BGP Routing Vocabulary: AS, Prefix, Advertisement, and Policies Explained
Master the essential BGP vocabulary for network engineers: autonomous systems, route advertisements, path attributes, BGP policies, and professional English usage for networking.
Border Gateway Protocol (BGP) is the routing protocol that makes the internet work — it is the language routers use to exchange reachability information between networks worldwide. For network engineers, BGP mastery is non-negotiable, but so is the ability to discuss, document, and explain BGP in clear professional English. This guide covers the essential vocabulary with definitions, examples, and usage patterns.
Foundational BGP Vocabulary
Autonomous System (AS)
An autonomous system (AS) is a collection of IP networks and routers under the control of a single organisation that presents a common routing policy to the internet.
Each AS is identified by an Autonomous System Number (ASN) — a globally unique 16-bit or 32-bit number assigned by a Regional Internet Registry (RIR) such as ARIN, RIPE, or APNIC.
“Our AS peers with three upstream providers. All inbound routing policy is configured at the AS boundary.”
Related terms:
- ASN (Autonomous System Number) — the unique identifier for an AS
- Stub AS — an AS with only one upstream provider and no transit traffic
- Transit AS — an AS that carries traffic between other ASes
- Multi-homed AS — an AS connected to two or more upstream providers
BGP Peer / BGP Neighbour
BGP routers form peering sessions with one another. Two routers that have established a BGP session are called BGP peers (or BGP neighbours).
eBGP (External BGP) — a peering session between routers in different autonomous systems iBGP (Internal BGP) — a peering session between routers within the same autonomous system
“The BGP session with our upstream ISP dropped last night — the eBGP peer went unreachable after the fibre cut.”
Prefix
A prefix is an IP address range represented in CIDR notation (e.g., 203.0.113.0/24). When a BGP router advertises to peers that it can reach a particular prefix, those peers update their routing tables.
“We announced the /24 prefix to both upstream providers. One provider accepted the full /24, the other deaggregated it.”
Related terms:
- Prefix length — the number after the slash; longer prefix = more specific route
- Aggregate / Supernet — a shorter prefix that covers a larger address block (e.g., /16 containing /24s)
- More-specific prefix — a prefix with a longer mask that represents a smaller block
- Deaggregation — splitting a large prefix into smaller, more specific announcements
Route Advertisement / Announcement
When a router tells its BGP peers that it can reach a particular prefix, this is called a route advertisement or route announcement.
“After we configured the null route, the /24 was advertised to all upstream peers within 30 seconds.”
Key operations:
- Advertise — to announce a prefix to a BGP peer
- Withdraw — to remove an announced prefix (when the route is no longer valid)
- Originate — to announce a prefix that your AS owns (as opposed to re-advertising a received route)
- Propagate — to pass an advertisement received from one peer to another
BGP Path Attributes
BGP uses path attributes to describe and evaluate routes. These attributes allow routers to implement routing policy.
AS-PATH
The AS-PATH attribute is the list of autonomous system numbers a route has passed through, from origin to destination. It prevents routing loops and influences path selection.
“The route via AS64512 has an AS-PATH length of 3; the route via AS64513 has a path length of 5 — we prefer the shorter AS-PATH.”
AS-PATH prepending — deliberately adding your own ASN multiple times to the AS-PATH to make a route appear less preferred from external peers.
NEXT-HOP
The NEXT-HOP attribute identifies the IP address of the router the packet should be sent to next. In eBGP, the NEXT-HOP is automatically set to the advertising router’s interface IP.
“The NEXT-HOP address unreachable error in the BGP table means iBGP is not redistributing the route to the loopback interface correctly.”
LOCAL PREF (Local Preference)
LOCAL_PREF is used within an AS to tell iBGP routers which exit point to prefer when leaving the AS. Higher value = more preferred.
“We set LOCAL_PREF of 200 on routes received from our primary upstream and 100 on the backup — this ensures all traffic leaves via the primary link unless it goes down.”
MED (Multi-Exit Discriminator)
MED is advertised to external peers to suggest the preferred entry point into your AS when multiple entry points exist. Lower MED = more preferred.
“We lowered the MED on our East Coast router to attract inbound traffic from our US peer through that link.”
Communities
BGP communities are tags attached to route advertisements to carry policy information. They are used to influence how routes are handled by peers.
“We tag all transit routes with community 64512:100 so our route maps can apply the correct policy automatically.”
Well-known communities:
- NO_EXPORT — do not advertise this route to eBGP peers
- NO_ADVERTISE — do not advertise this route to any peer
- BLACKHOLE — route this prefix to null (commonly used for DDoS mitigation)
BGP Policy Vocabulary
Route Map
A route map is a policy tool used to filter or modify BGP routes based on conditions. Route maps can match prefixes, communities, AS-PATH patterns, and then apply actions.
“The route map on the upstream-facing interface strips the LOCAL_PREF attribute before advertising routes to the peer — we don’t want our internal policy leaking externally.”
Prefix List
A prefix list is a filter that matches IP prefixes. Used in BGP to control which routes are accepted from or sent to peers.
“We applied a prefix list to the neighbour statement to ensure we only accept the customer’s assigned /22 and reject any more-specific advertisements beyond /24.”
BGP Filtering
Common reason to filter BGP routes:
- Inbound filtering — control what routes you accept from a peer
- Outbound filtering — control what routes you advertise to a peer
- Bogon filtering — block advertisements of RFC 1918 private addresses and other non-routable prefixes
“The inbound filter ensures we only accept prefixes from legitimate ASes — bogon filtering is applied to all eBGP sessions by default.”
BGP Session States
BGP sessions move through states during establishment:
| State | Meaning |
|---|---|
| Idle | BGP is not attempting to connect |
| Connect | TCP connection in progress |
| Active | TCP connection failed; retrying |
| OpenSent | OPEN message sent; waiting for response |
| OpenConfirm | OPEN received; waiting for KEEPALIVE |
| Established | Session is up; routes are being exchanged |
“The BGP session has been stuck in Active state for 10 minutes — check for a firewall blocking TCP port 179.”
TCP port 179 — the port BGP uses for all session establishment.
Troubleshooting Vocabulary
| Term | Meaning |
|---|---|
| Flap | A route or session going up and down rapidly |
| Route dampening | Suppressing flapping routes to protect routing stability |
| Soft reset | Refreshing BGP policy without tearing down the session |
| Hard reset | Tearing down and re-establishing a BGP session |
| NLRI | Network Layer Reachability Information — the prefixes in a BGP update |
| RIB (Routing Information Base) | The table of routes learned from BGP |
| FIB (Forwarding Information Base) | The table of active routes installed in the data plane |
“The route was in the BGP RIB but not installed in the FIB — it turns out the local preference policy was blocking it from being selected as best route.”
Practice
Reinforce your networking vocabulary with the Networking Advanced exercise set and explore all resources on the Network Engineer learning path.