Also getting this:

The image "production-dpams.railway-registry.com/REDACTED" could not be pulled from the registry. If the image is private, check that the registry credentials saved for this service are still valid.

same here

Same, on and off all day

I’m seeing the same issue on EU West.

Build succeeds, image push succeeds, but deployment fails at Create container because the freshly built image cannot be pulled from Railway’s own registry.

Railway Diagnose reports Infrastructure Error.

I’ve retried the same commit twice with the same result. Previous deployment remains active and healthy.

Deployment/service details available if a Railway team member wants to trace it - Web deployment ID: 5c9efae1

Same here.. not good

For everyone on this thread: this is on our side. We're seeing intermittent build and deploy failures in EU West today, and the "image could not be pulled from the registry" error at container creation is that issue, not your registry credentials or anything in your project.

We've opened an incident and you can track it here: https://status.railway.com/incident/TX3M5SWF

Your previous deployment stays live while this is going on. Redeploys may pass on and off, as some of you have seen, but it'll be reliable again once the incident is resolved.

Since the major outage on October 21st, the builds service has been noticeably less reliable than usual. I’d appreciate some transparency on:

• When we can expect the platform to return to a stable state?
• What changes were made that have contributed to the increased downtime and flakiness?

I’m fully supportive of the security improvements and want to continue using the platform. That said, I need clarity on whether these changes are coming at the cost of build performance and reliability.

We understand the frustration, and we appreciate the support.

We can't give a specific date for when builds will be fully reliable again. Today's EU West and US East incident is still under investigation. Its latest update says we have found the cause and are working with upstream providers to fix it.

As for what has changed, the other build incidents this week, the Oct 26 snapshot upload latency and the Oct 28 US West build failures, came from upstream storage and build infrastructure issues. Our updates on both stated they were unrelated to the October 21 security incident. The latest recovery information for the October 21 incident is in the addendum to our preliminary report.

Your running services are not affected while this is ongoing.

My systems are now in a mismatched state, my frontend deployment went through but the backend did not deploy. I am hitting errors because of missing apis. It has been 17 hours since the last update with a timeline.

Working with railway has become seemingly unstable and frustrating. I am a big fan but starting to consider switching.