Backend concept

Observability & Incident Triage

Metrics, logs, traces, alerts, service-level symptoms, and high-signal debugging during incidents.

Why this matters

Production systems need evidence fast; guessing during incidents burns time and confidence.

Choose signals that isolate customer impact, dependency failures, and bad deployments.

0 active misses 0 reviewed 0 games completed

No local review items for this concept yet.

Choose high-signal telemetry for common backend incidents.
Use metrics, logs, traces, deploy markers, and request IDs together.
Distinguish actionable alerts from noisy operational trivia.
Choose when to open, half-open, and close a circuit breaker.
Design fallbacks that preserve business correctness.
Combine timeouts, bounded retries, jitter, and bulkheads to reduce blast radius.

Relying only on average latency while p95 or p99 users suffer.
Diving into random logs before scoping by service, route, deploy, or request id.
Alerting on noisy resource blips instead of sustained symptoms or SLO burn.
Watching queue depth without message age, retry rate, or worker errors.
Using very long timeouts that hold threads and amplify outages.
Counting expected 4xx validation errors as dependency-health failures.

Start with the first game, then use local review history to revisit missed decisions.

Reliability Intermediate

Triage production incidents by choosing useful metrics, logs, traces, queue signals, database evidence, request ids, and alerting strategies.

Time: 6-9 minutes
Concept: Production observability, incident triage, metrics, logs, traces, and alerts

Reliability Intermediate

Diagnose dependency failures and choose circuit breaker, timeout, fallback, retry, half-open, and bulkhead strategies that reduce blast radius.

Time: 6-9 minutes
Concept: Circuit breakers, timeouts, retries, fallbacks, and dependency isolation

Scaling Intermediate

Route simulated traffic across backend servers using round robin, weighted round robin, least connections, and random strategies.