i am getting coredns error [ERROR] plugin/errors: 2 2235095750251820864.5183428657909233302. HINFO: read udp 10.244.1.17:53534->168.63.129.16:53: i/o timeout

Question

coredns plugin error on monday onwards [ERROR] plugin/errors: <personal info> timeout

Answer 1

@Mohammed Suhail

We reviewed CoreDNS and cluster networking telemetry for the timeframe 2026-01-24 19:55 UTC through 2026-01-28 20:55 UTC, selected to broadly cover the reported period of concern. Additionally checing for future guidance on how similar issues can be prevented in the near future and what proactive measures can be taken to avoid a recurrence of this type of problem.

What was verified

• CoreDNS was running with two pods continuously available during the entire timeframe, with no restarts or scaling activity.
• Both pods remained ready and responsive throughout.
• CoreDNS actively handled DNS traffic, processing approximately 1.77 million requests and responses, evenly distributed across both pods.
• No DNS response failures (such as SERVFAIL) were observed.
• CPU and memory usage remained low and stable, with no abnormal resource pressure.
• DNS latency (P50, P90, P99) stayed consistent, showing no signs of degradation.
• Cluster nodes-maintained connectivity to CoreDNS; only one isolated, transient connectivity event was recorded, which does not indicate a broader issue.
• No upstream DNS forwarding activity was observed, indicating DNS resolution remained internal to the cluster.

Current assessment: Based on the available data, there is no evidence of a CoreDNS or AKS platform-level DNS issue during the reviewed timeframe. CoreDNS was healthy and serving requests as expected.

If DNS resolution errors were observed by workloads, the data suggests the cause is likely outside the CoreDNS or AKS platform scope, such as workload-specific DNS behavior, client-side configuration, or application-level handling.

Moving forward if the issue reccurs, having exact timestamps, affected workloads, and representative error details will enable more targeted analysis. The following public resources may also be useful for initial investigation

Microsoft References:

1. DNS troubleshooting in AKS: https://learn.microsoft.com/en-us/troubleshoot/azure/azure-kubernetes/connectivity/dns/basic-troubleshooting-dns-resolution-problems
2. Collecting AKS system diagnostics: https://learn.microsoft.com/en-us/troubleshoot/azure/azure-kubernetes/logs/capture-system-insights-from-aks

Answer 2

We reviewed CoreDNS and cluster networking telemetry for the timeframe 2026-01-24 19:55 UTC through 2026-01-28 20:55 UTC, selected to broadly cover the reported period of concern. Additionally checing for future guidance on how similar issues can be prevented in the near future and what proactive measures can be taken to avoid a recurrence of this type of problem.

What was verified

CoreDNS was running with two pods continuously available during the entire timeframe, with no restarts or scaling activity.

Both pods remained ready and responsive throughout.

CoreDNS actively handled DNS traffic, processing approximately 1.77 million requests and responses, evenly distributed across both pods.

No DNS response failures (such as SERVFAIL) were observed.

CPU and memory usage remained low and stable, with no abnormal resource pressure.

DNS latency (P50, P90, P99) stayed consistent, showing no signs of degradation.

Cluster nodes-maintained connectivity to CoreDNS; only one isolated, transient connectivity event was recorded, which does not indicate a broader issue.

No upstream DNS forwarding activity was observed, indicating DNS resolution remained internal to the cluster.

Current assessment: Based on the available data, there is no evidence of a CoreDNS or AKS platform-level DNS issue during the reviewed timeframe. CoreDNS was healthy and serving requests as expected.

If DNS resolution errors were observed by workloads, the data suggests the cause is likely outside the CoreDNS or AKS platform scope, such as workload-specific DNS behavior, client-side configuration, or application-level handling.

Moving forward if the issue reccurs, having exact timestamps, affected workloads, and representative error details will enable more targeted analysis. The following public resources may also be useful for initial investigation

Microsoft References:

1. DNS troubleshooting in AKS: https://learn.microsoft.com/en-us/troubleshoot/azure/azure-kubernetes/connectivity/dns/basic-troubleshooting-dns-resolution-problems
2. Collecting AKS system diagnostics: https://learn.microsoft.com/en-us/troubleshoot/azure/azure-kubernetes/logs/capture-system-insights-from-aks