Updated on 2025-10-23 GMT+08:00

Setting a Diagnosis Threshold

You can set a diagnosis threshold to identify abnormal tables whose tablespace is above the threshold.

To use this function, you need to subscribe to Intelligent O&M first. For details, see Subscribing to Intelligent O&M.

Procedure

  1. Log in to the management console.
  2. Click in the upper left corner and select a region and a project.
  3. Click in the upper left corner of the page and choose Databases > Relational Database Service.
  4. On the Instances page, click the instance name to go to the Overview page.
  5. In the navigation pane, choose DBA Assistant > Real-Time Diagnosis.
  6. Click the Storage Analysis tab to view the table diagnosis results of your instance.
  7. Click next to Auto Diagnosis. In the displayed dialog box, configure a daily tablespace increase limit. The value ranges from 1 to 100,000,000, in MB.

    Figure 1 Configuring a daily tablespace increase limit

  8. Click OK. When the size of a single table is greater than the threshold, the system automatically filters the table and collects statistics on the number of such tables on the Tablespaces page. The system filters tables once a day.
  9. Click a number and view the details about abnormal tables on the Diagnosis Details page.

FAQ

  • Q: What can I do if there are tables whose tablespace has grown abnormally in the past day?

    A: Check tablespace fragmentation and reclaim fragmented space. Do not use DELETE to clear data. If you have any other questions, submit a service ticket.

  • Q: What is the impact of tables without primary keys on my DB instance?

    A: Tables without primary keys can cause slow SQL statements, affecting instance stability. You are advised to add primary keys to such tables to reduce the primary/standby replication delay.

  • Q: What is the impact of tables without indexes on my DB instance?

    A: Tables without indexes can cause slow SQL statements, affecting instance stability. You are advised to add indexes to table fields for more efficient query.