Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
A reverse proxy receives requests on behalf of your servers, terminating TLS and applying policy before anything reaches the application. It is also a parsing boundary worth watching.

What Is a Reverse Proxy?
A reverse proxy stands in front of one or more servers and accepts requests on their behalf. To a client it looks like the application; to the application it looks like the client. nginx, HAProxy, Envoy, Traefik, and the edge of every CDN all perform this role, and most internet-facing applications now sit behind at least one, often several.
Termination and consolidation. TLS terminates at the proxy, so certificate management, cipher policy, and protocol versions are configured in one place rather than on every backend. Enabling HTTP/3, HSTS, or a new cipher suite becomes a single change.
Origin concealment. Backends hold private addresses and accept connections only from the proxy, so an attacker cannot reach them directly. This only holds if the origin is genuinely restricted — a firewall rule or mutual TLS, not merely an unpublished address, because addresses leak through historical DNS, certificate transparency logs, and outbound connections.
Policy enforcement. A web application firewall, rate limiting, bot management, authentication, and geographic restrictions all apply at the proxy, which means they apply consistently and can be evidenced in one configuration.
Availability. Load balancing, health checks, and failover, plus the ability to serve a maintenance page or shed load when a backend is struggling.
Legacy coverage. An application that cannot be modified can still gain modern TLS, security headers, and request filtering by being placed behind a proxy. This is frequently the only practical option for software a vendor no longer updates.
Putting a parser between the client and the application creates a new class of problem: two systems reading the same bytes and disagreeing about what they mean.
HTTP request smuggling is the sharpest example. When the proxy and the backend resolve a conflict between Content-Length and Transfer-Encoding differently, an attacker can hide a second request inside the first. The front end sees one request, the back end sees two, and the smuggled one is prefixed onto the next user’s connection — poisoning their response, stealing their session, or bypassing the proxy’s access rules entirely. Research by James Kettle from 2019 onwards showed how widespread the conditions were. Mitigate it by using HTTP/2 end to end where possible, rejecting requests that carry both headers, and keeping both components current.
Header trust. The proxy adds X-Forwarded-For so the application can log the real client address. If the application trusts that header without checking that the request came from the proxy, a client can forge it — defeating address-based rate limiting, poisoning logs, and sometimes bypassing access control. Strip incoming instances of forwarding headers at the proxy and configure the application to trust them only from the proxy’s address.
Path confusion. Differences in how the proxy and the backend normalise paths, decode percent-encoding, or handle traversal sequences can allow an attacker to reach a route the proxy intended to block. Access control implemented only in proxy configuration is fragile for this reason; enforce authorisation in the application as well.
Cache behaviour. Where the proxy caches, the cache key must be defined deliberately, or cache poisoning and cache deception become possible.
Lock the origin to the proxy with network rules and, better, mutual TLS or a shared secret header so an attacker who discovers the address still cannot connect. Keep the proxy itself patched — nginx CVE-2021-23017 in its resolver and the HAProxy header parsing issues of recent years show these components are targets in their own right. Do not let the proxy run as root beyond binding privileged ports. Log the full request including the client address and ship it to your SIEM, since the proxy is now where you see real client addresses before NAT and cloud translation obscure them.
And resist the temptation to implement authorisation purely in location blocks. A proxy rule is an excellent additional layer and a poor sole control. Our Web Application Security Services test the proxy and the application as the pair they are, which is where smuggling and path confusion findings actually come from.
The roles overlap heavily. A load balancer distributes traffic across backends for availability and capacity. A reverse proxy also terminates connections and can inspect, modify, cache, and filter requests. Most modern products do both, so the distinction is about which capability you are using rather than about two different kinds of device.
An attack exploiting disagreement between a front-end proxy and a back-end server about where one request ends and the next begins, usually through conflicting Content-Length and Transfer-Encoding headers. The attacker hides a second request inside the first, which is then prefixed onto another user’s connection. Use HTTP/2 end to end, reject ambiguous requests, and patch both components.
Only from your own proxy. Strip any incoming instances of forwarding headers at the proxy so a client cannot inject its own, then configure the application to accept them solely from the proxy’s address. Applications that trust the header unconditionally allow attackers to forge client addresses, defeating rate limiting and sometimes access control.
It can, but only if the origin actually refuses connections from anywhere else. An unpublished address is not concealment, because addresses leak through historical DNS records, certificate transparency logs, mail headers, and outbound connections the origin makes. Enforce origin access with firewall rules plus mutual TLS or a secret header that the proxy supplies.
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.