Platform Event

Last updated: 2019-12-30 17:52:15

    Event Definition

    Platform events record changes in the underlying infrastructure and services of Tencent Cloud. These events are generated or caused by Tencent Cloud infrastructure and services. Customers cannot control these events, which instead can only be managed by Tencent Cloud.

    Each platform event record consists of the following attributes.

    Attribute Description
    Event The product+event name supported by the underlying infrastructure and services that caused the event
    Time The time when an event record is generated, accurate to second
    Event Type Event types are divided into problems, changes, operational announcements, and other subscriptions based on the causes and impacts of the event. Problem-type platform events are supported. Other subscription events, which are not generated by Tencent Cloud but affect underlying infrastructure, are external events for customers
    Associated Resource The resources associated with the infrastructure and services affected by events. Multiple resource instances of a single account may be affected
    Status The status indicates the event triggering and restoration status of a problem-type platform event, the event start and end status of a change-type platform event,e or no status for an announcement-type event
    Region (if applicable) The region of the underlying infrastructure and services affected by platform events

    Two consecutive event records are associated with an event that has a status. To better locate and understand the status of infrastructures and services in the console, the two associated event records are displayed together and the progress of a specific platform event is updated according to the Status attribute. The triggering and restoration event records differ by their start times and update times.

    Features

    The platform event module retrieves and records events generated by the platform infrastructure and underlying services, and provides the following features to support display rewind, alarm notification, data pull through APIs, and automated operation triggering:

    • Filtering Events: allows users to filter events by attributes, such as time span, event type, event, event status, status, and region. Multiple attributes can be selected for filtering. When multiple options of one attribute are used, the OR operation applies. When multiple attributes are used, the AND operation applies.
    • Searching for object-associated events: allows users to search for associated platform events by the IDs of affected objects.
    • Subscribing to platform events: allows users to subscribe to and promptly obtain information about platform events of specific infrastructure and services.

    The progress of platform event processing is updated in real time. Event pull via APIs and automated operation triggering capacities are supported and will be publicly available in the future.

    Scenarios

    • Planning resource changes and troubleshooting method through alarm triggering: You can subscribe to platform events associated with specific products to promptly learn about platform problems and information changes, thus quickly planning for resource changes and tracking the troubleshooting progress.
    • Rewinding to locate exceptions: You can refer to the processing records of platform events to locate events and determine their causes by working backward.
    • Automated operation triggering: The event sources of automated triggering rules support platform events, allowing you to implement self-recovery and automated scheduling to improve the efficiency of monitoring and OPS.

    Use Limits

    • You can only view event information from the past 6 months.
    • Event alarm messages are counted into the message quota.

    Operation Instructions

    Filtering Events

    Use the global event filtering capability to query information about platform events that match a specified attribute combination.

    1. Log in to Cloud Monitoring Console.
    2. In the left sidebar, click Event Center > Platform Event to access the platform event page.
    3. Click the global filtering search box in the upper-right corner, modify default filtering items Product Type and Event Name, and confirm. Alternatively, customize filtering items such as Region and Event Status, and click to complete filtering. By using the time filtering box in the upper-left corner, you can customize the time range filter to display events updated within the specified time range.
    4. Use the scroll bar to scroll up and down the event list page or select a specific page to jump to.

    Searching for events associated with affected objects

    View the information about events associated with specific affected objects.

    1. Log in to Cloud Monitoring Console.
    2. In the left sidebar, click Event Center > Platform Event to access the platform event page.
    3. Click the global filtering search box in the upper-right corner, select Affected Object, enter the resource object ID, and press Enter.
    4. Use the scroll bar to scroll up and down the event list page or select a specific page to jump to.

    Subscribing to platform events

    Subscribe to the recipient of information about the platform event associated with a specific product.

    1. Log in to Cloud Monitoring Console.
    2. In the left sidebar, choose Alarm Configuration > Platform Event Subscription to access the platform event subscription page.
    3. Hover over a subscription item, click Subscription Management, select a message reception mode and a recipient, and then click OK.
    4. View the latest reception modes and recipients in the subscription item list.
    5. View alarm records in Alarm List.

    Supported Events

    The following events are currently supported and other associated products will be gradually supported.

    Event Type Event Cause Impact
    Problem Cloud Virtual Machine (CVM) storage Caused by the CVM infrastructure storage module The I/O performance of the CVM instance deteriorates and data read-write exceptions occur
    Problem CVM network connection Caused by the CVM infrastructure network The speed of the CVM instance network slows down or the network is disconnected
    Problem CVM running exception Caused by CVM infrastructure The CVM instance bears a high load or crashes, causing services to be unavailable

    Troubleshooting suggestion: Tencent Cloud will fix the problem. You can authorize the affected resources according to the alarm information, or wait for Tencent Cloud to solve the problem. For more information, see the documentation about service scheduling and migration.

    Event Type Event Cause Impact
    Other subscription ISP network jitter Caused by ISP network problems The latency to access Tencent Cloud via a public network increases or the network is disconnected

    Was this page helpful?

    Was this page helpful?

    • Not at all
    • Not very helpful
    • Somewhat helpful
    • Very helpful
    • Extremely helpful
    Send Feedback
    Help