APMIA Disk Metrics Intermittently Dropping to Zero

Products

• DX SaaS

Issue/Introduction

Users of the APMIA agent (specifically version 25.10 on CentOS 7) may observe that disk-related metrics, such as Percent Available and Percent Used, intermittently drop to zero simultaneously. This behavior triggers false alarms (e.g. VM Available Disk Space has breached the CRITICAL threshold).

In a healthy state, if "Percent Available" decreases, "Percent Used" should increase. If both metrics drop to zero at once, it indicates a monitoring ingestion issue rather than a genuine disk capacity problem.

Environment

APMIA Agent (Version 25.10.x)

Cause

The issue is related to how the latest APMIA agent handles metric ingestion from the SystemEdge component. Internal investigations identified a configuration gap in the ingestion logic that causes intermittent zero-value reporting for specific storage groups.

Resolution

To resolve this issue, enable NASS metric ingestion in the APMIA agent configuration:

1. Locate the agent configuration file IntroscopeAgent.profile.
2. Add or update the following property:
   properties

    

   introscope.agent.nass.metric.ingestion.enabled=true

3. Restart the APMIA agent to apply the changes.

Additional Optimization (Optional)

If strange metric behavior persists, engineering suggests increasing the clamping limit for the corresponding storage group in the host monitoring extension.

• File: bundle.properties (located under /extensions/hostmonitor-.../)
• Properties to update:
  •  

    ntroscope.agent.hostmonitor.profiles.localhost.linux.group.storage.clamp=50
    introscope.agent.hostmonitor.profiles.localhost.windows.group.storage.clamp=50