16:00:02 <dougwig> #startmeeting neutron lbaas 16:00:03 <openstack> Meeting started Tue Jan 13 16:00:02 2015 UTC and is due to finish in 60 minutes. The chair is dougwig. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:04 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:06 <dougwig> #chair blogan 16:00:07 <openstack> The meeting name has been set to 'neutron_lbaas' 16:00:08 <openstack> Current chairs: blogan dougwig 16:00:11 <dougwig> #topic roll call and agenda 16:00:14 <johnsom> o/ 16:00:16 <dougwig> #link https://wiki.openstack.org/wiki/Network/LBaaS 16:00:18 <dougwig> hi folks 16:00:23 <dougwig> wake up wake up! 16:00:24 <blogan> here! 16:00:49 <ajmiller> good morning 16:01:06 <dougwig> #topic Announcements 16:01:07 <xgerman> o/ 16:01:28 <dougwig> Kilo-2 ends February 5, so we really need to get at least the first big part of lbaasv2 in by then (more on that in a minute.) 16:01:31 <dougwig> Any other announcements? 16:01:38 <rm_work> po/ 16:01:42 <rm_work> err, o/ 16:01:56 <dougwig> we translated morning adam. 16:02:06 <dougwig> ok, moving on to the fun part. 16:02:12 <a2hill> 0/ 16:02:18 <dougwig> #topic Mid-cycle meetup details 16:02:24 <dougwig> #link https://etherpad.openstack.org/p/lbaas-kilo-meetup 16:02:42 <dougwig> we have tried to list discrete tasks that need completing this mid-cycle. 16:03:08 <dougwig> please update your attendance info and signup for items. lbaas v2, tis, l7, cli, and tempest are things that we really need in order to be able to ship this thing in Kilo. 16:03:22 <dougwig> also add any items that you know of. 16:03:50 <dougwig> please take a moment to look at it/edit it now, and let's circle back with questions or comments. 16:04:24 <xgerman> also make sure to add your name to the color 16:04:58 <rm_work> ack my color is sbalukoff's color :/ 16:05:12 <dougwig> i think i have three colors on that etherpad now. 16:05:17 * TrevorV missed the link 16:05:35 <johnsom> #link https://etherpad.openstack.org/p/lbaas-kilo-meetup 16:05:37 <dougwig> #link https://etherpad.openstack.org/p/lbaas-kilo-meetup 16:06:13 <TrevorV> thanks guys 16:07:23 <xgerman> also I assume we will have a remote option... 16:07:59 <blogan> yes, but not for you 16:08:16 <sballe> morning 16:08:25 <blogan> im sure we can set up something similar to what we did in seattle 16:08:25 <xgerman> I figured that I need to send johnsom and ajmiller with spy tech ;-) 16:08:39 <rm_work> yeah, hopefully should have something up 16:08:42 <dougwig> send spy tech for everyone. 16:08:45 <sbalukoff> Morning, folks! 16:08:48 <sballe> blogan: that would be great! I cannot travel that week either 16:08:59 <rm_work> I can bring my camera and TrevorV can bring his studio mic setup :P 16:09:00 <dougwig> morning. for the newcomers, we're reviewing/editing this: https://etherpad.openstack.org/p/lbaas-kilo-meetup 16:09:08 <dougwig> please update your attendance info and signup for items. lbaas v2, tis, l7, cli, and tempest are things that we really need in order to be able to ship this thing in Kilo. 16:09:17 <TrevorV> rm_work don't volunteer my stream equipment... crazy head 16:09:24 <rm_work> TrevorV: too late just did :P 16:10:11 <xgerman> awesome!! 16:10:15 <blogan> agent is a really needed one, and pretty fun 16:11:01 <dougwig> yeah, the top ones that need owners: agent driver and CLI/client. 16:11:15 <dougwig> i'd like to see those merged by the end of the mid-cycle. 16:11:43 <rm_work> did ctracey get pulled off the client stuff from BBG? 16:11:49 <dougwig> yes. 16:11:54 <dougwig> he got busy with other things. 16:11:55 <rm_work> :( 16:12:00 <xgerman> ctracyes's stuff actually works - we played with it extensively 16:12:01 <blogan> his code is still up there somewhere 16:12:07 <rm_work> k 16:12:08 <dougwig> rohara is working on the octavia client. i'm not sure if he's also planning on neutron lbaas, but he's not on the attendance list. 16:12:21 <blogan> yeah i havent seen him around lately 16:12:25 <rm_work> is rohara HP? 16:12:28 <xgerman> we might be able to reach out and get him remotely 16:12:29 <blogan> red hat 16:12:29 <dougwig> client person can decide whether to go straight to openstack-client. it would sidestep the debates that sidelined ctracey 16:12:31 <rm_work> ah right 16:13:06 <rm_work> is openstack-client further along now? 16:13:36 <blogan> devstack is using it 16:13:38 <blogan> i believe 16:13:41 <rm_work> hmm 16:13:56 <dougwig> figuring out whether we use that or not is the first task in getting the client done. :) 16:14:14 <xgerman> maybe we can figure that out beforehand ;-) 16:14:27 <blogan> xgerman: are you volunteering for the client portion? 16:14:29 <blogan> :) 16:15:10 <dougwig> we will have three cores handy (myself and brandon local, mestery remote), so i'm hoping for some fast iteration. 16:15:13 <johnsom> I will work on the client if rohara isn't 16:15:18 <xgerman> I just think we can make the decision openstack vs/. newutron before the hackathon 16:15:36 <blogan> oh im sure we can 16:15:37 <dougwig> johnsom: great, can you note that on the etherpad? 16:15:51 <dougwig> let's have the client as a topic next week. 16:16:00 <xgerman> +1 16:16:02 <dougwig> unless someone has enough info to advocate today. 16:16:07 <johnsom> I thought we had some guidance that openstack client was the way of the future and we should head in that direction 16:16:36 <dougwig> i believe that guidance was from the openstack-client people. :) 16:16:52 <dougwig> my real question is, which one is most accessible to our users in the Kilo timeframe. 16:17:17 <blogan> i vaguely remember rohara saying writing for openstack-client was much easier than neutronclient 16:17:32 <rm_work> woo 16:17:39 <johnsom> I will dig around a bit and see what is going on with the openstack-client 16:17:40 <dougwig> otoh, we have code for the neutronclient 16:17:50 <blogan> very true 16:17:54 <xgerman> yep, which actually works quite well 16:17:54 <dougwig> ok, johnsom is on that one. 16:18:01 <sballe> would it make sense that johnsom write down the pros and cons and we decide at the next meeting? 16:18:03 <blogan> which will need a few modifications 16:18:10 <xgerman> sballe +1 16:18:34 <dougwig> #action johnsom research neutron client vs openstack client 16:18:39 <dougwig> any other comments/questions/other we need for logistics about the mid-cycle? 16:18:56 <SantoshNetScaler> Hi All , I am new neutron lbaas. I am curious to know if we can use mentioned neutron-lbass from git to devstack setup ? 16:19:17 <dougwig> SantoshNetScaler: hi, and welcome. can you ask that in open discussion in a few minutes? 16:19:30 <SantoshNetScaler> Sure 16:20:32 <dougwig> ok, we'll be bugging people to be owners/working with owners more this week, to try and get lbaas v2 ready for coding on the first day of the sprint. 16:20:34 <dougwig> moving on 16:20:41 <dougwig> #topic Open Discussion 16:20:58 <rm_work> if anyone has *zero* budget like I usually do -- I have a guest room. just FYI :P 16:21:02 <dougwig> SantoshNetScaler: neutron-lbaas is used automatically if you specify q-lbaas in your conf file. 16:21:08 <rm_work> … also a super long-hair cat 16:22:48 <blogan> btw this is what I believe is the final draft of the "improved" neutron lbaas v2 API 16:22:50 <blogan> #link https://review.openstack.org/#/c/144833/ 16:23:29 <rm_work> WIP? 16:23:33 <blogan> just the extension, but if you can read the RESOURCE_ATTRIBUTE_MAP (which is a task in itself) you can see it 16:23:47 <xgerman> did we ever settle the status question? 16:24:09 <blogan> WIP bc there is the possibility of change, like the statuses 16:24:23 <blogan> and the plugin code in neutron-lbaas is still a WIP 16:25:03 <blogan> xgerman: I believe we loosely decided that we would just add operating and provisioning statuses 16:25:09 <blogan> but thats not in the extension yet 16:25:22 <xgerman> so I guess we should talk that through bext weekm too 16:25:31 <xgerman> so we cna hit the ground running at the Hackathon 16:25:36 <blogan> agreed 16:25:41 <blogan> or we can talk about it nwo 16:25:43 <blogan> now 16:25:58 <dougwig> #topic Object status 16:26:03 <dougwig> it's magic. 16:26:11 <xgerman> the status? 16:26:20 <dougwig> no, the meeting room can hear us. 16:26:31 <sbalukoff> Heh! 16:26:39 <xgerman> I have trouble channeling my inner Radware and they had the most issues with status 16:26:57 <blogan> so statuses to be like what octavia already has? 16:27:12 <blogan> which woudl lead into another question 16:27:31 <blogan> how much do we care about migrating v1 to v2 16:27:46 <xgerman> I wouldn't care at all 16:27:49 <blogan> lol 16:27:55 <dougwig> i don't, but i also don't want to be in a nova-network scenario. 16:28:20 <xgerman> we never met the unicorn which was running LBaaS V1 16:28:27 <blogan> this would be 1% of nova-networks user base probably 16:28:37 <dougwig> we should ask the upgrade question in the next neutron meeting. 16:28:49 <blogan> good idea 16:28:53 <xgerman> +1 16:28:55 <sballe> dougwig: +1 16:28:57 <blogan> thats the monday afternoon one 16:29:04 <blogan> and before the next lbaas meeting 16:29:33 <dougwig> #action dougwig put lbaasv1 upgrade question on neutron on-demand agenda for next week 16:29:35 <sbalukoff> I think we would all like to avoid the v1->v2 upgrade. 16:29:40 <blogan> in the meantime, do we just want to shoot for the statuses with what we have in octavia? 16:29:45 <sballe> sbalukoff: +1 16:29:53 <dougwig> is that oper/prov separate fields? 16:30:04 <blogan> yeah but prov only on lb and listener 16:30:04 <sbalukoff> er... migration 16:30:26 <xgerman> sounds good - as I said Radware ahd the most trouble with that 16:30:31 <sballe> sbalukoff: I knew what you meant ;-) 16:30:36 <dougwig> blogan: odd, i'd think prov would be on everything and oper would be less. 16:31:14 <rm_work> doesn't look like we have anyone from radware today? don't see samuel 16:31:20 <blogan> dougwig: well if you change a child of lb or listener, the lb and/or listener's prov status gets put into PENDING_* 16:31:42 <dougwig> err, provisioning is the one that doesn't change based on the tree. 16:31:53 <dougwig> or, i guess that'd be driver specific. 16:32:26 <dougwig> let's take this one offline/into channel and circle back next week? 16:32:32 <blogan> sounds good 16:32:41 <dougwig> #topic Open Discussion 16:32:46 <dougwig> anything else today? 16:33:07 <SantoshNetScaler> I have enable neutron service with devstack setup. I want to use neutron-lbaas v2 API 16:33:40 <blogan> SantoshNetScaler: did you enable q-lbaas in devstack conf? 16:33:48 <SantoshNetScaler> yes 16:34:11 <xgerman> isn't v2 still in reviews and you would need to get it straight from git 16:34:16 <xgerman> ? 16:34:22 <blogan> SantoshNetScaler, then you'll have to make a few changes in the neutron.conf and restart neutron-server 16:34:32 <dougwig> xgerman: the core api is in. if you want to run it all the way, you need a review. 16:34:47 <blogan> SantoshNetScaler: the only driver available for it now though is the noopdriver 16:34:48 <xgerman> thanks 16:34:49 <dougwig> if you just want to hit the api with your own driver, what's in the repo will suffice. 16:34:57 <xgerman> cool 16:35:22 <dougwig> SantoshNetScaler: join us in #openstack-lbaas if you have questions outside of meeting time. 16:35:51 <dougwig> anything else for today? 16:36:21 <dougwig> alright, let's blast out of here. thanks, folks. 16:36:25 <sballe> bye 16:36:33 <a2hill> 3,2,1 16:36:37 <blogan> #endmeeting