Documentation
Home Page PolarDB PolarDB for Xscale PolarDB-X 1.0 (DRDS) Best Practice Migrate PolarDB-X 1.0 instances across zones Upgrade and migrate instances from a classic network architecture

Upgrade and migrate instances from a classic network architecture

Updated at:
Copy as MD
Product

PolarDB-X 1.0 instances on the classic network cannot communicate with VPC-based resources or use VPC security controls. This topic describes how to migrate a PolarDB-X 1.0 (formerly DRDS) instance from the classic network to a Virtual Private Cloud (VPC).

Prerequisites

Before you begin, make sure that:

  • The instance is a dedicated instance. If it is a shared instance, contact us to migrate it to a dedicated instance first.

  • The instance runs version 5.2.8 or later. If the version is earlier than 5.2.8, it is automatically upgraded to the latest version of 5.2.8 during migration.

  • If ApsaraDB RDS instances are attached to the PolarDB-X 1.0 instance, you must change their network type to VPC and fix database shard connections before proceeding. See Change the network type and Fix database shard connections.

Important

The migration causes a transient connection interruption lasting a few minutes. Make sure your application can automatically reconnect before you begin.

Migrate the instance

The migration follows two phases: first, add a VPC endpoint while keeping the classic network endpoint active (hybrid mode); then complete the cutover to a pure VPC architecture.

Phase 1: Switch to VPC

  1. Log on to the PolarDB-X 1.0 console.

  2. In the top navigation bar, select the region where the instance is deployed.

  3. In the left-side navigation pane, click Instances.

  4. Find the instance and click the instance ID to go to the Basic Information page.

  5. In the Common Actions section, click Switch to VPC.

  6. In the Switch to VPC dialog box, select the VPC and vSwitch.

  7. (Optional) Select Retain Original Classic Network Endpoint and configure the Expiration Time parameter.

    If you select Retain Original Classic Network Endpoint, the classic network endpoint is retained for a period of time after the VPC endpoint is generated. To ensure optimal security and performance, we recommend that you switch to the VPC endpoint at the earliest opportunity to prevent connection failures when the classic network endpoint expires.

  8. Click OK. The instance is now in hybrid mode with both a classic network endpoint and a VPC endpoint. Switch your business traffic from ECS instances on the classic network to ECS instances in the VPC, and connect to the PolarDB-X 1.0 instance using the VPC endpoint.

Phase 2: Complete the cutover

Migrate the instance from the hybrid architecture to a pure VPC architecture. For instructions, see Migrate an instance of the classic network and VPC hybrid type.

What's next

After the migration completes:

  • Verify that your application connects to the instance using the VPC endpoint.

  • Update any connection strings or configuration files that still reference the classic network endpoint.

  • If you retained the classic network endpoint temporarily, monitor the expiration time and switch to the VPC endpoint before it expires.

Previous：Migrate PolarDB-X 1.0 instances across zonesNext：Migrate an instance of the classic network and VPC hybrid type