EMR Serverless StarRocks records system events on instance resources and automatically syncs them to Cloud Monitor. Use events to track instance health, investigate anomalies, and respond before issues affect your business. Set alert rules in the Cloud Monitor console to get notified automatically.
Event types
| Type | Description | Examples |
|---|---|---|
| Health event | Abnormal changes in service health, not triggered by user actions. Covers threats and anomalies for service components installed on the instance. | Node system events, data storage bucket issues, FE (front-end) / BE (back-end) / CN (compute node) crashes or restarts |
| Operation event | Abnormal status changes caused by manual operations. Covers all operations on the instance. | Business operations: query timeouts, import timeouts. Operations and maintenance (O&M): scale-in failures, upgrade or downgrade failures, version upgrade failures |
Event statuses
| Status | Description |
|---|---|
| Pending | The event is abnormal and requires immediate attention. |
| Completed | The event was processed successfully. |
| Canceled | The corresponding object was deleted, or the event was ignored. |
| Scheduled | The event is scheduled for processing within the next 24 hours. |
Event levels
| Level | Impact | Action |
|---|---|---|
| Critical | Major impact on your business. | Handle immediately. |
| Warning | Some impact. | Monitor while the impact persists, or handle at a suitable time. |
| Info | Informational only. | Follow up as needed. |
Event list
Storage bucket events
The following events indicate that the data storage bucket for your StarRocks instance has reached a throttle threshold. Follow the suggestions in the event details, and contact the EMR Serverless StarRocks team to increase the bandwidth threshold or adjust related parameters.
| Event name |
|---|
| The downstream bandwidth of the data storage bucket for the StarRocks instance exceeds the throttle threshold |
| The upstream bandwidth of the data storage bucket for the StarRocks instance exceeds the throttle threshold |
| The queries per second (QPS) of the data storage bucket for the StarRocks instance exceeds the throttle threshold |
| The downstream bandwidth for the current user exceeds the throttle threshold |
| The upstream bandwidth for the current user exceeds the throttle threshold |
| The QPS for the current user exceeds the throttle threshold |
Node maintenance events
| Event name | Recommended action |
|---|---|
| The node needs to be restarted for system maintenance. | Restart the node immediately, or schedule a restart time that minimizes impact on your business. |
Handle events
On the Health Events or Operation Events tab, click Handle in the Actions column for the target event. Choose to handle the event immediately or schedule a specific time.