Could you check the app Logs and tell me what you see when one of them stops?

• A stack trace or "OOM / killed" → it's crashing (code/memory fix)
• Clean shutdown with no error → it's the compute lifecycle (keep-alive job fix)
• Nothing / app never fully started → deployment or dependency issue

If you can paste the last several log lines (redact anything sensitive), I can pinpoint the cause and, if it's the keep-alive route, help you write the small Databricks Job + API call to auto-restart both apps. What does the log show?