[Ns-bugs] [Bug 407] OLSR is missing HNA support

code@nsnam.ece.gatech.edu code at nsnam.ece.gatech.edu
Wed Jan 27 10:18:32 PST 2010


http://www.nsnam.org/bugzilla/show_bug.cgi?id=407





--- Comment #13 from Gustavo J. A. M. Carneiro <gjcarneiro at gmail.com>  2010-01-27 13:18:32 EDT ---
(In reply to comment #12)
> (> 
> > Perhaps two separate methods could be added:
> > 
> >   1. AddHostAssociation(Ipv4Address)
> > 
> >   2. AddNetworkAssociation(Ipv4Address, Ipv4Mask)
> 
> I would probably lean towards one method, because 1) can be supported by
> passing an all ones mask to method 2), and because the RFC tends to deprecate
> using HNA to advertise hosts (Sec 12.2):
> 
>    In HNA-messages, announcing
>    reachability to an address sequence through a network- and netmask
>    address is typically preferred over announcing reachability to
>    individual host addresses

I think you're right.

> > 
> > _however_, I normally like code that is more intelligent.  In OLSR you don't
> > have have AddOlsrInterface; it just uses all the available interfaces in the
> > node.  Likewise, perhaps an automatic approach could be used here as well?  It
> > would require the code to:
> >   1. Go to the Ipv4 interface of the node;
> >   2. Get the list of routing protocols;
> >   3. Iterate over the list and, for each "static ipv4 routing protocol":
> >     3a. Iterate over the static routing entries and generate Hna message(s)
> > based on them.
> 
> That would tend to generate HNAs for all interfaces, including OLSR active
> ones.  Is that what you want?
> 
> It seems from the RFC that HNA is intended for the non-OLSR interfaces.  If we
> were to add the suggested SetInterfaceNonOlsr() method, we could perhaps do it
> with a flag argument
> SetInterfaceNonOlsr (bool generateHna) // default true
> 
> and then have the code auto-generate HNA only for all non-OLSR configured
> interfaces unless disabled by the above argument.  
> 
> Then, we could keep the AddAssociation(address/mask) for an additional manual
> injection of an HNA.

Yes, my idea by default was to inject routing entries only if those entries are
associated with a non-OLSR interface.  I wouldn't even make it configurable.

-- 
Configure bugmail: http://www.nsnam.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.


More information about the Ns-bugs mailing list