Toshihiro Noguchi | 18 Apr 05:41 2014
Picon

Starting Openflow Service

Hi All!

I am a bit new to openflow but am trying to get it up and running on a WR1043ND. I believe that I have openflow installed (the service is located at /etc/init.d/openflow), but when I run:

/etc/init.d/openflow start

I get the error message

/sbin/ofdown: .: line 4: can't open '/etc/functions.sh'
/sbin/ofup: .: line 5: can't open '/etc/functions.sh

Any ideas would be greatly appreciated!

Regards,
Toshi


--
Toshihiro Noguchi
Cornell University
College of Engineering
<div><div dir="ltr">
<div dir="ltr">
<div>Hi All!</div>
<div><br></div>
<div>

I am a bit new to openflow but am trying to get it up and running on a WR1043ND. I believe that I have openflow installed (the service is located at /etc/init.d/openflow), but when I run:</div>
<div>

<br>
</div>
<div>
<span>/etc/init.d/openflow start</span><br>
</div>
<div><br></div>
<div>I get the error message</div>
<div><br></div>
<div>
<span>/sbin/ofdown: .: line 4: can't open '/etc/functions.sh'</span><br><span>/sbin/ofup: .: line 5: can't open '/etc/functions.sh</span><br>
</div>
<div><br></div>
<span>Any ideas would be greatly appreciated!</span><div>

<br>
</div>
<div>Regards,</div>
<div>Toshi</div>
</div>
<br clear="all"><div><br></div>-- <br><span><span><div>

Toshihiro Noguchi</div>
<div>Cornell University</div>
<div>College of Engineering</div>
<div><a href="mailto:tn229@..." target="_blank">tn229 <at> cornell.edu</a></div></span></span>
</div></div>
AHMAD AZMI ABO NASER | 16 Apr 14:30 2014
Picon

(no subject)

I am subscribed, but i can't confirm my subscription (expired confirmation string) nor can i receive from the mailing list, who can sort this out?

abo naser.

 

 

 

  Save a tree. Don't print this e-mail unless it's really necessary

 

 

<div>
<div>I am subscribed, but i can't confirm my subscription (expired confirmation string) nor can i receive from the mailing list, who can sort this out?
<div><br></div>
<div>abo naser.</div>
</div>

<p class="MsoNormal"><span><p>&nbsp;</p></span></p>
<p class="MsoNormal"><span><p>&nbsp;</p></span></p>
<p class="MsoNormal"><span><p>&nbsp;</p></span></p>
<p class="MsoNormal"><span>&nbsp; </span><span>Save a tree. Don't print this e-mail unless it's really necessary</span><span><p></p></span></p>
<p class="MsoNormal"><span>&nbsp;<p></p></span></p>
<p class="MsoNormal"><span><p>&nbsp;</p></span></p>

</div>
LIANG YANG | 15 Apr 09:04 2014
Picon

Understanding of Group in OpenFlow 1.3.1

Hi, Dear All:

We are implemented OF 1.3.1 on merchant silicon. When we design and implement the L3 Multicast, we have some uncertainties about the OF Spec.

According to spec: "This group is used for multicast or broadcast forwarding. The packet is cloned for each bucket; one packet is processed for each bucket of the group."

For L3 Multicast routing, can we support below group.

match :  DIP(CLASS D address),  DMAC,  VLAN, ETH-TYPE,
action:  bueckt 1 :  SMAC1 , VLAN1 , outport 1,2,3,4,5;
              bueckt 2:   SMAC2 , VLAN2, outport  6,7,8,9.10;
              bueckt 3 :  SMAC3 , VLAN3 , outport 11,12,13,14,15;

OR we only can support the below one?

match :  DIP(CLASS D address),  DMAC,  VLAN, ETH-TYPE,
action:  bueckt 1 :  SMAC1 , VLAN1 , outport 1;
              bueckt 1 :  SMAC1 , VLAN1 , outport 2;
              .........
              bueckt 1 :  SMAC1 , VLAN1 , outport 5;
              bueckt 2:   SMAC2 , VLAN2, outport  6;
              .....
              bueckt 2:   SMAC2 , VLAN2, outport  10;
              bueckt 3 :  SMAC3 , VLAN3 , outport 11;


Thanks,
Liang




<div><div dir="ltr">
<div>
<div>
<div>
<div>
<div>Hi, Dear All:<br><br>
</div>We are implemented OF 1.3.1 on merchant silicon. When we design and implement the L3 Multicast, we have some uncertainties about the OF Spec.<br><br>
</div>
According to spec: "This group is used for multicast or broadcast forwarding. The packet is cloned for each bucket; one packet is processed for each bucket of the group."<br>
</div>
<br>
</div>For L3 Multicast routing, can we support below group.<br><br>match : &nbsp;DIP(CLASS D address),
&nbsp;DMAC, &nbsp;VLAN, ETH-TYPE, 
<br>action: &nbsp;bueckt 1 : &nbsp;SMAC1
, VLAN1 , outport 1,2,3,4,5;
<br>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;
&nbsp; bueckt 2: &nbsp; SMAC2 , VLAN2, outport &nbsp;6,7,8,9.10;
<br>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;
&nbsp; bueckt 3 : &nbsp;SMAC3 , VLAN3 , outport 11,12,13,14,15;
<br><br>
</div>OR we only can support the below one?<br><br>match : &nbsp;DIP(CLASS D address),
&nbsp;DMAC, &nbsp;VLAN, ETH-TYPE, 
<br>action: &nbsp;bueckt 1 : &nbsp;SMAC1
, VLAN1 , outport 1;
<br>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;
&nbsp; bueckt 1 : &nbsp;SMAC1 , VLAN1 , outport 2;
<br>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;
&nbsp; .........
<br>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;
&nbsp; bueckt 1 : &nbsp;SMAC1 , VLAN1 , outport 5;
<br>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;
&nbsp; bueckt 2: &nbsp; SMAC2 , VLAN2, outport &nbsp;6;
<br>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;
&nbsp; .....
<br>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;
&nbsp; bueckt 2: &nbsp; SMAC2 , VLAN2, outport &nbsp;10;
<br>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;
&nbsp; bueckt 3 : &nbsp;SMAC3 , VLAN3 , outport 11;
<br><div><div><div><div>
<div>
<br><br>
</div>
<div>Thanks,<br>Liang<br>
</div>
<div>
<br><br><br><table class="" width="400" cellpadding="0" cellspacing="0"><tr>
<td width="400">				  			
	  
	  				<br><span class=""><a href="https://emailcherry.com/extension/redirect/9/aHR0cHM6Ly9lbWFpbGNoZXJyeS5jb20=" target="_blank" class=""></a>
		  			</span>	
	  		
	  		
	  				  			
	  		</td>
	  	</tr></table>
</div>
</div></div></div></div>
</div></div>
Rushit Parekh | 15 Apr 07:47 2014
Picon

Flow of openflow ..


Hello,
 i am confuse in flow of openflow, can any one help me understand

In Openflow floder if we look we have some folder..
>openflow\controller\ -> controller.c
>lib\ -> rconn.c , vconn.c ,netlink.c , netdev.c , vconn_ssl.c , vocnn_stream.c


if we look at controller.c
---------------------------------------------------------------------------
line 218 ..static do_switching(struct switch *sw)
{
unsigned int packets_sent;
struct ofpbuf *msg;

packets_sent = rconn_packets_sent(sw->rconn);// counts packet is sent

msg = rconn_recv(sw->rconn); // recvs msg.. rocnn_recv from rconn.c
if (msg) {
lswitch_process_packet(sw->lswitch, sw->rconn, msg);
ofpbuf_delete(msg);
}
rconn_run(sw->rconn);

return (!rconn_is_alive(sw->rconn) ? EOF
: rconn_packets_sent(sw->rconn) != packets_sent ? 0
: EAGAIN);
}
-------------------------------------------------------------------------
now in rconn.c for rconn_recv line 474 rconn_recv function is defined it calles vconn_recv
so in vconn.c there is vocnn_recv function defined at line 460 that call for do_recv which is defined on line 470..
but at line num 475

>retval = (vconn->class->recv)(vconn, msgp);

written so.. we need to find struct vconn, its object class and.. that class is having function 'recv'

where is that "recv" is defined..? i am not getting exactly process

**********************************************************************
and. .. difference between netlink.c and netdev.c
netlink.c uses family AF_NETLINK
nedev.c uses family PF_PACKET
what is difference between them

*********************************************************
when openflow will use vconn_stream.c , vconn_ssl.c??
--

With Regards,

Rushit Parekh
<div><div dir="ltr">
<div>
<div><br clear="all"></div>Hello,<br>
</div>&nbsp;i am confuse in flow of openflow, can any one help me understand<br><div><div>
<br>In Openflow floder if we look we have some folder.. 
<br>
&gt;openflow\controller\ -&gt; controller.c
<br>
&gt;lib\ -&gt; rconn.c , vconn.c  ,netlink.c , netdev.c  , vconn_ssl.c ,  vocnn_stream.c
<br><br><br>
if we look at controller.c
<br>
---------------------------------------------------------------------------
<br>
line 218 ..static do_switching(struct switch *sw)
<br>
{
<br>
    unsigned int packets_sent;
<br>
    struct ofpbuf *msg;
<br><br>
    packets_sent = rconn_packets_sent(sw-&gt;rconn);// counts packet is sent
<br><br>
    msg = rconn_recv(sw-&gt;rconn); // recvs msg.. rocnn_recv from rconn.c
<br>
    if (msg) {
<br>
        lswitch_process_packet(sw-&gt;lswitch, sw-&gt;rconn, msg);
<br>
        ofpbuf_delete(msg);
<br>
    }
<br>
    rconn_run(sw-&gt;rconn);
<br><br>
    return (!rconn_is_alive(sw-&gt;rconn) ? EOF
<br>
            : rconn_packets_sent(sw-&gt;rconn) != packets_sent ? 0
<br>
            : EAGAIN);
<br>
}
<br>
-------------------------------------------------------------------------
<br>
 now in rconn.c for rconn_recv line 474 rconn_recv function is defined it calles vconn_recv
<br>
so in vconn.c there is vocnn_recv function defined at line 460 that call for do_recv which is defined on line 470.. 
<br>
but at line num 475 
<br><br>
&gt;retval = (vconn-&gt;class-&gt;recv)(vconn, msgp);
<br><br>
written so.. we need to find struct vconn, its object class and.. that  class is having function 'recv'
<br><br>
where is that "recv"  is defined..? i am not getting exactly process
<br><br>
**********************************************************************
<br>
and. .. difference between netlink.c and netdev.c  
<br>
netlink.c uses  family AF_NETLINK
<br>
nedev.c uses family PF_PACKET 
<br>
what is difference between them
<br><br>
*********************************************************
<br>
when openflow will use vconn_stream.c , vconn_ssl.c?? 
<br>-- <br><br>With Regards,<br><br>Rushit Parekh<br>
</div></div>
</div></div>
jjf_neu_scu | 15 Apr 03:51 2014

question about flowvisor

Dear
 
I have a question about flowvisor: Does flowvisor1.4 controller support openflow1.3 protocol now? Hope someone can help me with that.
 
2014-04-15
jjf_neu_scu
<div>
<div>Dear 
</div>
<div>&nbsp;</div>
<div>I have a question about flowvisor: Does 
flowvisor1.4&nbsp;controller support openflow1.3 protocol now? 
Hope&nbsp;someone&nbsp;can help me with that.</div>
<div>&nbsp;</div>
<div align="left">2014-04-15</div>

<div><span>jjf_neu_scu</span></div>
</div>
Srini Seetharaman | 14 Apr 21:31 2014
Picon

SDN Hackathon at ONUG Spring 2014

ONUG and SDN Hub are organizing the first-ever SDN Hackathon in the US east coast. Our goal is to promote the open-source app developer community and jump-start the ecosystem of SDN applications useful to user groups, such as those represented at ONUG (including several Fortune 500 financial enterprises). Attendees can form form groups of maximum 5 people and build a few "bite-sized" SDN applications. For this hackathon, the enterprises themselves will propose seed ideas to hack on. 

Only 10 seats left. Register soon! Have fun coding and win awesome prizes!

Dates

May 3 (optional)

May 4 (optional)

May 5

Time

6-8pm EST

10am-5pm EST

10am-5pm EST

Venue

Welcome Dinner co Pizza

10gen

229 W 43rd St,

5th Floor., New York NY 10036

Citigroup,

399 Park Ave.

New York City NY

Demo

10 min slots between 4-5pm on Monday May 5, 2014 at Citigroup


Additional information:
  • Each member of the top three teams will win cash prize of $500 (1st place), $300 (2nd place), or $100 (3rd place).  
  • Please participate only if you are familiar with SDN platforms. 
  • The hackathon is only open to folks already registered or registering for the ONUG Spring 2014. 
  • More details, please visit: http://opennetworkingusergroup.com/agenda/hackathon/

<div><div dir="ltr">
<a href="http://opennetworkingusergroup.com">ONUG</a> and <a href="http://sdnhub.org">SDN Hub</a> are organizing the first-ever SDN Hackathon in the US east coast. Our goal is to promote the open-source app developer community and jump-start the ecosystem of SDN applications useful to user groups, such as those represented at ONUG (including several Fortune 500 financial enterprises). Attendees can form&nbsp;<span>form groups of maximum 5 people and build&nbsp;</span><span>a few "bite-sized" SDN applications.&nbsp;</span><span>For this hackathon,&nbsp;</span>the enterprises themselves will propose seed ideas to hack on.&nbsp;<div>
<br>
</div>
<div>
<div>Only 10 seats left. <a href="http://sdnhub.org/events/sdn-hackathon-onug-2014/register">Register soon</a>! Have fun coding and win awesome prizes!</div>
<div><br></div>
<table>
<tr>
<td>
<p dir="ltr"><span>Dates</span></p>
</td>
<td>
<p dir="ltr"><span>May 3 (optional)</span></p>
</td>
<td>
<p dir="ltr"><span>May 4 (optional)</span></p>
</td>
<td>
<p dir="ltr"><span>May 5</span></p>
</td>
</tr>
<tr>
<td>
<p dir="ltr"><span>Time</span></p>
</td>
<td>
<p dir="ltr"><span>6-8pm EST</span></p>
</td>
<td>
<p dir="ltr"><span>10am-5pm EST</span></p>
</td>
<td>
<p dir="ltr"><span>10am-5pm EST</span></p>
</td>
</tr>
<tr>
<td>
<p dir="ltr"><span>Venue</span></p>
</td>
<td><p dir="ltr">Welcome Dinner co Pizza</p></td>
<td>
<p dir="ltr">10gen</p>
<p dir="ltr">229 W 43rd St,</p>
<p dir="ltr">
5th Floor., New York NY 10036</p>
</td>
<td>
<p dir="ltr"><span>Citigroup, </span></p>
<p dir="ltr"><span>399 Park Ave. </span></p>
<p dir="ltr">
<span>New York City NY</span></p>
</td>
</tr>
<tr>
<td>
<p dir="ltr"><span>Demo</span></p>
</td>
<td colspan="3">
<p dir="ltr"><span>10 min slots between 4-5pm on Monday May 5, 2014 at Citigroup</span></p>
</td>
</tr>
</table>
<div><br></div>
<div>Additional information:</div>
<div>
<ul>
<li>
<span>Each member of the top three teams will win cash prize of $500 (1st place), $300 (2nd place), or $100 (3rd place). &nbsp;</span><br>
</li>
<li>Please participate only if you are familiar with SDN platforms.&nbsp;<br>
</li>
<li>The hackathon is only open to folks already registered or registering for the ONUG Spring 2014.&nbsp;<br>
</li>
<li>More details, please visit: <a href="http://sdnhub.org/events/sdn-hackathon-onug-2014/">http://opennetworkingusergroup.com/agenda/hackathon/</a><br>
</li>
</ul>
<div><br></div>
</div>
</div>
</div></div>
Rushit Parekh | 11 Apr 12:22 2014
Picon

TLS



How can i enable TLS option for southbound(OF controller-OF switch)communication?





--

With Regards,

Rushit Parekh
9426317847
<div><div dir="ltr">
<br><br>How can i enable TLS option for southbound(OF controller-OF switch)communication?<br><br><br><br><br clear="all"><div>
<br>-- <br><br>With Regards,<br><br>Rushit Parekh<br>9426317847<br>
</div>
</div></div>
שמעון אלפרט | 9 Apr 18:35 2014
Picon

Re: openflow-discuss Digest, Vol 66, Issue 6

Re:Question about section 4.7 View OpenFlow Messages for Ping in the open flow tutorial.
I have a problam with this pat in the tutorial, from some reason the only kind of messages i see in the wireshark are "Hello" , and i don't see al the types written in the tutorial,
thanks
shimon


Send openflow-discuss mailing list submissions to
        openflow-discuss-3+4lAyCyj6BegcJQxxnBRDe48wsgrGvP@public.gmane.org

To subscribe or unsubscribe via the World Wide Web, visit
        https://mailman.stanford.edu/mailman/listinfo/openflow-discuss
or, via email, send a message with subject or body 'help' to
        openflow-discuss-request-3+4lAyCyj6BegcJQxxnBRDe48wsgrGvP@public.gmane.org

You can reach the person managing the list at
        openflow-discuss-owner-3+4lAyCyj6BegcJQxxnBRDe48wsgrGvP@public.gmane.org

When replying, please edit your Subject line so it is more specific
than "Re: Contents of openflow-discuss digest..."


Today's Topics:

   1. Re: Quick experimenting with OVS on OpenWrt (Alison Chan)
   2. Re: Quick experimenting with OVS on OpenWrt (Nicholas Bastin)
   3. Re: Quick experimenting with OVS on OpenWrt (Ben Pfaff)
   4. Re: Release of libfluid: ONF Driver Competition Winner w/
      OpenFlow 1.0 and 1.3 (Christian Esteve Rothenberg)
   5. Re: Release of libfluid: ONF Driver Competition   Winner w/
      OpenFlow 1.0 and 1.3 (Murphy McCauley)


----------------------------------------------------------------------

Message: 1
Date: Tue, 8 Apr 2014 17:29:20 -0400
From: Alison Chan <chan7781 <at> kettering.edu>
To: Ben Pfaff <blp-l0M0P4e3n4LQT0dZR+AlfA@public.gmane.org>
Cc: dev-yBygre7rU0TnMu66kgdUjQ@public.gmane.org, openflow-discuss
        <openflow-discuss-3+4lAyCyj6BegcJQxxnBRDe48wsgrGvP@public.gmane.org>
Subject: Re: [openflow-discuss] Quick experimenting with OVS on
        OpenWrt
Message-ID:
        <CABcSDhdTUirSo9SevVz6a6Fxt4eT-BPEAhiJVy1Qv+WoiNaoWg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
Content-Type: text/plain; charset=UTF-8

Thanks Ben. I just subscribed to dev <at> ovs list.

I was able to take the router back up to the lab (I had been testing
it with laptops in my office). With two gigabit ethernet hosts on the
data plane, throughput (tcp iperf) is 450~460 Mbit/s. I will test
things a bit more tomorrow (e.g. adding 802.11 clients) and if all
goes well then I will move all five of our TP-Links to OVS.

I was not expecting throughput with OVS to be 11 times better than
Pantou throughput -- I had to double check things to make sure the
switch was still OpenFlow enabled and not just using hardware learning
switch!

Cheers,
Alison

2014-04-08 13:15 GMT-04:00 Ben Pfaff <blp-l0M0P4e3n4LQT0dZR+AlfA@public.gmane.org>:
> [adding ovs-dev since there might be interest there]
>
> On Tue, Apr 08, 2014 at 11:56:28AM -0400, Alison Chan wrote:
>> Hi everyone,
>>
>> I was able to get OpenVswitch running on one of our TL-WR1043ND
>> routers running OpenWrt Attitude Adjustment. I was expecting somewhat
>> better performance than Pantou, but not this much better:
>>
>> Pantou: 39.7 Mbit/s tcp iperf using static flows (in_port:1 output:2
>> and in_port:2 output:1)
>> OVS: 94.1 Mbit/s testing the same way
>>
>> At this point, the limiting factor is one of my laptops' 10/100
>> ethernet interface! I will try to find another computer with gigabit
>> ethernet and report back.
>>
>> Cheers,
>> --
>> Alison ChanI
>> chan7781-/TcrQwaiJtkRA9EQCv0AEA@public.gmane.org
>> SMS: +1 909 278 7753
>> Fax: +1 206 203 2793
>> _______________________________________________
>> openflow-discuss mailing list
>> openflow-discuss-3+4lAyCyj6BegcJQxxnBRDe48wsgrGvP@public.gmane.org
>> https://mailman.stanford.edu/mailman/listinfo/openflow-discuss



--
Alison Chan
chan7781-/TcrQwaiJtkRA9EQCv0AEA@public.gmane.org
SMS: +1 909 278 7753
Fax: +1 206 203 2793


------------------------------

Message: 2
Date: Tue, 8 Apr 2014 17:51:14 -0500
From: Nicholas Bastin <nick.bastin-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
To: Alison Chan <chan7781 <at> kettering.edu>
Cc: dev-yBygre7rU0TnMu66kgdUjQ@public.gmane.org, openflow-discuss
        <openflow-discuss-3+4lAyCyj6BegcJQxxnBRDe48wsgrGvP@public.gmane.org>
Subject: Re: [openflow-discuss] Quick experimenting with OVS on
        OpenWrt
Message-ID:
        <CADmMkWdqjcDrPbsGE=U8pWGhBk_Cc905FZXrk1ed_QPn-FCKag-JsoAwUIsXov1KXRcyAk9cg@public.gmane.orgl.com>
Content-Type: text/plain; charset="utf-8"

On Tue, Apr 8, 2014 at 4:29 PM, Alison Chan <chan7781-/TcrQwaiJtkRA9EQCv0AEA@public.gmane.org> wrote:

> I was able to take the router back up to the lab (I had been testing
> it with laptops in my office). With two gigabit ethernet hosts on the
> data plane, throughput (tcp iperf) is 450~460 Mbit/s. I will test
> things a bit more tomorrow (e.g. adding 802.11 clients) and if all
> goes well then I will move all five of our TP-Links to OVS.
>

There are two variants of this device - one with an AR9132 (boring white
case), and one with a QCA9558 (blue spaceship-looking thing).  The AR9132
version uses a separate switch ASIC (AR8316 - I have one of these), so your
max throughput through that device in any software forwarding (regardless
of CPU speed) will be a gigabit.  The 400Mhz ARM in that box is going to
limit you even further, so 450Mbits seems a bit suspect.

If you have the rev. 2 (blue) hardware, that is more of an unknown (I don't
have one of these to look at or tinker with) - the 9558 has a 720Mhz MIPS
core, so that's better, but I found images of a different AP that uses this
chip on the internet and it appears to use an AR8327 for the switching, so
likely there's no integrated ethernet switch on the SoC.  This means your
max throughput is still probably a gigabit, and you might be able to
actually get 450Mbits through that CPU.  The AR8327 actually has a 96-entry
ACL table and 64-entry VLAN translation table (as well as a bunch of more
specialized tables that would be more difficult to use well) that could be
used for hardware acceleration, but I doubt anyone has done the work to
program them.

--
Nick
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/openflow-discuss/attachments/20140408/2957a37b/attachment-0001.html>

------------------------------

Message: 3
Date: Tue, 8 Apr 2014 16:00:23 -0700
From: Ben Pfaff <blp-l0M0P4e3n4LQT0dZR+AlfA@public.gmane.org>
To: Nicholas Bastin <nick.bastin <at> gmail.com>
Cc: "dev-yBygre7rU0TnMu66kgdUjQ@public.gmane.org" <dev-yBygre7rU0TnMu66kgdUjQ@public.gmane.org>, openflow-discuss
        <openflow-discuss-3+4lAyCyj6BegcJQxxnBRDe48wsgrGvP@public.gmane.org>
Subject: Re: [openflow-discuss] Quick experimenting with OVS on
        OpenWrt
Message-ID:
        <CACjuMbzXnBsa3-xcAuRN3EuJuhEQjW3wTTNdgjW-qM-nGxmP2w-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
Content-Type: text/plain; charset=ISO-8859-1

On Tue, Apr 8, 2014 at 3:51 PM, Nicholas Bastin <nick.bastin-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> If you have the rev. 2 (blue) hardware, that is more of an unknown (I don't
> have one of these to look at or tinker with) - the 9558 has a 720Mhz MIPS
> core, so that's better, but I found images of a different AP that uses this
> chip on the internet and it appears to use an AR8327 for the switching, so
> likely there's no integrated ethernet switch on the SoC.  This means your
> max throughput is still probably a gigabit, and you might be able to
> actually get 450Mbits through that CPU.  The AR8327 actually has a 96-entry
> ACL table and 64-entry VLAN translation table (as well as a bunch of more
> specialized tables that would be more difficult to use well) that could be
> used for hardware acceleration, but I doubt anyone has done the work to
> program them.

Wow, the AR8327 seems to have full public documentation:
    https://lafibre.info/images/doc/201106_spec_AR8327.pdf
At a glance, it's not that general-purpose but certainly nice to
see a documented switching ASIC in a low-cost device.


------------------------------

Message: 4
Date: Tue, 8 Apr 2014 21:34:08 -0300
From: Christian Esteve Rothenberg <chesteve-z5i6iNftOEzjuJLLi9cJmF/I71DsQ//L@public.gmane.org>
To: Murphy McCauley <murphy.mccauley-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>,
        "openflow-discuss-xGejAJT2w6wUYqjhf3gUyfZ8FUJU4vz8@public.gmane.org"
        <openflow-discuss-xGejAJT2w6wUYqjhf3gUyfZ8FUJU4vz8@public.gmane.org>
Subject: Re: [openflow-discuss] Release of libfluid: ONF Driver
        Competition Winner w/ OpenFlow 1.0 and 1.3
Message-ID:
        <CAEj5p9TOj8gTy8Pk-WOObY9Zw9ov-Q+ws5U=50w3oTk=ThG07A-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
Content-Type: text/plain; charset=ISO-8859-1

Thanks Murphy,

one of my students working with POX wants to give a try and hopefullly
contribute to these efforts, can you share the pointers to that POX
branch. We can only say positive things about prototypiing with POX --
giving it clean and effective OF1.3 support would be a neat upgrade
beneficial to all parties :)

-Christian

>
>
> On Mon, Mar 24, 2014 at 10:17 PM, Murphy McCauley <murphy.mccauley-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>>
>> Congratulations.
>>
>> There's now POX branch that's been quickly hacked up to (sort of) use libfluid's Python bindings.  It also includes a minor patch for one of the swig .i files.
>>
>> -- Murphy
>>
>> On Mar 22, 2014, at 5:08 AM, Christian Esteve Rothenberg <chesteve-z5i6iNftOEzjuJLLi9cJmF/I71DsQ//L@public.gmane.org> wrote:
>>
>> Dear OpenFlow fellows,
>>
>> in case you are not aware about the public release of the winner implementation of the OpenFlow driver competition (https://www.opennetworking.org/competition) here is the pointer to the github repository:
>>
>> http://opennetworkingfoundation.github.io/libfluid/
>>
>> libluid may be interesting to developers of both OpenFlow switches and controllers. It features support of OpenFlow 1.0 and 1.3, high performance, bindings to Python and Java, easy port to different hardware architectures, etc.
>>
>> We welcome users and developers interested in building an open community to maintain libfluid as a useful, multi-purpose OpenFlow library to develop switch agents and controller implementations.
>>
>> -Christian (on behalf of the libfluid team)
>> _______________________________________________
>> openflow-discuss mailing list
>> openflow-discuss-3+4lAyCyj6BegcJQxxnBRDe48wsgrGvP@public.gmane.org
>> https://mailman.stanford.edu/mailman/listinfo/openflow-discuss
>>
>> --
>> Christian




--
Christia


------------------------------

Message: 5
Date: Tue, 8 Apr 2014 19:49:23 -0700
From: Murphy McCauley <murphy.mccauley-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
To: Christian Esteve Rothenberg <chesteve-z5i6iNftOEzjuJLLi9cJmF/I71DsQ//L@public.gmane.org>
Cc: "openflow-discuss-xGejAJT2w6wUYqjhf3gUyfZ8FUJU4vz8@public.gmane.org"
        <openflow-discuss-xGejAJT2w6wUYqjhf3gUyfZ8FUJU4vz8@public.gmane.org>
Subject: Re: [openflow-discuss] Release of libfluid: ONF Driver
        Competition     Winner w/ OpenFlow 1.0 and 1.3
Message-ID: <A69FEFAD-402B-41EC-A588-B38E3BDD2897-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Content-Type: text/plain; charset="iso-8859-1"

Sure, it's just the libfluid_experiment branch of the main fork:
http://noxrepo.org/git/pox/tree/libfluid_experiment

It's definitely very much an experiment. ;)  For starters, I was only messing with 1.0 and while I used libfluid to read off the wire, the responses are still generated with POX's OpenFlow library.  I just thought it was worth experimenting to get a sense of what would be involved.  If you guys are interested in doing much more work on this, we should talk!

-- Murphy

On Apr 8, 2014, at 5:34 PM, Christian Esteve Rothenberg <chesteve-z5i6iNftOEzjuJLLi9cJmF/I71DsQ//L@public.gmane.org> wrote:

> Thanks Murphy,
>
> one of my students working with POX wants to give a try and hopefullly
> contribute to these efforts, can you share the pointers to that POX
> branch. We can only say positive things about prototypiing with POX --
> giving it clean and effective OF1.3 support would be a neat upgrade
> beneficial to all parties :)
>
> -Christian
>
>>
>>
>> On Mon, Mar 24, 2014 at 10:17 PM, Murphy McCauley <murphy.mccauley-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>>>
>>> Congratulations.
>>>
>>> There's now POX branch that's been quickly hacked up to (sort of) use libfluid's Python bindings.  It also includes a minor patch for one of the swig .i files.
>>>
>>> -- Murphy
>>>
>>> On Mar 22, 2014, at 5:08 AM, Christian Esteve Rothenberg <chesteve-z5i6iNftOEzjuJLLi9cJmF/I71DsQ//L@public.gmane.org> wrote:
>>>
>>> Dear OpenFlow fellows,
>>>
>>> in case you are not aware about the public release of the winner implementation of the OpenFlow driver competition (https://www.opennetworking.org/competition) here is the pointer to the github repository:
>>>
>>> http://opennetworkingfoundation.github.io/libfluid/
>>>
>>> libluid may be interesting to developers of both OpenFlow switches and controllers. It features support of OpenFlow 1.0 and 1.3, high performance, bindings to Python and Java, easy port to different hardware architectures, etc.
>>>
>>> We welcome users and developers interested in building an open community to maintain libfluid as a useful, multi-purpose OpenFlow library to develop switch agents and controller implementations.
>>>
>>> -Christian (on behalf of the libfluid team)
>>> _______________________________________________
>>> openflow-discuss mailing list
>>> openflow-discuss-3+4lAyCyj6BegcJQxxnBRDe48wsgrGvP@public.gmane.org
>>> https://mailman.stanford.edu/mailman/listinfo/openflow-discuss
>>>
>>> --
>>> Christian
>
>
>
>
> --
> Christia

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/openflow-discuss/attachments/20140408/01b051dc/attachment.html>

------------------------------

Subject: Digest Footer

_______________________________________________
openflow-discuss mailing list
openflow-discuss <at> lists.stanford.edu
https://mailman.stanford.edu/mailman/listinfo/openflow-discuss


------------------------------

End of openflow-discuss Digest, Vol 66, Issue 6
***********************************************

<div>
<div dir="rtl">
<div dir="ltr">Re:<span>Question about section&nbsp;<a href="http://archive.openflow.org/wk/index.php/OpenFlow_Tutorial#View_OpenFlow_Messages_for_Ping"><span>4.7</span> <span>View OpenFlow Messages for Ping</span></a> in the open flow tutorial.</span>
</div>
<div dir="ltr"><span>I have a problam with this pat in the tutorial, from some reason the only kind of&nbsp;messages i see in the wireshark are "Hello" , and i don't see al the types written in the tutorial, </span></div>
<div dir="ltr"><span>thanks</span></div>
<div dir="ltr"><span>shimon</span></div>
</div>
<div class="gmail_extra">
<br><br><div class="gmail_quote">
<div dir="ltr">2014-04-09 4:49 GMT+02:00  <span dir="ltr">&lt;<a href="mailto:openflow-discuss-request@..." target="_blank">openflow-discuss-request@...</a>&gt;</span>:</div>
<blockquote class="gmail_quote">Send openflow-discuss mailing list submissions to<br>
&nbsp; &nbsp; &nbsp; &nbsp; <a href="mailto:openflow-discuss@...">openflow-discuss@...</a><br><br>
To subscribe or unsubscribe via the World Wide Web, visit<br>
&nbsp; &nbsp; &nbsp; &nbsp; <a href="https://mailman.stanford.edu/mailman/listinfo/openflow-discuss" target="_blank">https://mailman.stanford.edu/mailman/listinfo/openflow-discuss</a><br>
or, via email, send a message with subject or body 'help' to<br>
&nbsp; &nbsp; &nbsp; &nbsp; <a href="mailto:openflow-discuss-request@...du">openflow-discuss-request@...</a><br><br>
You can reach the person managing the list at<br>
&nbsp; &nbsp; &nbsp; &nbsp; <a href="mailto:openflow-discuss-owner@...">openflow-discuss-owner@...</a><br><br>
When replying, please edit your Subject line so it is more specific<br>
than "Re: Contents of openflow-discuss digest..."<br><br><br>
Today's Topics:<br><br>
&nbsp; &nbsp;1. Re: Quick experimenting with OVS on OpenWrt (Alison Chan)<br>
&nbsp; &nbsp;2. Re: Quick experimenting with OVS on OpenWrt (Nicholas Bastin)<br>
&nbsp; &nbsp;3. Re: Quick experimenting with OVS on OpenWrt (Ben Pfaff)<br>
&nbsp; &nbsp;4. Re: Release of libfluid: ONF Driver Competition Winner w/<br>
&nbsp; &nbsp; &nbsp; OpenFlow 1.0 and 1.3 (Christian Esteve Rothenberg)<br>
&nbsp; &nbsp;5. Re: Release of libfluid: ONF Driver Competition &nbsp; Winner w/<br>
&nbsp; &nbsp; &nbsp; OpenFlow 1.0 and 1.3 (Murphy McCauley)<br><br><br>
----------------------------------------------------------------------<br><br>
Message: 1<br>
Date: Tue, 8 Apr 2014 17:29:20 -0400<br>
From: Alison Chan &lt;<a href="mailto:chan7781@...">chan7781 <at> kettering.edu</a>&gt;<br>
To: Ben Pfaff &lt;<a href="mailto:blp@...">blp@...</a>&gt;<br>
Cc: <a href="mailto:dev@...">dev@...</a>, openflow-discuss<br>
&nbsp; &nbsp; &nbsp; &nbsp; &lt;<a href="mailto:openflow-discuss@...">openflow-discuss@...</a>&gt;<br>
Subject: Re: [openflow-discuss] Quick experimenting with OVS on<br>
&nbsp; &nbsp; &nbsp; &nbsp; OpenWrt<br>
Message-ID:<br>
&nbsp; &nbsp; &nbsp; &nbsp; &lt;<a href="mailto:CABcSDhdTUirSo9SevVz6a6Fxt4eT-BPEAhiJVy1Qv%2BWoiNaoWg@...">CABcSDhdTUirSo9SevVz6a6Fxt4eT-BPEAhiJVy1Qv+WoiNaoWg@...</a>&gt;<br>
Content-Type: text/plain; charset=UTF-8<br><br>
Thanks Ben. I just subscribed to dev <at> ovs list.<br><br>
I was able to take the router back up to the lab (I had been testing<br>
it with laptops in my office). With two gigabit ethernet hosts on the<br>
data plane, throughput (tcp iperf) is 450~460 Mbit/s. I will test<br>
things a bit more tomorrow (e.g. adding 802.11 clients) and if all<br>
goes well then I will move all five of our TP-Links to OVS.<br><br>
I was not expecting throughput with OVS to be 11 times better than<br>
Pantou throughput -- I had to double check things to make sure the<br>
switch was still OpenFlow enabled and not just using hardware learning<br>
switch!<br><br>
Cheers,<br>
Alison<br><br>
2014-04-08 13:15 GMT-04:00 Ben Pfaff &lt;<a href="mailto:blp@...">blp@...</a>&gt;:<br>
&gt; [adding ovs-dev since there might be interest there]<br>
&gt;<br>
&gt; On Tue, Apr 08, 2014 at 11:56:28AM -0400, Alison Chan wrote:<br>
&gt;&gt; Hi everyone,<br>
&gt;&gt;<br>
&gt;&gt; I was able to get OpenVswitch running on one of our TL-WR1043ND<br>
&gt;&gt; routers running OpenWrt Attitude Adjustment. I was expecting somewhat<br>
&gt;&gt; better performance than Pantou, but not this much better:<br>
&gt;&gt;<br>
&gt;&gt; Pantou: 39.7 Mbit/s tcp iperf using static flows (in_port:1 output:2<br>
&gt;&gt; and in_port:2 output:1)<br>
&gt;&gt; OVS: 94.1 Mbit/s testing the same way<br>
&gt;&gt;<br>
&gt;&gt; At this point, the limiting factor is one of my laptops' 10/100<br>
&gt;&gt; ethernet interface! I will try to find another computer with gigabit<br>
&gt;&gt; ethernet and report back.<br>
&gt;&gt;<br>
&gt;&gt; Cheers,<br>
&gt;&gt; --<br>
&gt;&gt; Alison ChanI<br>
&gt;&gt; <a href="mailto:chan7781@...">chan7781@...</a><br>
&gt;&gt; SMS: +1 909 278 7753<br>
&gt;&gt; Fax: +1 206 203 2793<br>
&gt;&gt; _______________________________________________<br>
&gt;&gt; openflow-discuss mailing list<br>
&gt;&gt; <a href="mailto:openflow-discuss@...">openflow-discuss@...</a><br>
&gt;&gt; <a href="https://mailman.stanford.edu/mailman/listinfo/openflow-discuss" target="_blank">https://mailman.stanford.edu/mailman/listinfo/openflow-discuss</a><br><br><br><br>
--<br>
Alison Chan<br><a href="mailto:chan7781@...">chan7781@...</a><br>
SMS: +1 909 278 7753<br>
Fax: +1 206 203 2793<br><br><br>
------------------------------<br><br>
Message: 2<br>
Date: Tue, 8 Apr 2014 17:51:14 -0500<br>
From: Nicholas Bastin &lt;<a href="mailto:nick.bastin@...">nick.bastin@...</a>&gt;<br>
To: Alison Chan &lt;<a href="mailto:chan7781@...">chan7781 <at> kettering.edu</a>&gt;<br>
Cc: <a href="mailto:dev@...">dev@...</a>, openflow-discuss<br>
&nbsp; &nbsp; &nbsp; &nbsp; &lt;<a href="mailto:openflow-discuss@...">openflow-discuss@...</a>&gt;<br>
Subject: Re: [openflow-discuss] Quick experimenting with OVS on<br>
&nbsp; &nbsp; &nbsp; &nbsp; OpenWrt<br>
Message-ID:<br>
&nbsp; &nbsp; &nbsp; &nbsp; &lt;CADmMkWdqjcDrPbsGE=<a href="mailto:U8pWGhBk_Cc905FZXrk1ed_QPn-FCKag@...">U8pWGhBk_Cc905FZXrk1ed_QPn-FCKag@...l.com</a>&gt;<br>
Content-Type: text/plain; charset="utf-8"<br><br>
On Tue, Apr 8, 2014 at 4:29 PM, Alison Chan &lt;<a href="mailto:chan7781 <at> kettering.edu">chan7781@...</a>&gt; wrote:<br><br>
&gt; I was able to take the router back up to the lab (I had been testing<br>
&gt; it with laptops in my office). With two gigabit ethernet hosts on the<br>
&gt; data plane, throughput (tcp iperf) is 450~460 Mbit/s. I will test<br>
&gt; things a bit more tomorrow (e.g. adding 802.11 clients) and if all<br>
&gt; goes well then I will move all five of our TP-Links to OVS.<br>
&gt;<br><br>
There are two variants of this device - one with an AR9132 (boring white<br>
case), and one with a QCA9558 (blue spaceship-looking thing). &nbsp;The AR9132<br>
version uses a separate switch ASIC (AR8316 - I have one of these), so your<br>
max throughput through that device in any software forwarding (regardless<br>
of CPU speed) will be a gigabit. &nbsp;The 400Mhz ARM in that box is going to<br>
limit you even further, so 450Mbits seems a bit suspect.<br><br>
If you have the rev. 2 (blue) hardware, that is more of an unknown (I don't<br>
have one of these to look at or tinker with) - the 9558 has a 720Mhz MIPS<br>
core, so that's better, but I found images of a different AP that uses this<br>
chip on the internet and it appears to use an AR8327 for the switching, so<br>
likely there's no integrated ethernet switch on the SoC. &nbsp;This means your<br>
max throughput is still probably a gigabit, and you might be able to<br>
actually get 450Mbits through that CPU. &nbsp;The AR8327 actually has a 96-entry<br>
ACL table and 64-entry VLAN translation table (as well as a bunch of more<br>
specialized tables that would be more difficult to use well) that could be<br>
used for hardware acceleration, but I doubt anyone has done the work to<br>
program them.<br><br>
--<br>
Nick<br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: &lt;<a href="http://mailman.stanford.edu/pipermail/openflow-discuss/attachments/20140408/2957a37b/attachment-0001.html" target="_blank">http://mailman.stanford.edu/pipermail/openflow-discuss/attachments/20140408/2957a37b/attachment-0001.html</a>&gt;<br><br>
------------------------------<br><br>
Message: 3<br>
Date: Tue, 8 Apr 2014 16:00:23 -0700<br>
From: Ben Pfaff &lt;<a href="mailto:blp@...">blp@...</a>&gt;<br>
To: Nicholas Bastin &lt;<a href="mailto:nick.bastin@...">nick.bastin <at> gmail.com</a>&gt;<br>
Cc: "<a href="mailto:dev@...">dev@...</a>" &lt;<a href="mailto:dev@...">dev@...</a>&gt;, openflow-discuss<br>
&nbsp; &nbsp; &nbsp; &nbsp; &lt;<a href="mailto:openflow-discuss@...">openflow-discuss@...</a>&gt;<br>
Subject: Re: [openflow-discuss] Quick experimenting with OVS on<br>
&nbsp; &nbsp; &nbsp; &nbsp; OpenWrt<br>
Message-ID:<br>
&nbsp; &nbsp; &nbsp; &nbsp; &lt;<a href="mailto:CACjuMbzXnBsa3-xcAuRN3EuJuhEQjW3wTTNdgjW-qM-nGxmP2w@...">CACjuMbzXnBsa3-xcAuRN3EuJuhEQjW3wTTNdgjW-qM-nGxmP2w@...</a>&gt;<br>
Content-Type: text/plain; charset=ISO-8859-1<br><br>
On Tue, Apr 8, 2014 at 3:51 PM, Nicholas Bastin &lt;<a href="mailto:nick.bastin@...">nick.bastin@...</a>&gt; wrote:<br>
&gt; If you have the rev. 2 (blue) hardware, that is more of an unknown (I don't<br>
&gt; have one of these to look at or tinker with) - the 9558 has a 720Mhz MIPS<br>
&gt; core, so that's better, but I found images of a different AP that uses this<br>
&gt; chip on the internet and it appears to use an AR8327 for the switching, so<br>
&gt; likely there's no integrated ethernet switch on the SoC. &nbsp;This means your<br>
&gt; max throughput is still probably a gigabit, and you might be able to<br>
&gt; actually get 450Mbits through that CPU. &nbsp;The AR8327 actually has a 96-entry<br>
&gt; ACL table and 64-entry VLAN translation table (as well as a bunch of more<br>
&gt; specialized tables that would be more difficult to use well) that could be<br>
&gt; used for hardware acceleration, but I doubt anyone has done the work to<br>
&gt; program them.<br><br>
Wow, the AR8327 seems to have full public documentation:<br>
&nbsp; &nbsp; <a href="https://lafibre.info/images/doc/201106_spec_AR8327.pdf" target="_blank">https://lafibre.info/images/doc/201106_spec_AR8327.pdf</a><br>
At a glance, it's not that general-purpose but certainly nice to<br>
see a documented switching ASIC in a low-cost device.<br><br><br>
------------------------------<br><br>
Message: 4<br>
Date: Tue, 8 Apr 2014 21:34:08 -0300<br>
From: Christian Esteve Rothenberg &lt;<a href="mailto:chesteve@...icamp.br">chesteve@...</a>&gt;<br>
To: Murphy McCauley &lt;<a href="mailto:murphy.mccauley@...">murphy.mccauley@...</a>&gt;,<br>
&nbsp; &nbsp; &nbsp; &nbsp; "<a href="mailto:openflow-discuss@...du">openflow-discuss@...</a>"<br>
&nbsp; &nbsp; &nbsp; &nbsp; &lt;<a href="mailto:openflow-discuss@...">openflow-discuss@...</a>&gt;<br>
Subject: Re: [openflow-discuss] Release of libfluid: ONF Driver<br>
&nbsp; &nbsp; &nbsp; &nbsp; Competition Winner w/ OpenFlow 1.0 and 1.3<br>
Message-ID:<br>
&nbsp; &nbsp; &nbsp; &nbsp; &lt;CAEj5p9TOj8gTy8Pk-WOObY9Zw9ov-Q+ws5U=50w3oTk=<a href="mailto:ThG07A@...">ThG07A@...</a>&gt;<br>
Content-Type: text/plain; charset=ISO-8859-1<br><br>
Thanks Murphy,<br><br>
one of my students working with POX wants to give a try and hopefullly<br>
contribute to these efforts, can you share the pointers to that POX<br>
branch. We can only say positive things about prototypiing with POX --<br>
giving it clean and effective OF1.3 support would be a neat upgrade<br>
beneficial to all parties :)<br><br>
-Christian<br><br>
&gt;<br>
&gt;<br>
&gt; On Mon, Mar 24, 2014 at 10:17 PM, Murphy McCauley &lt;<a href="mailto:murphy.mccauley@...">murphy.mccauley@...</a>&gt; wrote:<br>
&gt;&gt;<br>
&gt;&gt; Congratulations.<br>
&gt;&gt;<br>
&gt;&gt; There's now POX branch that's been quickly hacked up to (sort of) use libfluid's Python bindings. &nbsp;It also includes a minor patch for one of the swig .i files.<br>
&gt;&gt;<br>
&gt;&gt; -- Murphy<br>
&gt;&gt;<br>
&gt;&gt; On Mar 22, 2014, at 5:08 AM, Christian Esteve Rothenberg &lt;<a href="mailto:chesteve@...">chesteve@...</a>&gt; wrote:<br>
&gt;&gt;<br>
&gt;&gt; Dear OpenFlow fellows,<br>
&gt;&gt;<br>
&gt;&gt; in case you are not aware about the public release of the winner implementation of the OpenFlow driver competition (<a href="https://www.opennetworking.org/competition" target="_blank">https://www.opennetworking.org/competition</a>) here is the pointer to the github repository:<br>

&gt;&gt;<br>
&gt;&gt; <a href="http://opennetworkingfoundation.github.io/libfluid/" target="_blank">http://opennetworkingfoundation.github.io/libfluid/</a><br>
&gt;&gt;<br>
&gt;&gt; libluid may be interesting to developers of both OpenFlow switches and controllers. It features support of OpenFlow 1.0 and 1.3, high performance, bindings to Python and Java, easy port to different hardware architectures, etc.<br>

&gt;&gt;<br>
&gt;&gt; We welcome users and developers interested in building an open community to maintain libfluid as a useful, multi-purpose OpenFlow library to develop switch agents and controller implementations.<br>
&gt;&gt;<br>
&gt;&gt; -Christian (on behalf of the libfluid team)<br>
&gt;&gt; _______________________________________________<br>
&gt;&gt; openflow-discuss mailing list<br>
&gt;&gt; <a href="mailto:openflow-discuss@...">openflow-discuss@...</a><br>
&gt;&gt; <a href="https://mailman.stanford.edu/mailman/listinfo/openflow-discuss" target="_blank">https://mailman.stanford.edu/mailman/listinfo/openflow-discuss</a><br>
&gt;&gt;<br>
&gt;&gt; --<br>
&gt;&gt; Christian<br><br><br><br><br>
--<br>
Christia<br><br><br>
------------------------------<br><br>
Message: 5<br>
Date: Tue, 8 Apr 2014 19:49:23 -0700<br>
From: Murphy McCauley &lt;<a href="mailto:murphy.mccauley@...">murphy.mccauley@...</a>&gt;<br>
To: Christian Esteve Rothenberg &lt;<a href="mailto:chesteve@...amp.br">chesteve@...</a>&gt;<br>
Cc: "<a href="mailto:openflow-discuss@...">openflow-discuss@...</a>"<br>
&nbsp; &nbsp; &nbsp; &nbsp; &lt;<a href="mailto:openflow-discuss@...">openflow-discuss@...</a>&gt;<br>
Subject: Re: [openflow-discuss] Release of libfluid: ONF Driver<br>
&nbsp; &nbsp; &nbsp; &nbsp; Competition &nbsp; &nbsp; Winner w/ OpenFlow 1.0 and 1.3<br>
Message-ID: &lt;<a href="mailto:A69FEFAD-402B-41EC-A588-B38E3BDD2897 <at> gmail.com">A69FEFAD-402B-41EC-A588-B38E3BDD2897@...</a>&gt;<br>
Content-Type: text/plain; charset="iso-8859-1"<br><br>
Sure, it's just the libfluid_experiment branch of the main fork:<br><a href="http://noxrepo.org/git/pox/tree/libfluid_experiment" target="_blank">http://noxrepo.org/git/pox/tree/libfluid_experiment</a><br><br>
It's definitely very much an experiment. ;) &nbsp;For starters, I was only messing with 1.0 and while I used libfluid to read off the wire, the responses are still generated with POX's OpenFlow library. &nbsp;I just thought it was worth experimenting to get a sense of what would be involved. &nbsp;If you guys are interested in doing much more work on this, we should talk!<br><br>
-- Murphy<br><br>
On Apr 8, 2014, at 5:34 PM, Christian Esteve Rothenberg &lt;<a href="mailto:chesteve@...">chesteve@...</a>&gt; wrote:<br><br>
&gt; Thanks Murphy,<br>
&gt;<br>
&gt; one of my students working with POX wants to give a try and hopefullly<br>
&gt; contribute to these efforts, can you share the pointers to that POX<br>
&gt; branch. We can only say positive things about prototypiing with POX --<br>
&gt; giving it clean and effective OF1.3 support would be a neat upgrade<br>
&gt; beneficial to all parties :)<br>
&gt;<br>
&gt; -Christian<br>
&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; On Mon, Mar 24, 2014 at 10:17 PM, Murphy McCauley &lt;<a href="mailto:murphy.mccauley@...">murphy.mccauley@...</a>&gt; wrote:<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; Congratulations.<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; There's now POX branch that's been quickly hacked up to (sort of) use libfluid's Python bindings. &nbsp;It also includes a minor patch for one of the swig .i files.<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; -- Murphy<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; On Mar 22, 2014, at 5:08 AM, Christian Esteve Rothenberg &lt;<a href="mailto:chesteve@...">chesteve@...</a>&gt; wrote:<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; Dear OpenFlow fellows,<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; in case you are not aware about the public release of the winner implementation of the OpenFlow driver competition (<a href="https://www.opennetworking.org/competition" target="_blank">https://www.opennetworking.org/competition</a>) here is the pointer to the github repository:<br>

&gt;&gt;&gt;<br>
&gt;&gt;&gt; <a href="http://opennetworkingfoundation.github.io/libfluid/" target="_blank">http://opennetworkingfoundation.github.io/libfluid/</a><br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; libluid may be interesting to developers of both OpenFlow switches and controllers. It features support of OpenFlow 1.0 and 1.3, high performance, bindings to Python and Java, easy port to different hardware architectures, etc.<br>

&gt;&gt;&gt;<br>
&gt;&gt;&gt; We welcome users and developers interested in building an open community to maintain libfluid as a useful, multi-purpose OpenFlow library to develop switch agents and controller implementations.<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; -Christian (on behalf of the libfluid team)<br>
&gt;&gt;&gt; _______________________________________________<br>
&gt;&gt;&gt; openflow-discuss mailing list<br>
&gt;&gt;&gt; <a href="mailto:openflow-discuss@...">openflow-discuss@...</a><br>
&gt;&gt;&gt; <a href="https://mailman.stanford.edu/mailman/listinfo/openflow-discuss" target="_blank">https://mailman.stanford.edu/mailman/listinfo/openflow-discuss</a><br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; --<br>
&gt;&gt;&gt; Christian<br>
&gt;<br>
&gt;<br>
&gt;<br>
&gt;<br>
&gt; --<br>
&gt; Christia<br><br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: &lt;<a href="http://mailman.stanford.edu/pipermail/openflow-discuss/attachments/20140408/01b051dc/attachment.html" target="_blank">http://mailman.stanford.edu/pipermail/openflow-discuss/attachments/20140408/01b051dc/attachment.html</a>&gt;<br><br>
------------------------------<br><br>
Subject: Digest Footer<br><br>
_______________________________________________<br>
openflow-discuss mailing list<br><a href="mailto:openflow-discuss@...">openflow-discuss <at> lists.stanford.edu</a><br><a href="https://mailman.stanford.edu/mailman/listinfo/openflow-discuss" target="_blank">https://mailman.stanford.edu/mailman/listinfo/openflow-discuss</a><br><br><br>
------------------------------<br><br>
End of openflow-discuss Digest, Vol 66, Issue 6<br>
***********************************************≤br>
</blockquote>
</div>
<br>
</div>
</div>
Picon

Re: Release of libfluid: ONF Driver Competition Winner w/ OpenFlow 1.0 and 1.3

Thanks Murphy,

one of my students working with POX wants to give a try and hopefullly
contribute to these efforts, can you share the pointers to that POX
branch. We can only say positive things about prototypiing with POX --
giving it clean and effective OF1.3 support would be a neat upgrade
beneficial to all parties :)

-Christian

>
>
> On Mon, Mar 24, 2014 at 10:17 PM, Murphy McCauley
<murphy.mccauley@...> wrote:
>>
>> Congratulations.
>>
>> There's now POX branch that's been quickly hacked up to (sort of) use libfluid's Python bindings.  It also
includes a minor patch for one of the swig .i files.
>>
>> -- Murphy
>>
>> On Mar 22, 2014, at 5:08 AM, Christian Esteve Rothenberg
<chesteve@...> wrote:
>>
>> Dear OpenFlow fellows,
>>
>> in case you are not aware about the public release of the winner implementation of the OpenFlow driver
competition (https://www.opennetworking.org/competition) here is the pointer to the github repository:
>>
>> http://opennetworkingfoundation.github.io/libfluid/
>>
>> libluid may be interesting to developers of both OpenFlow switches and controllers. It features
support of OpenFlow 1.0 and 1.3, high performance, bindings to Python and Java, easy port to different
hardware architectures, etc.
>>
>> We welcome users and developers interested in building an open community to maintain libfluid as a
useful, multi-purpose OpenFlow library to develop switch agents and controller implementations.
>>
>> -Christian (on behalf of the libfluid team)
>> _______________________________________________
>> openflow-discuss mailing list
>> openflow-discuss@...
>> https://mailman.stanford.edu/mailman/listinfo/openflow-discuss
>>
>> --
>> Christian

--

-- 
Christia
Alison Chan | 8 Apr 17:56 2014

Quick experimenting with OVS on OpenWrt

Hi everyone,

I was able to get OpenVswitch running on one of our TL-WR1043ND
routers running OpenWrt Attitude Adjustment. I was expecting somewhat
better performance than Pantou, but not this much better:

Pantou: 39.7 Mbit/s tcp iperf using static flows (in_port:1 output:2
and in_port:2 output:1)
OVS: 94.1 Mbit/s testing the same way

At this point, the limiting factor is one of my laptops' 10/100
ethernet interface! I will try to find another computer with gigabit
ethernet and report back.

Cheers,
--

-- 
Alison Chan
chan7781@...
SMS: +1 909 278 7753
Fax: +1 206 203 2793
Namrah Tariq | 5 Apr 15:11 2014
Picon

Extract MAC header

Hi I'm working on wireless in openflow. I want to extract the mac header of 802.11 of the packets that arrive at the controller pox. I cannot find a way to do that, I can extract the Ethernet header but that only gives me the mac addresses and the protocol type, I need to extract the frame control bytes of the mac header. Can someone please help me with this. Any help is appreciated. Thank you in advance.
<div><div dir="ltr">Hi I'm working on wireless in openflow. I want to extract the mac header of 802.11 of the packets that arrive at 
the controller pox. I cannot find a way to do that, I can extract the 
Ethernet header but that only gives me the mac addresses and the protocol type, I
 need to extract the frame control bytes of the mac header. Can someone please help me 
with this. Any help is appreciated. Thank you in advance.</div></div>

Gmane