Fault Handling

The fault manager (ecu_fault) provides priority-based fault reporting, latching, retry policies, and degrade mode support.

• Fault Catalogue

Severity Levels

Level	Action	Behaviour
Info	None	Logged but no system action taken
Warning	Degrade	Reduce power, may auto-clear when condition resolves
Critical	Shutdown	Latched — requires safe shutdown sequence
Fatal	Shutdown	Immediate safe shutdown, cannot clear without reset

Degrade Mode

When a warning-level fault is active the system enters degrade mode:

1. Maximum throttle limited to 50 %

2. RPM limit reduced by 20 %

3. Additional sensor monitoring enabled

4. Telemetry sends fault notification to host

Recovery

• Warning faults — auto-clear when the triggering condition resolves.

• Critical faults with retry — up to 3 retry attempts (configurable via ecu_fault_set_retries()).

• Fatal faults — require a power cycle.

Fault Log

All reported faults are recorded in a circular log buffer (FAULT_LOG_SIZE = 32). Each entry includes the fault code, severity, action, and timestamp. The log can be read back via the telemetry TEL_CMD_READ_FAULTS command for post-mortem analysis.