Real-Time Alerts Setup

Real-Time Alerts keep you informed about your server’s health. You can receive notifications when an instance crashes, goes offline, or breaches performance thresholds like low TPS or high CPU usage.

---

1. Setting Up Notification Channels

Before you can receive notifications, you must configure one or more Notification Channels at the Organization level. Once channels are set up, they can be selected and assigned to individual alerts.

Discord Webhooks

1. Navigate to Organization Settings -> Notification Channels on the dashboard.
2. Click Add Channel and select Discord Webhook.
3. Enter a name for the channel and paste your Discord channel webhook URL.
4. Click Save and send a test notification to verify delivery.

Email Groups

1. Navigate to Organization Settings -> Notification Channels on the dashboard.
2. Click Add Channel and select Email Group.
3. Enter a name for the channel and input the email addresses of the team members who should be notified.
4. Click Save.

---

2. Integrated Uptime and Crash Notifications

Uptime and server crash alerts are handled automatically by an integrated ShardAnalytics backend job. You do not need to create custom alert rules for server crashes.

If an instance stops transmitting telemetry without publishing a clean shutdown event, the backend job detects the missing heartbeat and sends a notification.

How to receive crash and recovery notifications:

1. Configure your desired Notification Channels (Discord webhook or Email group) as described in Section 1.
2. Within the settings for your notification channel, toggle the Server Crash & Recovery option to Enabled.
3. (Optional) To modify the heartbeat checking interval or deactivate these integrated notifications entirely, navigate to Organization Settings -> Alerting Settings.

---

3. Performance Threshold Alerts

You can configure custom alert rules based on instance performance metrics (such as TPS, CPU, memory usage, or network latency).

Common Performance Alert Rules

Metric	Warning Threshold	Critical Threshold	Purpose
TPS (Ticks Per Second)	TPS < 18.0 for 2m	TPS < 15.0 for 1m	Detects tick rate drops and server lag.
CPU Usage	CPU > 90% for 5m	CPU > 95% for 3m	Flags runaway threads or high entity counts.
Memory Allocation	Memory > 85% for 10m	Memory > 95% for 5m	Spots potential memory leaks.
Network Latency	Latency > 150ms for 5m	Latency > 300ms for 3m	Warns about network routing degradation.

Creating a Custom Alert:

1. Navigate to Alert Config on the dashboard.
2. Click Create Alert Rule.
3. Select the source metric (e.g., Tick Rate).
4. Define the trigger condition (e.g., TPS < 18.0 for 5 consecutive evaluation cycles).
5. Select which Notification Channel(s) should receive alerts for this specific rule (configured per alert, allowing you to route different alerts to different channels).
6. Click Save Rule.