Project troubleshooting

Live. This area is documented as current, user-reliable behavior.

Goal

Shorten the path from “deployment failed” to the actual misconfiguration or runtime issue.

Check build status first, then deployment status, then runtime logs.

Verify service type, workload type, port expectations, and environment variables before assuming a platform problem.

Review domains separately from application runtime failures.

If the project is a worker service, lack of a public URL is expected behavior rather than a routing bug.
If the project is contract build, expect build output and logs, not a live runtime or healthcheck URL.
If boot is slow, increase startup timeout before assuming the image or platform is broken.

A rejected mount path usually means it targeted a reserved system directory or did not start with /.
If state disappears after redeploy, confirm persistent volume was enabled and mounted at the path the app actually uses.
If storage size changes do not appear immediately, remember that resizing takes effect with redeploy behavior.

You can classify a project failure accurately before retrying or rolling back.

⌘I