21:01:25 <mestery> #startmeeting networking
21:01:26 <openstack> Meeting started Mon Apr 14 21:01:25 2014 UTC and is due to finish in 60 minutes.  The chair is mestery. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:01:27 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:01:30 <openstack> The meeting name has been set to 'networking'
21:01:44 <mestery> #link https://wiki.openstack.org/wiki/Network/Meetings Agenda
21:02:00 <mestery> #topic Announcements
21:02:06 <Swami> hi all
21:02:23 <mestery> First of all, I'd like to say a huge thank you to markmcclain for being PTL the past two cycles!
21:02:34 <salv-orlando> thanks markmcclain and congratulations mestery
21:02:38 <mestery> I have some incredibly large shoes to fill, I hope I am up to the task!
21:02:46 <salv-orlando> I think he's 10 1/2
21:03:05 <mestery> salv-orlando: In that case, my feet are about to hurt, as I'm a 14. :)
21:03:09 <emagana> I second salv-orlando
21:03:12 <markmcclain> 12 :p (or 46 if you prefer euro sizing) :p
21:03:26 <mestery> That's a bit closer markmcclain, but still tight. ;)
21:03:40 <salv-orlando> I guessed it… 12 US size is pretty much 10 1/2 UK size ;) ok, back to things that matter
21:04:07 <mestery> So, a note on Icehouse: If there are critical icehouse issues, please let me and markmcclain know. He is the point of contact of Icehouse issues.
21:04:09 <armax> have I joined the right meeting? :)
21:04:17 <mestery> markmcclain: I don't think there are any at the moment, correct?
21:04:33 <markmcclain> to my knowledge nothing is blocking release
21:04:42 <mestery> markmcclain: thanks!
21:05:04 <mestery> #topic Bugs
21:05:28 <mestery> jilbosova: You have put this one in the agenda: https://bugs.launchpad.net/neutron/+bug/1280464
21:05:30 <uvirtbot> Launchpad bug 1280464 in neutron "grenade-dsvm-neutron Failure in upgrade-swift (dup-of: 1235149)" [Critical,Confirmed]
21:05:31 <uvirtbot> Launchpad bug 1235149 in grenade "Add neutron upgrade process to grenade project" [Undecided,Fix committed]
21:05:45 <mestery> jilbosova: Any updates on this one?
21:06:11 <mestery> jlibosva: ^^^
21:06:37 <jlibosva> mestery: salv-orlando started an email thread about dealing with service plugins
21:06:57 <mestery> salv-orlando jlibosva: This is related to this bug, or there is overlap? https://bugs.launchpad.net/neutron/+bug/1307344
21:06:59 <uvirtbot> Launchpad bug 1307344 in neutron "neutron migrations are not idempotent or invariant to enabled services" [Critical,Confirmed]
21:07:24 <salv-orlando> yes. However I'd want to anticipate that this is not being solved today, tomorrow or next monday
21:07:43 <salv-orlando> I still think we should strive to get grenade running - even if we know we have this problem.
21:07:49 <mestery> salv-orlando: Yes, understood, I've read your comments and this is a large issue for sure.
21:08:06 <salv-orlando> So I would propose to put grenade in a situation where previous and current revision run the same configuration
21:08:11 <jlibosva> mestery: salv-orlando ugly workaround could be generating specific migration in grenade
21:08:18 <mestery> #link http://lists.openstack.org/pipermail/openstack-dev/2014-April/032608.html Neutron Migrations Email Thread from salv-orlando
21:08:18 <jlibosva> just to have working grenade
21:08:34 <jlibosva> but similar issues may pop up later
21:08:41 <mestery> That sounds pretty ugly jlibosva. :)
21:08:42 <salv-orlando> jlibosva: I was just thinking to run the 'new' devstack in a way that the same plugins as the old one are executed
21:08:55 <salv-orlando> and frankly, this might sort by itself once we switch to juno
21:09:02 <jlibosva> salv-orlando: won't it break tempest tests for metering?
21:09:02 <salv-orlando> as new would be juno and old icehouse
21:09:52 <salv-orlando> jlibosva: I think it should be possible for grenade to give devstack a different localrc than the one executed in standard jobs. However, this is the kind of thing we can discuss tomorrow morning ;)
21:10:06 <jlibosva> salv-orlando: ack
21:10:25 <mestery> Thanks for looking into this salv-orlando and jlibosva.
21:10:51 <mestery> arosen has a list of nova bugs in the agenda which neutron cores should also be reviewing.
21:11:00 <mestery> arosen: Anything specific you wanted to add around these?
21:11:12 <arosen> mestery:  nope not really
21:11:35 <mestery> arosen: OK, thanks. So, please take some time to review these now.
21:11:36 <arosen> mestery:  these bugfixes just remove some of the unneeded calls from nova into neutron
21:11:47 <arosen> mestery:  thanks!
21:12:04 <mestery> arosen: Yes, I looked at them briefly this morning, hopefully these can be merged, as it appeared some nova folks had issues merging them at the end of icehouse.
21:12:25 <mestery> Are there other bugs anyone wants to bring up at this time?
21:13:25 <salv-orlando> Nope, just a friendly reminder that nova core recommends nova/neutron patches to be reviewed by neutron cores as well
21:13:49 <mestery> salv-orlando: Thanks for the reminder here. :)
21:14:14 <mestery> #topic Docs
21:14:22 <mestery> emgagana: Hi!
21:14:25 <emagana> mestery: here!
21:15:01 <emagana> Wiki has been updated with the commits from everybody!!! Huge thanks to all good citizens from the Docs World!
21:15:25 <mestery> Yes, thank you to everyone who is submitting docs patches!
21:15:33 <emagana> There are one or two bugs without owner:
21:15:42 <emagana> https://bugs.launchpad.net/openstack-manuals/+bug/1296997
21:15:43 <uvirtbot> Launchpad bug 1296997 in openstack-manuals "Add script to migrate ovs or lb db to ml2 db" [High,Confirmed]
21:15:53 <emagana> https://bugs.launchpad.net/openstack-manuals/+bug/1026745
21:15:55 <uvirtbot> Launchpad bug 1026745 in openstack-manuals "Quantum docs should describe XenAPI setup" [Medium,Confirmed]
21:16:09 <emagana> and https://bugs.launchpad.net/openstack-manuals/+bug/1283930
21:16:10 <uvirtbot> Launchpad bug 1283930 in openstack-manuals "Make metaplugin be used with a router service plugin" [Medium,Confirmed]
21:16:28 <emagana> volunteers???
21:16:42 <emagana> specially for 1296997
21:16:43 <markmcclain> The metaplugin is untested
21:16:54 <markmcclain> and technically deprecated in Icehouse
21:16:59 <mestery> markmcclain: That's a good point.
21:17:18 <emagana> should be removed all docs about Metaplugin?
21:17:42 <salv-orlando> should be metaplugin removed (cheeky comment)
21:17:52 <nati_ueno> why?
21:17:53 <salv-orlando> on a less cheeky note, I think now ML2 supersedes it
21:17:59 <mestery> Those are interesting questions at this stage.
21:18:00 <nati_ueno> Metaplugin has gating now
21:18:02 <nati_ueno> and tested
21:18:25 <salv-orlando> well if it implements any use case not supported my ml2 it's ok to keep
21:18:34 <salv-orlando> also we won't have a migration straegy, so I was joking
21:18:39 <mestery> nati_ueno: OK, cool, I didn't know that at this point.
21:18:40 <markmcclain> nati_ueno: link?
21:18:42 <salv-orlando> nati_ueno: pleasy don't be offended
21:18:49 <sweston> mestery: I'll volunteer for 1296997
21:18:53 <nati_ueno> let me grab link
21:19:04 <mestery> emagana: You've got a volunteer in sweston. :)
21:19:04 <emagana> Thanks sweston
21:19:18 <sweston> always :-)
21:19:25 <emagana> yeap, I will chase sweston to the end of the days!
21:19:44 <sweston> hehe, it's okay, I'm easy to catch
21:20:01 <mestery> While we wait for the metaplugin link from nati_ueno, any other docs updates emagana?
21:20:16 <emagana> nope!
21:20:35 <nati_ueno> https://review.openstack.org/#/dashboard/10121
21:20:53 <nati_ueno> Metapluing can support plugins which don't support ml2
21:21:01 <annegentle> emagana: how about https://bugs.launchpad.net/openstack-manuals/+bug/1226279? if Diane can have samples she can document
21:21:02 <uvirtbot> Launchpad bug 1226279 in openstack-manuals "document multiprovidernet extension" [Medium,Confirmed]
21:21:14 <nati_ueno> I'm thinking about how to migrate it for the ML2, but It is under progress.
21:21:28 <emagana> annegentle: Yes, I will ask directly arosen to help Diane
21:22:00 <mestery> nati_ueno: OK, but the bottom line is you're still supporting metaplugin at the moment?
21:22:06 <nati_ueno> mestery: ys
21:22:07 <emagana> annegentle: I will follow up by ML with Diane and arosen
21:22:08 <nati_ueno> mestery: yes
21:22:19 <mestery> nati_ueno: OK, thanks!
21:22:33 <mestery> OK, moving along to the next item on the agenda.
21:22:39 <marun> sweston: let me know if you have any questions that aren't answered by the script's docstring.
21:22:39 <mestery> #topic Juno Design Summit
21:22:59 <mestery> The deadline to submit Summit Sessions is April 20, which is coming up quickly.
21:23:16 <nati_ueno> let me take 1283930
21:23:17 <sweston> marun: thanks, I will take you up on that
21:23:45 <mestery> Remember when filing ideas to also think about community work and things which benefit the project as a whole (e.g. common core).
21:23:49 <emagana> nati_ueno: Thanks! Please, assign the bug to yourself
21:23:55 <nati_ueno> emagana: done :)
21:24:30 <mestery> I'm hoping to start reviewing Summit Session Ideas this week yet to get a head start on planning the sessions.
21:24:46 <mestery> core devs: I'll be reaching out to you to help as I go through the sessions.
21:25:27 <mlavalle> mestery: would you share the url where we can enter ideas for session?
21:25:43 <mestery> #link http://summit.openstack.org/ Summit Session Proposal Page
21:25:47 <mestery> mlavalle: ^^^
21:25:57 <mlavalle> mestery: thanks
21:26:08 <mestery> Any questions on the Juno Design Summit?
21:27:29 <mestery> #topic Neutron Tempest
21:27:34 <mestery> mlavalle: hi!
21:27:43 <mestery> mlavalle: Any updates on Tempest this week?
21:27:52 <mlavalle> mestery: we have continued making progress merging api tests
21:28:09 <mlavalle> we are tracking 28 patchsets. As of today we have merged 20
21:28:09 <mestery> mlavalle: Great!
21:28:31 <mestery> Thanks to mlavalle and everyone contributing Tempest patches!
21:28:32 <mlavalle> we have only 2 abandoned, which we will recover soon
21:28:35 <HenryG> mlavalle: where is the list?
21:29:01 <mlavalle> HenryG: the list of what has been merged is in the agenda
21:29:25 <HenryG> mlavalle: the remaining 8?
21:29:25 <mestery> HenryG: https://wiki.openstack.org/wiki/Network/Meetings#Neutron_Tempest_.28mlavalle.29
21:29:30 <mlavalle> and I can send a list of what remains to be merged to the ML
21:30:01 <HenryG> mlavalle: thanks, please do
21:30:05 <salv-orlando> mestery: my update is that as soon as Icehouse is release I will push an openstack-infra/config patch to make the full job voting. This also means you will be pinged a lotfrom infra team regarding gate failures!
21:30:06 <mlavalle> I am also planning with yfried to start a similar effort for scenario tests
21:30:16 <mlavalle> during Juno
21:30:32 <mestery> salv-orlando: Thanks ... I think. ;)
21:30:46 <mestery> mlavalle: Great!
21:31:08 <salv-orlando> mestery: a higher failure rate is unavoidable, but the early stages of the release cycle is probably the best time to deal with it.
21:31:18 <mestery> So, along with these tests, I know marun has been working on boot strapping our functional tests. All of these areas will help with Neutron stability and testability.
21:31:30 <Sukhdev> mlavalle: do you have any list for scenario tests?
21:31:33 <mestery> salv-orlando: Yes, agreed.
21:32:00 <mlavalle> Suhkdev: Not yet. We will send a message to the ML asking for proposals
21:32:23 <Sukhdev> mlavalle: cool…thanks
21:32:31 <mestery> mlavalle: So you're looking at new scenario tests, is that right?
21:32:42 <mlavalle> mestery: correct.
21:32:51 <mestery> mlavalle: That may be a good Design Summit Session to propose.
21:32:55 <mlavalle> we would like to discuss with the core team during the summit
21:33:12 <mlavalle> mestery: we were thinking the same :-)
21:33:30 <mlavalle> how can I get that session in the agenda?
21:33:31 <mestery> mlavalle: Perfect!
21:33:40 <marun> mlavalle: I hope to be involved in discussion of new scenario additions to ensure that we need to work at that level.
21:33:56 <Sukhdev> mlavalle: Be sure include some multi-node scenario tests as well
21:33:57 <mlavalle> marun: more than welcome
21:34:21 <mestery> mlavalle: Please propose your sessions here http://summit.openstack.org/ and we'll make sure it gets a slot.
21:34:40 <mlavalle> mestery: will do. that's all I have
21:34:51 <mestery> marun: Are you going to propose any functional testing sessions as well? I think those would be good to talk about as well.
21:35:00 <mestery> thanks mlavalle!
21:35:30 <marun> mestery: Yes, I will be proposing one or more sessions on testing, including functional testing.
21:35:43 <mestery> marun: OK, thanks!
21:35:49 <mestery> #topic L3
21:35:54 <mestery> carl_baldwin: hi!
21:35:59 <carl_baldwin> mestery: hi
21:36:07 <mestery> carl_baldwin: Looks like you have some updates for us?
21:36:33 <carl_baldwin> Continued progress in a few areas.  DVR is getting WIP in to gerrit.  L2 and L3 agent changes soon to come.
21:36:55 <carl_baldwin> Also have made progress with root wrap performance issue.  Also in review in gerrit.
21:36:56 <mestery> carl_baldwin: Fantastic! Those are critical to us closing the nova-network gap, so this is great to see!
21:37:40 <carl_baldwin> This week we began investigation in to multiple subnets on external network and discussion.  Will continue in this week's meeting.
21:37:52 <salv-orlando> carl_baldwin: is the plan still to address E-W traffic first, and then N-S?
21:38:04 <carl_baldwin> salv-orlando: both are being addressed.
21:38:08 <salv-orlando> ok cool
21:38:25 <carl_baldwin> Also continue dynamic routing use case discussion for possible summit topic.
21:38:41 <Sukhdev> carl_baldwin, mestery: any thought on ML3? - Perhaps a design session to open discussion?
21:39:02 <Swami> salv_orlando: We are planning to address both East-West and North-South
21:39:06 <mestery> Sukhdev: I owe carl_baldwin an agenda item for the L3 meeting to discuss this, yes.
21:39:08 <markmcclain> carl_baldwin: +1 to bgp speaker discussion
21:39:14 <carl_baldwin> Sukhdev: I'd love to know more about ML3.
21:39:18 <salv-orlando> Sukhdev: Isn't this what you're looking for: http://summit.openstack.org/cfp/details/81
21:39:51 <mestery> salv-orlando: Yes, that session from pcm_ is along those lines.
21:40:04 <marun> off topic, but I'd like to see us avoid ML2 for other functions
21:40:19 <marun> It should simply be a framework within neutron, as ml2 will eventually become
21:40:26 <salv-orlando> the outcome from icehouse summit re bgp was that no action was required for neutron ;)
21:40:29 <Sukhdev> salv-orlando: yes - something like this…thanks for the link - was not aware of this
21:40:34 <salv-orlando> I remember I led that session discussion
21:40:49 <mestery> salv-orlando: :)
21:41:08 <emagana> I believe we should move Neutron in a entire Services Framework  but it is just me and my crazy world!
21:41:16 <carl_baldwin> salv-orlando: Could you do read through of the use cases on the wiki?  Linked from the l3 agenda page.
21:41:25 <salv-orlando> but I think it's ok to discuss it again… after all we've been discussing what emagana says for the past two years
21:41:26 <marun> emagana: baby steps
21:41:55 <enikanorov_> true
21:41:55 <mestery> salv-orlando: Has it been two years? It feels older.
21:42:15 <salv-orlando> just the advanced services insertion steering chaining and whatever
21:42:16 <marun> can of worms?  apologies
21:42:17 <carl_baldwin> ML3 / L3 vendor plugins is something that we can add to the agenda.
21:42:24 <emagana> we are all getting old!
21:42:52 <mestery> carl_baldwin: Thanks! If you put that on your agenda this week, we should send email to the ML alerting folks that we'll discuss it there.
21:43:01 <carl_baldwin> I will do that.
21:43:13 <carl_baldwin> Thanks mestery that is all I have to report.
21:43:19 <mestery> thanks carl_baldwin!
21:43:20 <Sukhdev> carl_baldwin: please do….thanks
21:43:32 <mestery> #topic IPv6
21:43:37 <mestery> sc68cal: Any updates this week?
21:44:27 <markmcclain> mestery: sc68cal is offline, but I met with him today
21:44:33 <yamahata> uuuuuuuuuuuu
21:44:46 <yamahata> sorry
21:44:54 <mestery> markmcclain: OK, thanks! I wasn't sure if the IPv6 section was updated or not.
21:44:55 <enikanorov_> yamahata: cleaning keyboard? : )
21:45:12 <mestery> yamahata: Or perhaps a cat run across your keyboard!
21:45:19 <markmcclain> now that Juno is open resume work on host configuration backends
21:45:41 <mestery> OK, thanks for the update markmcclain!
21:45:55 * salv-orlando thinks that since it's full moon yamahata might as well be a werewolf
21:46:26 <Sukhdev> salv-orlando: :-)
21:47:13 <mestery> #topic Group Based Policy
21:47:17 <mestery> SumitNaiksatam: hi
21:47:22 <SumitNaiksatam> mestery: Hi
21:47:37 <SumitNaiksatam> can give a quick update
21:47:56 <mestery> SumitNaiksatam: OK, thanks!
21:48:04 <SumitNaiksatam> few of us here who are working on fleshing out the model and working on a PoC
21:48:12 <SumitNaiksatam> model is mostly fleshed out
21:48:32 <SumitNaiksatam> currently little more work needs to be done with respect to services’ insertion
21:48:32 <Sukhdev> SumitNaiksatam: are you in a position to share the model?
21:49:19 <SumitNaiksatam> Sukhdev: already shared - #link https://docs.google.com/a/noironetworks.com/document/d/1ZbOFxAoibZbJmDWx1oOrOsDcov6Cuom5aaBIrupCD9E/edit?pli=1#heading=h.x1h06xqhlo1n
21:49:31 <mestery> OK, thanks for the update SumitNaiksatam!
21:49:35 <SumitNaiksatam> the latest WIP is here #link https://docs.google.com/a/noironetworks.com/presentation/d/1Nn1HjghAvk2RTPwvltSrnCUJkidWKWY2ckU7OYAVNpo/edit#slide=id.g1c910cf8b_00
21:49:38 <Sukhdev> SumitNaiksatam: Thanks
21:49:44 <SumitNaiksatam> we have not merged it into the main document yet
21:49:49 <marun> any plans to transition those docs to the new gerrit repo?
21:50:01 <SumitNaiksatam> we are also working in paralled on the policy model and policy enforcement code
21:50:29 <marun> How are you planning on submitting this code to the tree?
21:50:29 <SumitNaiksatam> mestery: thats it from us, happy to answer questions
21:50:32 <mestery> marun: The new git repo for doc reviews is on my list of things to flesh out this week.
21:50:39 <mestery> #topic Open Discussion
21:50:47 <marun> ^
21:50:52 <SumitNaiksatam> marun: yes, pretty soon
21:51:34 <SumitNaiksatam> marun: it will be the regular gerrit review process
21:51:46 <marun> SumitNaiksatam: I get that
21:52:06 <marun> SumitNaiksatam: I'm wondering how the POC that has been worked on in github is going to be broken into reviewable chunks
21:52:20 <marun> SumitNaiksatam: Or is the intent that it be small enough to review as a single patch?
21:52:36 <SumitNaiksatam> marun: currently, its not that large
21:52:47 <marun> sorry, maybe should ask offline.  I'm sure there are better questions
21:52:47 <SumitNaiksatam> marun: but its very much WIP
21:53:00 <SumitNaiksatam> marun: so havent yet figured out how or when to break it up
21:53:01 <mestery> marun: No worries, this is Open Discussion after all. :)
21:53:09 <SumitNaiksatam> marun: earlier we were thinking two patches probably
21:53:17 <marun> SumitNaiksatam: Would like to see it WIP sooner than later, even if not final.
21:53:29 <marun> SumitNaiksatam: makes it more accessible for everyone to see
21:53:30 <SumitNaiksatam> marun: absolutely, that was the plan
21:53:37 <marun> SumitNaiksatam: +1
21:53:45 <SumitNaiksatam> marun: it has taken longer because the discussions on the model are still ongoing
21:54:26 <marun> SumitNaiksatam: I guess I'm not sure of the idea of trying to get to some 'good' state before review
21:54:53 <marun> SumitNaiksatam: Because then fewer artifacts are in openstack-maintained infrastructure
21:55:12 <marun> SumitNaiksatam: Not necessarily a concern just for group policy
21:55:24 <SumitNaiksatam> marun: agreed
21:55:31 <marun> SumitNaiksatam: rather I'd like to make sure we are working towards better process for introducing new capabilities
21:55:59 <mestery> marun: I agree with that completely. I think the revamping of the BP process to use gerrit is moving us along those lines to be honest.
21:55:59 <SumitNaiksatam> marun: probably a question for the PTL? ;-)
21:56:20 <markmcclain> I thought the plan was to follow Nova's spec repo
21:56:34 <marun> that's my hope too
21:56:38 <mestery> markmcclain: Yes, it is.
21:57:00 <mestery> markmcclain: I think we just need to make people aware of it and post some instructions for neutron folks to use.
21:57:31 <marun> …and then we can push back on patches whose bp's arent' maintained in the spec repo
21:57:32 <carl_baldwin> mestery: +1 for a few instructions like how the repository will be laid out.
21:57:44 <markmcclain> mestery: https://wiki.openstack.org/wiki/Blueprints
21:58:17 <mestery> markmcclain: Thanks for sharing. I'll update this for Neutron as well.
21:58:32 <mestery> And I'll send an email to the ML letting people know we've moved to this officially.
21:58:49 <carl_baldwin> markmcclain: That link looks like a good start.
21:59:15 <mestery> OK, we're almost out of time here for today.
21:59:31 <mestery> I'll stay on top of htis new BP process and send out instructions to the ML soon.
21:59:46 <mestery> Please keep filing session ideas for the Juno Summit!
21:59:53 <mestery> Thanks everyone!
21:59:55 <mestery> #endmeeting