Production deploy fails after Metal builder image push, before runtime
hassant147
FREEOP

15 days ago

Please investigate an existing Railway production service whose deployment reaches the assigned Metal builder and completes the Docker image build/push, but then fails before any deploy/runtime or HTTP log is emitted. Please inspect the builder-side failure and advise whether the deployment can be resumed or assigned to a healthy builder. Do not create or duplicate a service.

Exact scope

  • Workspace: ziyarat2026's Projects
  • Plan: active Hobby usage-based subscription; Agent Usage Limit is a separate exhausted quota and is not the requested diagnosis
  • Project: magnificent-compassion
  • Project ID: b181d298-e13f-424c-9d03-1fe577b22096
  • Environment: production
  • Environment ID: b95ee5c9-5760-47ca-999e-0573555b96bb
  • Service: ziyarat-api
  • Service ID: 83396a37-31a5-4b33-9478-b651455da046
  • Source: Hassant147/Ziyarat, branch main, root /Ziyarat-Backend
  • Diagnostic source commit: 31749f098b4e477f8787ebab8ed0e780943105dd
  • Assigned builder: builder-tvnmfy

Controlled reproduction

Only NO_CACHE=1 was added to the existing API service. The write used redeploy suppression so it did not start an extra deployment. Values of all variables were withheld; no secret or database value is included here.

Exactly one controlled redeployment was then triggered:

  • Deployment: ff8254bc-94f3-4147-86f0-d98c5e039b69
  • Commit: 31749f098b4e477f8787ebab8ed0e780943105dd
  • Result: FAILED
  • Terminal status timestamp: 2026-09-20T15:48:05.604Z

Observed output

  • Build stream: 286 entries. It recorded the Metal builder assignment, dependency installation, successful mysqlclient wheel build, Dockerfile copy steps, image export, and an image-push completion event.
  • Deploy stream: 0 entries.
  • HTTP stream: 0 entries.
  • No application container start, migration, readiness, or HTTP response was observed.
  • Earlier diagnostic deployments d8d3e26e-461a-44a6-92be-ebd18b636d0f and bf8a6cf1-f08b-467b-8021-8e9625745938 exposed only the builder scheduling line and also had zero deploy/HTTP entries.
  • Earlier normal deployment 5a8b76a0-a0aa-42df-b862-26b01e03740b failed at the old pre-deploy stage.

Billing context

The Hobby subscription is active and billing was visually verified. The project cost is approximately $0.0579; the separate Railway Agent Usage Limit is exhausted. No Hobby plan upgrade or Agent-limit increase should be treated as the deployment fix.

Safety boundary

No duplicate service, deletion, force push, DNS mutation, production-data operation, or secret disclosure occurred. Please return the builder-side error and a provider-supported next step without exposing variable values.

Solved

1 Replies

Railway
BOT

15 days ago

The project and workspace you referenced ("magnificent-compassion" in "ziyarat2026's Projects") cannot be reached from the account this thread is tied to. Your account here resolves to a different workspace with no active plan or subscription, which is why deployments on it would not proceed, but the resources you described sit under a separate account entirely. To investigate or redeploy that service, the request needs to come from the account that owns that workspace. You can open a thread from there if the deployment issue persists once you are in the right account.


Status changed to Awaiting User Response Railway • 15 days ago


Railway
BOT

8 days ago

This thread has been marked as solved automatically due to a lack of recent activity. Please re-open this thread or create a new one if you require further assistance. Thank you!

Status changed to Solved Railway • 8 days ago


Welcome!

Sign in to your Railway account to join the conversation.

Loading...