mansaxel at besserwisser.org
Sat Oct 24 22:40:49 UTC 2015
Subject: IGP choice Date: Thu, Oct 22, 2015 at 06:57:01PM +0200 Quoting marcel.duregards at yahoo.fr (marcel.duregards at yahoo.fr):
> Hi everyone,
> Anybody from Yahoo to share experience on IGP choice ?
> IS-IS vs OSPF, why did you switch from one to the other, for what reason ?
> Same question could apply to other ISP, I'd like to heard some international
> ISP/carriers design choice, please.
We use IS-IS in our network mostly because I was around when a bunch
of NREN switched to IS-IS some 15 years ago, and it stuck. It is, as
has been noted, mostly a matter of preference, but there is one or two
technical arguments for IS-IS that tip the scales for me;
- One IGP for both v6 and v4. Mostly interesting if you are running a
lot of traffic outside VRFen. But nevertheless a good thing to keep v6
and v4 in sync.
- No leakage. Not many external peers speak IS-IS on their peering
interfaces, so chances are that even if I do, nothing will fall over.
This of course also applies to access interfaces, where my hosts won't
even have an OSI stack and thus won't try to process the frames.
The argument for OSPF mostly is that there are several FOSS OSPF dæmons
for Posixly machines, making it a good choice for things like anycast
name servers or similar. We do run it for precisely this setup.
Do read the presentation Vijay Gill made and that people keep pointing to.
It is a very good account of how to purge OSPF in favour of IS-IS.
Måns Nilsson primary/secondary/besserwisser/machina
MN-1334-RIPE +46 705 989668
I'm also pre-POURED pre-MEDITATED and pre-RAPHAELITE!!
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 181 bytes
Desc: Digital signature
More information about the NANOG