An ASP.NET Core endpoint suddenly became slow in production. How would you investigate it?

Diagnose a slow ASP.NET Core API endpoint

Investigate a slow API by checking database queries, blocking async, dependency latency, serialization, logging, and runtime pressure.

1Say itAnswer out loud first

2CompareCompare with Core answer

3RefineRefine with Stronger and Advanced

TL;DR

I start by confirming the symptom with metrics instead of guessing. I check which endpoint is slow, whether it affects all users or one tenant, and whether latency changed after a deployment. Then I look at traces, logs, database time, downstream calls, CPU, memory, and thread pool behavior. The goal is to isolate the bottleneck before changing code.

Full answer below

Practice focus

production debugging, bottleneck isolation, observability, database/API/runtime reasoning, and mitigation planning

Answer first

Answer path

Core answer first, then checkpoints toward a stronger answer.

Start
Core answer
Start here. This is the direct spoken answer to practice first.
Why this question matters
Investigate a slow API by checking database queries, blocking async, dependency latency, serialization, logging, and runtime pressure. It matters in real backend work because load, cancellation, retries, and dependency latency decide whether the API stays predictable under pressure. The practical angle is production debugging, bottleneck isolation, observability, database/API/runtime reasoning, and mitigation planning, tied to a concrete production decision.
I start by confirming the symptom with metrics instead of guessing. I check which endpoint is slow, whether it affects all users or one tenant, and whether latency changed after a deployment. Then I look at traces, logs, database time, downstream calls, CPU, memory, and thread pool behavior. The goal is to isolate the bottleneck before changing code.
Checkpoint
Stronger answer
Optional upgrade with examples, trade-offs, and sharper phrasing.
A practical approach is comparing a slow request trace with a normal one. If most time is in SQL, I check query count, query plan, indexes, and N+1 patterns. If time is in HTTP calls, I check dependency latency, timeouts, retries, and connection exhaustion. The trade-off is that quick fixes like caching or raising timeouts can hide the problem; I prefer safe mitigation plus root cause analysis.

Review signals

Use these to catch weak phrasing, shallow reasoning, and risky claims.

Common mistakesCandidateCommon mistakesMistakes that weaken the answer or signal shallow understanding.Open details

Details

Optimizing random code without measuring
Looking only at average latency
Ignoring deployment history
Raising timeouts first
Checking only controller code

An ASP.NET Core endpoint suddenly became slow in production. How would you investigate it?

Practice focus

Answer path

Why this question matters

Review signals

Follow-up prompts

Continue the section

Tech stacks

Skill areas

Suggest an improvement