Docs

Application Performance Management Plus

FAQ

How to fix the possible network packet latency caused by the old server-agent on some images?

Issue description

Old server-agent (below v1.0.31) may cause host network packet latency on non‑default VKE images.

Cause analysis

On those images, server-agent enables many protection rules, increasing logs and CPU scheduling delays, which may occasionally impact packet transfer.

Solutions

The server-agent collects host metrics, logs, and traces. For performance, it must run as a privileged container, so existing agents need privilege elevation.

Select one of the two solutions below based on your actual server-agent deployment method.

Method 1: Manually modify the YAML configuration

If you deployed server-agent manually via YAML, edit its YAML file to set it as a privileged container.

Run the kubectl command in your cluster to open the server-agent editing page.

Add privileged: true under securityContext to make server-agent a privileged container.

Method 2: Upgrade the VKE APMPlus add-on

If you deployed server-agent via the VKE console, upgrading the add-on to the latest version can fix the issue.

tip

The add-on defaults to the latest version. This issue has been fixed in v1.0.31.

In the cluster management page, select Addons in the left navigation pane, and find the apmplus-server-agent add-on in the Monitoring tab.

Click Upgrade in the upper-right corner of the add-on card to upgrade the add-on to the latest version.

Validate results

Go to the VKE console and check whether server-agent has been successfully started in your cluster. Expected result: All instances of the server-agent daemon set are in the running state.

Go to the APMPlus console and check for anomalies in host, Docker, and process monitoring data. Expected result: Monitoring data returns to normal.

Last updated: 2026.08.18 15:18:15