[quagga-dev 6307] Re: rip version 1 bug?

James Carlson james.d.carlson at sun.com
Mon Dec 22 22:04:12 GMT 2008


Lennart Sorensen writes:
> On Wed, Dec 17, 2008 at 05:50:27PM -0600, Laverdiere,Phil wrote:
> > I have a customer who found a problem with rip version 1.  I know, they
> > should be using rip version 2, but for some bizarre reason they want to use
> > version 1.
> > 
> > Anyway, what happens is that ripd only advertises interface routes if the
> > address is on the same classfull network.
> 
> That would be because RIPv1 by spec only works with classfull networks.
> RIPv1 has no concept of netmask's in the protocol because they didn't
> exist at the time.  That is why we have RIP v2 is to make RIP work with
> netmasks rather than classfull networks.
> 
> So RIPv1 explicitly means classfull networks only.

There's a well-known RIP-1 "optimization" related to this that
generally allows the use of a single subnet mask within an
organization.

If you have a route destination that matches a locally defined
interface using the classful mask, then just assume that the netmask
configured on that interface is the same for all subnets in the same
organization.

For example, if you have (say) an interface with 10.8.1.1/24
configured, and you see a route with destination given as 10.9.1.0,
then because that matches the above interface under the classful mask
(net 10 is class A, or 255.0.0.0), you can assume that this another
/24 in the same organization.

Obviously, if you have interfaces that match under the classful mask
but have different subnet masks, that's an ambiguous situation.  It
would imply the use of VLSM within that organization, and it means
you're sunk.  (Just as obviously, seeing subnet routes at all outside
of an organization, where there are no matching local interfaces, is
an error.)

> So if the customer wants RIPv1, then they must also want classfull
> networks only on their network.

Not necessarily, at least for the BSD routed.

I wouldn't be surprised if Quagga RIP-1 doesn't do this, though.

-- 
James Carlson, Solaris Networking              <james.d.carlson at sun.com>
Sun Microsystems / 35 Network Drive        71.232W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.496N   Fax +1 781 442 1677



More information about the Quagga-dev mailing list