Chapter quiz · Production

Test what you learned.

8 questions on Production. Pass 70% to clear the chapter.

Quiz

Testing & Production — Chapter Quiz

Eight questions on supertest, pino logging, health/metrics endpoints, DEBUG=koa*, the missing-await-next bug, compression, cluster, app.proxy, and Docker deployment.

0/ 9

Question 1
1
How do you use supertest to test a Koa app without starting a real server?
Arequest(app).get(\"/\")Brequest(app.callback()).get(\"/\")Crequest(app.listen()).get(\"/\")Dsupertest.koa(app).get(\"/\")
app.callback() returns the raw (req, res) handler. Passing it to supertest lets it bind an ephemeral port in-process so you get fast, isolated HTTP tests without app.listen.
Question 2
2
What does setting app.proxy = true do?
AEnables HTTP/2 pushBTells Koa to trust X-Forwarded-For and X-Forwarded-Proto headers from a reverse proxyCEnables gzip compressionDProxies all requests to another server
When app.proxy = true, ctx.ip reads from X-Forwarded-For and ctx.secure / ctx.protocol read from X-Forwarded-Proto. Only set this when Koa sits behind a trusted reverse proxy (Nginx, Caddy, load balancer).
Question 3
3
Which environment variable enables Koa's internal debug logging?
ALOG_LEVEL=debugBDEBUG=koa*CKOA_DEBUG=1DNODE_ENV=debug
Koa uses the debug package namespaced under koa. Setting DEBUG=koa* in the shell emits detailed middleware dispatch and error logs, helpful for diagnosing routing and cascade issues.
Question 4
4
A route handler returns 404 for every request even though it clearly sets ctx.body. What is the most common cause?
AKoa requires ctx.status = 200 to be set explicitlyBAn upstream middleware forgot await next(), so the route handler never runsCctx.body must be set before await next()DThe router is not imported correctly
The classic Koa bug: if any middleware in the chain omits await next(), all downstream middleware and route handlers are skipped. Koa then responds with 404 because no body was ever set.
Question 5
5
Why is pino preferred over console.log for production logging in Koa apps?
Aconsole.log is not available in Node.js productionBpino emits structured JSON logs, is 5–10x faster than alternatives, and integrates with log aggregatorsCpino automatically masks PIIDconsole.log only works in development
pino uses a fast JSON serializer and an async transport to minimize I/O impact on the event loop. Structured JSON logs are easily parsed by tools like Datadog, Loki, and CloudWatch.
Question 6
6
A /healthz endpoint should return 200 only when the app and its critical dependencies (DB, cache) are reachable.
TTrueFFalse
A health endpoint that always returns 200 is useless for orchestrators (Kubernetes, ECS). A meaningful liveness/readiness check pings the DB and cache and returns 503 if any dependency is down.
Question 7
7
Use the koa-___ package to gzip/deflate responses and reduce payload sizes in production.
koa-compress negotiates Accept-Encoding with the client and compresses the response body using gzip, deflate, or brotli. It skips small payloads and already-compressed formats automatically.
Question 8
8
Which practices improve Koa app reliability and observability in production?
Select all that apply.
AUse Node.js cluster or a process manager (PM2) to utilise all CPU coresBExpose a /metrics endpoint in Prometheus format for scrapingCLog all errors with app.on(\"error\", handler) to catch unhandled Koa errorsDRun the app as root inside Docker for easier port binding
Clustering, structured metrics, and centralised error logging all raise production quality. Running as root inside a container is a security anti-pattern — use USER node or a non-root UID in your Dockerfile instead.