System events

Routine maintenance versus system events

ECS instances are the core component through which you can establish your applications. After you select and start ECS instances, initiate configuration, and start to deploy applications, the health of your ECS instance is crucial to your business. To guarantee the backend performance and security of your ECS instances, Alibaba Cloud perform routine maintenance for the physical servers. When Alibaba Cloud scans your instances for hardware and software faults, or for potential risks on the physical servers, your instances are migrated to healthy servers in real time if exceptions are detected. However, unlike system events, you do not receive any notifications. Your instances are not impacted while such a routine maintenance is in progress.

If system events occur, you are notified about the default actions and the time scheduled to perform these actions on your instances. For planned system events, information such as the impact of the event on the instance and the expected execution point is sent in advance. To prevent impacts to your services, we recommend that you back up your data and distribute incoming traffic before handling system events. After a system event is resolved, you can query your system events history for the past week, for further analysis of faulty diagnosis and repairs.

Limits

Phased-out instance types, including but not limited to sn2, sn1, t1, s1, s2, s3, m1, m2, c1, c2, c4, ce4, cm4, n1, n2 and e3, do not support system events. For more information, see instance type families.

Event types

The following table describes the types of ECS system events.

Event status

The following table describes the status of a system event during its lifecycle.

System event periods

System events observe the following two periods:

• User operation period: The period between initiation and scheduled time of system events. Normally, you will receive a notification between 24 and 48 hours before a system failure event is fixed, from 3 days before a Subscription instances is stopped, and within 1 hour before a Pay-As-You-Go instance is stopped. Instances are released 15 days later if no renewal or addition of funds to your account are made. During this period, you can choose the recommended method by which to handle system events in advance. You can also wait until the default actions are triggered.
• System action period: Generally, if you wait until the default action is triggered, system events are automatically fixed within 6 hours after the system action period begins at a scheduled time. After, you receive a report of the system events.
  
  
  Note Only scheduled system events have a user operation period. Unexpected system events that are caused by emergency failures or invalid operations do not have any user operation periods. If an unexpected system event occurs, you will receive notifications, but you cannot take any action. However, you can query the system events history for fault diagnosis, cause analysis, or data recovery.

View system events

If a system event is scheduled, the Pending Tasks button in the ECS console shows a highlighted tag to remind you to check the event.

1. Log on to the ECS console.
2. In the left-side navigation pane, select Overview.
3. Select Pending Tasks from the navigation pane on the right-side of the Overview page.
4. On the Pending Tasks page, you can see the list of instance ID, region, status, system tasks, recommended user operations, and buttons for operations. Alternatively, you can choose recommended user operations under the Actions column to handle the system events.

API operation: Call DescribeInstancesFullStatus to view system events.

View system events history

On the All events page, you can query the system events history within the past week for faulty diagnosis and faulty replay.

1. ECS console
2. On the left-side navigation pane, select Overview.
3. Select Pending Tasks from the navigation pane on the right-side of the Overview page.
4. Click All Tasks, and on the All Tasks page, click System Tasks > Instances to see the list of instance ID, event type, region, and task status.

API operation: Call DescribeInstanceHistoryEvents to view system events history.

System event suggestions

System events allow you to perceive the underlying components of Alibaba Cloud ECS. You can optimize the O&M of instances based on system events. We recommend that you use the following actions to handle system events accordingly.