● The switching causes a 30-second brief connection error. Make sure that your application is configured to automatically reconnect to your RDS instance.
● The switching causes changes to the virtual IP addresses (VIPs) of your RDS instance. We recommend that you connect your application to your RDS instance by using an endpoint.
● The VIP changes temporarily interrupt the connection to DRDS . You must update and view the connection information in the DRDS console.
● The VIP changes temporarily interrupt the connections to Data Management (DMS) and Data Transmission Service (DTS). After the switching is completed, these connections are restored to normal.
● If data is cached on a client, you can read data but cannot write data. After the switching is completed, we recommend that you immediately clear the cache on the client.
| Migration scenario | Description |
|---|---|
| Migration from one zone to another zone | The original zone where your RDS instance resides cannot ensure service performance due to issues such as heavy loads. |
| Migration from one zone to multiple zones | You want to implement disaster recovery across data centers. After the migration is complete, your RDS instance and its secondary RDS instance reside in different zones.The multi-zone deployment method delivers higher disaster recovery capabilities than the single-zone deployment method. If you select the single-zone deployment method, your database system can withstand server and rack failures. If you select the multi-zone deployment method, your database system can withstand data center failures. |
| Migration from multiple zones to one zone | You want to use specific features that are supported only for the single-zone deployment method. |