Routine inspection overview

This chapter describes the check items for routine inspections of OceanBase clusters, covering core metrics in performance, availability, resource usage, and data backup. To identify and resolve potential issues in OceanBase as soon as possible, we recommend that you pay close attention to the routine inspection results.

Overview

This chapter is divided into the following topics:

• Check cluster parameters

  OceanBase Database supports checking cluster parameters against best practices. During troubleshooting, you can check cluster parameters and modify the log level to obtain more parameter logs. After the problem is identified and resolved, restore the parameters to their original values.

• Check host status

  OceanBase Database supports checking host status, including but not limited to disk usage, host time, and host load.

• Check tenant resource usage

  OceanBase Database supports checking tenant resource usage, including but not limited to CPU, memory, number of replicas, and disk.

• Check cluster resource usage

  OceanBase Database supports checking cluster resource usage, including but not limited to CPU, memory, number of replicas, and disk.

• Check OBServer status

  OceanBase Database supports checking OBServer status, including whether the process exists and the number of replicas.

• Check NTP offset

  OceanBase Database supports checking the NTP offset of hosts. If the NTP offset of host nodes in an OceanBase cluster differs too much, leader election may occur abnormally.

• Deadlock detection

  OceanBase Database supports automatic distributed deadlock detection. When this feature is enabled, OceanBase Database can automatically detect and resolve deadlocks.

  Notice

  The check items listed here are only the ones that require particular attention, not a complete list of all inspection items.