[pox-dev] openflow, pox tplink

[Murphy McCauley]

WARNING:forwarding.l2_learning:Same port for packet from 38:60:77:3f:89:86
-> 00:22:fb:98:a9:f2 on 00-00-00-00-00-01.1.

Was this from doing a ping? What are those two MAC addresses?

Are you trying to ping between wireless devices?

Also, add openflow.keepalive onto your commandline. There's a bug in the
reference switch which causes it to disconnect a lot without this.

(Sorry for brevity; on my phone)

-- Murphy
On Feb 6, 2013 4:36 AM, "Shabbir Ahmed" <shabbir1282 at gmail.com> wrote:

> few more logs....
>
> openflow at openflowtutorial:~/pox$ ./pox.py log.level --WARNING
> --openflow=DEBUG  forwarding.l2_learning --transparent web.webcore
> --port=8888
> POX 0.0.0 / Copyright 2011-2012 James McCauley, et al.
> DEBUG:openflow.of_01:Listening on 0.0.0.0:6633
> DEBUG:openflow.of_01:[None 1] Socket error: Broken pipe
> INFO:openflow.of_01:[None 1] disconnected
> ERROR:openflow.of_01:[None 1] already disconnected!
> INFO:openflow.of_01:[None 1] closed
> WARNING:openflow.of_01:<class 'pox.openflow.PacketIn'> raised on dummy
> OpenFlow nexus
> WARNING:openflow.of_01:<class 'pox.openflow.PacketIn'> raised on dummy
> OpenFlow nexus
> WARNING:openflow.of_01:<class 'pox.openflow.PacketIn'> raised on dummy
> OpenFlow nexus
> INFO:openflow.of_01:[00-00-00-00-00-01 2] connected
> INFO:openflow.of_01:[None 3] closed
> INFO:openflow.of_01:[None 4] closed
> WARNING:forwarding.l2_learning:Same port for packet from 38:60:77:3f:89:86
> -> 00:22:fb:98:a9:f2 on 00-00-00-00-00-01.1.  Drop.
> WARNING:forwarding.l2_learning:Same port for packet from 00:22:fb:98:a9:f2
> -> 38:60:77:3f:89:86 on 00-00-00-00-00-01.1.  Drop.
> WARNING:forwarding.l2_learning:Same port for packet from 38:60:77:3f:89:86
> -> 00:22:fb:98:a9:f2 on 00-00-00-00-00-01.1.  Drop.
> INFO:openflow.of_01:[None 5] closed
> INFO:openflow.of_01:[00-23-20-25-ad-88 6] connected
>
>
>
> On Wed, Feb 6, 2013 at 4:49 PM, Shabbir Ahmed <shabbir1282 at gmail.com>wrote:
>
>> one more thing i haven't compiled openwrt with OVS, do i have to ? bcaz i
>> dont think we need to add ovs onto it
>>
>>
>>
>> On Wed, Feb 6, 2013 at 4:45 PM, Shabbir Ahmed <shabbir1282 at gmail.com>wrote:
>>
>>> root at OpenWrt:~# ps | grep of
>>>     3 root         0 SW   [ksoftirqd/0]
>>>  4367 root      1324 S    ofdatapath ptcp:6634 -i eth0.1 --no-local-port
>>> --pidfile -d 000000000001
>>>  4383 root      1064 S    ofprotocol tcp:127.0.0.1:6634 tcp:
>>> 192.168.1.2:6633 --fail=closed -D --pidfile=/var/run/ofprotocol.pid
>>> --out-of-band --listen=ptcp:
>>>  4391 root      1488 S    grep of
>>>
>>>
>>>
>>> On Wed, Feb 6, 2013 at 3:49 PM, Shabbir Ahmed <shabbir1282 at gmail.com>wrote:
>>>
>>>> >You can test this.  Try running POX without l2_learning.  Does the
>>>> ping work?  Then try running *with* >l2_learning.  Does the ping work?
>>>>
>>>> Connection is refused, or connects for some time and disconnects again
>>>> no error msgs other than closign connection
>>>>
>>>>
>>>> > wat logs do you want me to send.
>>>>
>>>> >Run POX.  Wait for it to connect.  Do a ping test.  Shut down POX.
>>>> Send the entire POX log.
>>>> starting listen on 0.0.0.0:6633
>>>> prompt POX>
>>>> then the same msg flooding.
>>>>
>>>>
>>>> > 1 more thing i can use enable_learning '0' in switch configuration on
>>>> openwrt but it doesnt effect any thing, >where i think of this statement as
>>>> disable default learning of the switch. the switch chip installed is
>>>> >'rtl8366rb'. I have a rb450g and i have ported the latest openwrt and
>>>> followed pantou but same error.
>>>>
>>>> I don't really know much about this configuration.  The
>>>> openflow-discuss list (and Yiannis in particular) will have better ideas.
>>>>
>>>> >One thought I'd give is that it looks like you've got it configured
>>>> for in-band OpenFlow.  I'd suggest that it >may be a good idea to try to
>>>> get out-of-band OpenFlow working first.
>>>>
>>>> same error on outofbound as well.
>>>>
>>>>
>>>> On Wed, Feb 6, 2013 at 2:49 PM, Murphy McCauley <
>>>> murphy.mccauley at gmail.com> wrote:
>>>>
>>>>> On Feb 6, 2013, at 1:39 AM, Shabbir Ahmed wrote:
>>>>>
>>>>> > the ping works but i think due to the switch's self learning process
>>>>> and not through OpenFlow bcaz in the absense of controller the ping should
>>>>> not work?
>>>>>
>>>>> You can test this.  Try running POX without l2_learning.  Does the
>>>>> ping work?  Then try running *with* l2_learning.  Does the ping work?
>>>>>
>>>>> > wat logs do you want me to send.
>>>>>
>>>>> Run POX.  Wait for it to connect.  Do a ping test.  Shut down POX.
>>>>>  Send the entire POX log.
>>>>>
>>>>> > 1 more thing i can use enable_learning '0' in switch configuration
>>>>> on openwrt but it doesnt effect any thing, where i think of this statement
>>>>> as disable default learning of the switch. the switch chip installed is
>>>>> 'rtl8366rb'. I have a rb450g and i have ported the latest openwrt and
>>>>> followed pantou but same error.
>>>>>
>>>>> I don't really know much about this configuration.  The
>>>>> openflow-discuss list (and Yiannis in particular) will have better ideas.
>>>>>
>>>>> One thought I'd give is that it looks like you've got it configured
>>>>> for in-band OpenFlow.  I'd suggest that it may be a good idea to try to get
>>>>> out-of-band OpenFlow working first.
>>>>>
>>>>> -- Murphy
>>>>
>>>>
>>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.noxrepo.org/pipermail/pox-dev-noxrepo.org/attachments/20130206/8e6be933/attachment-0001.htm>