20:00:39 <xgerman> #startmeeting octavia
20:00:39 <openstack> Meeting started Wed Apr 22 20:00:39 2015 UTC and is due to finish in 60 minutes.  The chair is xgerman. Information about MeetBot at http://wiki.debian.org/MeetBot.
20:00:40 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
20:00:43 <openstack> The meeting name has been set to 'octavia'
20:00:48 <rm_work> o/
20:00:52 <ajmiller> o/
20:00:52 <mwang2> o/
20:00:55 <johnsom> o/
20:01:01 <xgerman> Agenda: https://wiki.openstack.org/wiki/Octavia/Weekly_Meeting_Agenda#Meeting_2015-04-22
20:01:05 <xgerman> #link https://wiki.openstack.org/wiki/Octavia/Weekly_Meeting_Agenda#Meeting_2015-04-22
20:01:22 <xgerman> #topic  Announcements
20:01:50 <xgerman> Octavia/LBaaS/VPNaaS/FWaaS Midcycle: https://etherpad.openstack.org/p/LBaaS-FWaaS-VPNaaS_Summer_Midcycle_meetup
20:02:01 <xgerman> #link https://etherpad.openstack.org/p/LBaaS-FWaaS-VPNaaS_Summer_Midcycle_meetup
20:02:11 <sballe> o/ sorry for being late I was in the wrong channe;;;;;;;l
20:02:22 <xgerman> bo worries it's jus 120s
20:02:29 <Trevor_V> o/
20:02:35 <xgerman> please vote on date and location with +1
20:02:41 <fnaval> o/
20:02:55 <dougwig> the MN location was offered by mestery, fyi.
20:03:05 <blogan> im here!
20:03:09 <mestery> Lake of the Woods is beautiful in the summer, trust me.
20:03:24 <blogan> ive never been to minnesota
20:03:31 <rm_work> are we only voting if we can actually go to it?
20:03:41 <xgerman> I have a tent and sleeping bags ;-)
20:03:45 <blogan> we won't know if we can go to it until a while
20:03:47 <rm_work> because if so, RAX is going to have to hold off on any votes until about 2 weeks before
20:03:47 <dougwig> rm_work: i'd think yes.
20:03:53 <Aish> o/
20:03:58 <dougwig> all of tax?
20:03:59 <dougwig> rax?
20:04:02 <dougwig> sheesh, then vote.
20:04:13 <xgerman> you guys can't hust rent a minvan? Roadtrip?
20:04:17 <rm_work> I mean, we still won't know if any of us can go until about 2-3 weeks
20:04:19 <dougwig> someone go spin up a rax server, so they have a travel budget.
20:04:25 <rm_work> T_T
20:04:38 <blogan> those proceeds go to the marketing budget
20:04:38 <johnsom> lol
20:05:04 <rm_work> I'm only going to Vancouver because I am handling all of my own lodging and no per diem :P
20:05:15 <rm_work> but at least I'll be there
20:05:21 <blogan> panhandling
20:05:28 <sballe> rm_work: It will great to have you there
20:05:35 <xgerman> do I need to bring a tent, rm_work?
20:05:42 <xgerman> sballe +1
20:05:43 <rm_work> thanks sballe :P might just have to beg for food periodically
20:06:00 <xgerman> I will make sure to forward party invites to you
20:06:05 <rm_work> hehe
20:06:05 <sballe> I thought the food was included int he conference fee
20:06:14 <rm_work> sballe: LUNCH is :)
20:06:19 <johnsom> rm_work Sorry to hear that.  We can probably help get you some food and beer at the events
20:06:20 <sballe> and yeah we now have to go to all the parties
20:06:32 <rm_work> Ah! Thanks sballe, gave me a great idea -- I'll just bring tupperware -- lunch for dinner! :P
20:06:39 <sballe> -)
20:06:54 <rm_work> anywho
20:07:02 <xgerman> next announcement:
20:07:04 <xgerman> LBaaS Vancouver Topics: https://etherpad.openstack.org/p/lbaas-vancouver-planning
20:07:04 <rm_work> we will look at that list and vote :)
20:07:16 <xgerman> #link https://etherpad.openstack.org/p/lbaas-vancouver-planning
20:07:28 <xgerman> I assume dougwing and blogan started that
20:07:41 <xgerman> dougwig
20:07:45 <blogan> xgerman: yes
20:07:48 <dougwig> i was told to expect to working group sessions for lbaas.  one will be for lbaas feature planning, the other for "octavia as the ref driver".  fyi.
20:07:52 <dougwig> /to/two/
20:08:02 <blogan> excellent
20:08:05 <xgerman> awesome
20:08:12 <sballe> +1
20:08:26 <sballe> Do we know when they will be?
20:08:38 <dougwig> mestery?
20:08:49 * mestery reads backscrool
20:08:50 <dougwig> i'm not sure if it's final yet.
20:09:01 <mestery> I'm still working on the schedule, so not yet
20:09:05 <mestery> Hopefully by next Monday, but we'll see.
20:09:10 <xgerman> ok, cool
20:09:20 <xgerman> xo we need to add Octavia topics to that pad?
20:09:37 <sballe> mestery: +1
20:09:40 <dougwig> yes, but start a new section at the bottom or something.
20:09:46 <xgerman> ok, will do
20:10:07 <xgerman> next
20:10:08 <xgerman> Neutron Vancouver topics: https://etherpad.openstack.org/p/liberty-neutron-summit-topics
20:10:20 <xgerman> #link https://etherpad.openstack.org/p/liberty-neutron-summit-topics
20:10:28 <blogan> my eyes!
20:11:03 <xgerman> and lastly I think an RC2 is being prepared... blogan, dougwig, mestery any details?
20:11:29 <dougwig> i think that's all the details.  if you need a fix in rc2, propose a cherry-pick to stable/kilo
20:11:39 <mestery> RC is in the works, will be cut tomorrow.
20:11:50 <xgerman> so be quick with proposing
20:11:50 <mestery> I spent an hour with ttx yesterday, we're just doign rechecks to get the final patches in
20:11:55 <blogan> getting something in would have to be very important right?
20:12:03 <mestery> It's actually too late to propose now.
20:12:17 <mestery> Unless it's a legitimate, "holy crap the relase will burn with this!" type of bug
20:12:46 <dougwig> given the overall flakiness of openstack, that is a very high bar indeed.
20:13:08 <mestery> lol
20:13:25 <xgerman> chair blogan
20:13:30 <xgerman> #chair blogan
20:13:31 <openstack> Current chairs: blogan xgerman
20:13:43 <xgerman> #topic Brief progress reports
20:14:19 <xgerman> johnsom?
20:14:38 <dougwig> neutron-lbaas driver - initial coding done, not tested *at all*, needs update for writable ids.  WIP is here:
20:14:39 <dougwig> #link https://review.openstack.org/#/c/174114/
20:14:40 <johnsom> I have been testing out controller worker on devstack (beating head against gerrit outages, back pypi packages, etc.) merging patchsets, etc.
20:14:41 <blogan> 3 big things are controller worker, network driver, and amphora driver
20:15:09 <johnsom> Right now I have it up building through load balancers and plugging the networks
20:15:20 <xgerman> that's something
20:15:32 <blogan> so building the amphorae? and plugging the networks on those?
20:15:49 <xgerman> sound slike it
20:15:56 <johnsom> I want to test through config generation, then might propose a merge and many bug fixes from there
20:15:58 <blogan> are you using the devstack private network as the management network?
20:16:04 <johnsom> blogan, yes
20:16:31 <blogan> okay cool
20:16:39 <johnsom> blogan, yes, private for management, public for vip (just for giggles)
20:16:50 <xgerman> demo-quality
20:17:25 <blogan> well i think the vip should be on an isolated network, and then just point a flip to that port to get public accessibility (but the vip network is controlleed by the user)
20:17:31 <johnsom> not yet demo quality, we have some bugs and I'm doing some manual DB stuff, but getting there.
20:17:36 <blogan> so that should be relatively easy
20:18:27 <Trevor_V> Just to provide a little update from myself.  I tested the ssh_driver thus far with ubuntu 14.04, looks good on most recent patchset.  Please comment if you find otherwise
20:18:46 <blogan> Trevor_V: looks good for all methods?
20:18:52 <Trevor_V> Yep.
20:18:59 <blogan> great
20:19:11 <ptoohill> I am working/testing on patch for Al's devstack plugin to build the mgmt network and start the api. Also, tempest stuffs
20:19:24 <xgerman> cool
20:19:44 <blogan> and network driver should be ready to go for a non vrrp setup, it needs some work to get multiple vrrp instances to work
20:19:55 <xgerman> I updated the Amphora API spec and the whole REST stuff is coded... more testing needed though
20:20:29 <johnsom> Yeah, I'm really targeting non-vrrp at the moment, but it's cool that it is seeing progress too
20:20:40 <blogan> for vrrp, which would be icing on the cake for a demo, the amphora driverw ould need an update
20:21:00 <xgerman> well, we need something for Sunday night ;-)
20:21:12 <blogan> this sunday night?
20:21:23 <xgerman> Sunday night before the talk
20:21:27 <blogan> i know lol
20:22:06 <johnsom> blogan dougwig It would be handy to get this merged: https://review.openstack.org/#/c/172558  It's -1 dougwig
20:22:13 <blogan> okay then we should get controller worker tested out, reviewed, and merged, which would also need tobe tested out with the network driver and amphora drivers
20:22:19 <johnsom> The neutron driver depends on it
20:22:48 <blogan> dougwig: would you be okay with merging that in with a later review that would fix for vrrp?
20:23:01 <dougwig> yes.
20:23:54 <johnsom> Problem solved...
20:23:56 <blogan> that would leave the controller worker and amphora driver then
20:24:14 <blogan> and then we'd need to merge the stevedore loading of the handler ptoohill did
20:24:30 <xgerman> i thought I +2'd that
20:24:42 <xgerman> so blogan merge ahead
20:24:43 <blogan> lol well then i'll +A it
20:24:45 <blogan> lol
20:24:47 <ptoohill> You did, but adam only +1
20:24:58 <blogan> i need to test it out first
20:25:01 <xgerman> yeah, he's cautious
20:25:40 <xgerman> #topic Sticker - Pick a design
20:25:41 <blogan> oh it may need to add soemthing
20:25:47 <blogan> D
20:25:54 <ptoohill> its a super simple change, and stevedore does what its supposed to. Theres no real logic there
20:25:56 <xgerman> #link https://s3.amazonaws.com/uploads.hipchat.com/8522%2F409388%2FzzvhUeiZJPki7iZ%2Foctavia_logo_revs3.pdf
20:26:09 <xgerman> #link https://docs.google.com/drawings/d/1PvJj_8A6oFRdUN9xAgm9Z8joM33Z7CuTJUzCQn3GoCM/edit
20:26:13 <rm_work> alright, can +2, but i prefer to try RUNNING code like that :P
20:26:20 <ptoohill> what does it need to add blogan
20:26:27 <Trevor_V> NOT "your load is our job"
20:26:29 <blogan> add the queue producer to the setup.cfg
20:26:37 <ptoohill> oh
20:27:10 <xgerman> so the s3 link is our corporate artist
20:27:21 <Trevor_V> I did notice you guys completely disregarded our sticker ideas from last week.
20:27:25 <Trevor_V> That's fine I guess.
20:27:32 <madhu_ak> "Balance is our Business" +1
20:27:34 <Trevor_V> Oh pfff
20:27:40 <Trevor_V> I didn't see the link to our stuff
20:27:50 <blogan> wtg Trevor_V
20:27:53 <xgerman> I linked it
20:28:05 <xgerman> https://docs.google.com/drawings/d/1PvJj_8A6oFRdUN9xAgm9Z8joM33Z7CuTJUzCQn3GoCM/edit
20:28:07 <Trevor_V> I thought someone at HP was going to mock up our sticker stuffs with some tweaks of your own, my b
20:28:18 <blogan> lol @ "Your load is our job"
20:28:48 <Trevor_V> Not happening... We aint hookers
20:29:10 <johnsom> Or sewage treatment workers...
20:29:43 <fnaval> Drop your load.
20:29:51 <Trevor_V> Stop fnaval... just stop
20:29:59 <ptoohill> lol
20:30:43 <madhu_ak> lol
20:30:45 <xgerman> Ok, I think we need to decide on a design and then pick a slogan
20:31:12 <blogan> is it imperative to have this done today?
20:31:15 <Trevor_V> I like the Rax design for "D"
20:31:31 <xgerman> #vote Design: A, B, C, RAX-A, RAX-B, RAX-C, D, E, F, G
20:32:03 <madhu_ak> D
20:32:12 <Trevor_V> xgerman, we want rax-D in there
20:32:14 <ptoohill> #vote A
20:32:18 <xgerman> #startvote Design: A, B, C, RAX-A, RAX-B, RAX-C, D, E, F, G
20:32:19 <openstack> Unable to parse vote topic and options.
20:32:19 <Trevor_V> Or is that just "d"?
20:32:22 <johnsom> #startvote
20:32:23 <openstack> Only the meeting chair may start a vote.
20:32:25 <xgerman> yep, just D
20:32:28 <Trevor_V> got it
20:32:36 <ptoohill> #vote A
20:32:36 <madhu_ak> D
20:32:39 <ptoohill> A
20:32:51 <dougwig> #startvote Which design? A, B, C, RAX-A, RAX-B, RAX-C, D, E, F, G
20:32:51 <openstack> Only the meeting chair may start a vote.
20:32:59 <ptoohill> :P
20:33:08 <xgerman> #startvote Which design? A, B, C, RAX-A, RAX-B, RAX-C, D, E, F, G
20:33:09 <openstack> Begin voting on: Which design? Valid vote options are A, B, C, RAX-A, RAX-B, RAX-C, D, E, F, G.
20:33:11 <openstack> Vote using '#vote OPTION'. Only your last vote counts.
20:33:12 <johnsom> Yeah, what he said
20:33:14 <Trevor_V> #vote d
20:33:19 <ptoohill> #vote A
20:33:22 <mwang2_> #vote D
20:33:23 <blogan> #vote D
20:33:24 <madhu_ak> #vote D
20:33:24 <dougwig> #vote d
20:33:27 <ijw> #vote D
20:33:34 <johnsom> #vote a
20:33:36 <fnaval> #vote D
20:33:49 <dougwig> i actually prefer A, except for the two-tone.
20:33:51 <xgerman> #vote B
20:33:55 <ajmiller> #vote A
20:34:07 <bharath> #vote D
20:34:14 <ptoohill> which is D?
20:34:14 <jwarendt> #vote B
20:34:19 <Trevor_V> Its on our link ptoohill
20:34:22 <xgerman> D is on the RAX sheet
20:34:23 <ijw> ptoohill: the nice one
20:34:23 <ptoohill> oh
20:34:28 <Aish> #vote D
20:34:28 <Trevor_V> pff ijw
20:34:50 * ijw is only lurking in the background, really, but I'm trying to save you from yourselves ;)
20:35:05 <xgerman> thanks we need that
20:35:08 <xgerman> ijw
20:35:17 <xgerman> #endvote
20:35:18 <openstack> Voted on "Which design?" Results are
20:35:19 <openstack> A (3): johnsom, ajmiller, ptoohill
20:35:20 <openstack> B (2): xgerman, jwarendt
20:35:20 <dougwig> ijw: lol
20:35:21 <openstack> D (9): Aish, fnaval, ijw, dougwig, Trevor_V, madhu_ak, bharath, mwang2_, blogan
20:35:31 <xgerman> ok, D it is
20:35:33 <blogan> Octavia gets the D
20:35:38 <Trevor_V> Woot
20:35:45 <Trevor_V> Now time to drop a load
20:35:47 <rm_work> aww missed it
20:35:48 <rm_work> ah well
20:35:50 <rm_work> lol
20:35:51 <Trevor_V> (damnit fnaval)
20:35:52 <ijw> Trevor_V: tmi
20:36:05 <xgerman> so do we need a different tagline then "Openstack Loadbalancing"
20:36:05 <fnaval> ?? Trevor_V
20:36:05 <Trevor_V> it matched too well ijw
20:36:07 <ptoohill> nii
20:36:17 <Trevor_V> I honestly don't think so xgerman
20:36:35 <blogan> xgerman: i'm worried that if we tag it as openstack that might be a problem, then again it probably wont witht eh big tent approach now
20:36:40 <Trevor_V> Up to other suggestions for sure
20:37:54 <xgerman> ok, blogan has a valid point
20:38:27 <blogan> its a low probability chance
20:38:45 <blogan> i just want to use "Your load, our job" or whatever that was
20:38:52 <Trevor_V> Stop it blogan
20:38:54 <xgerman> lol
20:40:08 <dougwig> moving on?
20:40:08 <blogan> yeah i like openstack load balancing
20:40:14 <xgerman> #topic Open Discussion
20:40:15 <fnaval> Balancing at work.
20:40:41 <ajmiller> I've been doing bunches of devstack/vagrant stuff for both neutron-lbaas and octavia.  Would appreciate reviews on:
20:40:50 <ajmiller> #link https://review.openstack.org/173438
20:41:02 <ajmiller> #link https://review.openstack.org/175174
20:41:15 <blogan> possible topic: octavia under mestery's dictatorship (big tent discussion)?
20:41:17 <ajmiller> #link https://review.openstack.org/167796
20:41:24 <ptoohill> ajmiller i can adjust my vote now ;)
20:41:29 <blogan> ajmiller: is a pinata of reviews
20:41:39 <ajmiller> The first two are for neutron-lbaas and I think are ready for +2/A
20:41:42 <mestery> blogan: ++, I think that makes sense, see russellb's email about the other networking-foo projects today and my response
20:41:53 <ajmiller> The third is octavia.
20:42:20 <blogan> mestery: i do too, even with you being a ruthless dictator, but i'd like to guage everyone else's ideas
20:42:27 <dougwig> blogan: let's ship some code first.
20:42:34 <mestery> lol
20:42:40 * mestery makes a note to have blogan walk the plank
20:42:40 <xgerman> nah, let's be like congress
20:42:53 * blogan sinks like a witch
20:43:06 <russellb> i think projects should be considered "one of us" without having to be in production already, we need to change that outlook
20:43:18 <russellb> when a project is under development in our community, it's already a part of hte openstack community, IMO
20:43:32 <ptoohill> i like that thinking
20:43:32 <xgerman> nicely put russelb
20:43:50 <russellb> that's how the TC has been looking at projects as of late
20:43:51 <dougwig> hmm, i'd like to see a team gel and prove itself before we rush to inclusiveness.
20:43:59 <xgerman> ok, so move under mestery;s iron fist immediately
20:44:01 <blogan> russellb: so everything is openstack that is stackforge then?
20:44:09 <dougwig> but i've been wrong plenty of times before.
20:44:12 <russellb> blogan: most of it, i expect
20:44:24 <mestery> dougwig: party pooper
20:44:24 <mestery> ;)
20:44:27 <russellb> dougwig: yes agree that there should be some level of "yes, this appears to be a real thing"
20:44:56 <russellb> people seem to care, people are participating
20:44:58 <blogan> russellb: any discussion on how that will be done?
20:45:01 <russellb> and they are doing it in the openstack way
20:45:05 <russellb> blogan: lots of it :)
20:45:20 <russellb> http://governance.openstack.org/reference/new-projects-requirements.html
20:45:22 <blogan> russellb: lol i mean any solid solutions?
20:45:29 <dougwig> given this, do folks want to move earlier?  is mestery in favor of that?
20:45:31 <russellb> solid solutions to what
20:45:31 <blogan> ah thanks
20:45:37 <xgerman> dougwig +1
20:45:52 <mestery> +2
20:46:00 <blogan> russellb: to what quality the project has to be in before it can be considered openstack (not just code quality)
20:46:06 <blogan> i think that link will answer my questiosn though
20:46:13 <russellb> yeah, it should answer it
20:46:14 <russellb> hopefully
20:46:28 <russellb> and then we're defining higher optional bars for project to reach with "tags"
20:46:34 <blogan> russellb: will the ptl have to lead every meeting and manage the releases for every project under him?
20:46:38 <blogan> him/her
20:46:55 <russellb> like, a project *may* be given a tag indicating that its participation has a diverse affiliation -- http://governance.openstack.org/reference/tags/team_diverse-affiliation.html
20:47:01 <dougwig> blogan: they already don't.
20:47:06 <xgerman> +1
20:47:13 <xgerman> we can have a delegate
20:47:31 <xgerman> aka mestery can send an envoy
20:47:40 <blogan> okay, as long as mestery's soul isn't sucked out of him
20:47:45 <mestery> lol
20:48:08 <blogan> russellb: one more question then, what role would the ptl have in these projects then?
20:48:14 <dougwig> mestery absorbs open-source souls into his body, via an expanding empire, i think.
20:48:24 <russellb> that's project specific
20:48:27 <xgerman> so mestery is the vampire
20:48:31 <russellb> PTL can be involved or delegate as much as they'd like
20:48:31 <mestery> That's an image I can't scrub from my eyes now dougwig, thanks.
20:48:54 * russellb hand waves
20:49:07 <mestery> lol
20:49:08 * blogan noticed
20:49:10 <blogan> lol
20:49:54 <blogan> sounds like an agile process
20:50:02 <xgerman> ok, so do we need to vote when to move?
20:50:24 <blogan> i'd say give people some time to think about it
20:50:41 <xgerman> yeah + the earliest after the demo to minimize distractions
20:50:52 <dougwig> let's let russellb's ML thread play out, too. he can pave the way.  :)
20:51:06 * russellb just trying to help
20:51:18 <blogan> russellb: thanks for answering questions btw
20:51:22 <xgerman> +1
20:51:23 <russellb> dougwig: really nice comments in gerrit pointing out questions that need to be answered
20:52:04 <blogan> russellb: there's a spec for this?
20:52:09 <dougwig> russellb: thanks for taking the initiative on this.  i like it.
20:52:20 <xgerman> link to the spec please?
20:52:30 <xgerman> dougwig +1
20:53:05 <russellb> no spec
20:53:09 <russellb> i started a ML thread that's related
20:53:22 <russellb> http://lists.openstack.org/pipermail/openstack-dev/2015-April/062310.html
20:53:37 <dougwig> blogan: there was an early gerrit review, since abandoned: https://review.openstack.org/#/c/175952/3
20:53:45 <dougwig> the ML is where it's at now.
20:53:52 <russellb> yeah, my gerrit reviews were a little ahead of the bigger question
20:53:58 <russellb> so i dropped them for now
20:54:16 <blogan> dougwig, russellb: alright thanks!
20:54:46 <xgerman> ok, 6 minutes left - anything else?
20:55:15 <mwang2_> code review needed for tempest
20:55:20 <mwang2_> #link https://review.openstack.org/#/c/173542/
20:55:40 <ptoohill> would like to clarify the lb_network_name that is currently in the config
20:55:50 <madhu_ak> same here as well. #link: https://review.openstack.org/#/c/171832/
20:56:10 <madhu_ak> thanks blogan
20:56:15 <ptoohill> This question kinda ties in with Al's debstack plugin review and amp_network_list. But suppose we can talk about this later
20:56:17 <xgerman> ptoohill I put that in because for network)id you need some extension HP didn't have
20:56:18 <johnsom> ptoohill I think that is going away as I added a controller config for that
20:56:26 <johnsom> it was just bad timing
20:56:41 <ptoohill> so that option should live in the controller_worker group?
20:56:43 <ptoohill> as a list?
20:56:46 <russellb> blogan: you're quite welcome, thanks for letting me drop in :)
20:57:17 <johnsom> ptoohill yep
20:57:24 <johnsom> [controller_worker]
20:57:24 <johnsom> amp_network_list = 62a63520-7bf1-47e0-b90d-5918d4dc45d4
20:57:47 <ptoohill> yea, i seen that and was just curious. Then we should remove the option under networking group
20:57:48 <blogan> johnsom: is that the management network?
20:57:55 <ptoohill> this works for me, i was just confused
20:57:58 <johnsom> Yes
20:58:10 <blogan> okay, amp_network is probably a better name
20:58:27 <ptoohill> and i believe itll need to be updated in the compute worker also
20:58:45 <blogan> russellb: like dougwig said, thanks for leading the effort
20:59:26 <ptoohill> being thats a list im sure we need to define logic on which one to use from the list?
20:59:33 <blogan> all
20:59:34 <xgerman> all?
20:59:38 <blogan> i would assume all
20:59:52 <xgerman> or select at random
20:59:53 <ptoohill> thats works also
20:59:56 <johnsom> I welcome comments on the ocde
21:00:10 <dougwig> we
21:00:12 <blogan> right now all is a good bet, random would cause issues later
21:00:13 <dougwig> we're past time
21:00:23 <blogan> stop ruining meeting time!
21:00:23 <xgerman> #endmeeting