Host Smart Monitoring¶
Leveraging intelligent detection algorithms, the system periodically performs intelligent monitoring on host CPU and memory usage. When anomalies are detected in CPU or memory, it conducts in-depth root cause analysis to determine whether the host exhibits abnormal patterns such as sudden increases, sudden decreases, or sustained upward trends, thereby effectively monitoring the operational status and stability of the host.
Use Cases¶
Suitable for monitoring business hosts that require high stability and reliability, and supports providing analysis reports for generated anomaly events.
Detection Configuration¶
- Go to Monitoring > Smart Monitoring, create and select Host Smart Monitoring.
- Enter the monitor name.
- Adjust the detection conditions and delete unneeded rules.
- To filter hosts, expand the Detection Query (DQL) of a rule and modify the filtering conditions.
- After configuring event notifications, alert policy, and operation permissions, save.
View Events¶
In the monitor list, click View Related Events to view the anomaly events generated by the monitor.
Event Details Page¶
Click Events to view the detail page of an intelligent monitoring event, including event status, anomaly occurrence time, anomaly name, analysis report, extended fields, alert notifications, history, and related events.
-
Click Go to Monitor in the upper-right corner to view and adjust the Smart Monitoring configuration;
-
Click Export in the upper-right corner to choose Export JSON File or Export PDF File and obtain all key data corresponding to the current event.
Analysis Report
- Event content: Displays the event content configured for the monitor;
- Anomaly summary: Displays the hostname tag of the current anomaly, the anomaly analysis report details, and the anomaly value time series chart showing the anomaly trend;
- Anomaly analysis: An anomaly analysis dashboard where you can view basic information such as anomalous processes and CPU usage of the host;
- Host details: Displays the host integration runtime status, system information, and cloud provider information.
Note
When anomalies exist in multiple intervals, the Anomaly Summary > Anomaly Trend Chart and the Anomaly Analysis dashboard display the analysis for the first anomalous interval by default. You can click the Anomaly Trend Chart to switch; after switching, the Anomaly Analysis dashboard updates in sync.