18:00:24 <bh526r> #startmeeting gluon
18:00:25 <openstack> Meeting started Wed May 31 18:00:24 2017 UTC and is due to finish in 60 minutes.  The chair is bh526r. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:00:26 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
18:00:28 <openstack> The meeting name has been set to 'gluon'
18:00:39 <bh526r> #topic Roll Call
18:00:45 <bh526r> #info Bin Hu
18:00:50 <pcarver> hi
18:00:55 <pcarver> #info Paul Carver
18:00:55 <bh526r> Hi Paul
18:01:22 <jinli> Hi all
18:01:28 <jinli> #info JinLi
18:01:35 <bh526r> Hi Jin
18:02:10 <bh526r> Paul, I kept on receiving emails regarding ice cream social, pop corn social etc. in New Jersey :)
18:02:30 <bh526r> You guys are cool
18:03:20 <bh526r> Looks like only 3 of us
18:03:26 <pcarver> Yeah, apparently the new organization has an admin who sends that sort of stuff to everyone
18:03:26 <bh526r> #topic Admin Update
18:03:52 <bh526r> I wish she could ship some to me :)
18:04:29 <bh526r> #info No admin update
18:04:54 <bh526r> #topic Pike Release Update
18:05:28 <bh526r> Jin, I tried to merge Version management patch https://review.openstack.org/#/c/459894/
18:05:39 <bh526r> But failed in Jenkins
18:05:41 <jinli> yes, I am looking at it
18:05:58 <jinli> looks like both of my patches fail on the same error
18:06:09 <bh526r> That's right.
18:06:14 <bh526r> Recheck didn't work
18:06:32 <jinli> NoSuchOptError: no such option api in group [DEFAULT]
18:06:52 <bh526r> Was it because of some sort of deprecated lib in other projects, like oslo?
18:07:01 <bh526r> Very strange error
18:07:14 <jinli> don't know yet,
18:07:22 <jinli> need to look more into it
18:07:53 <jinli> I think it could be deprecated lib
18:08:13 <bh526r> or related to doc utils
18:08:34 <bh526r> Can you also ping Tin and see if he has clue?
18:08:39 <jinli> yes
18:08:42 <jinli> I will
18:08:55 <bh526r> Great
18:09:10 <jinli> I think he also rechecked my patches yesteray
18:09:31 <jinli> will talk to him later today
18:09:37 <bh526r> Yes, he did. But didn't work
18:09:49 <bh526r> sounds good.
18:09:53 <jinli> hopefully we can find out whats going on by end of day
18:10:06 <bh526r> that will be awesome :)
18:10:52 <bh526r> Not sure if Ian is here or not
18:11:39 <bh526r> We need his update regarding more details proposal of integrating with Neutron
18:11:58 <bh526r> And Sukhdev
18:12:38 <bh526r> Armando gave some comments on his 2 RFEs regarding Neutron improvement
18:13:04 <pcarver> I still don't really understand Ian's explanation of why we need a network type field that's separate from the already existing network_type field
18:13:04 <bh526r> We need Sukhdev's further action
18:13:45 <bh526r> His explanation is that semantically, we cannot mix 2 types of networks in one field
18:14:05 <bh526r> so existing network_type field is for different types of L2 networks
18:14:18 <pcarver> yeah, I understood the words but don't see what the reason is
18:14:29 <bh526r> And a new network type field for new L3 networks
18:14:34 <pcarver> I don't see where it is stated that network_type can only hold L2 values
18:14:51 <pcarver> other than Ian saying it I mean
18:14:52 <bh526r> It doesn't say, but current usage is for L2
18:15:11 <pcarver> I've hear Ian say it many times, but I haven't heard anyone else say it, nor seen it in any documentation
18:16:20 <bh526r> It is not documented, I guess because it is designed for L2 at the beginning, so I assume it didn't mention it because it was a no-brainer at that time
18:16:58 <bh526r> But the usage, and current values are all for L2
18:17:40 <bh526r> So if we put L3 values there, it could be confusing
18:18:05 <bh526r> So Ian thinks it is a cleaner method to have a new network type field dedicated to L3 networks
18:19:00 <krenczewski> Hi, sorry I am late
18:19:12 <bh526r> Hi Kamil, how are you?
18:19:19 <bh526r> Thank you for joining
18:19:23 <krenczewski> I was on week vacation
18:19:28 <krenczewski> Great, thanks :-)
18:19:33 <bh526r> How is your vacation?
18:19:34 <pcarver> I guess we should find out if any neutron cores are confused. The review comment seemed more confused by why we would want to add an additional network type field.
18:20:12 <bh526r> That's right.
18:20:29 <pcarver> Unless there are actual problems in the code, it would seem simpler to write some documentation that explains what the valid values of the existing field are and document that it can distinguish both L2 and L3 network types
18:21:03 <pcarver> Otherwise we need to write documentation explaining the difference between the two network type fields or that'll cause confusion as well.
18:21:21 <bh526r> Right.
18:21:44 <bh526r> Confusing one way or the other. So documentation is critical here
18:22:04 <bh526r> We need to see more details from Ian and Sukhdev regarding their new thoughts of how to integrate with Neutron.
18:22:27 <bh526r> Then we can revisit those 2 RFEs, and revise our approach
18:23:56 <bh526r> @krenczewski, welcome back, and hope you have had great time on your vacation
18:24:31 <bh526r> What is your plan for next week (after you are back :)?
18:24:36 <krenczewski> It was great, I am working now on documenting the whole Contrail installation process
18:24:45 <bh526r> Excellent.
18:24:49 <bh526r> Thank you
18:25:18 <krenczewski> It is not so easy to do everything right to deploy contrail so it works
18:25:57 <krenczewski> It may be that I forgot about something and need to look into same topic again
18:25:57 <bh526r> That's right, lots of complex configurations to deploy Contrail to make it work.
18:26:14 <krenczewski> Anyway I am making constant progress
18:26:24 <bh526r> :) It always happens. So documentation is very important
18:26:26 <krenczewski> And I can see the finish line in couple next days
18:26:49 <bh526r> Great, thank you.
18:26:49 <krenczewski> I hope I am correct with my estimations
18:26:56 <bh526r> :)
18:27:01 <pcarver> krenczewski: I'll try following your instructions and let you know if anything doesn't work
18:27:34 <krenczewski> OK Paul, I'll let you know when I finish
18:27:41 <bh526r> #info Kamil is working documenting the deployment process to make Contrail work with Gluon framework
18:28:26 <bh526r> #info Ian and Sukhdev are working on more details regarding how to integrate with Neutron
18:28:37 <bh526r> #info There are recent comments on 2 RFEs
18:29:14 <bh526r> #info After Ian and Sukhdev finish those details, we will revisit those 2 RFEs, and revise our approach if needed
18:29:34 <bh526r> #info Jin is debugging the Jenkins failure
18:30:17 <bh526r> Anything else from anybody?
18:31:16 <bh526r> If nothing else, we may finish earlier today, and give back everyone 30 minutes
18:31:36 <bh526r> Paul can enjoy ice cream, pop corn, or whatever social :)
18:32:04 <bh526r> #info Meeting adjourned
18:32:20 <bh526r> Thank you everyone, and hope we will have more to share and discuss next week
18:32:28 <krenczewski> Thanks
18:32:50 <bh526r> Bye all
18:32:54 <bh526r> #endmeeting