build failing on that one builder. Could someone check/recycle builder-bdnvyr or repin the project to a healthy builder?