On 2016/11/09 12:55, Martin Pieuchot wrote: > > > > I'm using that, and I think many people using an IGP will be too (you > > want services - e.g. ssh, snmp, ntp, bgp - to stay working even when a > > physical interface is down - and at least where the IGP is OSPF you > > want those addresses hanging off an IFF_LOOPBACK interface, vether > > won't do). I bet the majority of people doing this use exactly lo1. > > Can't you use lo99 for that purpose?
Yes, it's the sort of thing that people following OpenBSD development can do pretty easily. (Here it's only 12 machines and it's a change that can be made in advance so not too painful - though from past experience it would be wise to restart ospf{,6}d rather than just reload config for this change). It might be quite painful for some isp or ixp if they only discover about the change after upgrading a remote machine though. > > [...] > > Much of the diff would stand, but not the automatic interface creation. > > That's the whole point of the diff. In that case is anything needed at all other than a doc change? "If you want to connect to a local address in an rdomain, add an lo interface to that domain". (I misunderstood it as things getting confused and using lo0 when another interface should have been used.)