Viewing graphs and faults for anomalies

Last published : Jul 06, 2026
You can view the trend graphs and faults in real time in the UI and receive email notifications for the detected anomalies.
Viewing graphs
You can view the graphs for entropy changes and for the anomalies that are detected during the read, write, and rename operations performed on a file system.
To view graphs:
  1. In the Management Server console, go to the Server perspective and expandManage in the left pane.
  2. Search or locate the host in the left pane.
  3. Expand the host and expand Volumes.
  4. Click the volume that has the file system on which anomaly detection is enabled.
  5. Click Anomaly Watch.
The following trend graphs are shown:
  • Rename Count: Shows measure of rename operations analyzed over a time series.
    • Green contours: Represent the count of actual rename operations.
    • Blue contours: Represent the count of rename operations predicted by the ML engine.
    • Blue range: Represents the accepted range of rename operations predicted by the ML engine. Any count beyond this threshold is considered a deviation and shown as an anomaly indicated in red.
  • Read Count: Shows measure of read operations analyzed over a time series.
    • Green contours: Represent the count of actual read operations.
    • Blue contours: Represent the count of read operations predicted by the ML engine.
    • Blue range: Represents the accepted range of read operations predicted by the ML engine. Any count beyond this threshold is considered a deviation and shown as an anomaly indicated in red.
  • Write count: Shows measure of write operations analyzed over a time series.
    • Green contours: Represent the count of actual write operations.
    • Blue contours: Represent the count of write operations predicted by the ML engine.
    • Blue range: Represents the accepted range of write operations predicted by the ML engine. Any count beyond this threshold is considered a deviation and shown as an anomaly indicated in red.
  • High Entropy: Shows the entropy levels analyzed over a time series.
    • Green contour: Represents the entropy level for the actual data.
    • Blue range: Represents the accepted entropy range that is predicted by the ML engine.
Viewing faults and receiving email notifications
Faults are generated if the count for file system attributes deviates from the ML engine predicted range. If a fault occurs and rule is set with the fault topic event.alert.vom.fs.anomaly , email notification is sent to the recipients for the generated faults. Ensure that you provide the details for the SMTP settings before setting the email notification for a fault.
The following fault is generated if an anomaly is detected:
The monitored-attributes patterns observed on the file-system-name file system indicate a few anomalies. Verify the data on the file system. Note: This alert originates from an early-stage ML model and may produce false positives if it has not yet fully learned the relevant patterns.
Related information