Instance Availability
• When enabling a product instance, you are advised to fully consider the application scenarios of business use and keep the instance server with a certain amount of resource redundancy, such as disk, CPU, memory, etc. Ensure that the usage does not exceed 70% to avoid Out Of Memory, disk exhaustion, high CPU usage, and other exceptions.
• To ensure high availability, it is recommended to use the primary/standby database for the production system.
• It is recommended to enable periodic full backups and incremental backups to ensure that the backup machine space is sufficient and not smaller than the instance disk.
• You are advised to modify the data synchronization mode of the primary/standby instances as needed. To avoid loss of data, you can set the mode to the synchronization mode; to ensure a fast business response speed, you can set the mode to asynchronous mode.
• To ensure the availability of the instance, it is recommended to upgrade the instance configuration promptly according to the resource usage, such as disk expansion, series upgrade, and CPU/memory upgrade.
Instance Operation and Maintenance
• Promptly clean up replication slots not in use to prevent them from blocking and affecting log collection.
• Timely execute VACUUM operation, especially in large-batch data operation scenarios.
• When deleting and modifying records, you need to execute SELECT first. Do not submit for execution before confirmation.
• Promptly clean up idle connections not in use. Avoid that the connections are not released for a long time on the business side.
• It is recommended to regularly monitor the age of the database. Once the database age exceeds 2 billion, it will be forced to shut down.
Secure Access
• Do not use a public network to access the database as much as possible. If the public network connection is required, you must first bind the public network EIP and strictly manage the trustlist.