Caching
Cache-Control: free-recall review
Retrieval beats re-reading. For each prompt, say or write a full answer from memory before you open the model answer — the effort of recall is what makes the directive semantics stick.
Reconstruct the unit’s core distinctions — store vs revalidate, browser vs shared TTL, public vs private, immutable assets, and stale serving — without looking back at the lesson.
- 01Explain the difference between no-store and no-cache, and why conflating them causes real incidents.
- 02What is the relationship between max-age and s-maxage, and what breaks when you forget s-maxage behind a CDN?
- 03Why is the absence of private on an authenticated route a classic data-leak, and what is the senior reflex?
- 04Walk through why content-hashed filenames make public, max-age=31536000, immutable both aggressive and correct, and what must NOT be cached that way.
- 05What do stale-while-revalidate and stale-if-error do, and how do they differ from must-revalidate?
- 06What does Vary do, why is it a cache key rather than a safety directive, and where does it fail you?
If you could reconstruct each answer from memory, you hold the unit’s spine: no-store forbids storage while no-cache only forces revalidation, so the name of no-cache is the trap; max-age governs every cache and s-maxage overrides it for shared caches only, so forgetting s-maxage hands the CDN the wrong TTL; private keeps per-user responses off the CDN and its absence is the classic leak; content-hashed filenames make immutable one-year caching correct because a change means a new URL; stale-while-revalidate and stale-if-error trade staleness for latency and resilience against must-revalidate’s strictness; and Vary keys variants rather than excluding sharing, so safety must live in private/no-store.