Skip to content
JelinaJelina DocumentationTechnical documentation
Back to website

Watchdog

At unattended industrial sites, device failures can go undetected. JelinaOS’s built-in watchdog acts as an automatic inspector, continuously monitoring system status and intervening on faults to minimize their impact.

What the watchdog does

Section titled “What the watchdog does”
  • Continuous monitoring: Periodically checks system responsiveness and critical service status.
  • Automatic intervention: Handles detected faults according to predefined policies, such as restarting services.
  • System fallback: Triggers system-level recovery for serious faults to prevent devices from remaining unresponsive.
  • Logging: Records faults and recovery actions for later analysis.

Integration with disaster recovery

Section titled “Integration with disaster recovery”
Application failure ──▶ Automatic recovery by restart policy
    │
    └── Recovery fails ──▶ Watchdog intervenes
                                  │
                                  ├─ System-level recovery action
                                  └─ Still failing ──▶ A/B fallback (automatic repair on next boot)

The watchdog, A/B systems, and application restart policies form three levels of recovery:

  1. Application level: Restart applications automatically on faults.
  2. System level: The watchdog detects and intervenes.
  3. Partition level: The standby partition takes over after a system fault.

View status

Section titled “View status”
Terminal window
systemctl status jelina-watchdog

Check whether the watchdog service is running normally.

Configuration

Section titled “Configuration”

Monitoring and recovery policies, such as checking intervals and recovery actions, can be configured for site requirements. Defaults suit most deployments. Contact Advantech technical support for help with special requirements.

Use cases

Section titled “Use cases”
  • Unattended sites: Detect and handle system faults automatically.
  • Critical business devices: Recover automatically when business applications hang.
  • Remotely deployed devices: Reduce on-site intervention and maintenance costs.
Section titled “Related documentation”