Explore our new Cybersecurity Terms Glossary, explained by our experts.

View the Glossary View the Glossary
PRIVIA

Cybersecurity Terms Glossary

What Is BGP Hijacking?

BGP hijacking is the announcement of address space you do not own, pulling other people’s internet traffic onto your network. It has stolen cryptocurrency and taken YouTube offline.

Illustration for the cybersecurity term BGP Hijacking
Home / Blog / Cybersecurity Terms / What Is BGP Hijacking?
17 November 2026

What Is BGP Hijacking?

What Is BGP Hijacking?

The internet routes traffic on trust. Border Gateway Protocol, defined in RFC 4271, lets each autonomous system tell its neighbours which blocks of address space it can reach, and those neighbours pass the claim on. Nothing in the base protocol checks whether the claim is true. BGP hijacking is the act of announcing address space you have no authority over, so that traffic destined for somebody else arrives at you instead.

The Shapes It Takes

Three variants account for nearly everything seen in the wild.

An origin hijack is the simplest: an autonomous system announces a prefix it does not hold. Because routers prefer the most specific route available, announcing a /24 inside somebody else’s /16 wins globally even where the legitimate announcement is also present.

A path manipulation keeps the correct origin but forges the AS path, usually to make a route look shorter or to evade a filter that checks only the originating AS.

A route leak is the accidental cousin. A network that learns routes from one provider re-announces them to another, violating the usual customer-provider relationships and sucking in traffic it has no capacity to carry. RFC 7908 catalogues the types.

Incidents That Defined the Problem

On 24 February 2008, Pakistan Telecom announced a more specific prefix covering YouTube in order to block the service domestically. Its upstream provider propagated the announcement, and YouTube became unreachable for much of the world for roughly two hours.

On 24 April 2018, attackers hijacked prefixes belonging to Amazon’s Route 53 DNS service and ran their own DNS servers on them, redirecting visitors to the MyEtherWallet cryptocurrency site towards a phishing page with a self-signed certificate. Users who clicked through the browser warning lost around 150,000 US dollars in Ethereum. The hijack lasted about two hours and demonstrated the pattern clearly: routing is used to capture DNS, and DNS is used to capture credentials.

In April 2020, Rostelecom leaked routes for more than 8,000 prefixes belonging to major cloud and content providers. Incidents of this scale now occur several times a year, and most are misconfiguration rather than malice, which does not make the outage any shorter.

What a Hijack Lets an Attacker Do

Blackholing is the crudest outcome: traffic goes to a network that drops it, and the victim is simply offline. Interception is more serious, where the hijacker forwards traffic onwards after copying or modifying it, so the victim sees degraded latency rather than an outage. Fraudulent certificate issuance is the quietest risk of all, because a certificate authority performing domain validation over HTTP or DNS can be convinced by an attacker who controls the route to the validation target for a few minutes.

Defences That Actually Deploy

Resource Public Key Infrastructure is the control with real traction. A holder of address space publishes a signed Route Origin Authorisation stating which AS may originate which prefix, and networks performing Route Origin Validation drop announcements marked invalid. Signing your own space is a short task in the RIPE NCC portal for UK and European holders; the harder part is persuading your transit providers to filter. Major carriers and cloud providers now do. RPKI validates origin only, so path manipulation remains possible until ASPA or BGPsec see wider use.

Beyond that: publish and maintain your Internet Routing Registry objects, require your providers to build prefix filters from them, implement the maximum prefix limits and peer-lock mechanisms described by the MANRS initiative, and monitor. Services such as BGPmon, RIPE RIS, and Cloudflare Radar will alert you when somebody announces your space, and that alert is often the only warning you get. Treat it as an incident trigger with a named contact at every upstream provider, because remediation happens by phone call, not by configuration change on your side.

For organisations that depend on a handful of internet-facing services, pair routing monitoring with certificate transparency alerts and DNSSEC validation so that a short hijack cannot quietly turn into a persistent compromise. Our Network Security Services include a review of routing hygiene alongside the perimeter itself.

FAQ – Frequently Asked Questions

Can I Prevent Someone Hijacking My Prefixes?

You cannot stop another network making a false announcement, but you can make it ineffective. Publishing Route Origin Authorisations through your regional registry lets validating networks reject the announcement automatically. Keeping Internet Routing Registry objects accurate helps providers build correct prefix filters. Announcing your space at a reasonably specific length also limits the room for a more-specific hijack.

How Would I Know a Hijack Is Happening?

Usually from an external monitoring service rather than your own equipment, because your routers still see your own announcement as valid. Free and commercial tools including RIPE RIS, BGPmon, and Cloudflare Radar watch the global routing table and alert when an unexpected AS originates your prefixes. Sudden traffic loss from one region is another common first symptom.

Does HTTPS Protect Against BGP Hijacking?

It helps considerably. An attacker who redirects traffic cannot present a valid certificate for your domain without obtaining one, so users see a warning. The gap is that route control can be used to pass domain validation and obtain a genuine certificate. Certificate Authority Authorisation records and certificate transparency monitoring close most of that gap.

Is RPKI Widely Deployed in the UK?

Coverage is good and improving. Most large UK transit providers, the major cloud platforms, and the London Internet Exchange support Route Origin Validation, and RIPE NCC makes signing straightforward for UK address holders. The remaining weakness is the long tail of smaller networks worldwide that neither sign their space nor filter invalid routes.

Related Terms

You May Be Interested In These

What Is DNS and How Does It Work?

Cybersecurity Terms

17 November 2026

What Is DNS and How Does It Work?

Read More
What Are HTTP Security Headers?

Cybersecurity Terms

17 November 2026

What Are HTTP Security Headers?

Read More
What Is Cloud Backup and Recovery?

Cybersecurity Terms

17 November 2026

What Is Cloud Backup and Recovery?

Read More