Hi, > This is not implied at all. Should we add an explicit note to the draft that you can indeed set up more IPv6
> addresses on a single machine running the route server, and therefore it stays possible (and actually kinda desirable) > to keep two separate sessions? You can also run multiple virtual machines with the route server software, or even > multiple physical machines … I think there could still be some BGP implementations which allow only single sessions between the same pair of BGP Identifiers. So using VMs or VRFs or physical RS may be needed. > I wouldn’t support this approach, until we find out how to reliably detect the intent of de-peering without the session logic. Naturally the notion of peering or de-peering does not need to associated with the state of transport protocol at that time. But this topic is completely outside of the subject. I will review your draft and provide feedback if any. Thx, R. On Thu, Jul 23, 2026 at 10:49 AM Maria Matejka <[email protected]> wrote: > On Thu, Jul 23, 2026 at 10:35:25AM +0200, Robert Raszuk wrote: > > What you are proposing is transition to a single session - unless you run > v4 & v6 over separate v6 sessions. > > This is not implied at all. Should we add an explicit note to the draft > that you can indeed set up more IPv6 addresses on a single machine running > the route server, and therefore it stays possible (and actually kinda > desirable) to keep two separate sessions? > > You can also run multiple virtual machines with the route server software, > or even multiple physical machines … > > That would be clearly much more fragile as any session reset will > jeopardise the fun for both v4 and v6 reachable destinations. > > While not even a year ago, my home internet provider managed to ignore > their IPv6 misconfiguration in the local IXP for several days, causing all > my local IPv6 traffic go sideways, while IPv6 transit connections worked, > as well as all IPv4. > > Maybe if there was one session, there would notice earlier. > > I am not sure if this would be a good thing. > > I’m quite sure that joining the sessions would be better than now, > considering that even with split IPv4 and IPv6 sessions, they are still > often loosely tied if on the same machine … > > Yet, that’s not what the draft specifies. > > /* Maybe one day we will depart from session based BGP state bindings but > this moment does not seem to be happening in the near time. */ > > I wouldn’t support this approach, until we find out how to reliably detect > the intent of de-peering without the session logic. > > – > Maria Matejka (she/her) | BIRD Team Leader | CZ.NIC, z.s.p.o. >
_______________________________________________ GROW mailing list -- [email protected] To unsubscribe send an email to [email protected]
