Engine Watchdog Timeouts – nre:Engine Fault

Problem

Niagara station suddenly stops responding, displays "Watchdog Timeout" in logs, and may reboot or freeze. Dashboard shows offline devices downstream.

Root Cause

Application Director engine thread is blocked or starved by heavy CPU load, often caused by inefficient wire sheet logic, excessive polling, or memory pressure.

Solution

1. Check Application Director logs for thread dumps and the exact error line

2. Review the Watchdog policy setting (Log/Terminate/Reboot):

   • Set to Log for advanced diagnosis (collects dumps without restarting)
   • Set to Terminate for automatic recovery (engine restarts on hang)
   • Reboot is rarely needed unless system is unresponsive

3. Use the Station Copier tool or manually back up the config.bog file for recovery purposes

4. Identify CPU hogs:

   • Run BQL queries to find components with high execution time or blocked threads
   • Look for wire sheets with OnExecute sleeps, infinite loops, or excessive array iterations
   • Check if polling schedules are conflicting (multiple devices polling on same interval)

5. Refactor problematic logic:

   • Break large wire sheets into smaller, prioritized components
   • Increase polling intervals for non-critical devices
   • Use throttle/delay actions instead of busy-waiting

6. If the station runs out of memory, upgrade hardware or reduce historization/archiving scope