Restart a secondary instance of an RDS for SQL Server instance-ApsaraDB RDS(RDS)-阿里云帮助中心

RDS for SQL Server instances that run RDS Cluster Edition allow you to restart the secondary instance. If a secondary instance encounters issues such as connection exhaustion or performance bottlenecks, you can restart it to resolve them.

Prerequisites

The primary instance must meet the following requirements. You can verify this information on the Basic Information page:

Instance Version: 2017 Enterprise Cluster Edition, 2019 Enterprise Cluster Edition, 2022 Enterprise Cluster Edition, or 2025 Enterprise Cluster Edition
Series: RDS Cluster Edition
Storage Type: ESSD or Ultra Disk
Billing Method: subscription or pay-as-you-go. Serverless instances are not supported.
Status: Running

Impact

Restarting the secondary instance affects the secondary instance and how read-only requests are handled. We recommend that you perform this operation during off-peak hours. The impact varies depending on your configuration:
- If read/write splitting is disabled: The system directly restarts the secondary instance. During the restart, the secondary instance experiences a transient disconnection. The primary instance is not affected.
- If a unified read-only endpoint is enabled: During the restart, all read-only requests are forwarded to the primary instance. This prevents interruptions to read-only requests but increases the load on the primary instance.
- If a read-only endpoint for the secondary instance is enabled: The read-only endpoint for the secondary instance becomes unavailable during the restart, interrupting read-only requests.
After the restart begins, the instance status changes to Restarting. During this period, you cannot perform management operations, but connections to the primary instance are not affected.

Procedure

Log on to the ApsaraDB RDS console and go to the Instances page. In the top navigation bar, select the region in which the RDS instance resides. Then, find the RDS instance and click the ID of the instance.
In the upper-right corner of the page, click Restart Secondary Instance.
In the confirmation dialog box, click OK.
In the dialog box that appears, complete SMS or Multi-Factor Authentication (MFA) verification.

Related APIs

API	Description
RestartDBInstance	Restarts a secondary instance.

FAQ

Do the endpoints and IP addresses of the primary and secondary instances change after the secondary instance is restarted?

No. The restart operation does not involve network changes. The endpoints and IP addresses remain the same.

What are the main stages of a secondary instance restart, what factors affect the duration, and how can I reduce the restart time?

A secondary instance restart consists of two main stages:

Stop phase: The duration depends on factors such as the overall performance of the host and the instance, and the number of dirty pages in memory.
Start phase: The duration depends on factors such as the overall performance of the host and the instance, and the size of logs to be recovered. The log size is often related to the number of dirty pages at the time of shutdown and the presence of any long-running transactions. The duration also depends on the recovery interval parameter. We recommend that you use the default value. To change this parameter, see Configure the recovery interval.

To reduce the restart time, minimize long-running transactions or adjust the recovery interval parameter.