Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
A CDN serves your content from edge locations near the user. It also absorbs volumetric attacks, hides your origin, and becomes a dependency you must plan around.

What Is a Content Delivery Network?
A content delivery network is a distributed set of servers that hold copies of your content close to the people requesting it. A visitor in Manchester is answered from a London or Manchester edge location rather than from a server in Virginia, which removes most of the round-trip latency. That is the performance argument. The security argument, which is often the stronger one, is that the edge becomes the only part of your infrastructure the public can reach.
DNS does the steering. Your hostname resolves to an address belonging to the provider, selected by anycast or by geographic policy so the client reaches the nearest healthy location. The edge server checks its cache. A hit is answered immediately; a miss is fetched from your origin, stored according to the cache headers you set, and returned.
Caching rules are where most incidents begin. The Cache-Control header, the Vary header, and the cache key configuration together decide what is stored and who else may receive it. Get them wrong and a personalised page containing one customer’s details is served to the next visitor. Large providers have shipped this bug, and so have many of their customers.
Modern platforms do more than cache. TLS terminates at the edge, so the provider holds a certificate and key for your domain. A web application firewall, bot management, rate limiting, and serverless functions typically run there too, which makes the CDN an enforcement point rather than just a cache.
Concentration is the first. On 8 June 2021 a configuration change at Fastly took a large share of the web offline for about an hour, including UK government and major news sites. On 19 July 2024 an unrelated vendor failure demonstrated the same lesson about shared dependencies. Treat your provider as a single point of failure with a documented fallback, even if that fallback is only a static holding page served elsewhere.
Trust is the second. Terminating TLS at the edge means the provider sees your plaintext traffic. That is acceptable for most content and needs a considered decision for regulated data; your UK GDPR records of processing should name the provider and the regions its edge locations occupy, and international transfer safeguards must be in place.
Cache poisoning is the third. Web cache poisoning abuses unkeyed inputs, such as a header the origin reflects but the cache ignores when building its key, to store an attacker-controlled response for everyone. Cache deception does the reverse, tricking the cache into storing a private page under a static-looking path. Both are configuration problems rather than provider flaws, and both are routine findings in application testing.
Lock the origin to the provider by address range and, better, by mutual TLS or a shared secret header so that an attacker who learns the address still cannot connect. Set explicit cache rules per path rather than relying on defaults, and mark anything authenticated as private. Define the cache key deliberately, including only the headers and query parameters that genuinely vary the response. Keep logs flowing from the edge into your SIEM, because the edge is now where client addresses and attack patterns are visible. And confirm that your origin still enforces authorisation, since an edge rule is a convenience, not a substitute for server-side checks.
Our Web Application Security Services include cache and edge configuration review alongside the application itself.
It absorbs volumetric attacks well, because the provider has far more capacity than a single origin and spreads traffic across many locations. Application-layer floods that look like legitimate requests need rate limiting and bot management on top. A CDN also does nothing if an attacker finds and targets your origin address directly, so origin access control matters as much as the edge.
Yes, when TLS terminates at the edge, which is the normal arrangement. The provider decrypts, applies its rules, and re-encrypts to your origin. For regulated or sensitive data this is a processing decision that needs to appear in your records of processing and contracts. Some providers offer key management options that reduce, but do not remove, the exposure.
An attacker sends a request containing an input the origin reflects into its response but the cache does not include in its cache key, such as an unusual header. The poisoned response is stored and then served to every subsequent visitor. The fix is to define cache keys explicitly, avoid reflecting untrusted headers, and never cache responses that vary on unkeyed input.
Yes, and then verify it stays hidden. Check historical DNS records, certificate transparency logs, mail headers, and any service your origin connects to outbound, because all of these have leaked origin addresses in practice. Pair concealment with firewall rules that accept connections only from your provider, so a leak does not immediately become exposure.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.