I suspect discovery is fine and the problem is in topology, which is
(unfortunately) a piece of junk and I should just delete it until something
better can be written. I suggest that if you want to track topology, you just
watch discovery events. See POXDesk's tinytopo.py for an example.
You
On Nov 17, 2012, at 1:49 PM, Tmusic wrote:
However, if both topology and tinytopo are running, the graph in POXDesk
lacks some switches (same number as topology).
Does topology influence tinytopo when they are both running?
Yes. Topology is raising an exception when switches are
On Dec 11, 2012, at 9:57 AM, Christoph Koehler wrote:
Can Pox do wildcards on MAC addresses? Looks like the 1.1 spec allows for
masks on IP and MAC, but I don't see anything like that in Pox. Could you
clarify?
Current versions of mainline POX only support OpenFlow 1.0, so MAC wildcards
was reading a
paper (PortLand from Sigcomm 09) that says that OF can do longest prefix
matching on MAC addresses (end of section 3.2). Didn't think that was the
case, so I wanted the check.
Thanks again for the clarification!
On Tue, Dec 11, 2012 at 11:08 AM, Murphy McCauley murphy.mccau
On Dec 11, 2012, at 11:37 AM, rohit patil wrote:
Can anyone let me know the process of adding flows and deleting them via POX
controller in mininet.
Have you taken a look at the POX manual and the OpenFlow Tutorial?
https://openflow.stanford.edu/display/ONL/POX+Wiki
If you're still working on this, you might take a look at the new mac_blocker
component.
-- Murphy
On Dec 13, 2012, at 12:34 AM, rohit patil wrote:
Hello team,
Can anyone suggest me, in which module of POX I need to write / modify the
code by which I should be able to limit the hosts
On Jan 22, 2013, at 4:17 AM, Julius Bachnick wrote:
Works like a charm with the betta branch and your suggestion ;).
Glad to hear it!
-- Murphy
On Jan 22, 2013, at 8:12 AM, Ali Al-Khamis wrote:
if of.match.nw_proto == 6
This line needs a colon at the end.
-- Murphy
On Jan 23, 2013, at 3:23 AM, Ian Beatty-Orr wrote:
Here is the function that creates the of_stats_request message:
def _handle_ConnectionUp (self, event):
info_getter(event.connection)
bd = of.ofp_port_stats_request( port_no = OFPP_NONE )
pm = of.ofp_stats_request(
On Jan 28, 2013, at 9:17 AM, Julius Bachnick wrote:
thank you so much for your support so far! My current issue (and I admit it's
a bit urgent ;)) is to listen to Changes in a Port Status of an OVS, for
example if a port goes down or is disconnected.
According to the POX Wiki this can be
On Jan 30, 2013, at 5:34 PM, DON EHDED wrote:
i am new to pox but i have managed to install it on ubuntu 10.04 i also
connected 3 netfpga openflow switches to it with 2 hosts connected to to
switch 1, 2 reactively,...
i want to ping host 1 for host 2 and vice versa ,...
however i keep
On Feb 7, 2013, at 10:01 PM, Shabbir Ahmed wrote:
another reason could be that the switch in tp-link is not 100% openflow
compatible?
Well, the switch you're running is pretty much all software, so that's not
likely to be it.
can u guide me how to confirm if im doing the right thing
On Feb 14, 2013, at 3:39 PM, Murphy McCauley wrote:
Given that, it seems like the most reasonable way to implement it is by
keeping a timestamp of the most recent read in the Connection object (though
I am not entirely certain that I want to do this...).
I'm still not entirely certain I
On Feb 19, 2013, at 7:25 PM, Sam Russell wrote:
Great news! I suspect the work required to port my OF1.1 work to OF1.2 would
be similar to writing OF1.2 support from scratch based on the new
libopenflow. I'm happy to get going with that, but I'm not sure how to make
it interface with the
On Mar 19, 2013, at 10:35 AM, Saul St. John wrote:
Regarding pt 1: does running POX under PyPy allow for the program to take
simultaneous advantage of all the cores in a multiprocessor machine? IOW,
does PyPy not suffer from the single executing thread limitation that
CPython's GIL
On Mar 19, 2013, at 12:02 PM, Tmusic wrote:
I'm trying to run with pypy but it can't import some of my modules (all in
the /pox directory). For example myrouter.mylinkmonitor does not work, but my
myrouter.mypackethandler does.
python pox.py modules works like a charm
./pox.py modules
On Mar 22, 2013, at 12:09 PM, Brian Miller wrote:
Remove an extra ')'.
---
pox/openflow/libopenflow_01.py |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/pox/openflow/libopenflow_01.py b/pox/openflow/libopenflow_01.py
index 4905465..f81eb0c 100644
---
Responses inline.
On Mar 26, 2013, at 12:00 AM, Senthil wrote:
- I have started the pox with l2_learning - pox py forwarding.l2_learning,
my_app.
- Sent an HTTP packets from VM attached to the switch with dst -port 8000.
The HTTP server in the remote VM responds this request.
Working
On Mar 28, 2013, at 9:30 AM, Kouvakas Alexandros wrote:
Hello,
I am using openflow (openvswitch) on planetlab nodes.
I have connected a controller on a switch on my central node.
Despite the fact that POX is running very well I cannot have the command line
POX
to make some choices. I am
On Mar 29, 2013, at 11:40 PM, chenli wrote:
1. The meaning for registering
Is core.registerNew(Mycomponent) the function to register to core object, and
we can use the methods in core which defined in /pox/pox/core.py ?
If I register the Mycomponent to core, the other components can
On Mar 31, 2013, at 2:30 AM, Hong Wayne wrote:
I want to ask how to determine the orders for the multiple components be
executed on POX?
For example, If I want to let test2 be executed before test1, how can I do
that?
./pox.py forwarding.l2_learning openflow.of_01 --address=xxx
On Apr 1, 2013, at 3:46 AM, chenli wrote:
Thanks. Now I add some extra events and named it as Fail_server_invoke.py
like below:
from pox.core import core
import pox.openflow.libopenflow_01 as of
from pox.lib.revent.revent import EventMixin
log = core.getLogger()
class
On Apr 2, 2013, at 10:45 PM, yashwanth kp wrote:
I wanted to know if I can submit a patch to pox.
I have never submitted a patch before. So it will be really helpful if I get
to know the details on submitting a patch if I can submit.
Patch submissions are welcome. Fixes or minor additions
On Apr 4, 2013, at 1:46 PM, Balázs Németh wrote:
Hello all!
I'd like to send a OFPT_STATS_REQUEST message for a switch for debugging
reasons. In the OF specification it is said the switch must reply wit a
ofp_table_stats message. How can I catch this message in the controller?
Should I
this?
On Fri, Apr 5, 2013 at 5:36 AM, Murphy McCauley murphy.mccau...@gmail.com
wrote:
On Apr 4, 2013, at 4:08 PM, Sayed Qaiser Ali Shah Shah wrote:
The link which you provide was very helpful and I found match class in that
and I use this code
import pox.openflow.libopenflow_01 as of # POX
On Apr 8, 2013, at 2:57 PM, Mayumi Park Campos wrote:
I have implemented the pox controller with a openflow stanford's software
reference design openflow switch emulated on a pc and my problem is that when
a try to run pox as forwarding l2_learning behavior my “switch” do not work
as switch
I don't entirely follow your topology description. It looks like you're
testing this on a network with other hosts; you might try simplifying things
(e.g., removing devices for the purposes of testing).
In the log you sent, are you trying to communicate between your two hosts? Are
their
Does this happen reliably? If so, in /home/sdn/pox/pox/openflow/__init__.py
at line 282, can you insert a log or print statement like:
print str(value),' ',repr(value)
.. and send the new console output?
-- Murphy
On Apr 9, 2013, at 6:10 AM, 吴亚楠 wrote:
help!
sdn@IPL204:~/pox$ ./pox.py
On Apr 9, 2013, at 3:11 PM, Karthik Sharma wrote:
#Next we create tap devices on the host machine so we can bind them to
the guests VM's later.
for tap in `seq 1 5`; do
sudo ip tuntap add mode tap lan0-p$tap
done;
I wouldn't necessarily expect tap devices to work.
I responded to this message on pox-dev earlier today. I can see what's going
wrong, but I can't immediately figure out why it's happening.
It'd be helpful if you could add a print statement to openflow/__init__.py
right before line 282:
print str(value),' ',repr(value),' ',self
.. and send me
By default, switches usually send incomplete packets to the switch in case of a
table miss. For many protocols (including TCP), the incomplete packet is
usually enough to parse all the headers and it's just payload that gets chopped
off. In the case of DHCP, it often means that the DHCP
overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:4645 (4.6 KB) TX bytes:4645 (4.6 KB)
On 10 April 2013 15:25, Murphy McCauley murphy.mccau...@gmail.com wrote:
By default, switches usually send incomplete packets to the switch in case of
a table miss
April 2013 17:58, Murphy McCauley murphy.mccau...@gmail.com wrote:
There have only been a few users of the modern (betta) version, AFAIK. But
the messenger component is specifically JSON-oriented. If you want to do
binary messages, you can:
1) Write a plain old socket server thing
On Apr 12, 2013, at 2:30 AM, Kouvakas Alexandros wrote:
Note that only the SENDER (the OF switch) has the controller. By this I mean
that I run ovs-vsctl set-controller only on the SENDER node.
So each machine is running OVS, but only one of the OVS instances is connecting
to a controller?
13, 2013, at 4:12 PM, Sayed Qaiser Ali Shah Shah wrote:
Thanks for your reply. Is there anyway that dpid is taken automatically not
to be entered manually??
On Sun, Apr 14, 2013 at 4:00 AM, Murphy McCauley murphy.mccau...@gmail.com
wrote:
The event object also has a .dpid attribute which
On Apr 14, 2013, at 3:19 PM, Karthik Sharma wrote:
entry = SourcetoPort(src_address=packet.src , port_no=packet_in.in_port)
I am not looking too closely, but you seem to be adding the literal string
packet.src to your database. You probably want to be adding the actual
source address.
--
On Apr 14, 2013, at 3:47 PM, Karthik Sharma wrote:
What is the type of packet.src?
You could determine this with a print type(packet.src) statement. It's
probably an EthAddr. I don't know enough about your database to say, but it
probably makes sense to convert it to a string to store it
No, I think the latter --controller just overrides the former. I don't
actually think there's a decent way to do this in Mininet. In earlier
versions, I think there was missing infrastructure. In Mininet 2, I think
there's still some stuff missing and you could almost make it work except for
On Apr 17, 2013, at 2:20 PM, Tmusic wrote:
Sorry for the extremely late reply!
Thank you very much for the example code!!
No problem; glad it was helpful.
I've also finally cleaned up my profiling code and pushed it to github:
https://github.com/Timmmy/pox in the betta branch (sorry that
On Apr 17, 2013, at 3:50 PM, Mayumi Park Campos wrote:
I have a problem with the conection between the pox controller and
openvswitch.
Are you sure you have a problem?
And on the controller this is what I see:
WARNING:forwarding.l2_learning:Same port for packet from 00:0c:ce:1f:10:1d -
Do you have other things plugged into the hubs?
Even with the topology you have, it looks like something strange is happening.
What has the MAC 00:0c:ce:1f:10:1d and where is it connected?
-- Murphy
On Apr 18, 2013, at 8:44 AM, Mayumi Park Campos wrote:
Thanks for the answer I think I know
The OpenFlow support in ns-3 is based on a pre-1.0 version of OpenFlow and has
never been upgraded. POX doesn't support this version. One would need to do
some hacking on the ns-3 OpenFlow model to upgrade it to 1.0, which would be
GREAT, but nobody has done it.
(Alternately, one could hack
There are any number of ways you might do this. Two examples...
One would be to have your POX component just watch the config file to see if it
changes. When it changes, reconfigure.
POX also contains the messenger subsystem which is infrastructure for
communicating with outside applications
/messenger/example.py and in interactive prompt 'from pox.messenger import
*' then reading help(Channel) or what not. Wondering if there are other
documentations of example that I can get quickly up to speed on usage?
Eric
On Apr 22, 2013, at 9:27 PM, Murphy McCauley murphy.mccau
say 192.168.3.1/24) without
running OVS on them. I mean, can I substitute the tunnel from the host to the
OVS with something else?
2013/4/12 Murphy McCauley murphy.mccau...@gmail.com
On Apr 12, 2013, at 2:30 AM, Kouvakas Alexandros wrote:
Note that only the SENDER (the OF switch) has
It sounds like you're talking about monitoring and manipulating the contents of
queues, which is a very fine-grained operation that must operate at very small
timescales. This isn't addressed by OpenFlow and probably isn't suitable for a
remote controller. OpenFlow allows for assigning
Just as a quick sanity check -- are you sure you're running the right version
of Mininet (or at least the latest one)?
You may also find more help on the Mininet mailing list. There's obviously
crossover in users between the two, but this question isn't actually about POX
and is about
You do...
session.query(exists().where(SourcetoPort.src_address ==
str(packet.dst))).scalar()
Do exists() queries always return true or false? In that case, your if
statement would ALWAYS be true.
On the other hand, if dropping the exists...
session.query(where(SourcetoPort.src_address ==
still talking
about queues? If you mean queues, the answer is no. OVS and OpenFlow are not
meant for queue manipulation.
-- Murphy
On Fri, May 3, 2013 at 3:32 AM, Murphy McCauley murphy.mccau...@gmail.com
wrote:
It sounds like you're talking about monitoring and manipulating the contents
In a couple weeks if y'all want to ping me on this issue, maybe I can put
together a helpful example.
-- Murphy
On May 1, 2013, at 11:43 AM, kk yap wrote:
Hi Igor,
(cc-ing pox-dev)
I did not have much luck reusing the select loop in POX. I ended up
(in my mind) with the worst option
Ping sends packets from A to B, and B sends them back to A. Ping does this
continually until you stop it unless you specify a limit on the commandline.
Since you're installing flows, you'd actually expect them to start hitting the
flow table and stop hitting the switch, but the fewest you
:
sorry I had added a -c1 at the end of the command.
mininet h1 ping h3
On 3 May 2013 17:16, Murphy McCauley murphy.mccau...@gmail.com wrote:
Ping sends packets from A to B, and B sends them back to A. Ping does this
continually until you stop it unless you specify a limit
Whether you post it here or elsewhere, that error message is not that helpful.
If you get a stack trace, you should post the whole thing.
-- Murphy
On May 2, 2013, at 3:11 PM, Balázs Németh wrote:
Hi all!
I would like to set a link some options (bandwidth), on one of my topos.
I tried
It appears that something else is listening on the same port as POX (6633).
You can use lsof to find it. Or just check the process list for likely
candidates (other OpenFlow controllers or other instances of POX).
Or you could just switch which port POX listens to from 6633 to something
reach 40 then is there anyway to
delete entries from flow table by sending some sort of command from remote
controller. and how that command should be initiated?
On Fri, May 3, 2013 at 3:58 AM, Murphy McCauley murphy.mccau...@gmail.com
wrote:
On May 2, 2013, at 3:52 PM, Sayed Qaiser Ali Shah
If you look up __del__ in the Python reference, you'll see that Python gives no
assurance that destructors will be called for objects which exist when the
Python interpreter quits. POX follows this model and does not try to fix it.
I can think of two solutions off the top of my head:
1) Use
On May 15, 2013, at 2:14 AM, chenli wrote:
I did some test again, and it's so strange that the back flow for ssh
connection needs the field of nw_tos=16.
I don't think this is strange. I believe 16 is the TOS value for minimize
delay, which makes sense for ssh, since it's interactive. The
On May 15, 2013, at 8:13 AM, y...@gteccom.uff.br wrote:
I´m using mininet and POX and I have two questions:
II) How do I add a flow_mod without being involved in an event? I want to
install proactively rules in the switches, but I only have achieved this when
listening events. I didn't
On May 19, 2013, at 1:04 AM, Shabbir Ahmed wrote:
I have a question wat do we need to do run NAT when openflow is running,
i have tp-link but i its WAN port for controller connection and want to run
PAT on this port, this port is not part of Datapath in openflow.
If the port isn't part of
Most recently, I have done this by creating a veth pair where one end is on the
OpenFlow datapath. The other end, therefore, is just like a port to another
computer on a LAN; you can do normal Linux IP forwarding between it and your
WAN interface, complete with iptables-based NAT. And if you,
What you're doing wrong is not clear in the OpenFlow 1.0 spec. 1.0.1 clarified
it, and Open vSwitch's documentation is also fairly clear on the subject.
Matches have prerequisites. Basically, you can't match on something from a
deeper layer without matching on the appropriate thing in the
to a host on the same 'side' works now, and I've gotten rid of
all the errors. But one side still can't talk to the other side. Am I
doing ARM wrong?
On Fri, May 24, 2013 at 1:13 AM, Murphy McCauley
murphy.mccau...@gmail.com wrote:
At least one of your problems is the same as was recently
Are you using a switch that has the NXT_SET_PACKET_IN_FORMAT and
NXT_FLOW_MOD_TABLE_ID Nicira extensions?
I'm not sure which version of Open vSwitch is the first to support these...
maybe 1.8?
Glancing at the code, I actually don't immediately see where the new packet-in
format is actually
. I usually don't remember to
hit Reply All. :)
Alison Chan
(mobile)
On 28 May 2013 23:01, Murphy McCauley murphy.mccau...@gmail.com wrote:
On May 28, 2013, at 7:50 AM, Alison Chan wrote:
Thank you for pointing that out. The flow mods sent later, had the
previously set actions in them
:
Thanks Murphy.. It worked correctly..
On Wed, May 29, 2013 at 12:52 PM, Murphy McCauley murphy.mccau...@gmail.com
wrote:
I'd say the right way to do it is to remove the event listeners set by one
component, and then set them up using the other. l2_multi wasn't really
written
Are you sure you've got the upgraded version running? Just to make sure so we
don't spend time chasing down a dead end, try running this component:
http://www.noxathome.org/x/Murphy/switch_info.py
If that does show that you've got the new version running, capture the control
traffic using
The interfaces on s0 and s1 that connect to the other nodes in your topology
are part of an Open vSwitch datapath. Thus, packets on them do not go through
the normal Linux networking stack and do not have corresponding entries in the
Linux routing table. Instead, they are processed by OVS's
enough to help me with this.. Where I am going wrong ?
On Thu, May 30, 2013 at 10:31 AM, Murphy McCauley murphy.mccau...@gmail.com
wrote:
The interfaces on s0 and s1 that connect to the other nodes in your topology
are part of an Open vSwitch datapath. Thus, packets on them do not go
The WARNING:forwarding.l2_multi:Can't get from 96:bb:06:61:71:8d to
ae:b8:16:4d:a3:6b message indicates that your new version of _get_path
returned None rather than a path and has nothing to do with clearing the flow
tables. You need to fix your _get_path so that it returns a path in the
I assume you mean you want to add a flow to a specific table. This feature is
not available in OpenFlow 1.0. So you either need to use a higher version of
OpenFlow (which POX doesn't support yet and OVS only recently has gained
experimental support for), or using a Nicira extension.
There
Yes, each switch should have a unique dpid. I don't think I have ever seen
otherwise with Mininet. Are you using one of the standard Mininet topologies,
or could it be a bug in the script that's initializing Mininet?
Also, you will sometimes see the same DPID in a ConnectionUp more than once
Just pushed the fix to the carp branch. (It should backport to the betta
branch just fine and I'll accept a pull request if someone wants to do it. :) )
Thanks for the detailed report.
-- Murphy
On Jun 4, 2013, at 11:20 AM, Alison Chan wrote:
Hello,
When I attempt to suppress console logs,
On Jun 12, 2013, at 5:32 AM, hansinie vitharana wrote:
Hi,
I am trying to add flows to the POX controller and I want to give following
commands..
POXfrom pox.core import core
POXimport pox.openflow.libopenflow_01 as of
POXmsg = of.ofp_flow_mod()
POXmsg.priority = 10
I've added a section on the host_tracker component to the documentation --
mostly just by copy/pasting from the docstring in the code.
(Since the POX documentation is a wiki, anyone could have done this.)
Also, it's worth noting that the version of host_tracker in the carp branch has
at least
You can make the thread a daemon thread by setting the thread object's .daemon
property to True. This will stop it from keeping the process alive, but it can
make teardown ugly so you can get race conditions when shutting down which lead
to exceptions. This may not be particularly
It'll take some rewriting to make this work, for at least two reasons:
1) l3_learning uses OpenFlow 1.0 strictly which has no IPv6. To do IPv6,
you'll have to use Nicira extensions... at the minimum you'll need extensible
match. There's an example of this in l2_nx and l2_nx_self_learning.
2)
On Jul 2, 2013, at 7:25 AM, nibble nibble wrote:
I have a l2.learningSwitch and a network with a few switches. For every
packet I receive I want to determine packet was received from which one of
the switches. For this I used packet_in.in_port . But its does not seem that
it indicate
On Jul 4, 2013, at 7:42 AM, Windhya Rankothge wrote:
I want my controller to add a tag to the packet header.. Something similar to
the VLAN ID concept..
Can I do it ? Can i change the packet header using controller and add a
custom tag to the packet ?
Plese be kind to help me with
I think sendToDPID() just doesn't do what you think it does. It is for sending
OpenFlow commands to a switch -- the same as connection.send(), except it looks
up the connection for you based on DPID. I believe you're trying to use it to
send a packet out of some port on a switch. There's an
On Jul 8, 2013, at 9:13 AM, nibble nibble wrote:
to get the packet and process it I have the following code.
packet = event.parsed # This is the parsed packet data.
packet_in = event.ofp
I want to figure out which host the packet is coming from(either its ip or
its name).
I have used
On Jul 11, 2013, at 8:31 AM, adria sole wrote:
Thank you Murphy for the information, I solved my error.
Now I am trying to create my own action but I get:
ERROR:openflow.of_01:[00-23-20-12-b2-91 1] OpenFlow Error:
[00-23-20-12-b2-91 1] Error: header:
[00-23-20-12-b2-91 1] Error:
of.ofp_action_dl_addr.dst mac (that is the
same of old destination, because the host has change its IP but not the mac)
and of.ofp_action_output(port=prt) (also this is the same). Maybe I put it in
a wrong position
2013/7/23 Murphy McCauley murphy.mccau...@gmail.com
I assumed you were trying to look
I don't think you've provided enough information for an explanation to be
offered.
Please see the final question on the POX wiki's FAQ for guidance to help us
help you.
-- Murphy
On Jul 23, 2013, at 9:23 PM, Hong Wayne wrote:
Dear all:
I faced a problem when OpenFlow switch connect to
The POX wiki has a link to an extended version of the stats example by William
Emmanuel Yu which polls the stats periodically (Statistics Collector
Example). Start with that, and change the _handle_flowstats_received()
function to just print len(event.stats).
-- Murphy
On Jul 24, 2013, at
In the spec, ofp_flow_mod doesn't have a data parameter. POX has a special
case to support a common pattern, and your code isn't it.
You should just do a flow_mod followed by a packet_out (possibly to the
OFPP_TABLE special port).
-- Murphy
On Jul 26, 2013, at 9:00 AM, Keqiang He wrote:
I
: openflow.webservice
I am running the betta branch.
Is there any solution?
2013/4/22 Murphy McCauley murphy.mccau...@gmail.com
You should do a pull. This was fixed in the betta branch a month ago.
-- Murphy
On Apr 22, 2013, at 12:07 PM, ahmed A. wrote:
Hi,
I am trying
There are two aspects to intelligently handling vendor messages.
The first is to hook the vendor message unpacking process in libopenflow so
that it can unpack your vendor message into a custom class.
The second is hooking how core.openflow handles vendor messages. Usually this
will mean
component.
But isn't it true that the extra RTT delay that I see at the first ping
packet is because of going first to the controller ?
2013/7/31 Murphy McCauley murphy.mccau...@gmail.com
Answers inline.
On Jul 30, 2013, at 9:44 AM, Kouvakas Alexandros wrote:
Pretty nice
I think it's time to do some belated spring cleaning in the POX repo. There
are some temporary and feature branches which I think should just be deleted,
as well as some longer-lived ones that I at least want to address (I am fine
with things that are useful, but I don't want useless things
...@gmail.com wrote:
NAT is not yet possible, u can use iptables for NAT as one interface out of
of-datapath.
are you trying it on mininet or some other platform?
On Thu, Jul 18, 2013 at 11:41 PM, Murphy McCauley murphy.mccau...@gmail.com
wrote:
samples.mixed_switches demonstrates treating
I would like to ask you how can I get/retrieve the flow table of certain
switch
Best regards,
Amer
On ٢١/٠٨/٢٠١٣, at ٣:١٧ ص, Murphy McCauley murphy.mccau...@gmail.com wrote:
Sorry, there's currently no example for this.
There are a couple of prerequisites:
1) You must be using
to
my component.py but I don't like this idea...
Ty in advance!
2013/8/18 Murphy McCauley murphy.mccau...@gmail.com
macEntry.lastTimeSeen is just a float, not a function. All you need to do
are remove the parentheses from it on line 280.
This was fixed in the carp branch three months
Please see the final entry of the POX FAQ:
https://openflow.stanford.edu/display/ONL/POX+Wiki
-- Murphy
On Aug 30, 2013, at 2:19 PM, AMER AL-GHADHBAN amer7...@hotmail.com wrote:
Hello,
Hope you the best of wealth :)
Some times I am facing this error:
AttributeError: 'ipv4' object has
In general, you should never have reason to call _handle_PacketIn() yourself.
It is an event handler and is called by the OpenFlow component when a packet
has been sent to the controller from a switch.
If you want to keep track of more data, it seems like you should be able to do
it exactly
Is there a way of doing NATing other than what i post in my email; if there
is may you help me in how to do it.
Best regards,
Amer
في ٣١/٠٨/٢٠١٣، الساعة ١١:٣٧ م، كتب Murphy McCauley
murphy.mccau...@gmail.com:
On Aug 31, 2013, at 4:12 AM, AMER AL-GHADHBAN amer7...@hotmail.com wrote:
I am
That commandline should work.
Can you post more info? For example, run with logging at DEBUG level and post
the log. Also... why do you think it isn't working? Why do you think POX
isn't handling it? In general, please see the last section of the FAQ in the
manual, and post as much
Ah, I was initially wrong in saying it should work.
Note the conspicuous warning:
WARNING:core:Still waiting on 1 component(s)
And the informative message above it:
DEBUG:core:startup() in pox.forwarding.l2_multi still waiting for:
openflow_discovery
And the line in the l2_multi docstring:
On Sep 9, 2013, at 7:39 PM, Xu Zhongxing xu_zhong_x...@163.com wrote:
Does of.ofp_flow_mod(command=of.OFPFC_DELETE) support cookie mask now? I
would like to use cookie mask to filter cookies when deleting flows.
No. libopenflow_01 only speaks wire protocol version 0x01, which does not
Most (or all?) of the forwarding components POX ships with will cause
broadcast/flood storms when used on topologies with loops by themselves. This
isn't a property of POX, but of the specific forwarding components. It can be
resolved in multiple ways. One way is to disable flooding on some
This very likely has nothing to do with host_tracker at all and the real
problem is that you are running a forwarding component that's not prepared to
deal with loops.
See the other pox-dev thread from today involving loops, and/or check out the
loop question in the FAQ.
-- Murphy
On Sep 11,
1 - 100 of 402 matches
Mail list logo
