CoreDNS

CoreDNS Cache Prefetch Inefficiency

info
latencyUpdated Feb 9, 2026

Poorly configured prefetch settings fail to proactively refresh popular records before expiry, causing cache misses during high-traffic periods and increasing latency for frequently accessed services.

How to detect:

Monitor coredns_cache_prefetch_count versus coredns_cache_misses_count. If prefetch count is near zero but miss rate is high for known popular services, prefetch is not triggering. Check if frequently accessed records are expiring from cache.

Recommended action:

Configure aggressive prefetch for high-QPS clusters: 'prefetch 2 120s 30%' to refresh records requested 2+ times when 30% of TTL remains. For critical services, use 'prefetch 5 60s 20%'. Monitor coredns_cache_prefetch_count to verify prefetching is active. Combine with serve_stale 3600 for upstream failure tolerance.