Kubernetes API Connection Resets and Health Check Failures Caused by Application Polling

Products

• VMware Telco Cloud Automation
• VMware Tanzu Kubernetes Grid

Issue/Introduction

The application experiences health check failures across deployment sites due to intermittent connection resets when communicating with the Kubernetes/Kubelet API endpoint.

• Application-level health check reporting failures.

• Connection termination/resets between the application pod and the Kubernetes/Kubelet API server.

• application pod CPU and memory utilization remain within normal operational limits.

• No node-level failures, infrastructure anomalies, or pod evictions detected.

Environment

TCA 3.2

TKGm 2.5.4

Cause

The application generates an excessively high volume of unthrottled API requests to the Kubernetes/Kubelet API server, exceeding acceptable request thresholds and triggering API connection resets.

Resolution

• Review the application API polling mechanism and request frequencies.

• Update the application codebase or configuration to implement rate-limiting, request throttling, or exponential backoff logic when querying the Kubernetes/Kubelet API.

• Validate that application API polling request rates remain within sustainable API server limits.