21:00:59 <markmcclain> #startmeeting Networking
21:01:00 <openstack> Meeting started Mon Sep  2 21:00:59 2013 UTC and is due to finish in 60 minutes.  The chair is markmcclain. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:01:02 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:01:04 <openstack> The meeting name has been set to 'networking'
21:01:21 <enikanorov_> hi
21:01:41 <markmcclain> #link https://wiki.openstack.org/wiki/Network/Meetings
21:02:04 <markmcclain> It's modified agenda today, so that we can focus reviews
21:02:17 <markmcclain> #topic Announcements
21:02:44 <markmcclain> For those on holiday, thanks for joining today
21:03:19 <markmcclain> #info H3 Merge cutoff is End of Day Wednesday
21:03:28 <markmcclain> (September 4th)
21:03:53 <markmcclain> The extra day will be nice as we have lots of outstanding reviews
21:04:03 <markmcclain> #link https://launchpad.net/neutron/+milestone/havana-3
21:04:15 * salv-orlando yells in horror
21:04:34 * salv-orlando goes to gerrit and screams in pain
21:04:53 <markmcclain> haha
21:05:33 <markmcclain> We were very good on getting the code submitted on time, but several patches have required extra review cycles which is typical this time of year as everyone raced to beat the deadline
21:05:43 <markmcclain> #topic bugs
21:05:53 <markmcclain> #link https://bugs.launchpad.net/neutron/+bugs?search=Search&field.importance=Critical&field.status=New&field.status=Confirmed&field.status=Triaged&field.status=In+Progress
21:06:06 <markmcclain> Only open bug: https://bugs.launchpad.net/neutron/+bug/1208661
21:06:09 <uvirtbot> Launchpad bug 1208661 in neutron "nova show command sometimes don't includes ip address" [Critical,In progress]
21:06:19 <markmcclain> nati_ueno: any update
21:06:19 <markmcclain> ?
21:06:32 <nati_ueno> markmcclain: sorry no update yet
21:06:47 <markmcclain> nati_ueno: no worries
21:06:58 <markmcclain> Any other bugs the team should be tracking that are critical for H3?
21:07:27 <salv-orlando> none that I know of. However, is that bug really critical? We had more serious bugs marked as high or medium
21:07:34 <salv-orlando> I don't want to downplay, just asking
21:07:46 <markmcclain> salv-orlando: it can break the tempest gating at times
21:08:08 <salv-orlando> markmcclain: ok, that's sufficient as explanation.
21:08:32 <markmcclain> #topic docs
21:08:33 <nati_ueno> markmcclain: It looks like 1208661 isn't breaking now, so we can set high.
21:08:47 <nati_ueno> markmcclain: it looks like this one breaking now.. https://bugs.launchpad.net/neutron/+bug/1211915
21:08:48 <uvirtbot> Launchpad bug 1211915 in neutron "Connection to neutron failed: Maximum attempts reached" [High,Confirmed]
21:09:06 <nati_ueno> Rechecks: 33
21:09:10 <nati_ueno> http://status.openstack.org/rechecks/
21:09:24 <markmcclain> #undo
21:09:25 <openstack> Removing item from minutes: <ircmeeting.items.Link object at 0x312e290>
21:09:40 <markmcclain> nati_ueno: ok will reprioritize
21:09:45 <nati_ueno> markmcclain: Thanks
21:10:30 <emagana> On Docs:  SumitNaiksatam added FWaaS section
21:10:40 <markmcclain> #topic docs
21:10:41 <emagana> Mellanox plugin was included
21:10:49 <markmcclain> great news
21:11:07 <emagana> ML2 is still missing
21:11:13 <salv-orlando> regarding docs - now that you guys are doing a great job on the admin guide, is there any hope you could contribute api sections for the fw and vpn apis?
21:11:29 <emagana> is assigned to rkukura but I can help if needed
21:11:51 <markmcclain> mestery or rkukura: do you have anyone to write the ml2 docs?
21:11:54 <rkukura> no progress on ml2 docs yet
21:12:09 <rkukura> focusing on coding and reviews still
21:12:11 <emagana> salv-orlando: I can review it and propose changes
21:12:12 <markmcclain> salv-orlando: good point we need to make sure the API docs are complete too
21:12:35 <markmcclain> Any other doc items?
21:12:45 <nati_ueno> salv-orlando: Paul is proposing api doc https://review.openstack.org/#/c/41702/ , but we should work admin doc
21:13:01 <markmcclain> nati_ueno: cool
21:13:14 <salv-orlando> nati_ueno: ok, so let me understand. so for vpn we don't have admin docs?
21:13:15 <emagana> nothing else from my side, anyone else is welcome to comment
21:13:29 <salv-orlando> I should have checked gerrit better - I thought it was the exact opposite.
21:13:43 <nati_ueno> salv-orlando: it should have, but we didn't started yet.
21:13:59 <salv-orlando> ok so we don't have them ;)
21:14:06 <nati_ueno> salv-orlando: yep
21:14:06 <salv-orlando> no worries, but better raise priority on that item
21:14:19 <nati_ueno> salv-orlando: I agree
21:14:21 <salv-orlando> the admin doc has higher priority than the api doc, as the latter is mostly for devs
21:14:46 <emagana> salv-orlando: +1
21:14:47 <markmcclain> yeah.. deployers will need admin docs
21:14:48 <nati_ueno> salv-orlando: we have wiki, so we can translate it https://wiki.openstack.org/wiki/Neutron/VPNaaS/HowToInstall
21:14:58 <nati_ueno> salv-orlando: I'll work after 9/4 ends
21:15:09 <markmcclain> nati_ueno: sounds good
21:15:18 <markmcclain> #topic Reviews
21:15:22 <emagana> nati_ueno: I opened the bp in docs for that and I assigned to you, let me know if you need help!
21:15:32 <nati_ueno> emagana: Thanks!
21:15:38 <markmcclain> This is bulk of what we need to accomplish the next two days
21:15:43 <markmcclain> #link https://wiki.openstack.org/wiki/Network/Meetings#Reviews
21:16:06 <markmcclain> I've gone through and broke the review out by current status
21:16:17 <markmcclain> This does not include all reviews, just the major ones
21:17:10 <markmcclain> The first group of reviews are fairly good shape.  They might have some comments to address, but if the proposers stay responsive the should merge.
21:17:28 <salv-orlando> I don't think it misses any outstanding item, except the midonet blueprint probably, which is also a good candidate to merge.
21:17:39 <markmcclain> salv-orlando: good catch
21:18:12 <markmcclain> I'll add it
21:18:15 <nati_ueno> IMO l2 population is also a good candidate
21:18:19 <rkukura> I don't see at least 2 of the ml2 blueprint implementations listed
21:18:50 <rkukura> the Cisco mech driver and the multi-segment API
21:18:55 <markmcclain> rkukura: which ones did I miss
21:19:09 * markmcclain gerrit interface is overflowing
21:19:27 <rkukura> https://blueprints.launchpad.net/neutron/+spec/ml2-multi-segment-api
21:19:54 <rkukura> https://blueprints.launchpad.net/neutron/+spec/ml2-md-cisco-nexus
21:20:15 <markmcclain> the cisco nexus one is the closer of the two
21:20:29 <markmcclain> but needs to split into the general and vendor specific patchsets
21:21:02 <markmcclain> the multi-segment failed basic gating
21:21:15 <markmcclain> if it gets fixed up I don't see why it couldn't merge
21:21:23 <markmcclain> nati_ueno: on l2 population
21:21:40 <rkukura> markmcclain: I had advised them to include the manager fix in the merge
21:22:31 <markmcclain> rkukura: understandable, but at this time of year want to make sure that should we need to revert for stability
21:22:43 <markmcclain> the general and vendor code is not mixed
21:22:51 <markmcclain> should be a trival git fix up
21:23:02 <rkukura> markmcclain: No problem.
21:23:02 <markmcclain> the code itself is in good shape
21:23:40 <markmcclain> nati_ueno: looks like the l2 population is nearly ready to go too (I had missed kyle's comments)
21:24:01 <markmcclain> Now onto the 2nd grouping
21:24:01 <salv-orlando_> hi. internet died at my place. switched to mobile.
21:24:28 <markmcclain> salv-orlando_: really multiple personalities again this week? :p
21:24:40 <salv-orlando> yeah it must be the curse of the monday
21:24:47 <salv-orlando> or perhaps I need a psychiatrist
21:25:13 <markmcclain> we haven't seen salv3 yet
21:25:34 <markmcclain> so we've got 5 in the 2nd group which are the most at risk for not merging
21:26:14 <markmcclain> arosen1: do you think the allowed address pairs will be able to be revised in time?
21:26:39 <arosen1> markmcclain:  yea should be. I think it should be good to go in
21:26:50 <arosen1> after another patch per salvs comments.
21:26:59 <salv-orlando> I did not have any concern with cannot be addressed in a single review cycle
21:27:07 <salv-orlando> with -> which
21:27:13 <markmcclain> ok.. I had in the 2nd group since it had not be revised in a few days
21:27:36 <arosen1> i'll push a new patch first thing tomorrow morning revising it.
21:27:40 <markmcclain> cool
21:27:56 <markmcclain> The vArmour needs another round of reviews
21:28:04 <markmcclain> https://review.openstack.org/#/c/43034/
21:28:23 <markmcclain> I've done extensive reviews on it, but we need 2nd core
21:28:35 <markmcclain> SumitNaiksatam and arosen1 have both looked
21:28:40 <arosen1> markmcclain:  i'll help with it.
21:28:45 <markmcclain> arosen1: thanks
21:28:59 <markmcclain> ML2 PXE: https://review.openstack.org/#/c/44564/
21:29:12 <markmcclain> Looks like it is missing some unit tests
21:29:17 <markmcclain> amotoki has reviewed
21:29:30 <markmcclain> nati_ueno as well
21:29:40 <dkehn> yes it is it will need to be reworked since the merge but it is working
21:29:41 <markmcclain> so we've got 2 cores for it
21:29:45 <salv-orlando> on the 2nd group I can give an update on the Nicira patches.
21:29:53 <salv-orlando> before we move to lbass service types
21:29:55 <markmcclain> salv-orlando: go ahead
21:30:02 <salv-orlando> a very quick update - we have 6 patches
21:30:09 <salv-orlando> very confident on 3 of them.
21:30:13 <salv-orlando> 1 on the fence.
21:30:34 <salv-orlando> 2 are probably going to be a ffe request (lb, fw service patches)
21:30:49 <salv-orlando> that's all
21:31:14 <markmcclain> ok.. the code is limit to nvp plugin which should make it easier to get an ffe approved
21:31:26 <markmcclain> FFE = Feature Freeze Exception
21:31:53 <salv-orlando> all those patches have zero (0) lines in paths which do not start with /neutron/plugins/nicira or /etc/neutron/plugins/nicira
21:32:16 <markmcclain> good to know
21:32:38 <markmcclain> hopefully many of them will merge.. if not we can work offline to request the FFEs
21:32:57 <markmcclain> The last patch on the fence is:
21:33:03 <markmcclain> https://review.openstack.org/#/c/38120/
21:33:14 <markmcclain> LBaaS Service Type Framework
21:33:20 <enikanorov> yeah
21:33:31 <enikanorov> so we're going back and forth on design regarding using the noop
21:33:44 <enikanorov> actually from coding perspective it's a minor item
21:34:09 <enikanorov> will you have like 5-10 spare minutes after the meeting to discuss it and we'll make a final call?
21:34:31 <markmcclain> depending on what time we finish.. I'll might have a few minutes
21:34:45 <enikanorov> so i
21:35:11 <enikanorov> wasn't actually planning to dive in a deep discussion, just want to verify the steps
21:35:40 <enikanorov> which i thing would be 1) to fix provider at creation 2) stop service if provider is removed, but resources are not
21:36:28 <enikanorov> in such approach noop driver is not involved
21:36:42 <enikanorov> and we could discuss noop later during the summit
21:36:46 <enikanorov> what so you think?
21:36:49 <enikanorov> *do
21:36:56 <markmcclain> I think that makes sense for now
21:37:09 <markmcclain> and we can discuss no-ops in Hong Kong
21:37:14 <enikanorov> ok. I'll make corresponding code changes.
21:37:15 <enikanorov> yeah
21:37:43 <markmcclain> the last group are reviews that are doubtful
21:38:17 <markmcclain> it isn't to say that these won't make the merge, but they each have several hurdles to clear
21:38:25 * salv-orlando I pretty much think what has just been agreed was labeled as insufficient on the mailing list a while ago… but I will shut my mouth and keep a low profile
21:40:10 <markmcclain> ML3: https://review.openstack.org/#/c/20693/
21:40:33 <markmcclain> This review was rebased very late and I'm concerned about ensuring stability this late in the cycle
21:40:43 <markmcclain> (it is a big patch that touches many files)
21:40:46 <markmcclain> Thoughts?
21:42:03 <salv-orlando> I'm pretty sure Bob has done an awesome work. However, this kind of patch might inject bugs that are not easy to track down. not to mention potential impact on non-gated plugins (of which we do not care a lot anyway)
21:42:51 <salv-orlando> we're expecting already a huge bug intake in the RC period. I tend to think we might prefer to keep code churn contained at this stage of the release cycle
21:43:04 <markmcclain> agreed
21:43:42 <markmcclain> The last patch set is the Juniper Contrail
21:44:13 <markmcclain> There are several review comments outstanding it really all depends on how fast they are addressed
21:44:25 <emagana> who is the core dev supporting Juniper?
21:44:31 <markmcclain> emagana: me
21:45:20 <emagana> markmcclain: Thanks!
21:45:48 <markmcclain> I think we've hit all of the major reviews any I skipped that we've not discussed yet?
21:46:05 <nati_ueno_2> markmcclain: could you add service type support for vpn ?
21:46:14 <dkehn> one more https://review.openstack.org/#/c/30447/
21:46:18 <nati_ueno_2> markmcclain: I'll rebase my code asap after lbaas finished
21:47:06 <markmcclain> nati_ueno_2: does it actually depend on lbaas code?
21:47:22 <nati_ueno_2> markmcclain: yes, it's using common code
21:47:47 <enikanorov_> nati_ueno_2: going to push new version today
21:48:05 <salv-orlando> dkehn: deadlines do not apply to python-neutronclient
21:48:16 <amotoki> I also have another patch on nec plugin https://review.openstack.org/#/c/43572/ . I believe it is ready to go after removing test-requirements. it was accidently included :-(    I hope anyone helps me.
21:48:17 <dkehn> salv-orlando, k
21:48:17 <markmcclain> enikanorov-w: can we split the lbaas service types into the general code and lbaas specific patchsets?
21:48:32 <markmcclain> would make it easier to independently review the two changes
21:48:41 <nati_ueno_2> sounds great
21:49:07 <markmcclain> amotoki: correct… I figured it would get the 2nd +2 pretty quickly
21:49:19 <enikanorov_> yes we can, but common part is actually quite small.
21:49:28 <enikanorov_> Nachi could just copy it
21:49:40 <markmcclain> enikanorov_: ok.. that's a good thing
21:49:55 <markmcclain> we should be able to review and merge fast so you and nachi can work faster
21:49:58 <salv-orlando> amotoki: I'll review it before bed
21:50:17 <markmcclain> #topic
21:50:23 <markmcclain> #undo
21:50:24 <openstack> Removing item from minutes: <ircmeeting.items.Topic object at 0x2f85ed0>
21:50:26 <markmcclain> #topic client
21:50:37 <nati_ueno_2> markmcclain: enikanorov_: akihiro: Thanks!
21:50:47 <markmcclain> We've had a few jobs that have been hung in the gate
21:51:25 <markmcclain> once they finally finish, I'll spin a 2.3.0 release for horizon
21:52:10 <amotoki> do we have the corresponding milestone in launchpad?
21:52:24 <markmcclain> amotoki: no yet.. I'll create one and clean-up the bugs
21:52:48 <markmcclain> #topic horizon
21:52:57 <markmcclain> amotoki: any updates on Horizon for H3?
21:53:18 <amotoki> fwaas has good progress.
21:53:39 <amotoki> vpnass support is beging under working. it may be FFE candidates.
21:54:00 <amotoki> please see wiki about other review status.
21:54:09 <markmcclain> amotoki: thanks for updating
21:54:15 <markmcclain> #topic Open Discussion
21:54:17 <markmcclain> We
21:54:56 <markmcclain> We used a modified agenda today.  Does anyone have any items to discuss that we skipped?
21:55:13 <dkehn> amotoki, did you see my question concerning the dhcp_option delete?
21:55:43 <amotoki> dkehn: which review?
21:55:57 <dkehn> amotoki, it was in IRC
21:58:19 <markmcclain> I'm super excited the way the Havana release is shaping up.  Let's get through the next two days of reviews
21:58:38 <nati_ueno_2> yeah me too!
21:58:51 <markmcclain> Thanks to those who logged in on a holiday.
21:58:54 <markmcclain> #endmeeting