observability-and-instrumentation

작성자: addyosmani

코드를 계측하여 프로덕션 동작을 확인하고 진단할 수 있도록 합니다. 로깅, 메트릭, 트레이싱 또는 알림을 추가할 때 사용합니다. 프로덕션에서 실행되는 기능을 출시할 때, 해당 기능이 제대로 작동한다는 증거가 필요할 때 사용합니다. 프로덕션 문제가 보고되었지만 사용 가능한 데이터로는 무슨 일이 일어났는지 알 수 없을 때 사용합니다.

npx skills add https://github.com/addyosmani/agent-skills --skill observability-and-instrumentation

Observability and Instrumentation

Overview

Code you can't observe is code you can't operate. Observability is the ability to answer "what is the system doing and why?" from the outside, using the telemetry the code emits. Instrumentation is not a post-launch add-on — it's written alongside the feature, the same way tests are. If a feature ships without telemetry, the first user-reported bug becomes archaeology instead of a query.

When to Use

  • Building any feature that will run in production
  • Adding a new service, endpoint, background job, or external integration
  • A production incident took too long to diagnose ("we couldn't tell what happened")
  • Setting up or reviewing alerting rules
  • Reviewing a PR that adds I/O, retries, queues, or cross-service calls

NOT for:

  • Diagnosing a failure happening right now — use the debugging-and-error-recovery skill (observability is what makes that skill fast next time)
  • Profiling and optimizing measured slowness — use the performance-optimization skill
  • Launch-day monitoring checklists and rollback triggers — see the shipping-and-launch skill; this skill covers the instrumentation that feeds them

Process

1. Define "working" before instrumenting

Telemetry without a question is noise. Before adding any instrumentation, write down 2–4 questions an on-call engineer will ask about this feature:

FEATURE: checkout payment retry
QUESTIONS ON-CALL WILL ASK:
1. What fraction of payments succeed on first attempt vs after retry?
2. When a payment fails permanently, why? (provider error? timeout? validation?)
3. Is the payment provider slower than usual?
→ Every signal below must help answer one of these.

If you can't name the questions, you're not ready to instrument — you'll log everything and learn nothing.

2. Pick the right signal for each question

SignalAnswersCost profileExample
Structured log"What happened in this specific case?"Per-event; grows with trafficpayment_failed with provider error code
Metric"How often / how fast, in aggregate?"Fixed per series; cheap to queryp99 latency of provider calls
Trace"Where did time go across services?"Per-request; usually sampledOne slow checkout, broken down by hop

Rule of thumb: metrics tell you that something is wrong, traces tell you where, logs tell you why.

3. Structured logging

Log events, not prose. Every log line is a JSON object with a stable event name and machine-readable fields:

// BAD: string interpolation — unqueryable, inconsistent
logger.info(`Payment ${id} failed for user ${userId} after ${n} retries`);

// GOOD: stable event name + structured fields
logger.warn({
  event: 'payment_failed',
  paymentId: id,
  provider: 'stripe',
  errorCode: err.code,
  attempt: n,
}, 'payment failed');

Log levels — use them consistently:

LevelMeaningOn-call action
errorInvariant broken; someone may need to actInvestigate
warnDegraded but handled (retry succeeded, fallback used)Watch for trends
infoSignificant business event (order placed, job finished)None
debugDiagnostic detailOff in production by default

Correlation IDs are mandatory. Generate (or accept) a request ID at the system boundary and attach it to every log line, span, and outbound call. Without it, you cannot reconstruct a single request from interleaved logs:

// Express: child logger per request, ID propagated downstream
app.use((req, res, next) => {
  req.id = req.headers['x-request-id'] ?? crypto.randomUUID();
  req.log = logger.child({ requestId: req.id });
  res.setHeader('x-request-id', req.id);
  next();
});

Never log secrets, tokens, passwords, or full PII. This is a hard rule from the security-and-hardening skill — telemetry pipelines are a classic data-leak path. Allowlist fields; don't log whole request bodies.

4. Metrics

For request-driven services, instrument RED on every endpoint and every external dependency: Rate (requests/sec), Errors (failure rate), Duration (latency histogram, not average). For resources (queues, pools, hosts), use USE: Utilization, Saturation, Errors.

As with tracing, the vendor-neutral path is the OpenTelemetry metrics API (same SDK and context as step 5). The example below uses Prometheus' prom-client — one common backend choice, not the only one; the RED/USE and cardinality rules are identical either way.

import { Histogram } from 'prom-client';

const httpDuration = new Histogram({
  name: 'http_request_duration_seconds',
  help: 'HTTP request duration',
  labelNames: ['method', 'route', 'status_class'],  // '2xx', not '200'
  buckets: [0.05, 0.1, 0.25, 0.5, 1, 2.5, 5],
});

Cardinality is the failure mode. Every unique label combination is a separate time series. Labels must come from small, fixed sets (route template, status class, provider name). Never use user IDs, raw URLs, error messages, or other unbounded values as labels — that belongs in logs and traces.

OK as label:    route="/api/tasks/:id"   status_class="5xx"   provider="stripe"
NEVER a label:  user_id, email, request_id, full URL, error message text

Track averages never, percentiles always: an average hides the 1% of users having a terrible time. Use histograms and read p50/p95/p99.

5. Distributed tracing

Use OpenTelemetry — it's the vendor-neutral standard, and auto-instrumentation covers HTTP, gRPC, and common DB clients with near-zero code:

// tracing.ts — must be imported before anything else
import { NodeSDK } from '@opentelemetry/sdk-node';
import { getNodeAutoInstrumentations } from '@opentelemetry/auto-instrumentations-node';

const sdk = new NodeSDK({
  serviceName: 'checkout-service',
  instrumentations: [getNodeAutoInstrumentations()],
});
sdk.start();

Add manual spans only around meaningful internal units of work (e.g., applyDiscounts, chargeProvider) and attach the attributes on-call will filter by. Propagate context across every async boundary — HTTP headers, queue message metadata — or the trace dies at the gap. Sample head-based at a low rate by default; keep 100% of errors if your backend supports tail sampling.

6. Alerting

Alert on symptoms users feel, not on causes:

SYMPTOM (page-worthy):           CAUSE (dashboard, not a page):
error rate > 1% for 5 min        CPU at 85%
p99 latency > 2s                 one pod restarted
queue age > 10 min               disk at 70%

Cause-based alerts fire when nothing is wrong and miss failures you didn't predict. Symptom-based alerts fire exactly when users are hurt, regardless of the cause.

Rules for every alert you create:

  1. It must be actionable. If the response is "ignore it, it self-heals", delete the alert.
  2. It links to a runbook — even three lines: what it means, first query to run, escalation path.
  3. It has a threshold and duration justified by the SLO or by historical data, not by a guess.
  4. Use two severities only: page (user-facing, act now) and ticket (degradation, act this week). A third tier becomes noise that trains people to ignore everything.

7. Verify the telemetry itself

Instrumentation is code; it can be wrong. Before calling the work done, trigger the paths and look at the actual output:

  • Force an error in staging → find it in the logs by requestId, confirm fields are structured (not [object Object])
  • Send test traffic → confirm metric series appear with the expected labels and sane values
  • Follow one request across services in the tracing UI → no broken spans
  • Fire each new alert once (lower the threshold temporarily) → confirm it reaches the right channel and the runbook link works

Common Rationalizations

RationalizationReality
"I'll add logging after it works""After" becomes "after the first incident", which is the most expensive moment to discover you're blind. Instrument as you build.
"More logs = more observability"Unstructured noise makes incidents slower, not faster. Three queryable events beat three hundred prose lines.
"console.log is fine for now"Unstructured output can't be filtered, correlated, or alerted on. The structured logger costs five extra minutes once.
"We can just look at the dashboards when something breaks"Dashboards built without defined questions show you everything except the answer. Start from on-call questions.
"Alert on everything important, we'll tune later"A noisy pager trains people to ignore it. The tuning never happens; the missed real page does.
"User ID as a metric label makes debugging easier"It also makes your metrics backend fall over. High-cardinality lookups belong in logs and traces.
"Tracing is overkill for our two services"Two services already means cross-service latency questions logs can't answer. Auto-instrumentation makes the cost trivial.

Red Flags

  • A feature PR with retries, queues, or external calls and zero new telemetry
  • Log lines built by string interpolation instead of structured fields
  • No correlation/request ID — each log line is an orphan
  • Metrics labeled with user IDs, raw URLs, or error message text (cardinality bomb)
  • Latency tracked as an average with no percentiles
  • Alerts that fire daily and get acknowledged without action
  • Alerts on causes (CPU, memory) paging humans while user-facing error rate is unmonitored
  • Secrets, tokens, or full request bodies appearing in logs
  • "It works on my machine" as the only evidence a production feature is healthy

Verification

After instrumenting a feature, confirm:

  • The on-call questions for this feature are written down, and each signal maps to one
  • All log output is structured (JSON), with stable event names and a correlation ID on every line
  • No secrets, tokens, or unredacted PII in any log line (spot-check actual output)
  • RED metrics exist for every new endpoint and every external dependency, with bounded label sets
  • Latency is a histogram; p95/p99 are queryable
  • A single request can be followed end-to-end in the tracing UI without broken spans
  • Every new alert is symptom-based, has a runbook link, and was test-fired once
  • An induced failure in staging was located via telemetry alone, without reading the source

For the at-a-glance version of this list, including the pre-launch instrumentation gate, see references/observability-checklist.md.

addyosmani의 다른 스킬

accessibility
addyosmani
WCAG 2.2 지침에 따라 웹 접근성을 감사하고 개선합니다. "접근성 개선", "a11y 감사", "WCAG 준수", "스크린 리더 지원", "키보드 탐색", "접근 가능하게 만들기"와 같은 요청이 있을 때 사용하세요.
developmenttestingcode-review
web-quality-audit
addyosmani
성능, 접근성, SEO 및 모범 사례를 포괄하는 종합적인 웹 품질 감사입니다. "내 사이트 감사", "웹 품질 검토", "라이트하우스 감사 실행", "페이지 품질 확인", "내 웹사이트 최적화" 요청 시 사용하세요.
developmenttestingresearch
seo
addyosmani
검색 엔진 가시성과 순위를 최적화합니다. "SEO 개선", "검색 최적화", "메타 태그 수정", "구조화된 데이터 추가", "사이트맵 최적화", 또는 "검색 엔진 최적화"를 요청받을 때 사용하세요.
marketingresearchdevelopment
performance
addyosmani
웹 성능을 최적화하여 더 빠른 로딩과 더 나은 사용자 경험을 제공합니다. "사이트 속도 높이기", "성능 최적화", "로딩 시간 줄이기", "느린 로딩 수정", "페이지 속도 개선", "성능 감사" 요청 시 사용하세요.
developmenttesting
code-review-and-quality
addyosmani
다축 코드 리뷰를 수행합니다. 변경 사항을 병합하기 전에 사용하세요. 자신, 다른 에이전트 또는 사람이 작성한 코드를 검토할 때 사용하세요. 코드가 메인 브랜치에 들어가기 전에 여러 차원에서 코드 품질을 평가해야 할 때 사용하세요.
developmentcode-review
frontend-ui-engineering
addyosmani
프로덕션 품질의 접근 가능하고 반응형 사용자 인터페이스를 구축합니다. 인터페이스나 페이지를 만들거나 수정할 때, 컴포넌트를 생성할 때, 레이아웃을 구현할 때, WCAG 접근성 요구사항을 충족할 때, 상태를 관리할 때, 또는 결과물이 AI 생성물처럼 보이지 않고 프로덕션 품질처럼 보여야 할 때 사용하세요.
developmentdesign
security-and-hardening
addyosmani
코드를 취약점으로부터 강화합니다. 사용자 입력, 인증, 데이터 저장소 또는 외부 통합을 처리할 때 사용하세요. 신뢰할 수 없는 데이터를 받거나, 사용자 세션을 관리하거나, 타사 서비스와 상호작용하는 모든 기능을 구축할 때 사용하세요.
spec-driven-development
addyosmani
코딩 전에 명세를 작성합니다. 새 프로젝트, 기능 또는 중요한 변경을 시작할 때 아직 명세가 없는 경우 사용합니다. 요구사항이 불명확하거나, 모호하거나, 막연한 아이디어로만 존재할 때 사용합니다.
developmentdocumentproject-management