Code RoomDNS split-brain after resolver migration
HardPrep Room Coding #2450

DNS split-brain after resolver migration

On-callNetworking & APIsSenior–Staff~35 min

Half your fleet suddenly can't reach an internal service db-primary.internal. Affected pods log 'no such host' / NXDOMAIN; the other half resolve it fine. Dashboards: error rate on the orders API is ~50% and bouncing; the unaffected half of pods are healthy. Recent context: 20 minutes ago an infra change migrated internal DNS from a self-hosted CoreDNS to a managed resolver, and someone also bumped record TTLs from 30s to 3600s last week. No application deploy. How do you triage and mitigate?

What a strong answer looks like

Stop the bleeding first (mitigate), then form hypotheses from real signals. Separate root cause from symptom, communicate status as you go, and close with what prevents a repeat.

0:00 of about 35 min
Which questions mattered is sealed until you submit. Telling you now would just be handing over the edge cases.