General description of scheduled database events

Event Notification

Scheduled event notifications—such as text messages and emails—are typically sent to subscribers 1 to 3 days in advance based on your configurations in the Message Center (Message Receiving Management > Basic Receiving Management > ApsaraDB Failure Or O&M Notification). Ensure your Alibaba Cloud account is configured with correct recipients and that you have subscribed to the text message and email channels. After receiving a notification titled "Database Scheduled O&M Event Notification", you can go to the scheduled events page in the Event Center (or Event Management) of the corresponding product console to view details such as the event type, region, reason, cancellation risk, and the list of affected instances. You can also adjust the scheduled switchover time for an instance to off-peak hours as needed.

Notes

• Events are categorized into two types based on their urgency:

  • S0 Emergency Level: Risk Remediation: These typically address unexpected scenarios requiring prompt remediation to prevent failures. Notifications for such events may be sent 3 days or more in advance, with a limited window for modifying the scheduled switchover time. Typical scenarios include urgent version replacement upgrades, host exceptions, and SSL Certificate expiration upgrades.

  • S1 Planned Level: System Maintenance: These typically involve low-risk issue remediation or planned software and hardware upgrades. Notifications are usually sent more than 3 days in advance, and users can cancel these events.

• To ensure you receive scheduled notifications for O&M events, you must log on to the Message Hub and ensure the notification method check boxes for ApsaraDB failure or O&M notifications are selected and message recipients are set (we recommend assigning them to database O&M personnel). Otherwise, you will not receive event notifications.Notification methods include text message, email, and internal message. We recommend selecting both text message and email to improve delivery success rates.

  

  Figure 1. Message Center notification settings entry

  

  Figure 2. ApsaraDB notification settings

• To receive immediate updates about O&M events or to build custom O&M automation using an event-driven approach, you can configure system event subscriptions through the Cloud Monitor platform. ApsaraDB pushes CloudMonitor system events for the lifecycle of O&M events, such as subscription, start, completion, and cancellation. For detailed instructions, see Manage Event Subscriptions (Recommended). For a list of subscribable CloudMonitor events, see Appendix 1: CloudMonitor-Related System Events.

  Expand to view a Cloud Monitor event format example.

  {
    "eventId": "c864b30b-7f69-5f04-b0e7-8dfb0eabcfd9", // Event ID. The ID is the same for the same event.
    "product": "RDS",                                  // Product code
    "reason": "Host software/hardware upgrade",        // Event reason
    "extra": {
      "impactZh": "Transient instance disconnection",                           // Event impact
      "impactEn": "Transient instance disconnection",  // Event impact
      "eventCode": "rds_apsaradb_transfer",            // O&M event type code
      "eventNameEn": "Instance migration",             // O&M event name
      "eventNameZh": "Instance migration",                        // O&M event name   
      "switchTime": "2024-09-15T01:30:00+08:00",       // Scheduled switchover time. If a switchover occurs, this is the time of transient instance disconnection.
      "startTime": "2024-09-14T21:30:00+08:00",          // Scheduled start time. The event enters the scheduling queue and waits for execution.
      "cancelCode": "OutOfGoodPerfBySoftHardwareUpgrade", // Cancellation risk code. See Appendix 2: Detailed Reason Codes and Cancellation Risks.
      "detailCode": "HostSoftHardwareUpgrade",            // Detailed reason code. See Appendix 2: Detailed Reason Codes and Cancellation Risks.
      "instanceInfo": ""
    },
    "instanceId": "rm-2ze9d66o65q1g02g6",             // Instance ID
    "eventType": "Maintenance",
    "instanceComment": "rm-2ze9d66o65q1g02g6",        // Instance alias
    "instanceType": "Instance",
    "publishTime": "2024-09-10T16:01:47+08:00"
  }

Procedure

1. Log on to the management console of each product.

2. In the navigation pane on the left, click Event Center (formerly Event Management) > Scheduled Events. Then, select a region at the top of the console.

3. On the Planned Events page, you can view event details. By default, Scheduled events are displayed. To view historical completed or canceled events, click Completed or Canceled. The following describes event properties in detail:

   Property	Example	Description
   Event type	Risk remediation	Events are categorized into "Risk remediation" and "System maintenance" based on their urgency.
   Running status	Waiting for execution	The scheduling status of the event. The statuses to note are as follows: Waiting for time setting: The event execution time is empty. Set the time based on your business needs. If you do not set the time by the latest operation time, the system automatically cancels the event and does not execute it. Waiting for execution: The event waits until the scheduled start time to enter the scheduling phase. Executing: The event enters scheduled execution. Manual intervention is not possible at this stage. To terminate it urgently, submit a ticket (non-standard operations may carry unknown risks). Successfully completed: Execution successful. Canceled: Execution failed or canceled. Common cancellation reasons: Customer initiated cancellation (UserCancel): The user cancels in the console or through OpenAPI. Customer response timeout (UserResponseTimeout): The event requires a specified time, but the time is not set by the deadline. The event is automatically canceled. Database control cancellation (SupervisorCancel): The event initiator actively cancels. Avoidance cancellation (AvoidCancel): The risk is resolved, or the instance's current status no longer requires this event, such as when the instance is already the latest version and does not need an upgrade. System automatic cancellation (AutoCancel): The system periodically checks scheduled events. If an instance does not meet the conditions for event execution, the event may be canceled, for example, if the current instance status is abnormal and actions cannot be issued. Execution timeout (ExecuteTimeout): The event enters the execution queue but does not complete within the expected time. Execution failed (ExecuteFail): An unknown exception occurs during event execution.
   Event type	Minor version upgrade	For more information, seeEvent types and impacts.
   Event reason	-	For more information, see Appendix 2: Detailed Reason Codes and Cancellation Risks.
   Business impact	Transient instance disconnection	The business impact varies depending on the event. For more information, see Event types and impacts.
   O&M recommendation	Confirm that your business applications have an automatic database reconnection mechanism and monitor business impact	O&M recommendations vary by event type. See Appendix 1: Cloud Monitor–Related System Events.
   Scheduled start time	-	The time when the event starts to enter the scheduling queue. Before the start time, this event has no impact on the instance. After the start time, you can still access the database normally, but you cannot perform instance-level operations (such as changing configurations or migrating zones). This time is empty when the status is "Waiting for time setting".
   Scheduled switchover time	-	The time of primary/secondary or link switchover (if any), which usually refers to the time when the instance connection experiences a transient disconnection. This time is an estimated value. A switchover occurring around this time is expected. In extreme cases, such as a failback to a zone, a second switchover may occur. Note Considering factors such as event scheduling and Data Preparation time, a certain preparation period is usually required before the switchover. Therefore, there is an interval between the start time and the switchover time. This interval may vary for different database products and events.
   Latest operation time	-	The latest time you can set for the switchover. The adjusted switchover time cannot be later than this time.
   Cancellable	Yes	To block this event, you can cancel it. This feature is usually available for "System Maintenance" events. Important Planned events are typically issued by the ApsaraDB control system during regular inspections. Canceling an event may cause a new one to be issued in the next inspection cycle, and frequent cancellations can also lead to risk escalation. We recommend that you execute the event at a suitable time based on your business needs instead of canceling it. For information about the risks of cancellation, see Appendix 2: Detailed Reason Codes and Cancellation Risks.
   Is time modification allowed?	Yes	Most events allow you to adjust the execution time. In rare scenarios, urgent remediation for high-risk issues may not allow time adjustments due to insufficient adjustment windows.

4. Modify scheduled events (optional).

   Select the records for which you want to adjust the execution time. Click Modify Scheduled Event to enter the switchover time setting interface. Two modification methods are supported:

   • Execute immediately: The task start time is set to the current time, and the task immediately enters the execution queue.

   • Specify switchover time: Select an appropriate time point within the configurable switchover time range for execution. The start time is automatically calculated based on the switchover time, but the new start time cannot be earlier than the current time; otherwise, you cannot modify it.

5. Modify periodic time window (optional).

   Click "Periodic Time Window Configuration" in the upper-right corner of the event list to enter the periodic time window configuration page.

   The execution time of scheduled events is usually automatically calculated based on the instance's maintenance window (see Set maintenance window for RDS|Tair/Redis|MongoDB|PolarDB). You can also customize the periodic time window based on your O&M needs. When ApsaraDB initiates new events, it prioritizes calculating the execution time based on your configured time window.

   You can set the window by month or week. For example, if the periodic switchover time is set to 02:00 to 03:00 on Monday and Tuesday each week, and the cloud platform's scheduled event window is from this Tuesday to next Sunday, the event's switchover time will fall within 02:00 to 03:00 on this Tuesday and 02:00 to 03:00 on next Monday. Typically, the system prioritizes switching on this Tuesday.

   Important

   1. This configuration applies only to future new events. To adjust the time for events currently in the event list, click "Set Execution Time".

   2. This configuration serves only as an auxiliary setting for calculating execution time and is effective only for "System Maintenance" event types. The actual calculated execution time is based on the time in the event list.

   3. This is an account-level configuration. After configuration, it takes effect synchronously for all database products that support periodic times.

6. Cancel scheduled events (optional).

   Select the event records you want to cancel. Click Cancel Scheduled Event to enter the cancellation interface. After understanding the cancellation risks, click Confirm to initiate the cancellation.

Event Types and Impacts

Appendix

Appendix 1: Cloud Monitor System Events

For more information, see Supported Cloud Products and Their System Events.

Appendix 2: Detailed Reason Codes and Cancellation Risks

Appendix 3: Event Types