All Products
Search
Document Center

ApsaraDB RDS:Limits

Last Updated:Oct 23, 2023

This topic describes the limits and usage notes of ApsaraDB RDS.

After you create an ApsaraDB RDS instance, you do not need to perform basic database O&M operations, such as high availability configuration and security patching. However, you must take note of the following items.

Item

Description

Specification change

When you change the specifications of your RDS instance, the system may migrate the data of your RDS instance to a new RDS instance. The incremental data that is generated during the migration continues to be synchronized to the new RDS instance. After the migration is complete, the system switches your workloads over to the new RDS instance at the switching time that you specify. During the migration, service interruptions or transient connections may occur. The degree of impact varies across different RDS editions. If service interruptions or transient connections occur, you are unable to perform most of the operations that are related to databases, accounts, and network settings.

Important
  • If you use RDS Basic Edition, no secondary RDS instance is provided as a hot standby for your RDS instance. Therefore, your database service becomes unavailable for 30 minutes or longer when you change the specifications of the RDS instance. We recommend that you change the specifications of the RDS instance during off-peak hours to minimize the impact on your business.

  • If you use RDS editions other than RDS Basic Edition, your database service may experience a transient connection that lasts approximately 30 seconds when the specification change takes effect. This only slightly impacts your workloads. However, we still recommend that you change the specifications of your RDS instance during off-peak hours or make sure that your application is configured to automatically reconnect to your RDS instance.

  • In most cases, if your RDS instance is equipped with cloud disks and does not run RDS Basic Edition, no transient connections occur when you increase the storage capacity of the RDS instance.

  • If you use RDS Cluster Edition for ApsaraDB RDS for MySQL and your ApsaraDB RDS for MySQL cluster has only one secondary node, the read-only endpoint of the RDS cluster is inaccessible before the specification change takes effect. We recommend that you change the specifications during off-peak hours to ensure business continuity.

Version upgrade

When you upgrade the major engine version or update the minor engine version of your RDS instance, service interruptions or transient connections may occur. The degree of impact varies across different RDS editions. For example, you update the minor engine version of the RDS instance that runs MySQL 8.0 from 20230324 to 20230630 or upgrade the RDS instance from MySQL 5.5 to MySQL 5.6. If service interruptions or transient connections occur, you are unable to perform most of the operations that are related to databases, accounts, and network settings.

Important
  • If you use RDS Basic Edition, no secondary RDS instance is provided as a hot standby for your RDS instance. Therefore, your database service becomes unavailable for 30 minutes or longer when you upgrade the major engine version or update the minor engine version of the RDS instance. We recommend that you upgrade the major engine version or update the minor engine version of your RDS instance during off-peak hours to minimize the impact on your business.

  • If you use RDS editions other than RDS Basic Edition, your database service may experience a transient connection that lasts approximately 30 seconds when the specification change takes effect. This only slightly impacts your workloads. However, we still recommend that you upgrade the major engine version or update the minor engine version of your RDS instance during off-peak hours or make sure that your application is configured to automatically reconnect to your RDS instance.

  • If you use RDS Cluster Edition for ApsaraDB RDS for MySQL and your ApsaraDB RDS for MySQL cluster has only one secondary node, the read-only endpoint of the RDS cluster is inaccessible before the version upgrade of the secondary node takes effect. We commend that you upgrade the version of the secondary node during off-peak hours to ensure business continuity.

Failover

RDS High-availability Edition and RDS Cluster Edition provide a secondary RDS instance or secondary node that serves as a hot standby. If your primary RDS instance or primary node unexpectedly fails, your database system can fail over to the secondary RDS instance or secondary node within 30 seconds. A long period of time is required for the failover if data inconsistencies exist between the primary RDS instance or primary node and the secondary RDS instance or secondary node . During the failover, a transient connection that lasts approximately 30 seconds occurs. We recommend that you configure your application to automatically reconnect to your database system. This prevents interruptions to your workloads.

Network type change

When you change the network type of your RDS instance between the classic network and virtual private cloud (VPC), the server on which your application runs is disconnected from your RDS instance, and a new IP address is assigned to your RDS instance. After the migration is complete, we recommend that you immediately update the endpoint configuration on your application.

Data restoration

Before you restore the data of your RDS instance, we recommend that you back up the important data on your RDS instance to prevent data loss. We also recommend that you restore the data of your RDS instance by copying your data to a temporary RDS instance or a cloned RDS instance. After you verify the accuracy of the restored data, you can migrate the data back to your original RDS instance.

Storage capacity

If the storage capacity of your RDS instance is exhausted, the RDS instance is automatically locked and cannot provide services. We recommend that you check the storage usage on a regular basis.

Note

If the storage usage is abnormally high, troubleshoot the issue based on the description in What do I do if my ApsaraDB RDS for MySQL instance is locked because its storage capacity is exhausted?

Performance improvement

  • You must check whether the number of cores, memory capacity, input/output operations per second (IOPS), storage capacity, and maximum number of connections for your RDS instance are appropriate for your use case. If the specifications do not meet your requirements, you can optimize or upgrade your RDS instance.

  • We recommend that you routinely check whether your RDS instance runs as expected. This way, you can troubleshoot identified issues to ensure optimal performance. For example, you can check whether slow queries exist, whether SQL statements need to be optimized, or whether specific indexes are unnecessary or missing.

Limits on database engines