Manage events
E-MapReduce (EMR) records system events that occur in clusters and synchronizes them to CloudMonitor. Events capture resource exceptions, operation outcomes, and status changes — giving you a centralized view of cluster health and activity.
Supported cluster types
Only events from DataLake, Dataflow, online analytical processing (OLAP), DataServing, and custom clusters on the EMR on ECS page appear in the EMR console. For Hadoop clusters, see Manage Event Center (Hadoop clusters).
Event types
EMR generates three types of events: health events, operation events, and expense events.
Health events
Health events are triggered when a service component's health status changes abnormally. When a health event occurs, go to the Overview page to identify the cause and resolve it before it affects your business.
Health events follow this state lifecycle:
| Status | Description |
|---|---|
| Unhandled | The service component is abnormal. Manually restore its health status to good. |
| Complete | The service component's health status is restored to good. |
| Canceled | The service component is disabled or released. Its status is unknown. |
The following figure shows the state transitions for health events.
Operation events
Operation events are triggered when failures or other exceptions occur during cluster operations. When an operation event occurs, go to the Operation History page of the cluster to view the cause and take action.
The Pending tab shows only Critical and Warning events from the previous three days. Clicking Ignore or Ignore All Events removes the events from this view. To see all events, click the All tab on the Events page.
Expense events
Expense events are triggered when a subscription cluster expires or is about to expire. When an expense event occurs, go to the renewal page to check the expiration time and renew the cluster promptly to avoid business disruption.
A renewal operation takes effect in about 10 minutes. Refresh the page after 10 minutes to verify the updated event status.
Event levels
EMR classifies system events into three levels based on their impact:
| Level | Impact | Required action |
|---|---|---|
| Critical | May result in instance unavailability | Handle immediately |
| Warning | Has some impact on your business | Monitor closely or handle at an appropriate time |
| Info | Records event processes | View or ignore based on your needs |
View events
View events in the EMR console
Log on to the EMR console. In the left-side navigation pane, click EMR on ECS.
In the left-side navigation pane, click Events.
View unhandled or pending events. A number next to Health Events, Operation Events, or Expense Events indicates how many unhandled or pending events exist in that category. Filter events by event level, event name, or time range.
To view event details, find the event and click Details in the Actions column.
To view all events, click the All tab.
View system events in the CloudMonitor console
Log on to the CloudMonitor console and select the EMR service to view system events. For more information, see View system events.
What's next
Set up event-triggered alert rules to receive notifications and respond to issues as soon as specified events occur in EMR. You can also test system event-triggered alert rules to verify your alert configuration. For more information, see Create an event-triggered alert rule.