このページは日本語には対応しておりません。随時翻訳に取り組んでいます。翻訳に関してご質問やご意見ございましたら、お気軽にご連絡ください。

The Database Monitoring AlwaysOn Clusters view enables you to detect data synchronization issues, understand availability group behavior, and identify cluster bottlenecks in SQL Server availability groups.

To access the AlwaysOn Clusters view, navigate to the APM > Database Monitoring > Databases tab and select AlwaysOn Clusters.

Determine the health of your nodes

Use the AlwaysOn Clusters view to evaluate the health of your SQL Server availability groups. When selected, the page shows a color-coded visualization based on the current status of the primary (P) and secondary (S) nodes in each availability group.

To identify groups that are experiencing issues, use the status filters to show groups with nodes that are Reverting, Not Synchronizing, and so forth. You can also use the timeseries graphs to spot unusual performance activity in your clusters based on log, redo, and secondary lag time metrics.

View SQL Server AlwaysOn groups

Analyze historical metrics

To evaluate how node synchronization states have fluctuated over time, select an availability group to open the details side panel. The Historical Synchronization States graph at the top of the panel shows the status of each node over the selected timeframe.

View additional information about the secondary nodes and their associated databases on the Secondary Nodes tab. You can also use the timeseries graphs on the Metrics tab to spot abnormal behavior in individual nodes and databases based on view send, redo, and lag metrics.

View SQL Server AlwaysOn groups

Further Reading