Networking & Protocols
Stale-while-revalidate and cache stampede
At exactly T+3600 seconds, your most popular article’s cache entry expires across all edge POPs simultaneously. One thousand users request that page in the next second. Every one of them gets a cache miss. Every one triggers an origin fetch. Your origin sees 1000× normal traffic and starts timing out — and because some of those requests timed out, the CDN stored the 503 response as the new cached entry. Now every user sees a 503 for the next 3600 seconds.
The cache stampede problem
A cache stampede (also called thundering herd) happens when:
- A popular cached response expires.
- Many concurrent requests arrive simultaneously after expiry.
- All miss the cache and each independently fetches from origin.
- Origin is overwhelmed; some requests time out.
- The CDN caches the error responses — making things worse.
Without mitigation, the amplification factor is (requests/sec at expiry) × (origin response time). A page receiving 500 req/s with 200 ms origin response time can generate 100 simultaneous in-flight origin requests — 100× normal origin load.
stale-while-revalidate (SWR)
RFC 5861 defines stale-while-revalidate=<seconds>:
Cache-Control: public, max-age=60, stale-while-revalidate=604800After max-age=60 seconds expire:
- Serve the stale response immediately to every incoming request.
- Send one background revalidation request to origin.
- Cache becomes fresh again after origin responds.
- The staleness window is
max-age + stale-while-revalidate= 60 s + 7 days.
All 1000 concurrent users at T+60 still get a response in ~20 ms (stale edge hit), while origin sees exactly one revalidation request. The stampede never happens.
The trade-off: users may see content up to stale-while-revalidate seconds out of date. For a news article body (max-age=300, swr=3600) this means content can be 1 hour stale after max-age expires. For a breaking-news ticker, this is unacceptable — use a short SWR or no SWR.
- News article body (acceptable staleness 1h)
- max-age=300, stale-while-revalidate=3600
- Product listing (acceptable staleness 10 min)
- max-age=60, stale-while-revalidate=600
- Breaking news ticker (freshness critical)
- max-age=5, stale-while-revalidate=10
- Static asset (content-hashed URL)
- max-age=31536000, immutable — no SWR needed
- User-specific data (bank balance)
- no-store — no caching at all
stale-if-error: graceful degradation on origin failure
RFC 5861 also defines stale-if-error=<seconds>:
Cache-Control: public, max-age=3600, stale-if-error=86400When origin returns a 5xx or is unreachable, serve the stale cached response for up to stale-if-error seconds (1 day here) rather than returning an error to users. This is the CDN equivalent of a circuit breaker.
Use cases: marketing pages, documentation, article pages — anything where a 1-day-stale version is better than a 503. Not for checkout, payment, or any operation that must reflect real-time state.
The four stampede mitigations
| Strategy | How it works | Best for |
|---|---|---|
| Origin shield | Collapses all edge misses in a region to one origin request | All cache tiers |
| stale-while-revalidate | Serves stale immediately, one background revalidation | Mutable content, tolerable staleness |
| Request coalescing (singleflight) | Application-level: first miss starts origin fetch; others wait for the same result | Origin application layer |
| Probabilistic early expiration (PER / XFetch) | Stochastically refresh slightly before TTL, spreading the load over time | High-traffic caches |
Why this works
Why origin shield is the first line of defense. Without an origin shield, every CDN edge POP has a separate cache. When the same URL expires across 200 POPs in a region, all 200 independently fetch from origin. With an origin shield, all 200 edges route their misses through one shield node. The shield has its own cache (larger than any single edge); it makes at most one origin request per URL per region. SWR adds a second layer: even when the shield misses, users still see the stale response while one origin request is in flight. Both layers together mean a popular URL expiry generates exactly one origin request globally, not one per edge or one per concurrent user.
A news site experiences a 10× traffic spike from a viral article. Origin load alarm fires despite CDN being in front. Diagnose.
Why is stale-while-revalidate important for cache stampede defence?
Which RFC defines stale-while-revalidate and stale-if-error Cache-Control extensions?
Diagnose: users in two regions see different versions of the same page 2 hours after a deployment.
- 01Explain the cache stampede problem and why stale-while-revalidate prevents it.
- 02Under what conditions should you NOT use stale-while-revalidate?
- 03What does stale-if-error do and how does it differ from stale-while-revalidate?
The cache stampede problem: a popular cache entry expires; many concurrent users generate simultaneous origin requests; origin is overwhelmed and may start returning errors; those errors get cached. The four mitigations are: (1) origin shield, which collapses all edge misses in a region to one origin request; (2) stale-while-revalidate, which serves the stale response to all users while sending one background revalidation; (3) application-level request coalescing (singleflight), which prevents concurrent origin requests at the application layer; (4) probabilistic early expiration, which spreads revalidations across time. stale-if-error (RFC 5861) adds graceful degradation: on origin failure, serve the last cached version for up to N seconds instead of propagating errors. Match staleness windows to content correctness requirements — a news article can tolerate 10 minutes stale; a checkout price cannot tolerate 10 seconds.
appears again in162
- The journey of a request: seven stops from socket to responsejunior
- Accept and parse: from kernel queue to a typed requestmiddle
- Routing and middleware: choosing what runs, and in what ordermiddle
- Handler and response: from business logic to bytes on the wiremiddle
- Streaming and backpressure: when the client reads slower than you writesenior
- Timeouts and tail latency: budgets, deadlines, and the fan-out trapsenior
- Middleware and DI: the two patterns that shape every backendjunior
- Writing middleware: signatures, next(), and the three framework modelsmiddle
- Inversion of control: how dependencies reach a classmiddle
- DI scopes and lifecycles: singleton, request, transientmiddle
- DI as a testing seam: fakes, mocks, and the boundary that matterssenior
- DI containers in production: resolution graphs, circular deps, and when not tosenior
- Blocking vs non-blocking I/O: two ways to waitjunior
- The event loop: one thread, ordered phasesmiddle
- What blocks the loop: CPU work and sync callsmiddle
- Offloading CPU work: worker threads and the libuv poolmiddle
- Backpressure and bounded concurrencysenior
- Throughput under load: tail latency and saturationsenior
- Why pool: the cost of creating a connectionjunior
- Pool sizing: why bigger is not fastermiddle
- Acquisition and timeouts: the wait queue is the real latency dialmiddle
- Retry strategies: backoff, jitter, and thundering herdmiddle
- Observability, production failures, and global-scale designsenior
- Tasks, microtasks, and scheduler.yield()middle
- Timer accuracy, throttling, and idle workmiddle
- Node.js event loop: phases, nextTick, and loop lagsenior
- Rendering strategies: SSG, SSR, ISR, streaming, and hydrationjunior
- SSG, SSR, ISR, streaming, and RSC — how each worksmiddle
- Hydration cost: selective, progressive, islands, resumabilitymiddle
- Core Web Vitals: what LCP, INP, and CLS measurejunior
- LCP: four phases, one dominant costmiddle
- INP: input delay, processing, presentationmiddle
- Lab vs field: why the two disagree and how to use eachmiddle
- Metric tradeoffs, RUM attribution, and the CI+field loopsenior
- The full picture: URL to LCP to INP as a relay racejunior
- Eight layers traced: from the service worker to the second navigationmiddle
- Five canonical breaks: where production reliably diessenior
- The three-track method: reading traces and building a monitored systemsenior
- What an index is and how it speeds up queriesjunior
- The leading-column rule and composite index designmiddle
- Partial, expression, and covering indexesmiddle
- Index types: GIN, GiST, BRIN, Hash, Bloom, and HOT updatesmiddle
- Index-only scans, the Visibility Map, and INCLUDEsenior
- Production failure modes and the index audit playbooksenior
- Index design exercise: full-text search strategysenior
- EXPLAIN and execution plans: what the planner decides and whyjunior
- Scan types: Seq, Index, Bitmap, Index-Onlymiddle
- Join algorithms and the row-estimate cascademiddle
- pg_statistic, ANALYZE, and production observabilitymiddle
- Extended statistics: fixing correlated-column estimate failuressenior
- Plan cache, cost-constant tuning, and planner internalssenior
- Production failure modes and plan stabilitysenior
- Connection pools: amortising the cost of a Postgres backendjunior
- PgBouncer session, transaction, and statement modesmiddle
- Pool sizing: the (cores × 2) + spindles formula and the two-layer stackmiddle
- Pool exhaustion and idle-in-transaction: the 3 AM failure modemiddle
- Migrating to transaction mode: rollout playbook and PgBouncer 1.21 prepared statementsmiddle
- The Postgres process model and why raising max_connections degrades throughputsenior
- Pooler landscape 2026, serverless connection storms, and the full failure-mode taxonomysenior
- ADD COLUMN: instant in PG 11+ vs rewrite in older Postgresjunior
- The lock-queue failure mode: why instant DDL can freeze the databasemiddle
- Safe DDL patterns: NOT VALID, CONCURRENTLY, and unsafe-op fixesmiddle
- Migration failure taxonomy and production disciplinesenior
- Shard-key selection: hash, range, list, and directory strategiesmiddle
- Co-location and Citus: the invariant that makes sharding usablemiddle
- The hot-shard failure mode: detection, isolation, and durable policymiddle
- Online resharding, 2PC, and the operational cost of shardingsenior
- The seven acts: from CREATE TABLE to Citusjunior
- Acts 1–3 in depth: schema, indexes, and planner statisticsmiddle
- Acts 4–6 in depth: MVCC bloat, connection pooling, and safe migrationsmiddle
- Act 7 in depth: sharding, co-location, and the seven-tier tradeoff cascademiddle
- Observability, anti-patterns, and production triagesenior
- What the three signals are: logs, metrics, and tracesjunior
- Metrics and cardinality: the cost model of a time-series databasemiddle
- Logs and volume: the cost model of structured loggingmiddle
- Traces and sampling: the cost model of distributed tracingmiddle
- Join keys and exemplars: making the three signals composemiddle
- Observability 2.0: wide events and the cost shiftsenior
- Failure modes and engineering practice: cardinality budgets, PII, and samplingsenior
- Why structured logs exist: the diary vs the spreadsheetjunior
- The production log schema: fields every line must carrymiddle
- Log levels and alert routingmiddle
- Sampling strategies and log costmiddle
- PII redaction and log injectionsenior
- Trace context propagation in logssenior
- OTel Logs Data Model and audit logs as a subsystemsenior
- OTel signals, Semantic Conventions, and the OTLP wire formatmiddle
- Auto-instrumentation and manual spans: the 80/20 of OTelmiddle
- The OTel Collector: receivers, processors, exporters, and deployment patternsmiddle
- Sampling strategies: head, tail, and parent-basedmiddle
- Vendor neutrality, eBPF instrumentation, the Operator, and browser/serverless OTelsenior
- Operating the OTel Collector: reliability, version skew, failure modes, and governancesenior
- RED and USE: two checklists, one triage disciplinejunior
- Instrumenting RED in Prometheus: counters, histograms, and cardinality disciplinemiddle
- USE on Linux: CPU, memory, disk, network, and PSImiddle
- Golden signals, dashboard layout, and service mesh auto-REDmiddle
- Cardinality as a cost driver: labels, PII, exemplars, and samplingmiddle
- Native histograms, SLO tie-in, and production failure patternsmiddle
- Choosing SLIs and SLO targets: ratios, not feelingsmiddle
- Multi-window multi-burn-rate alerting: why AND beats ORmiddle
- Error budget policy, latency SLOs, and composite journeysmiddle
- Iceberg SLIs, composite SLO math, and SLA vs SLOsenior
- Flame graphs: reading the picture that shows where time goesjunior
- Sampling vs instrumentation profiling: why 99 Hz wins in productionmiddle
- Profile types: CPU, memory, off-CPU, mutex — which one to reach formiddle
- Continuous profiling: always-on flame graphs with eBPF and trace-id correlationmiddle
- How flame graphs are built from samples, and the production workflows that use themmiddle
- Linux perf, eBPF internals, PGO, and the limits of samplingsenior
- Profiling in production: security, war stories, OTel profiles, and the infrastructure designsenior
- The debugging funnel: SLO → RED → trace → profilejunior
- OTel architecture: one SDK, four signals, one wire formatmiddle
- Cost discipline: keeping observability under 5% of infra spendmiddle
- Scale, security, and the ROI of observable systemssenior
- Why profile first: measure where time actually goesjunior
- Amdahl''''s law and self-time: the ceiling on every speedup you can shipmiddle
- The measurement loop: microbench, macrobench, prod profile, observer effectmiddle
- Reading flame graphs: shapes, per-language profilers, and the 60-second scanmiddle
- Statistical baselines: why one run is not a measurementmiddle
- Profiler history and microbenchmark pitfalls: Knuth to GWPsenior
- Hardware counters, cold-start profiles, and profile securitysenior
- Continuous profiling at scale: costs, CI gates, trace correlation, and anti-patternssenior
- What makes a hot path: symptom vs causejunior
- Five shapes of hotspot: CPU, alloc, cache, lock, syscallmiddle
- Reading parent and child chains: where to apply the fixmiddle
- JIT deopt, the fix-and-verify loop, and PR-time profilingmiddle
- Hardware counters and Intel TMA: sub-category diagnosissenior
- False sharing and native-bridge hot pathssenior
- Hot paths in production: security, tail latency, and tooling lineagesenior
- Memory hierarchy: why the same O(N) loop can be 17x slowerjunior
- Row-major vs column-major: access order and the 9x gapjunior
- Branch prediction and branchless codemiddle
- Hardware prefetcher, TLB, and memory-level parallelismsenior
- GC basics: what the runtime taxes you forjunior
- GC algorithms: generational, concurrent, and per-runtimemiddle
- GC tradeoffs: pause, throughput, heap — and object poolingmiddle
- GC tuning: pacing, heap shape, and allocation observabilitymiddle
- GC internals: tri-color invariant, write barriers, and per-runtime deep-divessenior
- GC in production: observability, security, edge cases, and fleet governancesenior
- N+1: one logical operation, many round-tripsjunior
- Fix families: JOIN, IN, preload, and DataLoadermiddle
- Detecting N+1: query logs, APM traces, and CI gatesmiddle
- DataLoader: batching across resolver treesmiddle
- Cross-protocol N+1: HTTP fan-out and Redis MGETmiddle
- N+1 at scale: pool exhaustion, plan changes, and denormalisationsenior
- Batching: amortize fixed cost per operationjunior
- The batching window: size and wait timemiddle
- Batching in Kafka and Postgresmiddle
- io_uring and observability of batchingmiddle
- From Nagle to io_uring: evolution of batchingmiddle
- Backpressure, failure isolation, and batch security in productionsenior
- What a bundle actually costs: download, parse, compile, executejunior
- Core Web Vitals: LCP, INP, and CLSmiddle
- Code splitting: route-level, component-level, vendor splittingmiddle
- Tree shaking and compression: removing what you don''''t usemiddle
- Third-party scripts: the silent budget killermiddle
- CI enforcement and RUM: making budgets stickmiddle
- V8 JIT pipeline, HTTP priorities, and bundle securitysenior
- The performance loop: discipline, not a projectjunior
- Classify and fix: matching bottleneck families to remediesmiddle
- Observability stack and CI gates: catching regressions before they shipmiddle
- Incident to enforcement: SLO burn to verified fix in 35 minutesmiddle
- Culture, economics, and org-scale performancesenior