Restart a secondary database for SQL Server Cluster Edition to resolve performance issues - ApsaraDB RDS

Restart the secondary database of an ApsaraDB RDS for SQL Server instance on Cluster Edition to recover from exhausted connections or performance degradation — without affecting the primary database.

When to use

Restart the secondary database when you need to resolve issues on the secondary node:

Connections to the secondary node are exhausted
Query performance on the secondary node has degraded
The secondary node is slow or unresponsive

Prerequisites

Before you begin, confirm on the Basic Information page that the instance meets the following requirements:

Instance version: 2017 Enterprise Cluster Edition, 2019 Enterprise Cluster Edition, 2022 Enterprise Cluster Edition, or 2025 Enterprise Cluster Edition
Instance edition: Cluster Edition
Instance storage class: enterprise SSD (ESSD)
Billing method: subscription or pay-as-you-go (Serverless instances are not supported)
Instance status: Running

Potential impacts

The restart affects the secondary node and read-only request processing. The exact impact depends on your endpoint configuration:

Endpoint configuration	Impact on read-only requests	Impact on primary database
Read/write splitting not enabled	Transient disconnection on the secondary database	Not affected
Unified read-only endpoint enabled	Read-only requests are forwarded to the primary database during the restart	Extra traffic pressure
Secondary node read-only endpoint enabled	Read-only endpoint becomes unavailable; read-only requests are interrupted	—

After you start the restart, the instance status changes to Restarting. Management operations are blocked during this time, but connections to the primary database remain unaffected.

Warning

Perform this operation during off-peak hours to minimize impact on your workloads.

Restart the secondary database

Log on to the ApsaraDB RDS console. In the top navigation bar, select the region where the instance resides. Find the instance and click its ID.
In the upper-right corner of the page, click Restart Secondary Instance.
In the confirmation dialog, click OK.
Complete text message or multi-factor authentication (MFA) in the dialog box that appears.

FAQ

Will the endpoints and IP addresses of the primary and secondary instances change after the restart?

No. The restart does not involve network changes. Endpoints and IP addresses remain unchanged.

What affects the restart duration, and how can I reduce it?

A secondary database restart has two phases:

Stopping: Duration depends on the host, overall instance performance, and the state of dirty pages in memory.
Starting: Duration depends on the host, overall instance performance, and the size of transaction logs to recover. Recovery log size is related to the dirty pages state at shutdown and whether long-running transactions exist. It is also affected by the recovery interval parameter — use the default value unless you have a specific reason to change it. To modify it, see Configure the recovery interval.

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

API reference

API	Description
RestartDBInstance	Restarts a secondary database.