Running Your Own Autonomous System with BGP and FRR

For SRE teams managing global infrastructure, relying on a single upstream provider creates a critical single point of failure. Running your own Autonomous System (AS) with Border Gateway Protocol (BGP) gives you control over IP address announcements, enables multi-homing across providers, and provides resilience against upstream outages. FRR (Free Range Routing) makes this accessible with a mature, open-source routing suite.

What is an Autonomous System?

An Autonomous System is a collection of IP networks under a single administrative domain that presents a unified routing policy to the internet. Every AS has a unique number (ASN) assigned by a Regional Internet Registry. When you run your own AS, you can:

• Announce your own IP prefixes to multiple upstream providers
• Control inbound and outbound traffic paths with BGP policies
• Maintain connectivity even when one upstream fails
• Implement traffic engineering for latency optimization

Why FRR?

FRR is a fork of Quagga that has become the de facto standard for open-source routing. Key advantages include:

• Protocol support: BGP, OSPF, IS-IS, MPLS, and more
• Active development: Regular releases with modern features
• Production proven: Used by major cloud providers and network operators
• vtysh interface: Familiar Cisco-like CLI for network engineers

Basic BGP Configuration

After installing FRR, configure BGP peering in /etc/frr/frr.conf:

router bgp 65001
 bgp router-id 192.0.2.1
 neighbor 198.51.100.1 remote-as 65002
 neighbor 198.51.100.1 description upstream-provider-a
 !
 address-family ipv4 unicast
  network 203.0.113.0/24
  neighbor 198.51.100.1 soft-reconfiguration inbound
  neighbor 198.51.100.1 prefix-list EXPORT out
 exit-address-family

Define prefix lists to control announcements:

ip prefix-list EXPORT seq 10 permit 203.0.113.0/24

Operational Tips

• Use route filtering: Always apply prefix lists to prevent accidental route leaks
• Monitor sessions: Track BGP session state with show bgp summary and alert on flaps
• Implement BFD: Bidirectional Forwarding Detection speeds up failover detection
• Document policies: Maintain clear records of your routing policies for incident response
• Test changes: Use looking glass servers to verify your announcements appear correctly

Getting Started

Before diving in, you will need an ASN and IP allocation from your Regional Internet Registry (RIPE, ARIN, APNIC, etc.), plus agreements with at least two upstream providers for true multi-homing. Start with a test lab using private ASNs (64512-65534) before going live.

Conclusion

Running your own AS with BGP and FRR transforms your infrastructure from a tenant to a peer on the internet. The operational complexity is real, but the resilience and control benefits are substantial for organizations with global reach.

---

Curious how Akmatori can help monitor your BGP sessions and routing infrastructure? Check out our AI-powered platform for intelligent alerting on routing anomalies. Built by engineers who understand networking, powered by Gcore's global infrastructure.