ApsaraDB RDS announces O&M events—such as instance migration, database version upgrades, and minor engine version upgrades—in the console and via voice call, email, or internal message. For each event, you can view its type, region, cause, business impact, and affected instance IDs. You can also change the scheduled switchover time.
Prerequisites
Before you begin, ensure that you have:
Pending O&M events (hover over the
icon in the upper-right corner of the console to check)
Event urgency levels
Events fall into two categories based on urgency:
| Level | Name | Deferrable | Advance notice | Typical scenarios |
|---|---|---|---|---|
| S0 | Emergency — Risk fixes | No | 3 days or less | Emergency version replacements, host failure repairs, SSL certificate expiration upgrades |
| S1 | Scheduled — System maintenance | Yes | More than 3 days | Low-risk issue fixes, planned software and hardware upgrades |
The Recurring Time Window Configuration applies only to proactive O&M events. S0 emergency events execute regardless of any configured window.
Set up notifications
To receive event alerts, configure notifications in Message Center:
Select the check box for cloud database failure or O&M notifications.
Set the message recipients. Assign database O&M personnel as recipients.
Select email as the notification method. Email offers better delivery reliability than internal message.
Figure 1. Entry point for notification settings in Message Center
Figure 2. ApsaraDB notification settings
Monitor events in real time with Cloud Monitor
For real-time O&M event updates or event-driven automated O&M, subscribe to Cloud Monitor system events. Cloud database services push system events related to the lifecycle of O&M events, such as subscription, start, completion, and cancellation.
For setup instructions, see Manage event subscriptions (Recommended). For a list of subscribable events, see Appendix 1: Related Cloud Monitor system events.
<details> <summary>Sample Cloud Monitor event payload</summary>
{
"eventId": "c864b30b-7f69-5f04-b0e7-8dfb0eabcfd9",
"product": "RDS",
"reason": "Host software/hardware upgrade",
"extra": {
"impactZh": "Transient instance disconnection",
"impactEn": "Transient instance disconnection",
"eventCode": "rds_apsaradb_transfer",
"eventNameEn": "Instance migration",
"eventNameZh": "Instance migration",
"switchTime": "2024-09-15T01:30:00+08:00",
"startTime": "2024-09-14T21:30:00+08:00",
"cancelCode": "OutOfGoodPerfBySoftHardwareUpgrade",
"detailCode": "HostSoftHardwareUpgrade",
"instanceInfo": ""
},
"instanceId": "rm-2ze9d66o65q1g02g6",
"eventType": "Maintenance",
"instanceComment": "rm-2ze9d66o65q1g02g6",
"instanceType": "Instance",
"publishTime": "2024-09-10T16:01:47+08:00"
}Key fields:
| Field | Description |
|---|---|
eventId | Unique event ID. The same event always has the same ID. |
extra.switchTime | Scheduled switchover time (ISO 8601). This is when the transient disconnection occurs if a switchover is performed. |
extra.startTime | Time the event enters the scheduling queue. |
extra.cancelCode | Cancellation risk code. See Appendix 2 for details. |
extra.detailCode | Detailed reason code. See Appendix 2 for details. |
</details>
View and reschedule events
Log on to the ApsaraDB RDS console.
In the left navigation pane, click Event Center.
If any events require you to schedule a time, a dialog box appears automatically. Schedule those events as soon as possible.
(Optional) On the Scheduled Events tab, configure a recurring time window for proactive O&M events.
Click Recurring Time Window Configuration.
In the panel that appears, set the recurring time and click OK.
The Recurring Time Window Configuration is a global setting for proactive database O&M events, but it does not apply to emergency threat remediation events. Without a recurring time window, proactive O&M events are automatically scheduled within the instance's maintenance window.
On the Scheduled Events tab, view event details. To change the switchover time for an event in the Scheduled state, select the instance and click Configure Scheduled Time.
The information displayed varies by event type. The information shown on the page takes precedence.
In the Configure Scheduled Time dialog box, set the switchover time and click OK.
To use the earliest available time, select Set Earliest Execution Time. The system fills in the earliest possible date and time, and the instance status changes to pending.
To set a custom time, clear Set Earliest Execution Time and enter the date and time manually.
The Scheduled Switchover Time cannot be later than the Latest Operation Time.
Event types and impacts
The information that is actually displayed on the page takes precedence.
Events that cause an instance switchover
The following events trigger an instance switchover. When the scheduled switchover time is reached, the event executes during the first maintenance window after that time.
Impact: The instance is in a read-only state for up to 30 seconds during the switchover to allow for full data synchronization. If a switchover is required, perform it during off-peak hours and make sure your application has a reconnection mechanism. For more information, see Impacts of an instance switchover. Usage of DMS and DTS is briefly interrupted; service resumes automatically after the switchover.
| Event type | Trigger |
|---|---|
| Instance migration | Host threats, expired hardware warranty, or operating system upgrades. The system migrates the instance—including non-HA and read-only instances—to a new server node. |
| Minor version upgrade | ApsaraDB periodically releases minor versions to add features or fix bugs. |
| Proxy minor version upgrade | ApsaraDB periodically releases minor versions for proxy nodes to improve proxy features or fix bugs. |
| Network upgrade | Network hardware upgrade to improve performance and stability. |
For minor version upgrades: Version differences vary between minor versions. Review the differences between the new version and your current version before the upgrade. See Update log of minor engine versions for ApsaraDB RDS for SQL Server.
For network upgrades: Some network upgrades involve a cross-zone migration, which changes the virtual IP address (VIP) of the instance. If your client connects using the VIP, the connection is interrupted. To prevent this, use the domain name provided by the instance as the endpoint and disable the DNS cache on your application and its server.
Events with other impacts
| Event type | Trigger | Impact |
|---|---|---|
| HA switchover | Host threats, expired hardware warranty, or operating system upgrades | Primary-secondary node switchover. Applies only to HA instances. |
| Host software or hardware upgrade | Host threats or software or hardware upgrades | Primary-secondary node switchover. |
| SSL certificate update | SSL certificate nears expiration | Maintains instance security and stability. |
| Backup mode upgrade | Switches the instance from logical backup to physical database and table backup | Enables faster recovery. |
Instance shutdown migration
Instance shutdown migration is triggered by host threats, expired hardware warranty, or operating system upgrades. The system migrates non-HA instances to a new server node.
After the scheduled switchover time:
Instance settings remain unchanged, including instance name, access port, tags, database accounts, and endpoints.
Service is typically unavailable for about 10 minutes. The actual duration depends on transaction volume before the switchover and total data volume.
If active DTS tasks exist, plan ahead. After migration, manually restart the affected tasks.
FAQ
Are scheduled events executed automatically? Can I cancel them?
Scheduled events are executed automatically to maintain instance stability and cannot be canceled. To minimize disruption, change the scheduled switchover time to off-peak hours and make sure your application has a reconnection mechanism. For instructions, see View and reschedule events above.