[quagga-dev 5449] Re: [PATCH] RFC 2328, chap 8.1:

paul at clubi.ie paul at clubi.ie
Tue Jun 3 16:55:21 BST 2008


On Tue, 3 Jun 2008, Andrew J. Schorr wrote:

> Currently, with the existing code base, it is fine to have shared 
> addresses (i.e. same local address used on multiple interfaces), as 
> long as the PtP interfaces have unique peer addresses.  This works 
> fine currently (case 'a'), so I see no problem.  Why is 'a' a 
> requirement if it is working fine today with existing code without 
> this detection?

It works in your setup, sure. Possibly cause you took care to avoid 
non-working configurations ;).

But in other setups, it won't, such as where there are interfaces 
where the shared-address is the identifying address (e.g. subnet 
assigned on PtP, or the shared address is shared with some other 
broadcast interface).

> But if by "shared addresses" you mean PtP links with both the same 
> local and peer addresses, then yes, that is a new situation not 
> currently supported, and auto-detection might be nice.

Well, put it this way, if the implementation auto-detects the shared 
address then that *expands* the administrative options we can offer, 
from "auto" to "configure".

Exactly what the administrative defaults should be is debatable - but 
shouldn't affect implementatoin too much..

> I don't see why 'a' requires any configuration at all.  Most likely 
> I'm confused.

Sorry: If we decide on "or not" for 'a', then it must require 
configuration.

I'm unclear on what people's positions are on 'a'.

> But maybe I'm just thinking of this from a linux perspective that is
> not appropriate for other platforms...

Doesn't subnet addressing also work on Linux PtP? ;)

regards,
-- 
Paul Jakma	paul at clubi.ie	paul at jakma.org	Key ID: 64A2FF6A
Fortune:
God is real, unless declared integer.



More information about the Quagga-dev mailing list