thanks a lot william.. That worked..
Now my task is to build a component in pox that would in the end facilitate
the communication between the 2 networks.

What exactly happens when h0 tries to ping h2.. The switch s0 sends the
packet to the controller which drops it because it is not connected to that
particular network.. Is this right?

If yes, How does the controller know that it is not connected to that
network?

If no, where is the information about network connections?
How does one find out information about network connections?


On Wed, Nov 7, 2012 at 12:14 PM, William Emmanuel Yu <[email protected]> wrote:

> I noticed the default routes configured on the mininet hosts use
> 0.0.0.0. Normally, interfaces are defined within their own networks
> defined by the network address.
>
> h0 route del -net 0.0.0.0
> h1 route del -net 0.0.0.0
> h2 route del -net 0.0.0.0
> h3 route del -net 0.0.0.0
> h0 route add -net 192.168.1.0 netmask 255.255.255.192 h0-eth0
> h1 route add -net 192.168.1.0 netmask 255.255.255.192 h1-eth0
> h2 route add -net 192.168.1.64 netmask 255.255.255.192 h2-eth0
> h3 route add -net 192.168.1.64 netmask 255.255.255.192 h3-eth0
>
> In my setup, I get the following (expected) results:
>
> mininet> pingall
> *** Ping: testing ping reachability
> h0 -> h1 X X
> h1 -> h0 X X
> h2 -> X X h3
> h3 -> X X h2
> *** Results: 66% dropped (8/12 lost)
>
> Good luck!
>
> On Tue, 2012-11-06 at 16:46 +0530, yashwanth kp wrote:
> >
> >
> >
> >
> > What i am planning to do is implement a research paper by Ricardo
> > Bennesby, Paulo Fonseca, Edjard Mota and Alexandre Passito which talks
> > about a development of a inter AS routing component for Software
> > Defined Networks(SDN).
> >
> >
> > For that they have used mininet for implementation.
> >
> >
> > I am planning to do the same.
> >
> >
> > So i need to create the initial topology with 2 or more autonomous
> > systems which are unable to communicate with one another.
> >
> >
> > I thought having the above example with 2 controllers in different
> > subnets will mean 2 AS.
> > But they are able to communicate with one another..
> >
> >
> > On Tue, Nov 6, 2012 at 4:37 PM, yashwanth kp
> > <[email protected]> wrote:
> >         What i am planning to do is implement a research paper
> >         by Ricardo Bennesby, Paulo Fonseca, Edjard Mota and Alexandre
> >         Passito which talks about a development of a inter AS routing
> >         component for Software Defined Networks(SDN).
> >
> >
> >         For that they have used mininet for implementation.
> >
> >
> >         I am planning to do the same.
> >
> >
> >         So i need to create the initial topology with 2 or more
> >         autonomous systems which are unable to communicate with one
> >         another.
> >
> >
> >         I thought having the above example with 2 controllers in
> >         different subnets will mean 2 AS.
> >         But they are able to communicate with one another..
> >
> >
> >
> >
> >         On Tue, Nov 6, 2012 at 4:33 PM, yashwanth kp
> >         <[email protected]> wrote:
> >                 What i am planning to do is implement a research paper
> >                 by Ricardo Bennesby, Paulo Fonseca, Edjard Mota and
> >                 Alexandre Passito which talks about a development of a
> >                 inter AS routing component for Software Defined
> >                 Networks(SDN).
> >
> >
> >                 For that they have used mininet for implementation.
> >
> >
> >                 I am planning to do the same.
> >
> >
> >                 So i need to create the initial topology with 2 or
> >                 more autonomous systems which are unable to
> >                 communicate with one another.
> >
> >
> >                 I thought having the above example with 2 controllers
> >                 in different subnets will mean 2 AS.
> >                 But they are able to communicate with one another..
> >
> >
> >                 (I have attached the research paper with this mail)
> >
> >
> >                 On Tue, Nov 6, 2012 at 4:16 PM, Murphy McCauley
> >                 <[email protected]> wrote:
> >                         On Nov 6, 2012, at 2:07 AM, yashwanth kp
> >                         wrote:
> >                         > But now since the hosts (h0,h1) and
> >                         hosts(h2,h3) are in different subnets, can
> >                         they still communicate?
> >                         > I tried and they are able to ping each
> >                         other..
> >                         > But i thought since they are in different
> >                         subnets they will not be able to communicate..
> >                         > So how do i simulate different AS
> >                         interconnected by switches?
> >                         > can different SDN based AS interconnect by
> >                         switches communicate with one another without
> >                         the need for an additional component?
> >
> >
> >
> >                         Subnets are an aspect of traditional L3
> >                         routing.  But you have an all-OpenFlow network
> >                         -- it has no traditional L3 routers in it, so
> >                         there aren't really any subnets unless you
> >                         write code to make them.
> >
> >                         Using the simple learning switch components
> >                         (e.g., l2_learning), you can stick one
> >                         OpenFlow island with a controller next to
> >                         another OpenFlow island with a different
> >                         controller, and things will just work.  This
> >                         even works with l3_learning.  But these work
> >                         by learning where every individual destination
> >                         is.  To oversimplify things, this doesn't
> >                         scale that well, so large networks (like the
> >                         internet) don't work that way and instead work
> >                         by by aggregating addresses at several levels
> >                         (e.g., subnets).  You can implement similar
> >                         things in OpenFlow, but you have to write the
> >                         code for it.
> >
> >                         I'm not clear on exactly what you mean by an
> >                         "SDN based AS".  Are you trying to reinvent
> >                         interdomain routing using OpenFlow?  Implement
> >                         an AS in OpenFlow?  Link an OpenFlow network
> >                         in one AS with an OpenFlow network in another
> >                         AS via a tunnel?  Something else?  SDN is
> >                         flexible; you can make networks do all sorts
> >                         of things if you're willing to write the
> >                         code...
> >
> >                         Good luck.
> >
> >                         -- Murphy
> >
> >
> >
> >
> >                 --
> >                 Regards,
> >
> >
> >                 Yashwanth K P
> >                 Final Year , B.Tech
> >                 NITK, Surathkal
> >                 Contact: +91-9538403606
> >
> >
> >
> >
> >
> >
> >         --
> >         Regards,
> >
> >
> >         Yashwanth K P
> >         Final Year , B.Tech
> >         NITK, Surathkal
> >         Contact: +91-9538403606
> >
> >
> >
> >
> >
> >
> > --
> > Regards,
> >
> >
> > Yashwanth K P
> > Final Year , B.Tech
> > NITK, Surathkal
> > Contact: +91-9538403606
> >
> >
>
> --
> -------------------------------------------------------
> William Emmanuel S. Yu, Ph.D. (杨怀义)
> Department of Information Systems and Computer Science
> Ateneo de Manila University
> email  :  wyu at ateneo dot edu
> blog   :  http://hip2b2.yutivo.org/
> web    :  http://CNG.ateneo.edu/cng/wyu/
> phone  :  +63(2)4266001 loc. 4186
> GPG    :  http://CNG.ateneo.net/cng/wyu/wyy.pgp
>
> Confidentiality Issue:  This message is intended only for the use of the
> addressee and may contain information that is privileged and
> confidential. If you are not the intended recipient, you are hereby
> notified that any use or dissemination of this communication is strictly
> prohibited.  If you have received this communication in error, please
> notify us immediately by reply and delete this message from your system.
>
>


-- 
Regards,

Yashwanth K P
Final Year , B.Tech
NITK, Surathkal
Contact: +91-9538403606

Reply via email to