Edge cache verification

Measure cache hits, 304 revalidation and Brotli behavior

Try

This sends about 10 header-only requests with serial sampling, conditional requests, compression negotiation and Range. Use a real static asset; dynamic HTML normally should not be cached. Limit: 4 runs per minute.

What this tool can and cannot do

Each item below is physically impossible to conclude from a single header capture. That is the boundary between CDN inspection (configuration) and this tool (behavior).

CDN inspection →
Does the cache really hit?
Send serial samples and see when responses converge to HIT. A first miss is normal cold start; repeated misses indicate a configuration issue.
Does Age increase?
Age should advance on the same shared copy. A hit with a flat Age often means requests landed on different nodes.
Does a conditional request save bandwidth?
Send If-None-Match after receiving an ETag and expect 304. Only a real follow-up request can verify the saving.
Is Brotli supported or simply unused?
Negotiate br, gzip, zstd and identity separately to distinguish server capability from client preference.
Does Range support partial delivery?
Send a Range request and expect 206. Video seeking and large-file resume depend on it.
Where does the edge node land?
Extract airport codes from common response headers to check routing. Unknown codes are shown as codes, never invented cities.

What this tool can and cannot do

Single egress
All requests originate from one region, so a flat Age may reflect that egress using multiple nodes rather than a global issue.
Measured rankings →
Limited sampling
Three samples separate cold start from persistent misses but cannot calculate hit rate. When Age precision is insufficient, the tool withholds a conclusion.
Point to a static asset
Dynamic HTML should not be cached. Verify JS, CSS, images, fonts or video segments that are expected to hit.

Edge cache verification: test CDN hits, 304, Brotli and Range behavior

Send serial requests to a real static asset and verify CDN cache hits, Age, conditional 304 responses, compression negotiation, Range delivery and edge routing.

What is edge cache verification?

Edge cache verification tests what a CDN actually does rather than only reading its configuration. It requests static assets such as JS, CSS, images, fonts or video segments and observes warm-up, Age, conditional requests and compression behavior.

How does edge cache verification work?

The tool sends a serial, read-only request set from one fixed egress and preserves evidence from every response.

  1. 01

    Sample serially

    Request the same static asset repeatedly to separate a normal cold start from persistent MISS responses and compare header timing.

  2. 02

    Verify protocol behavior

    After receiving an ETag or Last-Modified, send a conditional request, negotiate br, gzip, zstd and identity, and test whether Range returns 206.

  3. 03

    Explain the boundary

    Use Age, Cache-Control, Vary and edge headers to make a recommendation. The sample describes one egress, not a global hit rate.

Edge cache verification FAQs

Why send multiple requests instead of one?

One response cannot separate a cold start from a persistent miss. Serial samples reveal HIT convergence, Age progression and latency changes.

Should I test HTML or a static file?

Use JS, CSS, images, fonts or video segments that are expected to be shared by the edge. User-specific dynamic HTML generally should not be cached.

Does the result represent every region?

No. Requests originate from one region and describe that egress. Use multi-node probes or measured rankings for global differences.

Why can a response be HIT while Age stays flat?

Requests may have landed on different edge nodes, or the object may be repeatedly invalidated. Check node identifiers, TTL and more regions.

Diagnose your CDN