Self-hosted Rowy Run backend crash-loops on every derivative column evaluation (unhandled error calling rowy.events/monitor)

Describe the bug The self-hosted Rowy Run backend (deployed on Cloud Run) crashes and restarts continuously whenever a derivative column is evaluated, most consistently reproduced via "Evaluate All". As a result, no derivative column can currently be evaluated on my project. Tracing this through Google Cloud Logging shows the crash is caused by an unhandled error from an outbound telemetry call the backend itself makes to https://rowy.events/monitor. This endpoint currently appears unreachable (TLS handshake failure), and since the error isn't caught, it takes down the entire Node.js process instead of just failing that one call.

To Reproduce

1. Open a Rowy table containing at least one derivative column.
2. Select the column and click "Evaluate All" (a single-row "Evaluate" can also trigger it).
3. The browser request fails with TypeError: NetworkError when attempting to fetch resource.
4. In Google Cloud Logging for the Cloud Run backend service, filtering around the same timestamp shows: an outbound POST to https://rowy.events/monitor fails (isAxiosError: true, response: undefined), immediately followed by:

Node.js v22.21.1
error Command failed with exit code 1

Cloud Run then auto-restarts a new instance, which crashes again the same way shortly after — a continuous crash loop.

Expected behavior A failure of the internal/optional telemetry call to rowy.events should not be able to crash the entire backend process. Derivative column evaluation should keep working (or fail gracefully with a clear, isolated error) even if rowy.events is unreachable.

Screenshots Attaching screenshots of: the browser-side NetworkError, the Cloud Logging entry showing the failed /listSecrets request (503, "malformed response or connection error"), and the raw stdout/stderr log showing the unhandled axios error and process crash.

Rowy Run version Deployed via the gcr.io/rowy-run/rowy-backend:latest image tag on Cloud Run. I could not find an explicit semantic version number in my project settings — happy to check a specific location if you can point me to it.

Desktop (please complete the following information):

• OS: Windows
• Browser: Firefox
• Version: 142.0 (issue reproduced consistently; also independently confirmed rowy.events is unreachable using curl.exe on the same Windows machine, outside the browser entirely)

Smartphone: Not applicable — issue is backend-side (Cloud Run) and reproduces regardless of client device.

Additional context I independently confirmed https://rowy.events is unreachable from two separate networks (my local machine, outside of GCP):

• curl.exe -v --max-time 5 -X POST https://rowy.events/monitor → schannel: failed to receive handshake, SSL/TLS connection failed (curl error 35)
• Loading https://rowy.events directly in Firefox → PR_END_OF_FILE_ERROR

This strongly suggests rowy.events is currently down or has a broken TLS setup on Rowy's end, which is then crashing every self-hosted rowy-run backend that has an unhandled rejection on this call.

I already emailed [email protected] about this 6 days ago with no response, which is why I'm filing this issue directly — this is currently fully blocking my ability to work with my data. Any help is very much appreciated.