Share via

Intermittent 7% CheckFailedPercent Failures in Azure Connection Monitor – Possible AMA Agent/Internal Issue?

Kedar Mane 0 Reputation points
2025-04-16T02:19:35.54+00:00

Scenario: I’m using Azure Connection Monitor to test connectivity from multiple VMs (including a SOAP VM) to Azure Firewall and on‑premises endpoints. All resources are in the East US region.

Problem:

Connection Monitor’s CheckFailedPercent metric shows a consistent ~7% failure rate over 30‑minute windows, then snaps back to 0%.

PsPing tests from the same VMs show occasional latency spikes (>3000 ms), but our ISP and on‑prem network have no issues.

When polling from other VMs (not the SOAP VM), failure rates drop significantly—suggesting the issue isn’t purely network‑side.

Tests created entirely within Azure (VM → Azure Firewall) still fail at ~7%, indicating this may be an internal Azure or AMA agent behavior rather than external latency.

What I’ve Tried:

Removed DNS checks from Connection Monitor.

Increased alert thresholds (5% → 15%) and extended lookback to 30 minutes—this reduces noise but doesn’t explain the root cause.

Engaged Microsoft support; they recommended threshold adjustments or a new VNet case, but I need to understand the underlying issue.

Question:

  • Has anyone experienced similar intermittent failures in CheckFailedPercent without external network issues?

Could this be a known AMA agent polling/caching behavior or another internal Azure component?

What diagnostic steps or configurations should I use to pinpoint and resolve these 7% packet‑loss‑like failures?

Tags: azure-connection-monitor azure-monitor azure-networking azure-monitor-agent

Azure Network Watcher
Azure Network Watcher
An Azure service that is used to monitor, diagnose, and gain insights into network performance and health.

1 answer

Sort by: Most helpful
Most helpful Newest Oldest
  1. G Sree Vidya 4,005 Reputation points Microsoft External Staff Moderator
    2025-04-16T07:53:21.3633333+00:00

    Hello Kedar Mane

    The Intermittent 7% CheckFailedPercent Failures in Azure Connection Monitor might be several caused such as listed below

    AMA Agent Synchronization, Delays or misconfigurations in the AMA agent can cause periodic polling failures.

    The SOAP VM might have unique application-level or agent-level issues (e.g., SOAP service time-outs)

    We request you to check below details:

    1. Validate AMA Agent Health:

    2. Review Connection Monitor Configuration

    • Go to Azure Network Watcher > Connection Monitor > <your-test> > Test configuration.
    • Ensure the test frequency (e.g., 30 seconds) and timeout (default 100 ms) align with your SOAP VM’s response time. Increase timeout if needed (e.g., 200 ms).
    • Check for the existence of the AMA Agent Troubleshooter directory on the machine to be diagnosed to confirm the installation of the agent troubleshooter using below document.

    Refer: https://learn.microsoft.com/en-us/azure/azure-monitor/agents/troubleshooter-ama-windows?tabs=WindowsPowerShell#prerequisites

    Hope this information is helpful! If you have any other questions or are still running into more issues, please let me know.

    If it was helpful, please click "Upvote and Accept Answer" on his post to let us know.

    We're here to help, so if you have any further questions, don't hesitate to ask.

    Thank You.

    0 comments

Your answer

Answers can be marked as Accepted Answers by the question author, which helps users to know the answer solved the author's problem.