#openstack-meeting log

21:04:24 <markmcclain> #startmeeting Networking
21:04:25 <openstack> Meeting started Mon Aug 26 21:04:24 2013 UTC and is due to finish in 60 minutes.  The chair is markmcclain. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:04:26 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:04:28 <openstack> The meeting name has been set to 'networking'
21:04:51 <markmcclain> #link https://wiki.openstack.org/wiki/Network/Meetings
21:05:16 <markmcclain> #topic Announcements
21:05:43 <markmcclain> The proposal feature freeze is in effect
21:06:06 <markmcclain> Looks like nearly everything made it, but a few things were granted exceptions
21:06:24 <markmcclain> when they're posted I will be adding a note so that the reviews don't get penalized for being late
21:06:43 <markmcclain> We're going to have a busy week reviewing code
21:07:00 <markmcclain> as the H3 deadline is the end of day Sept 3rd
21:07:49 <markmcclain> new feature code submitted now will be held out until Icehouse opens around Sept 26th
21:08:52 <markmcclain> At this point we should turn our focus to polishing the code to be release ready and ensuring we have a good set of docs
21:09:20 <markmcclain> the team will be on the lookout for features that are masquerading as bugs
21:09:39 <markmcclain> Any questions about the feature freeze?
21:10:25 <markmcclain> #topic Bugs
21:10:39 <markmcclain> https://bugs.launchpad.net/neutron/+bugs?search=Search&field.importance=Critical&field.status=New&field.status=Confirmed&field.status=Triaged&field.status=In+Progress
21:10:45 <markmcclain> we
21:10:58 <markmcclain> we've got two critical bugs open
21:10:59 <markmcclain> https://bugs.launchpad.net/neutron/+bug/1216866
21:11:01 <uvirtbot> Launchpad bug 1216866 in neutron "Race condition in Neutron initialization (policy loading)" [Critical,Confirmed]
21:11:05 <markmcclain> https://bugs.launchpad.net/neutron/+bug/1208661
21:11:07 <uvirtbot> Launchpad bug 1208661 in neutron "nova show command sometimes don't includes ip address" [Critical,In progress]
21:11:21 <markmcclain> salvatore and enikanorov-w having been working on 1216866
21:11:38 <salv-orlando2> Bug 1216866 has been trusted
21:11:40 <uvirtbot> Launchpad bug 1216866 in neutron "Race condition in Neutron initialization (policy loading)" [Critical,Confirmed] https://launchpad.net/bugs/1216866
21:11:45 <salv-orlando2> Triaged
21:11:58 <salv-orlando2> Fix will be pushed shortly
21:12:18 <markmcclain> that's good to hear
21:12:38 <nati_ueno> 1208661 is in still investigations. It looks like 15 rechecks now
21:13:22 <markmcclain> nati_ueno: thanks for updating
21:13:27 <markmcclain> any other bugs the team should be tracking?
21:13:29 <wolfdreamer> As far as bugs masqeruading as features, has anybody seen this : https://bugs.launchpad.net/neutron/+bug/1211513
21:13:30 <uvirtbot> Launchpad bug 1211513 in neutron "neutron.api.extensions should resolve full module path when importing extensions so that proper base class comparison is done" [Undecided,In progress]
21:14:20 <nati_ueno> https://bugs.launchpad.net/neutron/+bug/1211915 gets high failure rate in gating also
21:14:21 <uvirtbot> Launchpad bug 1211915 in neutron "Connection to neutron failed: Maximum attempts reached" [Undecided,New]
21:14:42 <markmcclain> wolfdreamer: I looked at it briefly, but have not had time to check into it further
21:15:01 <wolfdreamer> markmcclain, k thanks
21:16:10 <markmcclain> nati_ueno: I'll add that the list of important bugs we're tracking
21:16:17 <nati_ueno> markmcclain: Thanks
21:16:40 <markmcclain> Any other bugs?
21:17:03 <markmcclain> #topic docs
21:17:08 <markmcclain> emagana: hi
21:17:17 <emagana> markmcclain: hi
21:17:41 <emagana> long list of bugs to fix but what is more important to start documenting the new features
21:19:09 <emagana> please, file Blueprints for any feature included in Havana
21:19:17 <markmcclain> yeah.. we're tracking fwaas, vpnass, and ml2
21:19:30 <markmcclain> we can work offline to make sure we've got a more complete list
21:19:46 <emagana> markmcclain: sure! nothing else to report
21:20:06 <emagana> markmcclain: Almost forget about it...
21:20:14 <markmcclain> We've got two items currently in review.  https://review.openstack.org/#/q/status:open+project:openstack/netconn-api,n,z
21:20:32 <markmcclain> when folks have change look at salv-orlando2 and pcm_ proposed docs
21:20:41 <emagana> We may need to create a new section for all services, the way it is now could be confusing for readers
21:20:51 <pcm_> yes please do; https://review.openstack.org/#/c/41702/
21:21:21 <markmcclain> emagana: let's discuss offline with annegentle and see what direction we should take
21:21:22 <annegentle> also please review https://review.openstack.org/#/c/42624/
21:21:29 <annegentle> I'm here
21:21:36 <markmcclain> annegentle: hi
21:21:43 <emagana> markmcclain: sounds good!
21:21:43 <annegentle> also happy to talk later
21:22:15 <markmcclain> annegentle: ok.. let chat real quick after the meeting is over if you're around
21:22:49 <markmcclain> Skipping the API this week because there's nothing to report
21:23:15 <markmcclain> #topic VPN
21:23:18 <markmcclain> nati_ueno: hi
21:23:31 <nati_ueno> OK client is approved.
21:23:38 <markmcclain> cool
21:24:04 <markmcclain> tried to ping you earlier because I wanted to confirm that if we select strongswan we can get rid of the wrapper correct?
21:24:07 <nati_ueno> there is four patches. Driver impl, status update, OpenSwan support, service type framework support
21:24:22 <markmcclain> or is it openswan that does not need the wrapper
21:24:32 <nati_ueno> markmcclain: its openswan
21:25:10 <nati_ueno> markmcclain: so you prefer to choose openswan and remove the ns wrapper?
21:25:23 <markmcclain> my preference is to support only openswan
21:25:29 <markmcclain> so that we get RHEL and centos
21:25:34 <markmcclain> and don't have a wrapper
21:25:42 <salv-orlando2> I think at this stage we should release Havana with the same f2f eat choice
21:25:53 <salv-orlando2> Which looks like it's openswan
21:25:56 <nati_ueno> ok if so,  I'll merge drvier impl and status update and openswan, and remove strongswan support
21:26:16 <markmcclain> I think that is the simplest approach
21:26:22 <nati_ueno> gotcha
21:26:31 <pcm_> There minor changes to limits in plugin: https://review.openstack.org/#/c/43559/1
21:26:34 <markmcclain> maybe in the next 6mos we can get someone to contribute an upstream fix for strongswan
21:26:50 <salv-orlando2> Safest often is simplest too
21:27:26 <nati_ueno> OK Thanks. That's all from vpnaas
21:28:00 <markmcclain> nati_ueno: thanks for the update
21:28:19 <markmcclain> #topic nova
21:28:22 <markmcclain> garyk: hi
21:28:35 <garyk> markmcclain: hi
21:28:50 <garyk> no updates on this front
21:29:26 <markmcclain> ok.. now that we're in a stability focus should we get some team members to focus on the bugs you've listed?
21:29:43 <markmcclain> bug 1192287
21:29:44 <uvirtbot> Launchpad bug 1192287 in nova "Creating server did not fail when exceeded Quantum quota limit" [Undecided,Confirmed] https://launchpad.net/bugs/1192287
21:30:31 <markmcclain> nati_ueno: what is the status of bug 1112912
21:30:32 <uvirtbot> Launchpad bug 1112912 in neutron "get_firewall_required should use VIF parameter from quantum" [High,In progress] https://launchpad.net/bugs/1112912
21:30:36 <garyk> arosen had ideas regarding this one but if i recall correctly the guys in the community did not like moving the checks.
21:30:47 <markmcclain> ok
21:31:02 <nati_ueno> markmcclain: Ah sorry it is slipped on my task list. I'll rebase the code
21:31:03 <garyk> i think that nati_ueno has a patch for that (unless i am mistaken)\
21:31:17 <markmcclain> nati_ueno: ok
21:31:18 <nati_ueno> garyk: yes. I rebased it in two week ago,
21:31:31 <nati_ueno> markmcclain: Ah but it needes nova side support
21:31:31 <markmcclain> Any other nova items?
21:31:40 <garyk> markmcclain: nope
21:32:06 <markmcclain> nati_ueno: do we have proposed code for the nova side or do we need to have someone work on it
21:32:34 <nati_ueno> markmcclain: I'll write the nova side code, but I should check nova side FFE policy. Do they think this is a bug or not
21:32:42 <nati_ueno> markmcclain: I'll check this with Russel
21:33:11 <markmcclain> nati_ueno: sounds good
21:33:21 <markmcclain> garyk: thanks for the update
21:33:24 <markmcclain> #topic FWaaS
21:33:27 <markmcclain> SumitNaiksatam: hi
21:33:29 <SumitNaiksatam> hi
21:33:32 <SumitNaiksatam> CLI patch for commit operation was posted by RajeshMohan (https://review.openstack.org/#/c/43249/)
21:33:42 <SumitNaiksatam> the API patch (https://review.openstack.org/#/c/41353) is waiting for more reviews
21:34:10 <SumitNaiksatam> (one of the three salvatores here can review it? :-))
21:34:18 <markmcclain> haha
21:34:22 <SumitNaiksatam> there was some review activity on the devstack patch (https://review.openstack.org/#/c/37147/) and is ready for approval in my opinion
21:34:34 <SumitNaiksatam> the Horizon patch (https://review.openstack.org/#/c/38724/) hasn't had any review activity in the past week
21:34:45 <SumitNaiksatam> I also submitted a FWaaS patch for Heat (https://review.openstack.org/#/c/43357/)
21:34:49 <salv3> I Havana not yet performed a full review but I have been . Looking for similar
21:35:15 <salv3> Approaches in restful PAId
21:35:16 <SumitNaiksatam> salv3: thanks
21:35:20 <salv3> Apis
21:35:42 <SumitNaiksatam> other than that FWaaS team is running tests, filling and fixing bugs
21:35:51 <markmcclain> garyk: would you mind taking a look at the devstack patch SumitNaiksatam referenced?
21:36:04 <SumitNaiksatam> markmcclain: garyk is on it
21:36:12 <markmcclain> great!
21:36:34 <salv3> So far I could not find anything similar. Probably because keeping state across requests is not restful at all. But I am not sure if I am saying something stupid here
21:36:46 <garyk> SumitNaiksatam: i have +2'ed. you addresed my concerns. Thanks!
21:36:53 <SumitNaiksatam> garyk: thanks!
21:36:58 <SumitNaiksatam> salv3: ok
21:37:01 <salv3> But I will perform a full code review so we can resumes the discussion from there
21:37:10 <SumitNaiksatam> salv3: thanks
21:37:13 <markmcclain> salv3: thanks for looking at it
21:37:29 <SumitNaiksatam> any other thoughts/comments/questions on FWaaS (else I am done)
21:37:37 <amotoki> Related to Horizon (or Heat), we need to release CLI.  Any release plan?
21:38:00 <markmcclain> amotoki: I can release contact me offline
21:38:08 <markmcclain> SumitNaiksatam: thanks for the update
21:38:22 <markmcclain> #topic LBaaS
21:38:24 <amotoki> markmcclain: thanks.
21:38:29 <enikanorov_> hi
21:38:43 <enikanorov_> anything we've planned for lbaas is on review
21:38:59 <enikanorov_> two major patches are blocked by policy loading issue
21:39:00 <markmcclain> service types is the major item remaining right?
21:39:07 <nati_ueno> markmcclain: amotoki: could you please wait vpnaas client is merged? (It should be merged in few hours)
21:39:19 <enikanorov_> but others are ready for review and are not very big
21:39:24 <enikanorov_> markmcclain: right
21:39:40 <markmcclain> nati_ueno: yes… I'm also going to look in the backlog to see if there are any other items to include as well
21:39:46 <enikanorov_> vpnaas patch is also blocked by the same issue as it is dependent on my patch
21:39:48 <nati_ueno> markmcclain: Thanks
21:40:01 <markmcclain> enikanorov_: I'll take a fresh look at the service types later today
21:40:10 <markmcclain> Any questions for LBaaS?
21:40:22 <markmcclain> #topic ML2
21:40:24 <salv-orlando> hi folks. I changed again my identity!
21:40:59 <rkukura> the key BPs made the deadline: portbinding, mult-segment-api, several mechanism drivers
21:41:06 * markmcclain wonders if the salv-orlando I'm sitting next to is real or a fake
21:41:17 <rkukura> we've got work to do to get them all merged
21:41:32 <markmcclain> rkukura: looks like we had a few vendors drivers submitted too
21:41:36 <markmcclain> that's good news
21:41:43 <rkukura> markmcclain: definitely!
21:42:10 <rkukura> one new BP/patch today may be candidate for an exception
21:42:18 <rkukura> https://blueprints.launchpad.net/neutron/+spec/ml2-mechanism-driver-subnet-calls
21:42:44 <markmcclain> ok.. I'll start a thread on the ML, but I'd like to make ML2 the default for open source plugin for Havana
21:42:57 <rkukura> This was a TODO that I though would get treated as a bug - its pretty much boiler plate competing the driver API
21:43:34 <rkukura> Big question is how to go about getting ml2 into gating?
21:43:42 <emagana> markmcclain: +1
21:43:51 <markmcclain> once ml2 becomes the default for devstack
21:44:01 <markmcclain> it will be the gate
21:44:13 <markmcclain> let's solicit input from the wider community this week
21:44:20 <rkukura> so do we want to just make it the default, or do we want to start gating on it 1st, then switch default?
21:44:30 <markmcclain> and then next week we can finalize the switch details
21:44:35 <rkukura> OK
21:45:02 <markmcclain> might make sense to gate first, but I don't want to change until after Tuesday
21:45:18 <markmcclain> we can work with the infra folks to ensure the change goes smoothly
21:45:24 <markmcclain> Anything else for ML2?
21:45:33 <rkukura> not from me
21:45:45 <markmcclain> rkukura: thanks for the update
21:45:51 <markmcclain> #topic Horizon
21:45:53 <markmcclain> amotoki: hi
21:45:58 <amotoki> hi
21:46:09 <markmcclain> How are things on the Horizon front?
21:46:30 <amotoki> fwaas, vpnaas are good progress.
21:46:41 <amotoki> i am reviewing the last round.
21:46:46 <markmcclain> good news
21:46:55 <markmcclain> we just need a client released for them
21:46:56 <amotoki> i have no concern abou them.
21:47:09 <amotoki> I will check quota bugs from now.
21:47:09 <markmcclain> ok
21:47:18 <markmcclain> Anything else to add?
21:47:27 <amotoki> nothing from me.
21:47:35 <markmcclain> amotoki: thanks for updating
21:47:37 <markmcclain> #topic Open Discussion
21:47:40 <gongysh> can someone help review multihost patch? https://review.openstack.org/#/c/37919/
21:47:47 <enikanorov_> I have a question about https://review.openstack.org/#/c/38230/ - "allowed address pairs"
21:48:08 <markmcclain> gongysh: looks like we've got 4 cores on it
21:48:12 <enikanorov_> some folks from my company expect this feature to be added
21:48:17 <emagana> gongysh: I will be review it
21:48:24 <markmcclain> and there are outstanding concerns on the current version
21:48:36 <gongysh> emagana has an opinion which needs community to address.
21:48:40 <enikanorov_> but corresponding bp is not approved yet. Can we expect it to land in havana?
21:49:00 <emagana> gongysh: however, I want to know if the team is ok having this as an extension or a default operation like in nova-network
21:49:47 <gongysh> emagana: even in nova-network, admin needs to specify the --multihost=true in api to create a multihosted network.
21:50:39 <markmcclain> I'm concerned about changing a well established API
21:50:41 <emagana> gongysh: let's move the discussion to ML
21:50:50 <markmcclain> +1 to moving to ML
21:50:54 <gongysh> emagana: I am not sure if I follow your idea.
21:50:57 <gongysh> ok to ML
21:51:11 <markmcclain> Any other open discussion items?
21:51:17 <enikanorov_> yeah ^^
21:51:29 <dkehn> amotoki, https://review.openstack.org/#/c/30447/
21:52:01 <amotoki> dkehn: this week is a review week :-)
21:52:12 <dkehn> amotoki, ok
21:52:23 <markmcclain> enikanorov_: you have something new or ok with moving discussion to ML?
21:52:41 <salv-orlando2> Using yet another nick
21:52:46 <nati_uen_> Which multi process patch we will use?
21:52:49 <enikanorov_> well, it's not quite a discussion, it's a question about the blueprint, https://blueprints.launchpad.net/neutron/+spec/allowed-address-pairs
21:53:00 <salv-orlando2> I suffer from multiple personalities syndrome
21:53:03 <markmcclain> enikanorov_:  that's what this time is for
21:53:06 <markmcclain> ask away
21:53:14 <enikanorov_> it's not approved, but the code was proposed a month ago
21:53:29 <enikanorov_> so I'm asking is there chances it will land in H-3
21:53:57 <markmcclain> enikanorov_: oops.. it's now approved :)
21:54:18 <enikanorov_> cool. That's wat i've expected :)
21:55:02 <markmcclain> enikanorov_: the BP admin UI is not the greatest
21:55:45 <markmcclain> Ok.. we did a really good job of getting the code submitted on time
21:56:23 <markmcclain> Have a great week and let's get some code reviewed
21:56:27 <markmcclain> #endmeeting