[Bug]: Cloud deployment fails: "did not become healthy before the readiness deadline" (custom domain serves Railway default cert)

[princepal9120]

Summary

A new Roomote Cloud deployment (trial account) stays in a failed state with: "Deployment did not become healthy before the readiness deadline." The deployment URL never serves the app.

Environment

• Roomote Cloud (cloud.roomote.dev), hosted deployment, trial plan
• Deployment name: testing → https://testing.roomote.ai
• Dashboard shows the deployment with a red status dot, "1 user · $49/month", and billing shows "0 deployments"

Observed

• testing.roomote.ai CNAMEs to jq9y5a4v.up.railway.app.
• TLS handshake serves Railway's default cert (CN=*.up.railway.app, SAN *.up.railway.app, up.railway.app), so curl https://testing.roomote.ai/ fails with SSL: no alternative certificate subject name matches target host name.
• With -k, the response is HTTP 404 (Railway edge, no routed service).

Likely cause (inference, not confirmed)

The custom domain testing.roomote.ai was never attached/verified on the Railway service, so no cert was issued and the health check against the deployment URL could not pass. It could also be the app service not starting; I can't see Railway logs from the outside.

Expected

Deployment provisions, custom domain gets a valid cert, readiness check passes.

Repro

Sign up at cloud.roomote.dev → Create deployment (name testing) → wait for readiness → error above.

Ask

Please check the provisioning/Railway logs for this deployment and whether the custom-domain attach step failed. Also worth surfacing the underlying failure reason in the dashboard instead of the generic readiness message.

[roomote-community]

I'd like to help with this issue, but I need one maintainer clarification before I can produce a code-level implementation plan grounded in the correct system.

I traced the current repository. It documents that a custom domain must be attached to the Railway web service and that R_APP_URL must use the canonical hostname, but it does not contain the hosted Cloud control plane that provisions Railway resources, attaches domains, polls readiness, persists provisioning failures, renders Cloud deployment status, or calculates the Cloud billing count. The reporter has already provided enough reproduction detail; no additional reporter-side diagnostics appear necessary.

Could a maintainer please clarify:

• Which repository or service owns the cloud.roomote.dev deployment provisioning/readiness workflow and its dashboard/billing state, and can this issue be routed there or that source be made available for planning?
• With authorized provider access, which provisioning phase failed in this case: service startup, custom-domain attachment/verification, certificate issuance, or application health? A sanitized failure code or phase is sufficient; please do not post credentials or private logs.

Once that source and failure phase are identified, the plan should cover idempotent attachment of the hostname to the web service, separate domain/TLS and application-health readiness checks, persistence and display of the first actionable failure, retry behavior, and consistent deployment-count/billing semantics.

Please tag @roomote in the response, and I can finish the repository-grounded implementation plan.

[mrubens]

Hey @princepal9120 thanks for reporting, taking a look now!

[mrubens]

Can you try again now?

[princepal9120]

Yes, it is working now