Your cron job logs show the task completing successfully, but a "Crashed" status means the process exited with a non-zero exit code or was force-terminated due to open resources (like database connections) preventing a clean shutdown. Per our cron job docs, cron services must close all connections and exit cleanly after the task finishes, and explicitly exiting with code 0 (e.g., process.exit(0)) at the end of your script will ensure Railway recognizes the completion as successful rather than a crash.

This did not solve my problem, but I found a solution. I use Neon for DB as a service. It automatically scales down to zero when not in use. When I created a connection pool I had a timeout of about 2 seconds, which sometimes was not enough time to connect to the DB given, network latency and cold start times. I increased the connection timeout time.

Glad you found it. Neon's scale-to-zero cold starts can definitely exceed a 2-second connection timeout, so increasing that makes sense.