18:01:00 <bh526r> #startmeeting gluon
18:01:01 <openstack> Meeting started Wed Nov 16 18:01:00 2016 UTC and is due to finish in 60 minutes.  The chair is bh526r. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:01:02 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
18:01:05 <openstack> The meeting name has been set to 'gluon'
18:01:18 <bh526r> Hi guys
18:01:26 <krenczewski> Hello
18:01:33 <bh526r> Hi Kamil
18:01:44 <bh526r> #topic Roll Call and Introduction
18:01:53 <bh526r> #info Bin Hu
18:01:59 <Jeffreyc42> #info Jeff Collins
18:01:59 <krenczewski> #info Kamil Renczewski
18:02:53 <bh526r> #topic Admin Update
18:03:09 <bh526r> #info No admin update
18:03:34 <bh526r> #topic Gluon Tasks
18:04:04 <bh526r> #info Tasks for Ocata and beyond is listed on wiki:
18:04:09 <bh526r> #link https://wiki.openstack.org/wiki/Gluon/Tasks-Ocata
18:05:06 <bh526r> #info Etherpad of discussion of tasks:
18:05:11 <bh526r> #link https://etherpad.openstack.org/p/ocata-gluon-work-plan
18:05:57 <bh526r> #info some work in progress in patches:
18:06:06 <bh526r> #link https://review.openstack.org/#/dashboard/self
18:07:03 <bh526r> #info current patches are primarily focusing on (1) service binding model - 2 patches pending discussion and reslution
18:07:33 <bh526r> #info (2) Testing (3) Documentation for Gluon Developers
18:08:17 <bh526r> #info (3) documentation is also pending on the resolution of (1) service binding model so as to document it more accurately
18:09:11 <bh526r> #info other tasks are still in progress, and pending on relevant patches
18:09:39 <bh526r> #topic Status Update
18:09:51 <bh526r> anyone wants to share something?
18:10:41 <bh526r> @Kamil, how is everything going on your side?
18:11:04 <krenczewski> I was absent for some days
18:11:14 <krenczewski> Now I am trying to catch up
18:11:15 <bh526r> I see
18:11:40 <krenczewski> Still have some problems with BGP peering between Nuage and Juniper node
18:12:34 <bh526r> I see. settings? RT/RD value doesn't match?
18:12:52 <krenczewski> And I have implementation of ML2 mechanism driver for Contrail
18:13:13 <krenczewski> as Nachi suggested
18:13:27 <bh526r> Ok.
18:13:56 <krenczewski> I think the problem is more basic as I don't have real experience with BGP
18:14:08 <bh526r> Did you get a chance to integrate your Contrail ML2 with Gluon's ML2 Plugin that Tom developed?
18:14:30 <krenczewski> not yet
18:14:41 <bh526r> Ok.
18:15:17 <bh526r> Regarding BGP peering, last time in June, what Nachi did was to configure those settings in Contrail so as to peer with other SDN Controllers
18:15:43 <bh526r> Of course, other SDN Controllers need to set up the same matching values respectively
18:16:25 <bh526r> And perhaps some controller-specific tricks
18:16:56 <krenczewski> OK, I'll ask him for details
18:18:51 <bh526r> And you may also want to integrate your Contrail's ML2 driver with current Gluon's Neutron Plugin so that it can be completed by working with Contrail too.
18:19:38 <bh526r> #info Kamil is working on BGP peering of Contrail with Nuage
18:19:47 <krenczewski> ok
18:20:15 <bh526r> #info Kamil will complete the integration of Contrail's ML2 driver with Gluon's Neutron ML2 Plugin
18:20:26 <bh526r> Thank you Kamil
18:20:41 <bh526r> anyone else have update?
18:21:31 <bh526r> Looks like everyone is working on the review of patches
18:22:56 <bh526r> #info Ian has discussed our patches 390513 with Nova team, and got useful feedback.
18:23:43 <bh526r> anything else?
18:24:29 <bh526r> #topic AOB
18:24:58 <bh526r> Because of Thanksgiving Holiday in US next week, I suggest to cancel our IRC meeting next week.
18:25:34 <bh526r> Any other opinion?
18:26:37 <trevormc> Agreed. I have something to disuss in AOB.
18:26:49 <bh526r> good. go ahead
18:27:12 <trevormc> I had a call  Monday about integrating att's VFd into gluon.
18:27:35 <trevormc> we were wondering if that is something gluon wanted to do?
18:27:52 <trevormc> https://github.com/att/vfd
18:29:00 <trevormc> the call was about implementing a sr-iov VFd agent to pass information to a backend. We originally aimed for an ML2 plugin in neutron but we are now exploring the possibility of integrating it with gluon. Thoughts?
18:29:24 <bh526r> How VFD works? What is the role in the overall NFV Architecture?
18:30:13 <bh526r> Does it work with a SDN controller?
18:30:26 <trevormc> we were hoping we could discuss that with you, I just wanted to know if the team was on board with that. Yes it does work with an SDN controller
18:31:06 <bh526r> If it works with an SDN controller, it should be the south bound of Gluon
18:31:53 <bh526r> And it is on the dataplane, if I am not mistaken
18:32:30 <bh526r> Gluon and Proton are more of the control plane
18:33:23 <trevormc> It works with SR-IOV and that I believe is on Layer 2 is that what you're calling the dataplane?
18:33:41 <bh526r> Then how it works should be that VFd agent needs to be integrated with an SDN controller, e.g. ODL or Contrail, for dataplane acceleration
18:34:34 <trevormc> It's designed to extract information from Layer 2 to be passed to a backend like an SDN but not specifically SDNs
18:34:48 <bh526r> SR-IOV, DPDK, VPP etc. are different types of dataplane acceleration methods
18:35:16 <trevormc> thats my understanding of it anyway. I was hoping I could schedule a call to get you in touch with people who can explain it better.
18:35:36 <trevormc> and any other gluon contributors here
18:35:52 <bh526r> Ok, let's have a call to help us understand it better
18:36:44 <bh526r> Extracting information from L2 and pass to backend all seem like dataplane functions
18:37:10 <bh526r> Anyway let's schedule a call to discuss
18:37:27 <trevormc> ok great, I will schedule something soon.
18:38:16 <bh526r> #info Trevor will schedule a followup conversaton to discuss https://github.com/att/vfd
18:39:01 <bh526r> anything else?
18:40:15 <bh526r> #info all agreed to cancel the IRC meeting next week (11/23) to observe Thanksgiving Holiday week
18:40:59 <bh526r> If nothing else, I suggest to adjourn the meeting and give back everyone 20 minutes
18:41:12 <bh526r> #info meeting adjourned
18:41:21 <bh526r> Thank you everyone, and have a great day
18:41:28 <bh526r> #endmeeting