PHP fatal error causes White Screen of Death before watchdog alert in Drupal

Goal: Configure Drupal to surface critical failures through watchdog without duplicating admin UI notifications or flooding privileged users. When a PHP fatal error aborts Drupal bootstrap, the White Screen of Death prevents the message system from executing, so watchdog entries may not be written through the normal Drupal path. Separately, system_status_message displays only to users with the access admin permission, and the Rules module default action writes to the same message queue as core, which can produce duplicate notifications for one event. Watchdog flushing is asynchronous and can be throttled, but there is no unified policy for separating low-impact logs from critical failures. It is unclear whether critical watchdog entries should be routed to external monitoring instead of the Drupal UI, and how to suppress duplicate alerts when Rules and core messages fire for the same event. Questions: - Should critical watchdog entries bypass the Drupal message queue and integrate directly with external monitoring? - How can duplicate notifications be prevented when Rules and core system_status_message handle the same event? - What severity threshold separates actionable alerts from noise without suppressing genuine failures?