Compromised customer → false‑origin prefix hijack¶
Redirect or disrupt traffic for a target IPv4 prefix by abusing trust in the BGP control plane.
Phase 1: Getting into position¶
This phase matters because it explains why the later BGP behaviour is accepted.
Target a small or poorly secured customer
AS.Typical targets:
Small ISPs
Hosting providers
Enterprises with their own
AS
Motivation: weaker operational security, weaker monitoring.
Gain control over routing configuration. This does not require router exploitation. Common paths:
Compromised customer portal credentials
Access to automation pipelines
Insider access
Inherited access after acquisition or contract change
Influence which prefixes the AS announces. The AS is legitimately allowed to send BGP updates to its upstream.
Phase 2: The BGP control‑plane attack¶
The AS originates a prefix it does not own
The prefix belongs to another organisation
The origin AS is now wrong
From BGP’s perspective: this is a normal UPDATE
Upstream accepts the announcement. Why this happens:
Prefix‑lists are missing or outdated
Trust‑based customer relationship
No strict origin validation
“This customer has never caused problems before”
The route propagates further.
Other networks learn the false origin
Some select it as the best path
No alarms necessarily trigger
This is a textbook BGP control‑plane failure: the protocol is doing exactly what it was designed to do.
Phase 3: Network‑wide effects¶
Traffic shifts towards the attacker‑controlled path. Depending on topology:
Partial traffic redirection
Regional impact
Full hijack
Impact can vary without changing the attack:
Blackholing (drop traffic)
Interception (forward selectively)
Instability (route flapping if corrected and re‑announced)
Detection is ambiguous:
Looks identical to a misconfiguration
Origin
ASappears “legitimate”Blame is unclear in early stages
This ambiguity is why these attacks last longer than people expect.
Why this chain works so well (teaching value)¶
No protocol violation occurred
No exploit was used
No packet payloads were touched
Every step is operationally plausible
Responsibility is diffuse and slow to converge
In other words: the attack succeeds because the control plane is built on trust, not proof.
This chain demonstrates that:
BGP control‑plane attacks often start outside BGP
The attack itself is a single, boring announcement
Detection depends more on governance and monitoring than on clever tooling
Intent and accident are indistinguishable at first glance
Which is precisely why this attack still works in 2026.
Related¶
BGP hijacking & route leaks - General, IPv4 context
IPv4 prefix hijacking - Specific mechanics