AwanCerahBGP
Home/Colombia/AS262191
Autonomous System

LIBERTY NETWORKS DE COLOMBIA S.A.S

AS262191ColombiaLACNIC · since 2011NSPPeering: Selective

AS262191

B84/100Good

BGP report

Ranked 117 of 441 networks in Colombia, 40 of 60 ISPs and transit networks. World rank 19,389.

  • RPKI ROA28/30
  • IRR route objects16/20
  • Upstream redundancy7/15
  • Global visibility10/10
  • IPv610/10
  • PeeringDB10/10
  • Prefix aggregation3/5

Computed from the route collector snapshot, IRR, RPKI and PeeringDB. How it is scored

Score history

RPKI ROA

Needs work
28/30

286 of 308 prefixes have a matching ROA (valid). 22 without a ROA.

Many large networks drop RPKI-invalid routes, and prefixes without a ROA are easier to hijack.

How to fix
  1. Create a ROA for every prefix in the portal of the RIR or NIR that holds it: MyAPNIC, the IDNIC member portal, the RIPE NCC LIR Portal, ARIN Online, MyAFRINIC or MiLACNIC.
  2. The ROA origin must be the AS that actually announces the prefix. Set maxLength to the announced length; a loose maxLength leaves room for sub-prefix hijacks.
Prefixes without a ROA (22)

IRR route objects

Needs work
16/20

149 prefixes have an exact route object, 121 are covered by a larger one, 33 only have objects with another origin, 5 have none.

Upstreams and IXPs build prefix filters from the IRR. Prefixes without a route object can be dropped silently.

How to fix
  1. Create route (IPv4) and route6 (IPv6) objects with your AS as origin in your RIR's IRR (for example APNIC via MyAPNIC, or IDNIC), or in RADB.
  2. Group your customers in one as-set and list it in PeeringDB so upstreams can build filters automatically.
No route object (5)
Route object with another origin (33)
Covered by a larger object (121)
2.152.8.0/242.152.9.0/242.152.10.0/242.152.11.0/242.152.12.0/242.152.13.0/242.152.14.0/242.152.15.0/24103.133.208.0/24103.133.209.0/24103.133.210.0/24103.133.211.0/24131.0.168.0/23131.0.170.0/24131.0.171.0/24138.121.12.0/24138.121.13.0/24138.121.14.0/24138.121.15.0/24168.197.68.0/24168.197.69.0/24168.197.70.0/24168.197.71.0/24168.227.244.0/24168.227.245.0/24168.227.246.0/24181.41.156.0/24181.41.157.0/24190.109.0.0/24190.109.1.0/24190.109.2.0/24190.109.3.0/24190.131.192.0/24190.131.196.0/24190.131.197.0/24190.131.198.0/24190.131.199.0/24190.131.200.0/24190.131.201.0/24190.131.202.0/24190.131.203.0/24190.131.204.0/24190.131.205.0/24190.131.206.0/24190.131.207.0/24190.131.208.0/24190.131.212.0/24190.131.213.0/24190.131.214.0/24190.131.215.0/24190.131.216.0/24190.131.217.0/24190.131.218.0/24190.131.219.0/24190.131.220.0/24190.131.221.0/24190.131.224.0/24190.131.225.0/24190.131.226.0/24190.131.227.0/24190.131.228.0/24190.131.229.0/24190.131.230.0/24190.131.231.0/24190.131.232.0/24190.131.234.0/24190.131.235.0/24190.131.236.0/24190.131.237.0/24190.131.238.0/24190.131.239.0/24190.131.240.0/24190.131.241.0/24190.131.242.0/24190.131.243.0/24190.131.244.0/24190.131.245.0/24190.131.246.0/24190.131.247.0/24190.131.248.0/24190.131.250.0/24190.131.251.0/24190.131.252.0/24190.131.253.0/24190.131.254.0/24190.131.255.0/24200.112.192.0/242800:b70:27::/482800:b70:28::/482800:b70:2a::/482800:b70:2e::/482800:b70:32::/482800:b70:34::/482800:b70:35::/482800:b70:36::/482800:b70:37::/482800:b70:3b::/482800:b70:3c::/482800:b70:3d::/482800:b70:3e::/482800:b70:41::/482800:b70:43::/482800:b70:44::/482800:b70:45::/482800:b70:47::/482800:b70:4c::/482800:b70:4d::/482800:b70:4e::/482800:b70:4f::/482800:b70:50::/482800:b70:51::/482800:b70:52::/482800:b70:53::/482800:b70:54::/482800:b70:55::/482800:b70:56::/482800:b70:57::/482800:b70:111::/482800:b70:402::/482800:b70:1000::/482800:b70:2001::/48

Upstream redundancy

Problem
7/15

1 upstream, 0 peer, 65 downstream.

With a single upstream, any outage there cuts your whole network off the internet.

How to fix
  1. Add a second upstream from a different provider, and join an IXP to peer directly with other networks.
Upstream (1)

Global visibility

Needs work
10/10

296 of 308 prefixes are seen by at least half of the full-table feeds worldwide.

Prefixes seen only partly are usually caught by filters (no ROA or route object), too specific, or announced to domestic IXPs only.

How to fix
  1. Check the ROA and route object of those prefixes, and keep them no longer than /24 for IPv4 or /48 for IPv6.
Prefixes with low visibility (12)

IPv6

Good
10/10

71 IPv6 prefixes announced.

Most mobile users and large services already run on IPv6.

PeeringDB

Good
10/10

Profile exists, listed at 1 IX.

Prospective peers and upstreams look up your contacts, policy and as-set in PeeringDB.

Prefix aggregation

Needs work
3/5

237 IPv4 prefixes could be summarized into 90.

Deaggregation grows the global routing table. Splitting for traffic engineering is still reasonable.

How to fix
  1. Announce the aggregate when the split is not needed. If it is intended, give every piece its own ROA and route object.

Dropping invalid routes (ROV)

Note

Seen passing on 11 RPKI-invalid routes of other networks in sampled paths. This suggests the AS does not drop all invalid routes, for example those from its customers.

Not part of the score. It is inferred from sample AS paths, so the routes may come from exempted customers.

How to fix
  1. Run an RPKI validator (for example Routinator or rpki-client) and reject invalid routes on your BGP sessions. MikroTik RouterOS 7 supports this through /routing rpki and the rpki-verify filter.
Example invalid routes (4)

Put the badge on your website

BGP report AS262191The badge updates with every snapshot.
HTML
<a href="https://bgp.awancerah.cloud/en/AS262191/rapor"><img src="https://bgp.awancerah.cloud/en/AS262191/rapor.svg" alt="BGP report AS262191"></a>
Markdown
[![BGP report AS262191](https://bgp.awancerah.cloud/en/AS262191/rapor.svg)](https://bgp.awancerah.cloud/en/AS262191/rapor)