View and subscribe to Milvus system events-Vector Retrieval Service for Milvus(Milvus)-阿里云帮助中心

Events record and notify you about cloud resource information, including resource exceptions, operation execution results, and resource status changes. Alibaba Cloud Milvus records system events that occur on instances and automatically synchronizes them to CloudMonitor. You can view events in the Milvus console for centralized review, or subscribe to events in CloudMonitor to receive notifications promptly and handle exceptions.

Event types

Alibaba Cloud Milvus supports the following system event types. You can filter events by event type in the event list.

Event type	Description
Instance lifetime	Records the execution results of operations such as instance creation, deletion, start, stop, upgrade, downgrade, version upgrade, and configuration change. The system generates a corresponding event when each operation succeeds or fails.
Health check	Records status changes such as Milvus component health exceptions or health recovery.

Instance lifetime events are triggered by actual instance operations. The Event Center displays a corresponding event only after you perform an operation such as creation, deletion, start, stop, upgrade, downgrade, version upgrade, or configuration change.

Event levels

System events are classified into the following levels based on the degree of impact on normal instance operations.

Event level	Description
Critical	Has a major impact and requires immediate attention. Otherwise, business operations may be affected.
Warning	Has some impact and requires attention during the impact period, or can be handled at an appropriate time.
Information	Displays event details. You can decide whether to pay attention to it.

Note

Event levels are displayed in Chinese in the Milvus console. After events are synchronized to CloudMonitor, event levels are displayed as corresponding English values, such as INFO and WARN.

View events in the Milvus console

Log on to the Alibaba Cloud Milvus console.
In the left-side navigation pane, click Event Center.

Set the filter conditions, and then click Search.

Filter condition	Description
Instance ID	Enter the target instance ID in the input box at the top of the page.
Time range	Set the start date and end date for event time at the top of the page.
Event type	Click the filter icon in the Event Type column header and select Instance Lifetime or Health Check. Multiple selections are supported.
Event level	Click the filter icon in the Event Level column header and select Critical, Warning, or Information. Multiple selections are supported.

View the events that match the filter conditions in the event list. The list displays the instance name/ID, event type, instance status, event name, event level, and event time.

Click Details in the Actions column of the target event to view its complete information.

Information group	Included fields
Basic information	Event ID, event name, event type, event level, event status, and Action. The event status is either ACTIVE (in progress) or RECOVERED (recovered). Action is the unique identifier of the event and corresponds to the Action in the events supported in this topic.
Instance information	Instance name and instance ID.
Time information	Event time, start time, and end time. If the event has not ended, the end time is displayed as -.
Trace information	Trace ID, request ID, and operator ID. For events triggered automatically by the system, the operator ID is system.
Details	Supplementary information about the event. For example, health check events describe the number of abnormal components, the components involved, and the current status of the instance.

View and subscribe to events in CloudMonitor

Milvus events are automatically synchronized to CloudMonitor. In CloudMonitor, you can query events in a unified manner and use event subscriptions to consolidate and reduce notification noise, customize notification methods, or push events to target channels.

View events

Log on to the CloudMonitor console.
In the left-side navigation pane, choose Event Center > System Events.
On the Event Monitoring tab, click the product filter box and select Vector Retrieval Service for Milvus.
Set the event level, event name, and time range as needed, and then click Search.
Find the target event in the event list, click Details in the Actions column, and view the event content.

Note

The event list in CloudMonitor displays the time, product name, event name, event level, region, resource, and content. The event name displays both the Action and the Chinese name, and the event level is displayed as an English value.

Subscribe to events

In the left-side navigation pane of the CloudMonitor console, choose Event Center > Event Subscription.
On the Subscription Policies tab, click Create Subscription Policy.

In the Subscription Scope section of the Configure Subscription step, set Product to Vector Retrieval Service for Milvus, and then set the subscription conditions as needed.

Subscription condition	Description
Event type, event level	Subscribe by event type and event level. For the values, see the event types and event levels in this topic. If not set, all are subscribed.
Event name	Subscribe by specific event, such as instance creation failure or instance downgrade failure. For the values, see the supported events in this topic. If not set, all event names are subscribed.
Application group	Limit the subscription scope by CloudMonitor application group. If not set, all application groups are subscribed.
Event content, event resource	Filter by keyword or resource. Separate multiple values with commas (,).

Configure noise reduction, notification settings, and push channels as needed to complete the creation of the subscription policy.

For more information about querying and subscribing to events, see View system events and Manage event subscriptions (recommended).

Supported events

Action	Event name	Event type	Event level	Recommended action
Cluster:Cluster.Create:Successful	Instance created	Instance lifetime	Information	The instance has been created. No action is required.
Cluster:Cluster.Create:Failed	Instance creation failed	Instance lifetime	Warning	View the event details and operation records, modify the configuration based on the failure cause, and then retry.
Cluster:Cluster.Delete:Successful	Instance deleted	Instance lifetime	Information	The instance has been deleted. No action is required.
Cluster:Cluster.Delete:Failed	Instance deletion failed	Instance lifetime	Warning	View the event details and operation records, check the instance status or associated resources, and then retry.
Cluster:Cluster.Stop:Successful	Instance stopped	Instance lifetime	Information	The instance has been stopped. No action is required.
Cluster:Cluster.Stop:Failed	Instance stop failed	Instance lifetime	Warning	View the event details and operation records, check the instance status, and then retry.
Cluster:Cluster.Start:Successful	Instance started	Instance lifetime	Information	The instance has been started. No action is required.
Cluster:Cluster.Start:Failed	Instance start failed	Instance lifetime	Warning	View the event details and operation records, handle the issue based on the failure cause, and then retry.
Cluster:Cluster.Upgrade:Successful	Instance upgraded	Instance lifetime	Information	The instance has been upgraded. No action is required.
Cluster:Cluster.Upgrade:Failed	Instance upgrade failed	Instance lifetime	Warning	View the event details and operation records, check the target specifications and instance status, and then retry.
Cluster:Cluster.Downgrade:Successful	Instance downgraded	Instance lifetime	Information	The instance has been downgraded. No action is required.
Cluster:Cluster.Downgrade:Failed	Instance downgrade failed	Instance lifetime	Warning	View the event details and operation records, check the target specifications and instance status, and then retry.
Cluster:Cluster.UpgradeVersion:Successful	Instance version upgraded	Instance lifetime	Information	The instance version has been upgraded. No action is required.
Cluster:Cluster.UpgradeVersion:Failed	Instance version upgrade failed	Instance lifetime	Warning	View the event details and operation records, handle the issue based on the failure cause, and then retry.
Cluster:Cluster.UpdateConfig:Successful	Instance configuration changed	Instance lifetime	Information	The instance configuration has been changed. No action is required.
Cluster:Cluster.UpdateConfig:Failed	Instance configuration change failed	Instance lifetime	Warning	View the event details and operation records, check the configuration parameters, and then retry.
Cluster:Health.ComponentUnhealthy	Component health exception	Health check	Warning	View the abnormal components, impact scope, and recovery suggestions in the event details, and handle the issue promptly.
Cluster:Health.ComponentUnhealthy:Maintenance	Component health exception (during change)	Health check	Information	A component health exception occurred during an instance change, and the system records it as a low-level health event. Monitor the change progress. If the component does not recover after the change is complete, handle the issue based on the event details.
Cluster:Health.ComponentRecovered	Component health recovered	Health check	Information	The component has recovered to a healthy state. No action is required.

Manage events