We're sorry for the trouble. An active incident, "Dashboard and API unavailable, builds paused", is affecting the dashboard, API, CLI and builds in all regions, which is why restarts and rollbacks are blocked for you right now. You can follow it at the incident page.

On your data, your Postgres service still shows its last successful deployment with its volume attached, and the API service in Production has no volume of its own, so stopping it did not touch any stored data.

Thanks for the update. We still need help from a support engineer to restore our existing Railway production API.

Update: Production and Test are accessible again through a temporary external workaround, but service is still degraded: background jobs and larger uploads are not fully restored.

One clarification: our active production database and file storage are hosted on Supabase. They are unchanged, and we have made a fresh encrypted backup and successfully tested a database restore in an isolated environment. No production schema migrations were performed during this recovery.

The previous working Railway production deployment is now marked REMOVED. Earlier restart and rollback attempts were rejected with "Deploys have been paused temporarily".

Since the latest incident update says the API is back, can an engineer confirm whether the previous production deployment can now be restored from its existing image without a new build? If not, is there a manual recovery option, and when do you expect restoration of existing deployments to be possible?

Please preserve the existing production configuration and data. We will keep the current workaround serving traffic until the Railway service is verified working. Please keep this issue open, as full service has not yet been restored.

Thank you.

The "Deploys have been paused temporarily" message you saw is part of the ongoing incident, "Dashboard and API unavailable, builds paused", and its latest update says builds are still paused even though the API is back. You can follow it at the incident page.

A REMOVED deployment can still be brought back with Rollback from that deployment's menu in your service's Deployments tab. Rollback restores the original Docker image and that deployment's variables rather than rebuilding your code, as long as the deployment is within your plan's image retention period. While the pause is in place, rollback and redeploy requests are blocked the same way your earlier attempts were, and we have no time yet for when that will lift.

A rollback deploys your existing image and configuration and does not touch your Supabase database or file storage.