Carlos, you should try to become an ONF Research Associate by submitting your application. I think you can easily get valuable endorsements to support your association.
BTW, which OF software switch are you using for OF1.4 prototyping? ONF EXT welcomes feature prototyping work and certainly lessons learned from those like you and Zoltán that realize potential shortcomings on the spec when getting the hands dirty.... Cheers, Christian On Tue, May 6, 2014 at 9:45 AM, Carlos Ferreira <carlosmf...@gmail.com> wrote: > Thank you for your answer Zoltan. I'm actually working with OF1.4. > > I don't see a reason to not add acknowledgment messages. Increase of > complexity is not really a good excuse simply because, yes the protocol may > be simpler but the software implementation complexity increases. > > I hop this changes in the future. I would also like to contribute with > proposals but I don't know how, without paying to become a ONF member. > > > On 6 May 2014 13:07, Zoltán Lajos Kis <zoltan.lajos....@ericsson.com> wrote: >> >> Hi Carlos, >> >> >> >> I definitely agree with you that a positive acknowledgement is needed in >> OpenFlow; especially since the introduction of groups (and meters). In fact >> I made a proposal more than two years ago in ONF for this feature. Some were >> objecting it, as it would add complexity to the protocol, or as barriers can >> already provide similar function. So this feature never got into the >> specification. >> >> >> >> My guess is that most people are still using OpenFlow in an OF 1.0 style, >> and so are fine without acknowledgements of any sort. Now that we are >> getting more and more OF 1.3 features supported in both controllers and >> switches I hope people will start using those features and realize the need >> for a light-weight positive acknowledgements. >> >> >> >> For now you can use barrier request/reply after messages to get the >> acknowledgment, or in OF 1.4 you execute one or more messages within a >> bundle. >> >> >> >> Regards, >> >> Zoltan >> >> >> >> >> >> From: openflow-discuss >> [mailto:openflow-discuss-boun...@lists.stanford.edu] On Behalf Of Carlos >> Ferreira >> Sent: Monday, May 05, 2014 7:35 PM >> To: openflow-discuss@lists.stanford.edu >> Subject: [openflow-discuss] OFP_FLOW_MOD response from OF Switch to >> Controller. >> >> >> >> Hi! >> >> Can anyone explain me why is that in the specs, there are error messages >> specified to respond to a bad configuration but, there is no specification >> for a simple OK of ACK message for a good configuration. >> >> Is there a good reason or a good programming principle for not specifying >> OK responses for successful configuration operations? >> >> >> >> -- >> >> >> >> Carlos Miguel Ferreira >> >> Researcher at Telecommunications Institute >> >> Aveiro - Portugal >> >> Work E-mail - c...@av.it.pt >> >> Skype & GTalk -> carlosmf...@gmail.com >> >> LinkedIn -> http://www.linkedin.com/in/carlosmferreira > > > > > -- > > Carlos Miguel Ferreira > Researcher at Telecommunications Institute > Aveiro - Portugal > Work E-mail - c...@av.it.pt > Skype & GTalk -> carlosmf...@gmail.com > LinkedIn -> http://www.linkedin.com/in/carlosmferreira > > _______________________________________________ > openflow-discuss mailing list > openflow-discuss@lists.stanford.edu > https://mailman.stanford.edu/mailman/listinfo/openflow-discuss > -- Christian _______________________________________________ openflow-discuss mailing list openflow-discuss@lists.stanford.edu https://mailman.stanford.edu/mailman/listinfo/openflow-discuss
