Platform For AI:View ServiceInstance events in CloudMonitor

View ServiceInstance events

In the CloudMonitor console

To view ServiceInstance events in the CloudMonitor console, you can perform the following steps:

1. Log on to the CloudMonitor console.
2. In the left-side navigation pane, choose Event Monitoring > System Event.

3. On the Event Monitoring tab, select PAI from the product selection drop-down list and click Search to view system events of EAS.

4. Find the system event that you want to view and click Details in the Actions column.

   The following figure shows an example of the event details.

   The following table describes the parameters.

   Parameter	Description
   Product	The code of the service. For example, the code of Machine Learning Platform for AI (PAI) is learn.
   Name	The name of the event. For information about PAI system events, see the Event name column of the table in the Supported ServiceInstance events section.
   Level	The severity level of the event. Valid values: INFO WARN CRITICAL
   Status	The status of the event. For information about the status of PAI system events, see the Event status column of the table in the Supported ServiceInstance events section.
   RegionId	The region ID of the service. For example, the ID of the China (Shanghai) region is cn-shanghai.
   ResourceId	The ID of the resource. For more information, see Policy description.
   InstanceName	The name of the service instance.
   Time	The time at which the event occurred. The timestamp follows the UNIX time format. It is the number of milliseconds that have elapsed since 00:00:00 Thursday, January 1, 1970.
   GroupId	The CloudMonitor application group to which the EAS service belongs. By default, this parameter is empty.
   Content	The content of the event. The value is in the JSON format. For more information, see Fields of the Content parameter.

   Fields of the Content parameter

   Field	Description
   serviceName	The service name of the instance.
   serviceId	The service ID of the instance.
   serviceGroup	The service group to which the instance belongs.
   resourceType	The type of the resource group to which the instance belongs. Valid values: PublicResource: public resource group. DedicatedResource: dedicated resource group.
   instanceType	The instance type.
   cpu	The number of CPUs used by the instance.
   memory	The memory usage of the instance. Unit: MB.
   gpu	The number of GPUs used by the instance.
   gpuMemory	The GPU memory usage of the instance. Unit: GB.
   nvidiaName	The name of the GPU used by the instance.
   role	The service role of the instance. Valid values: Queue: the queue service. DataLoader: the offline service. Standard: the standard service.
   isBurst	Specifies whether auto scaling is enabled for the resource group of the instance. Valid values: false true
   isSpot	Specifies whether the instance is a preemptible instance. Valid values: false true
   callerUid	The UID of the Alibaba Cloud account that is used to deploy the EAS service.
   timestamp	The time when the event occurred. The timestamp is in the UTC format.
   restartCount	The number of times for which the instance restarted.
   exitCode	The exit status code of the instance. By default, this parameter is empty.
   status	The status of the instance. For information about the valid values, see the Event status column of the table in the Supported ServiceInstance events section.
   reason	The reason why the event occurred.
   message	The information about the event.

API operation

You can also call an API operation to view ServiceInstance events. For more information, see DescribeSystemEventAttribute.

Create and enable an event-triggered alert rule

In the CloudMonitor console

1. Create a system event-triggered alert rule. For more information, see Create a system event-triggered alert rule. Take note of the following parameters:

   • Product Type: Select PAI.

   • Event Type: Select ServiceInstance. You can set this parameter only to ServiceInstance. This value specifies the type of events that are related to EAS service instances.

   • Event Level: Select one or more severity levels based on your business requirements.

   • Event Name: Select the name of the event that you want to monitor. The available names are listed in the Event name column of the table in the Appendix section. You can select one or more event names.

   • Keyword Filtering: Specify the keywords and condition that are used to filter the events.

   

2. Enable the system event-triggered alert rule. For more information, see Enable system event-triggered alert rules.

API operation

You can also call an API operation to create and enable an event-triggered alert rule. For more information, see Create a system event-triggered alert rule and Enable system event-triggered alert rules.

Appendix: Supported ServiceInstance events

The following table describes the ServiceInstance events that are defined by EAS based on the lifecycle of a service instance.