Hi, On Tue, Jan 07, 2025 at 04:55:59PM +0100, Tobias Fiebig wrote: > Moin, > > Okay, different assumptions at play. I assume that people connected > > to an IXP have some basic understanding about BGP prefix filters, and > > will not "accept /26 by default, just because one IXP is using one". > > > > Networks not connected to an IXP are not going to care either way > > anyway. > > Well, technically it is _their whole cone_ that has to accept the /26 > from upstream as soon as one upstream is on an IXP with a /26 (or, more > commonly: a /64) if you do not want to break PMTUD. ;-) > > It is essentially a slippery slope argument, and my point is 'we should > not get on that slope'.
Are you seeing IXPs with MTU steps today where PMTUD will be actually a
thing? You'd need to send a packet into the IXP that the next-ISP router
will not be able to forward *in* the ISP network to trigger a ICMP packet
with an IXP-owned source address...
But I'm not sure I understand what you are argueing for, here?
- announce and accept the /26, to not break PMTUD
- *not* announce the /26, because, RIB/FIB will explode?
consider me confused
Gert Doering
-- NetMaster
--
have you enabled IPv6 on something today...?
SpaceNet AG Vorstand: Sebastian v. Bomhard, Ingo Lalla,
Karin Schuler, Sebastian Cler
Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann
D-80807 Muenchen HRB: 136055 (AG Muenchen)
Tel: +49 (0)89/32356-444 USt-IdNr.: DE813185279
signature.asc
Description: PGP signature
_______________________________________________ GROW mailing list -- [email protected] To unsubscribe send an email to [email protected]
