Sporadic DNS Failures Due to Resource Contention in CoreDNS Pods

Sporadic DNS resolution failures occurred due to resource contention in CoreDNS pods, which were not allocated enough CPU resources.

Find this helpful?

What Happened

CoreDNS pods were experiencing sporadic failures due to high CPU utilization. DNS resolution intermittently failed during peak load times.

Diagnosis Steps

1Used kubectl top pod to monitor resource usage and found that CoreDNS pods were CPU-bound.
2Monitored DNS query logs and found a correlation between high CPU usage and DNS resolution failures.

Root Cause

CoreDNS pods were not allocated sufficient CPU resources to handle the DNS query load during peak times.

Fix/Workaround

• Increased CPU resource requests and limits for CoreDNS pods.
• Enabled horizontal pod autoscaling for CoreDNS to scale during high demand.

Lessons Learned

CoreDNS should be adequately resourced, and autoscaling should be enabled to handle varying DNS query loads.

How to Avoid