[quagga-dev 7212] Re: [PATCH 01/17] zebra: Initial RIB cleanup, and eliminate lots of redundant IPv4/IPv6 code.
joakim.tjernlund at transmode.se
Thu Aug 20 19:40:20 BST 2009
paul at jakma.org wrote on 20/08/2009 18:18:05:
> On Thu, 20 Aug 2009, Joakim Tjernlund wrote:
> > Exactly. In OSPF a nexthop is always an interface + optional
> > nexthop IP address. Would be good if you could express this too
> > when needed.
> Sure, but that's an implementation detail. The route itself doesn't
> need that restriction added, as such.
It is not a restriction, it is a feature to be used when needed.
> >> Also, A must be on a multi-access link, otherwise you could just say
> >> <I>.
> > How about IP aliases on a link? There interface isn't enough.
> On a PtP link? No gateway IP is needed. You just send the packet on
> the link - there's only one other host it can reach.
I was thinking ethernet, although one can probably do construct something
on a PtoP link too.
I wonder, what will happen if you do have 2 ethernet interfaces
on the same subnet and start OSPF on both of them?
> > Possibly, but how to know that this fits all uses? Why should not
> > the user have a choice? Perhaps the user just don't want to bother
> > with recursive routes, just turn on OSPF and it should work, right?
> If we can pass on the route that way to the OS, yes. I'll keep the
> possibility in mind if I tinker on Zserv more or people supply
Sure you can, which OS does NOT support this?
> I'm not going to address it now though.
Well, I think you at least should keep the interface.
> > :), then you get it the way you want, thanks.
> It seems easier to split it up myself. ;)
> > How about the more controversial stuff: nexthop_calculation()?
> I've applied and pushed the unnumber nexthop-calc, but not the
> oi-indexed-by-lsa-pos part.
I see(crossing fingers).
ospfd: Tighten up the connected check for redistribution
should not live there though, should go to master directly
More information about the Quagga-dev