openclaw - 💡(How to fix) Fix Fleet/dashboard: alerts not auto-resolved on recovery, causing contradictory online status + active offline alert [1 comments, 2 participants]

Fix Plan

To resolve the zombie alert issue, we need to implement alert auto-resolution when the gateway comes back online.

Step-by-Step Solution:

1. Modify the alert creation logic: Update the alert system to listen for both offline and online events.
2. Add a recovery condition: When a gateway comes back online, check if there's an active "Gateway Offline" alert and auto-resolve it.
3. Update the alert status: Change the alert status to "Recovered" and add timestamps for when the alert was triggered and resolved.

Example Code (Python):

def create_alert(gateway_status):
    if gateway_status == "offline":
        # Create "Gateway Offline" alert
        alert = {"status": "active", "type": "Gateway Offline", "triggered_at": datetime.now()}
        alerts.append(alert)
    elif gateway_status == "online":
        # Auto-resolve "Gateway Offline" alert
        for alert in alerts:
            if alert["type"] == "Gateway Offline" and alert["status"] == "active":
                alert["status"] = "recovered"
                alert["resolved_at"] = datetime.now()
                break

def update_dashboard(gateway_status, alerts):
    # Update dashboard with current gateway status and alerts
    if gateway_status == "online":
        online_alerts = [alert for alert in alerts if alert["status"] == "active"]
        recovered_alerts = [alert for alert in alerts if alert["status"] == "recovered"]
        # Display online status and active/recovered alerts
    else:
        # Display offline status and active alerts
        pass

Verification

To verify the fix, follow these steps:

• Trigger a "Gateway Offline" alert by taking a gateway offline.
• Verify the alert is displayed on the dashboard.
• Bring the gateway back online and verify the alert is auto-resolved and moved to the "Recovered / Recent Incidents" section or updated in-place with timestamps.

Extra Tips

• Ensure the alert system is properly handling concurrent updates to prevent race conditions.
• Consider implementing a timeout or retry mechanism to handle cases where the gateway takes longer than expected to come back online.