From 1e3c57341a04b91f87877dd07b02eb6afde03e74 Mon Sep 17 00:00:00 2001
From: uvok cheetah
Date: Mon, 29 Apr 2024 19:02:07 +0200
Subject: Update AS learnings

---
 _posts/2024-03-17-further-learnings-from-running-my-as.md | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

(limited to '_posts')

diff --git a/_posts/2024-03-17-further-learnings-from-running-my-as.md b/_posts/2024-03-17-further-learnings-from-running-my-as.md
index b1a3839..416d4e2 100644
--- a/_posts/2024-03-17-further-learnings-from-running-my-as.md
+++ b/_posts/2024-03-17-further-learnings-from-running-my-as.md
@@ -2,6 +2,7 @@
 layout: post
 title: Further learnings from running my AS
 date: 2024-03-17 14:42 +0100
+last_modified_at: 2024-04-29 18:55 +02:00
 lang: en
 categories: ["tech"]
 ---
@@ -91,3 +92,17 @@ Futher info:
 - [DN42 wiki](https://dn42.eu/howto/Bird2#route-origin-authorization), scroll
   down for RPKI example configuration.
 - [APNIC: Deploying RPKI at Comcast](https://blog.apnic.net/2022/04/08/deploying-rpki-at-comcast/)
+
+**Update 2024-04-29**
+
+- I can actually "safely" discard everything with an AS path length >15.
+  I took a look at the routes, and there are (only/mostly?) routes with
+  *lots of* path prepending in that size range. Maybe I could go
+  *even lower*, but I got tired of making the birdc filter
+  (`show route where bgp_path !~ [...]`) longer and longer, and then scrolling
+  through the route output.
+- Another "oops!". I thought giving my hosting provider a higher BGP local preference
+  was a good idea, as all the other transit providers are connected via GRE tunnel.
+  Only I wondered why *on my other VPS* suddenly a huge amount of routes preferred
+  to go via IBGP. I facepalmed when I realized what I did. (BGP local pref attribute
+  is sent via IBGP sessions).
-- 
cgit v1.2.3