Code Room
On-callMedium
Question
Your service's latency dashboard shows requests getting slow and some returning 504 Gateway Timeout. Drilling in, the slowness is concentrated on endpoints that call one downstream service; endpoints that don't call it are fast. The downstream's own status page shows it's degraded. You didn't deploy. Your service's threads/connections are filling up waiting on it. How do you triage and protect your service?
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.
Learn the concepts
Loading whiteboard…
Run or narrate your approach, then ask the coach.