17 days ago
Hi, Since this morning (UTC) deploys of our production service fail at the image build step with "Failed to build an image". The diagnosis says "Infrastructure Error". The build stops at "unpacking archive", before any Dockerfile step runs, and no error is logged.
The last successful deployment was earlier today, and our Dockerfile and railway.json have not changed since then. A local Docker build of the same code completes successfully. The Railway status page currently shows "Degraded Performance" for the October 21 incident recovery.
I have sent the deployment IDs and build details to support privately. Could someone from the Railway team check the build infrastructure for this service?
4 Replies
17 days ago
Thank you for letting us know! We've informed the Railway team, will update as we get more info. Apologies for the trouble 🙏
17 days ago
Apologies for the wait, the team opened an incident, you can track the issue here: https://status.railway.com/incident/ILLVW9WG
Status changed to Awaiting User Response Railway • 17 days ago
17 days ago
Sorry for the trouble. The build failures you're seeing ("build daemon returned an error" before any build step runs) are caused by an issue with our build infrastructure in US West, not by anything in your project or code. You don't need to change anything on your side.
We're working on a fix and tracking it here: https://status.railway.com/incident/ILLVW9WG
Your running services aren't affected. Redeploying may work in the meantime, but it can fail again until the fix is in. Once the incident is resolved, redeploying the same commit should build normally.
Your most recent attempt logged this error right after "unpacking archive", and the earlier attempts that stopped at the builder start line are the same failure. This is a new incident, separate from the October 21 one.
17 days ago
The issue causing builds to fail in US West has been resolved. Builds are now completing normally. If a build failed during this incident, redeploying it should now succeed.