Events are used to record and notify the information about cloud resources, such as the resource exceptions, the execution status of operations, and resource status changes. E-MapReduce (EMR) records system events that occur in clusters and synchronizes the events to CloudMonitor. You can view the events of a cluster in the EMR console and configure event alert rules in the CloudMonitor console. This topic describes event types, event levels, and how to view events in the EMR console.
Limits
Only events that occur in DataLake, Dataflow, online analytical processing (OLAP), DataServing, and custom clusters on the EMR on ECS page are displayed in the EMR console. For information about how to view the events of other types of clusters, see Manage Event Center (Hadoop clusters).
Event types
Health events
Health events are generated when abnormal changes in the health status of a service component are detected. Health events help you understand the risks and exceptions of the current service components. If a health event is generated, you must go to the Overview page to view the cause of the exception and handle the event at the earliest opportunity to prevent impacts on the upper-layer business.
The following table describes the event states.
Status | Description |
Unhandled | The service components are abnormal. You must manually restore the health status of the service components to good. |
Complete | The health status of the service components is restored to good. |
Canceled | The service components are disabled or released. The status of the service components is unknown. |
The following figure shows the relationship of event states.
Operation events
Operation events are generated when failures or other exceptions occur during operations on clusters. Operation events help you understand the operation risks and exceptions of clusters. If an operation event is generated, you can go to the Operation History page of the current cluster to view the cause of the operation failure and handle the event.
You can view only the critical and warning events that are generated in the previous three days on the Pending tab. If you click Ignore or Ignore All Events, the events are no longer displayed.
You can view all events on the All tab of the Events page.
Event levels
EMR defines the following levels for system events based on their impacts on the normal running of instances:
Critical: System events at this level may result in instance unavailability and must be handled at the earliest opportunity.
Warning: System events at this level have some impacts on your business. Pay close attention to these events or handle them at an appropriate point in time.
Info: System events at this level record event processes. You can view or ignore the events based on your business requirements.
Query events
Log on to the EMR console. In the left-side navigation pane, click EMR on ECS.
In the left-side navigation pane, click Events.
Query unhandled or pending system events.
If a number is displayed next to Health Events or Operation Events, the specific number of unhandled or pending events of the category are triggered. You can filter system events by event level, event name, or time range.
Find a desired event and click Details in the Actions column to view the details of the system event.
Click the All tab to view all system events.
References
Only events that occur in DataLake, Dataflow, OLAP, DataServing, and custom clusters on the EMR on ECS page are displayed in the EMR console. For information about how to view the events of other types of clusters, see Manage Event Center (Hadoop clusters).
You can create event-triggered alert rules and test system event-triggered alert rules. Event-triggered alert rules allow you to receive alert notifications and handle exceptions at the earliest opportunity after the specified events are generated in EMR. For more information, see Create an event-triggered alert rule.