OceanBase Autonomy Service integrates the routine O&M capabilities of OceanBase Cloud Platform (OCP). You can view the loads and health status of all clusters managed by OCP on the OceanBase Autonomy Service page.
Applicability
This topic applies only to OCP Enterprise Edition. OCP Community Edition does not support this feature.
Exception events
Cluster exception events that occurred in the last three days are displayed on the homepage of OceanBase Autonomy Service. You can scroll the page to view the exception events, which are listed in descending order of occurrence time.
View the information of exception events in the following columns: Cluster Name, Object, Name, Behavior, Current Status, Occurrence Time, Recovery Time, Duration, and Actions.
Click the name of a cluster to go to the overview page of the cluster.
Click SQL Diagnostics in the Actions column of an event to go to the real-time diagnostics page of the corresponding cluster. On this page, you can filter for the top SQL statements by the occurrence time and recovery time of the exception events.
Diagnostic information
The Diagnostic Information section displays the SQL, transaction, and session diagnostic information and optimization history.
SQL Diagnostics: displays the number of clusters with suspicious SQL statements, clusters with slow SQL statements, and clusters with high-risk SQL statements in the last 6 hours.
Transaction Diagnostics: displays the number of clusters with long-running transactions and clusters with suspended transactions in real time.
Session Diagnostics: displays the number of clusters with deadlocks in the last seven days.
Optimization History: displays the number of clusters with optimization history in the last seven days.
You can click a metric name or the number of clusters to filter clusters. The clusters that meet the filter condition are displayed in the list in the Cluster Details section. As shown in the following figure, if you click Clusters with Suspicious SQL Statements, only the clusters with suspicious SQL statements are displayed in the list in the Cluster Details section.
Cluster details
The Cluster Details section displays the SQL diagnostic information, real-time transactions, sessions, optimization history, and resource usage of each cluster.
You can click a statistical number in the column of an SQL type to go to the corresponding details page.
You can sort clusters by the number of suspicious SQL statements, slow SQL statements, high-risk SQL statements, long-running transactions, suspended transactions, deadlock sessions, active sessions, or optimization records or the total number of sessions in ascending or descending order.
You can click the name of a cluster to go to the autonomy service overview page of the cluster.
In the search window in the upper-right corner, you can search for a cluster by its name.