Project ID: bfc24f18-5da9-4d61-a95c-1cf4fded9e12

hello? anything?

Hey there,

I noticed this has been sitting without a response for over 20 days, so I wanted to share my perspective on what might be happening, though I should note upfront that I'm not Railway staff and can't provide an official solution or detailed fix.

That said, you are right... this doesn't look like a typical DNS issue. The fact that all three Railway ingress domains changed simultaneously is the key indicator.

These domains are Railway-generated and typically remain stable unless there's been some kind of project-level reprovisioning, service recreation, environment migration, or internal re-association event.

The secondary symptoms you're experiencing strongly support this theory.

This pattern suggests the deployments themselves are probably healthy, but the project metadata and integrations may have become partially desynced.

Regarding the GitHub issue: There was indeed a recent GitHub incident involving app installation token authentication failures (manifesting as "Bad credentials" and "repository not authorized" errors). This could definitely explain the GitHub-related symptoms, missing repos and broken deployment connections. However, I'm not certain whether that alone would trigger Railway to regenerate all three ingress domains. My best guess is the GitHub outage may have contributed to or triggered a separate reprovisioning/reassociation event on Railway's side.

Why you won't find a public explanation: Railway generally doesn't publish detailed changelogs for low-level infrastructure or routing migrations unless there's a broader platform-wide incident. Because of this, nobody outside Railway staff can tell you the exact trigger with certainty. I checked the public changelog, Discord, and GitHub issues/discussions but didn't find anything matching your situation exactly.

Before updating DNS, verify project stability:

• Test the new Railway URLs directly, you should confirm each app loads and functions correctly
• Check the Domains section for each service, verify custom domains are still properly attached
• Re-authorize the GitHub integration if repos are still showing as missing
• Verify project members/roles are now visible and correct

Then update your DNS CNAMEs to point to the new Railway targets

If the new Railway domains work as expected, updating DNS should restore production traffic normally.

One final observation: The old domains were short random IDs, while the new ones follow Railway's newer human-readable naming convention. This suggests these services were re-issued new ingress endpoints during some kind of project-level migration or reprovisioning event, rather than being manually edited.

Some links that may help you:

https://www.githubstatus.com/incidents/k5z4d1v1tqmt & https://status.railway.com/incident/RMR32LTK - GitHub's and Railway's incident reports about errors with GitHub app installation token authentication, currently an active incident, but there may have been similar in the past

https://station.railway.com/questions/bad-credentials-error-despite-valid-gi-82f45582 - a very similar thread concerning the GitHub issue

Since your issue has been ongoing for 20+ days and involves project-level changes, you'll ultimately need Railway support to investigate the root cause, but hopefully this context helps you understand what might have happened and verify stability before updating DNS.

Best of luck getting everything back online!