Hi Team,
I have a 7 machine setup in which one machine is running openflow
controller,one machine is running openflow switch (1.0.0) and other five
machine are connected to the openflow switch machine on
eth1,eth2,eth3,eth4,eth5.Open flow controller is connected on the eth0 of
the openflow swith. I am using user space switch. Following are the commands
i am executing to run the open flow controller and switch
Controller : sudo ./controller ptcp:6633 -v
OpenFlow Switch:
./udatapath/ofdatapath punix:/var/run/dp0.sock -i
eth1,eth2,eth3,eth4,eth5 --no-local-port
./secchan/ofprotocol unix:/var/run/dp0.sock tcp:192.168.5.50:6633
--out-of-band
After that i tried to ping all the machine from each other and these
machines are pinging each other properly. But i am seeing following error
messages from ofdatapath
|netdev|WARN|error sending Ethernet packet to eth4 :Message too long.
&
|netdev|WARN|Dropped <number> message in last <number> seconds due to
excessive rate.
All the 7 machin's i created on VM machine and followed the below links to
setup
http://www.openflow.org/wk/index.php/UserspaceSwitch_OnVMware
http://yuba.stanford.edu/foswiki/bin/view/OpenFlow/Deployment/HOWTO/LabSetup#OpenFlow_Controller_NOX_0_6
I checked the CPU usage of the VMWare machine which is running this software
openflow switch and its between 10-15% and also network usage are on average
50kbps.
MTU for all the eth{1-5} is set to default 1500.
Can you please let me know if is there any known issues or any pointer on
how to figure out why openflow switch is throwing these errors.
Thanks
Anil
_______________________________________________
openflow-discuss mailing list
[email protected]
https://mailman.stanford.edu/mailman/listinfo/openflow-discuss
