UIM cdm probe Physical Memory Usage Alerts Not Triggering

Products

• DX Unified Infrastructure Management (Nimsoft / UIM)

Issue/Introduction

• Critical server Physical Memory usage exceeds thresholds (e.g., >95%), but no alarms are visible in the Infrastructure Manager (IM) or Operator Console (OC).
• Review of the cdm probe logs shows that alarms are being generated internally (e.g., Average (# samples) Physical memory usage is now ##%).

Cause

This issue typically occurs due to one of the following:

1. NAS Pre-processing Rules: Rules in the nas probe may be configured to exclude alarm messages, Example containing the string "Average samples".
2. Configuration Mismatch: The alarm messages for Physical Error are not directly exposed in the standard cdm GUI. If the message token is incorrectly configured, the alarm may be suppressed or misidentified.

Resolution

To resolve this issue, manually update the alarm message mapping using the Raw Configure utility to ensure the correct tokens are used.

1. Open Infrastructure Manager.
2. Hold the Shift key on your keyboard, right-click the cdm probe, and select Raw Configure.
3. In the Raw Configure window, navigate to the following path in the left pane tree:
   • For Memory: memory -> alarm -> physical_errorOn the right-hand side, locate the key named message.
4. Double-click the message key to edit its value.
5. Change the value to match the exact Token/Name of the message you wish to use (e.g., change from PhysicalErrorProcesses to PhysicalError).
6. Click Apply, then click OK to save the changes and restart the probe.

Validation

Check the nas probe's Pre-processing tab to ensure no active rules are discarding messages based on the updated text string.