14:00:30 <mestery> #startmeeting networking
14:00:30 <openstack> Meeting started Tue Sep 22 14:00:30 2015 UTC and is due to finish in 60 minutes.  The chair is mestery. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:31 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:00:33 <hichihara> hi
14:00:34 <openstack> The meeting name has been set to 'networking'
14:00:34 <amotoki> hi
14:00:37 <johnsom> o/
14:00:43 <mestery> #link https://wiki.openstack.org/wiki/Network/Meetings Agenda
14:00:46 <emagana> mestery: A new one is coming.. ;-)
14:00:47 <xgerman> hi
14:00:47 <Piet> Morning
14:00:49 <dougwig> o/
14:00:59 * mestery waits a minute to let the stragglers wander in
14:01:01 <pc_m> hi
14:01:02 <ihrachys> mestery: time to vote for another tyrant
14:01:06 <haleyb> hi
14:01:15 <mestery> ihrachys: But can anyone be as tyrranical as I was? ;)
14:01:22 <russellb> hi
14:01:23 <mestery> I feel like a dinosaur or something :)
14:01:29 <scheuran> hi
14:01:34 <ihrachys> mestery: definitely! new ones are always worse then previous ones!
14:01:38 <mestery> lol
14:01:47 <mestery> #topic Announcements
14:01:58 <ajo> o/ :(
14:02:02 <ajo> :)
14:02:02 <mestery> #info We're going to cut liberty-rc1 tomorrow (likely)
14:02:08 <russellb> yay
14:02:11 <regXboi> mesteriasauras?
14:02:12 * ihrachys dances rumba
14:02:14 <mestery> #info We're tracking the final bugs (4 at last count), more later in the meeting
14:02:21 <emagana> last minutes commits.... now!!
14:02:30 <mestery> #link https://wiki.openstack.org/wiki/ReleaseNotes/Liberty#OpenStack_Liberty_Release_Notes Liberty Release Notes
14:02:42 <mestery> #info Please focus on making our release notes the best release notes out there! Thanks to those who have already jumped in there!
14:02:59 <mestery> #info Please vote in the Neutron PTL election!
14:03:01 <mestery> #link https://wiki.openstack.org/wiki/PTL_Elections_September_2015
14:03:14 <mestery> We have 3 great candidates, I for one am exçited to see who our new overlord will be later this week! :)
14:03:32 <mestery> And a last announcement ...
14:03:43 <mestery> #info THE PECAN WSGI BRANCH HAS BEEN FOLDED INTO MASTER! MOAR PECAN!
14:03:45 <mestery> I had to go all caps
14:03:50 <dougwig> Yay
14:03:52 <mestery> Because it's finally in! :)
14:03:57 * regXboi is in shock
14:03:58 <pc_m> nice!
14:03:59 <ryanpetrello> \o/
14:04:07 <regXboi> I'm shocked, SHOCKED to see all caps :)
14:04:10 * russellb offers pecan pie in celebration
14:04:11 <mestery> #info Thanks to kevinbenton salv-orlando blogan and a cast of thousands :)
14:04:14 <ajo> :-)
14:04:15 <mestery> russellb: lol
14:04:28 <mestery> Now to make it stable in Mitaka! :)
14:04:32 <mestery> Any other announcements for the team?
14:04:50 <ihrachys> me as an example of an ignorant packager could use some docs help on how we expect to run them, whether they need to conflict with current neutron-server service etc. do we have such docs?
14:05:02 <Piet> Let me know when you want to talk about the survey
14:05:04 <ihrachys> those new servers I mean
14:05:15 <kevinbenton> ihrachys: no, not yet.
14:05:23 <mestery> ihrachys: We do not yet
14:05:40 <mestery> kevinbenton: We need a release note indicating it's experitmental
14:05:43 <mestery> etc.
14:05:50 <gongysh_> kevinbenton: pecan version  is ready to create a network?
14:06:00 <kevinbenton> ihrachys: they don't work quite right with service plugins that have their own extension controllers
14:06:11 <kevinbenton> gongysh: Yeah, should work for that
14:06:21 <ihrachys> kevinbenton: that's fine, ignorant packagers just want to package it and maybe set deps between services
14:06:23 <gongysh_> kevinbenton: congrat.
14:06:31 <emagana> ihrachys and kevinbenton we could add a chapter into the networking guide indicating that its experimental
14:06:54 <mestery> Well, given it doesn't fully work yet, maybe we don't want to go that far in documenting it. What thinks you kevinbenton?
14:07:21 <emagana> mestery: the sooner the better.. it will help to get more people testing it
14:07:22 <mestery> Anyways, lets keep the agenda moving.
14:07:24 <gongysh_> where to find the doc, even devref to set it up?
14:07:27 <mestery> emagana: True dat :)
14:07:55 <mestery> Since Piet has joined us ...
14:07:59 <dougwig> Maybe add +q-pecan to devstack.
14:08:02 <mestery> #topic OpenStack UX team survey update
14:08:10 <mestery> Piet: Floor is yours for an update!
14:08:21 <Piet> The OpenStack UX project has created a survey to understand some of the challenges associated with moving from Nova Network to Neutron.
14:08:33 <Piet> It was based on a series of interviews conducted by a User Researcher with HP
14:08:42 <Piet> Please, please, please complete the survey and distribute to anyone with experience with Networks in OpenStack.
14:09:00 <Piet> https://www.surveymonkey.com/r/osnetworking
14:09:01 <mestery> Piet: Did you send a mail to the ML yet?
14:09:04 <mestery> Ah, even better
14:09:08 <mestery> #link https://www.surveymonkey.com/r/osnetworking
14:09:14 <mestery> #info Please spread the survey far and wide
14:09:21 <Piet> We're trying to get the sample to about 100 respondents
14:09:23 <mestery> Thanks Piet! Any questions from anyone for Piet and team?
14:10:06 <mestery> Thank you Piet and team for working on this. Looking forward to the results. :)
14:10:09 <mestery> OK, lets move along.
14:10:10 <Piet> Feel free to reach-out to me on IRC if you have any questions
14:10:15 <mestery> Thanks Piet!
14:10:20 <Piet> I can also email you the formal invitiation
14:10:29 <mestery> #topic liberty-rc1
14:10:36 <mestery> #link https://launchpad.net/neutron/+milestone/liberty-rc1
14:10:47 <mestery> The last BP, which I expect to defer, is this one: https://blueprints.launchpad.net/neutron/+spec/lbaas-ref-octavia
14:10:52 <mestery> dougwig: What says the LBaaS team on this?
14:10:59 <mestery> Defered to Mitaka?
14:11:05 <xgerman> dougwig rudely stopped us with a -2
14:11:10 <mestery> lol
14:11:14 <mestery> xgerman: Well, if he hadn't, I was gonna :)
14:11:16 <mestery> ;)
14:11:24 <xgerman> tyrant :-)
14:11:33 <mestery> rofl
14:11:36 <dougwig> the code is ready, the gate jobs are running into issues because it's VMs inside VMs.  johnsom has a patch up that might resolve tha.
14:11:36 <dougwig> that
14:11:41 <dougwig> we'll make the final call today, with you.
14:11:41 <sc68cal> gonna run mestery out on a rail
14:11:46 <mestery> dougwig: Ack, sounds good
14:11:51 <johnsom> I think the patches we merged yesterday will have resolved the gate issues.
14:11:57 <xgerman> +1
14:12:00 <mestery> #info We'll make the final call on Octavia later today, 9-22
14:12:09 <mestery> johnsom: Lets make sure to stay in sync on IRC today
14:12:13 <mestery> And see what we come up with by end of day
14:12:25 <johnsom> Sounds good
14:12:30 <mestery> Excellent!
14:12:35 <mestery> Now, on to the final four bugs on the dashboard
14:12:36 <mestery> #link https://bugs.launchpad.net/bugs/1453350
14:12:38 <openstack> Launchpad bug 1453350 in neutron "race between neutron port create and nova" [High,In progress] - Assigned to Kevin Benton (kevinbenton)
14:12:59 <mestery> kevinbenton: What's happening with this one? How is it coming along? And even better, do you see this as an RC blocker?
14:13:25 <kevinbenton> I will have the latest patch up for that today and we can determine if it's too risky for release
14:13:31 <mestery> kevinbenton: Ack
14:13:41 <mestery> #info for 1453350, we'll make a call later in the day.
14:14:01 <mestery> Next up
14:14:10 <mestery> #link https://bugs.launchpad.net/bugs/1497450
14:14:11 <openstack> Launchpad bug 1497450 in neutron "DNS lookup code in get_ports needs to be optimized" [High,In progress] - Assigned to Miguel Lavalle (minsel)
14:14:19 <mestery> mlavalle: Same thing here?
14:14:24 <mestery> Looks like we're expecting a patch later today
14:14:32 <mlavalle> mestery: we will have a patchset today
14:14:47 <mlavalle> I think it is looking good
14:15:06 <mestery> mlavalle: Thanks! regXboi make sure you review that patch when mlavalle posts it.
14:15:08 <kevinbenton> mlavalle: can you ping me with the review when it's ready?
14:15:17 <mestery> kevinbenton: thanks for your review as well :)
14:15:18 <mestery> Next up
14:15:20 <mlavalle> kevinbenton: will do
14:15:21 <mestery> #link https://bugs.launchpad.net/bugs/1488996
14:15:22 <openstack> Launchpad bug 1488996 in neutron "QoS doesn't work when a mechanism driver which doesn't implement bind_port explicitly supports it" [Medium,In progress] - Assigned to Miguel Angel Ajo (mangelajo)
14:15:23 <regXboi> mestery: ack - probably run it through the test scenario for an hour
14:15:26 <ihrachys> that one is in gate
14:15:31 <mestery> ihrachys ajo: This one looks good?
14:15:33 <mestery> nice!
14:15:36 * mestery ^5s ihrachys
14:15:42 <mestery> Easy peasy ihrachys, just how I like it ;)
14:15:51 <mestery> And the last one
14:15:52 <mestery> #link https://bugs.launchpad.net/bugs/1495628
14:15:53 <ajo> mestery, it does, it's ihrachy's work, I just pushed a last patch
14:15:54 <openstack> Launchpad bug 1495628 in neutron "In DHCP agent's enable_dhcp_helper, its good to call safe_configure_dhcp_for_network" [Medium,In progress] - Assigned to Sudhakar Gariganti (sudhakar-gariganti)
14:15:59 <ajo> thanks mestery  :)
14:16:00 <mestery> ajo: Ack sir
14:16:07 <mestery> This one went from a 1 line patch to something much more
14:17:11 <mestery> Anyways, we'll make the call on this one later today, doesn't seem to be a release blocker however
14:17:32 <mestery> OK, does anyone want to make a plea for anything else in Liberty at this point?
14:17:34 <mestery> :)
14:18:05 <ajo> love and liberty ;D
14:18:11 <mestery> ajo: Love and Liberty indeed :)
14:18:13 <regXboi> how about egalite? fraternite?
14:18:22 <mestery> :)
14:18:28 <regXboi> (sorry, can't find the proper glyphs on this keyboard)
14:18:30 <mestery> #topic Mitaka Design Summit
14:18:30 <mestery> #link https://etherpad.openstack.org/p/neutron-mitaka-designsummit
14:18:32 <pc_m> mestery: Don't think endpoint groups will be reviewed in time,
14:18:39 <mestery> pc_m: Ack :(
14:18:48 <mestery> So, we have a lot of items proposed on the etherpad which is great!
14:19:01 <mestery> Our next PTL will be responsible for planning this thing out, but I encourage folks to keep submitting ideas
14:19:05 <mestery> KEep in mind we only have 12 slots
14:19:17 <mestery> So I expect the next PTL to focus on community items that really need face to face discussion
14:19:41 <ajo> I suggested a cross neutron/nova one, but not sure if we have slots for cross-project stuff
14:19:43 <mestery> That's about all I have for Mitaka
14:19:50 * regXboi feels mestery weilding clue brick :)
14:19:55 <mestery> ajo: There is a cross-project track, but we could use one of our slots.
14:19:57 <regXboi> er wielding
14:19:59 <mestery> ajo: Anything specific for there?
14:20:09 <mestery> ajo: I would think perhaps a "brick" like library for nova to use for neutron interactions?
14:20:10 <ajo> mestery: the neutron/nova QoS integration
14:20:13 <mestery> That was discussed in Vancouver
14:20:17 <mestery> ajo: Ah, that too :)
14:20:20 <ajo> we have scheduling bits to be accounted from the nova side
14:20:39 <ihrachys> ajo: haven't we discussed it on one of the qos meetings with nova folks?
14:20:45 <ajo> (more than accounted, taken care of..) we already are discussing in meetings, but slow progress
14:20:47 <ihrachys> ah scheduler. ok.
14:21:03 <ajo> ihrachys: yes, but still no resolution on that part, it seemed to need a broader discussion
14:21:29 <ihrachys> I would suggest not to wait for summit though and start on ML
14:21:34 <mestery> OK, so this seems like a good candidate ajo and ihrachys  :)
14:22:02 <mestery> #topic Open Discussion
14:22:05 <mestery> Shortest. Meeting. Ever.
14:22:08 <mestery> I just jinxed myself :)
14:22:24 <regXboi> mestery: yes you did - I can hear murphy laughing all the way to the time clock
14:22:24 <ajo> mestery: (about the brick) yes we talked about a nova pluggable thing on the scheduler we could provide from neutron, but it seems that's far, and may be needing a short term solution until that's ready
14:22:51 <ajo> regXboi :)
14:23:06 <mestery> ajo: Ack sir, makes sense
14:23:31 <sc68cal> anything that touches nova-scheduler enters a warp in space time, where percieved time is slower than real time
14:23:46 <ajo> sc68cal lol
14:23:49 <xgerman> gantt?
14:23:53 <mestery> rofl
14:23:54 <sc68cal> ^ this
14:24:10 * sc68cal got a huge physics refresher from regXboi at the QA sprint. his mind is blown still
14:24:15 <mestery> :)
14:24:16 <ajo> they have low gravity
14:24:32 <emagana> mestery: ajo and myself are working on the QoS chapter. Thanks to ihrachys for the reviews but more eyes on it will not hurt!
14:24:33 <regXboi> sc68cal: that was only a few years of physics ...
14:24:41 <mestery> emagana: Excellent!
14:24:49 <ihrachys> emagana: link!
14:24:57 <emagana> #link https://review.openstack.org/#/c/221990/
14:25:00 <ihrachys> no links == no eyes
14:25:09 <rossella_s> :D
14:25:17 <ajo> :)
14:25:26 <emagana> ihrachys: how many eyes per times the link is posted  :-)
14:25:28 <mestery> cool
14:25:30 <mestery> rofl
14:25:51 <mestery> OK, well, next week we'll have a new PTL running this meeting. Exciting!
14:25:55 * ihrachys smiles
14:25:59 <emagana> Goos luck to the candidates!!!
14:26:01 <jckasper> mestery: I added Horizon topic to bottom of On Demand Agenda late yesterday
14:26:08 <ajo> good luck!! :)
14:26:11 <emagana> Goos = Good
14:26:11 * regXboi mutters *almost got it that time*
14:26:14 <mestery> jckasper: Well, that ruins my huge exit, but pleae go ahead :)
14:26:26 <emagana> LOL
14:26:29 <mestery> :D
14:26:31 <emagana> doh
14:26:43 <jckasper> Horizon updated the network topology diagram last week
14:26:51 <jckasper> https://blueprints.launchpad.net/horizon/+spec/curvature-network-topology
14:26:59 <mestery> #link https://blueprints.launchpad.net/horizon/+spec/curvature-network-topology
14:27:30 <jckasper> New diagram quite a bit diff than old design.   some items lacking (in my opinion)
14:27:45 <mestery> jckasper: You're looking for review, rights?
14:28:06 <jckasper> mestery: wanting to see if anyone else has concerns over new design
14:28:07 <emagana> curvature things sounds familiar to me
14:28:20 * mestery looks
14:28:24 <fawadkhaliq> emagana: +1
14:28:31 <dougwig> got a screenshot handy?
14:28:36 <robcresswell> It was demo'd in a summit about 2 years ago. It's actually been merged into Horizon at this point, but would be great to hear feedback.
14:28:53 <amuller_> screenshot: http://imgur.com/a/qddWB
14:29:16 <ajo> https://www.youtube.com/watch?v=oFTmHHCn2-g&feature=youtu.be
14:29:20 <mestery> jckasper: So, we're moving to the curvature version instead of the normal one?
14:29:27 <neiljerram> screenshot looks nice!
14:29:31 <jckasper> mestery: already moved
14:29:36 <amotoki> it is a replacement of the legacy one
14:29:42 <mestery> jckasper: Yikes, so your concerns are somewhat valid.
14:29:51 <xgerman> neiljerram +1
14:30:07 <mestery> OK, well, it will be interesting!
14:30:19 <neiljerram> So what specificatlly is lacking?
14:30:21 <mestery> Curvature itself is pretty cool from what I remember back in Portland, glad to see it got merged into Horizon!
14:30:28 <jckasper> mestery: new layout does not show IP address or CIDRs unless you drill down
14:30:36 <emagana> mestery: +1
14:30:39 * regXboi will likely have comments later today - especially after testing the DNS patch :)
14:31:03 <regXboi> since I'll be spinning 100 tenants in a cloud for that DNS patch, I'll see what I think of the net topo under scale :)
14:31:14 <robcresswell> Drop into #openstack-horizon if oyu have any concerns and we'll try to address them.
14:31:17 <robcresswell> you*
14:31:18 <neiljerram> jckasper: if that's the only thing, sounds acceptable to me.
14:31:33 <jckasper> flyover pop-ups seem to be missing from old layout. but that is minor
14:31:35 <emagana> jckasper: is this enabled by default in horizon?
14:31:36 <mestery> I'd like to see regXboi's scale testing as well
14:31:43 <xgerman> jckasper does it show load balancers?
14:31:48 <jckasper> yes.  it is enabled by default.
14:31:50 <regXboi> mestery: I'll try and do a screen shot when I'm done
14:32:21 <jckasper> xgerman: not sure how they are represented
14:32:46 <jckasper> xgerman: would be a good thing to try out.
14:33:04 <xgerman> yep - along with Firewall
14:33:07 <jckasper> regXboi: I'm also wondering how DVR will be displayed too
14:33:15 <mestery> And VPN xgerman :)
14:33:34 <regXboi> jckasper: that's not something I'm looking at with respect to the DNS patch, so I won't be able to say today
14:33:39 <amotoki> johnsom: even you use DVR, it is one logical router
14:33:52 <amotoki> jckasper: ^^
14:33:53 <doug-fish> were these things (load balancer, vpn, firewall) represented in the previous network topology UI?
14:34:10 <robcresswell> Horizon core invasion :p
14:34:11 <xgerman> I am always hoping for new features
14:34:15 <mestery> lol
14:34:23 <doug-fish> I don't think they were, but wanted to double check
14:34:30 <dougwig> doug-fish:they were not
14:34:32 <amotoki> doug-fish: no they were not.
14:34:42 <jckasper> amotoki: but DVR typically can have multiple router interfaces.  not sure how that is represented
14:34:43 <doug-fish> ok cool
14:35:30 <amotoki> jckasper: i got you point. need to check the detail, but IIRC there is no difference from legacy ones.
14:36:47 <mestery> OK, thanks horizon folks for bringing this to our attention!
14:36:53 <mestery> And with that, lets shut this thing down. :)
14:36:59 <mestery> Keep focused on those remaining RC bugs
14:37:00 * dougwig waves
14:37:02 <mestery> And if new ones are found
14:37:05 <mestery> Please let me know ASAP
14:37:09 <mestery> I'll be on the hunt as well :)
14:37:13 <mestery> Adieu!
14:37:15 <mestery> #endmeeting