> On Aug 27, 2015, at 10:44 PM, Yusuke Iwase <[email protected]> wrote:
> 
> If OVS work so, I think it is a bug of OVS and tester.py should not wait OVS.

I believe you are right.   In the process of investigating this I captured all 
the OpenFlow packets between Ryu and the switches and looked at them with 
Wireshark.  There seems to be no problem with how tester.py uses the OpenFlow 
protocol.

The flaw is that the OF agent that OVS uses will update its own data base 
(tables of flows) then signal via the Barrier Request Reply that this is done.  
It should wait until the flow tables all the way into the cache are updated but 
it does not.   However it appears this will be very hard to implement due to 
the architecture of OVS which separates the data base (ovsdb) from the actual 
switching logic (openvswitchd and the dkms kernel module.)

However since so many OF products are based on OVS I wonder if tester.py should 
have an additional command line option to invoke this work around.

----------------------------
Alan Deikman
ZNYX Networks, Inc.



------------------------------------------------------------------------------
_______________________________________________
Ryu-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ryu-devel

Reply via email to