Same issue. They've killed multiple production builds today. Super frustrating.

We're currently experiencing degraded build infrastructure that is causing builds to queue and delay, including for Pro plan users. You can track the status and updates at https://status.railway.com/incident/ZLI8N955. If any builds remain stuck after the incident is resolved, cancelling and redeploying should pick them back up.

Can you clarify what you meant by knocking offline a productive build?

Same issue here... frustrating...

Are you saying you had a service knocked offline, or build delays?

build delays

during last "incident" which led to 4 hour postgres downtime i migrated my DB to neon and never regretted the decision. 3 weeks flying - zero downtime. The migration took me whole night but was well worth it. Now I am sitting here without possibility to fix critical bugs in production for almost 1 hour while my users are suffering and thinking like: Should i start migrating services out of railway one by one? Seriously guys, it's been really frustrating journey, for which I (and lots of other customers paid money)

We hear you, and the frustration is completely valid. Repeated incidents impacting production workloads for paying customers is not acceptable, and we understand the urgency when you can't ship critical fixes. The current build issue is still being actively investigated at https://status.railway.com/incident/ZLI8N955. Your existing deployments should remain running during this, only new builds are delayed. I understand that doesn't help if you have updates to make. We are working as fast as we can to fix and to put in place the systems to avoid this in the future.

Anyone has an idea how long this takes usually? My production build was removed and can't push the new one.

The build infrastructure issue has been identified and a fix has been deployed. Builds should be progressing now - you can track the full status at https://status.railway.com/incident/ZLI8N955. If your build is still stuck, cancelling and redeploying should pick it back up. Regarding your production build being removed, that's not expected behavior during build delays. If your previous deployment is no longer running after the incident stabilizes, please share more details so we can investigate further.