My fork, which I attempted to update for modern POX: https://github.com/MurphyMc/riplpox
It might run on carp, but it was done for dart (and will probably work on eel). See my pull request against Brandon's repo for more info: https://github.com/brandonheller/riplpox/pull/5 (I think the short of it was that my updated riplpox works fine, but there seemed to be some incompatibilities with ripl and current versions of Mininet.) -- Murphy On Apr 16, 2015, at 3:31 AM, david jhon <djhon9...@gmail.com> wrote: > Thank you very much for your response Murphy, > > I have decided to use pox carp branch and running it with the > brandonheller(git://github.com/brandonheller/ripl.git, > git://github.com/brandonheller/riplpox.git) version of ripL and ripLpox but > getting following error on pox controller: > > Traceback (most recent call last): > File "/root/pox/pox/lib/revent/revent.py", line 231, in raiseEventNoErrors > return self.raiseEvent(event, *args, **kw) > File "/root/pox/pox/lib/revent/revent.py", line 278, in raiseEvent > rv = event._invoke(handler, *args, **kw) > File "/root/pox/pox/lib/revent/revent.py", line 156, in _invoke > return handler(self, *args, **kw) > File "/root/riplpox/riplpox/riplpox.py", line 344, in _handle_PacketIn > self._handle_packet_reactive(event) > File "/root/riplpox/riplpox/riplpox.py", line 240, in > _handle_packet_reactive > self._install_reactive_path(event, out_dpid, out_port, packet) > File "/root/riplpox/riplpox/riplpox.py", line 160, in _install_reactive_path > IDLE_TIMEOUT) > File "/root/riplpox/riplpox/riplpox.py", line 90, in install > self.connection.send(msg) > File "/root/pox/pox/openflow/of_01.py", line 692, in send > data = data.pack() > File "/root/pox/pox/openflow/libopenflow_01.py", line 2320, in pack > self.flags) > error: integer out of range for 'L' format code > INFO:.root.riplpox.riplpox:route: ['0_1_1', '0_2_1', '0_0_1'] > > could you please let me know if it is the right version to be used? I > couldn't find RipL and RipLPOX at POX git repository. Or could you please > suggest me a fix for this problem? Thanks a lot for your help/suggestions. > > Regards, > David > > > > > On Thu, Apr 16, 2015 at 12:47 PM, Murphy McCauley <murphy.mccau...@gmail.com> > wrote: > Is the version of POX you're using more than two years old? I think more > recent versions are unlikely to have this happen. > > If your POX is old, you could update it (in which case, you'd probably want > to run my fork of ripl/riplpox), or you could make the relevant change in > your local POX (I think it was a change to HELLO message processing in > of_01.py in early 2013). > > Another way to fix this would be to configure OVS to use OpenFlow 1.0 only. > This should be something like... > ovs-vsctl set bridge br0 protocols=openflow10 > > Except instead of br0, you'll want to do it for every switch that Mininet > creates. There may be an easy way to get Mininet to do this, but I don't > know it off the top of my head; you might try the Mininet list. Or maybe > there's a way to make it the default for OVS; I don't know a way to do that > off the top of my head either; you might try the OVS list. > > -- Murphy > > On Apr 16, 2015, at 12:32 AM, david jhon <djhon9...@gmail.com> wrote: > > > Hi everyone, > > > > I am working with openvswitch 2.3.90 and made some changes in vswitch code. > > I am trying to use ripl and riplpox along with the pox controller but pox > > throws this error: > > > > WARNING:openflow.of_01:Bad OpenFlow version (4) on connection [Con 19/None] > > INFO:openflow.of_01:[Con 19/None] closing connection > > > > and does not process any packets. I am building and running ovs by > > compiling from its source available at its git repository. > > > > P.S. I am using following command to start the topology: sudo mn --custom > > ~/ripl/ripl/mn.py --topo ft,4 --controller=remote --mac > > > > and ~/pox/pox.py riplpox.riplpox --topo=ft,4 --routing=hashed > > --mode=reactive to start the controller. > > > > It seems POX does not support 2.3.90 version of openvswitch or what else? > > Pls let me know if there exist a work around for this. Many thanks! > > > > Regards, > > Jhon > > > > > >
