21:00:51 #startmeeting networking 21:00:52 * haleyb wonders if this is a 3-hour tour on the good ship neutron 21:00:52 Meeting started Mon Oct 17 21:00:51 2016 UTC and is due to finish in 60 minutes. The chair is armax. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:00:53 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:00:55 The meeting name has been set to 'networking' 21:00:56 o/ 21:01:00 o/ 21:01:07 hi 21:01:14 o/ 21:01:41 welcome everyone 21:02:10 o/ 21:02:21 thanks for joining this neutron meeting! 21:02:25 the agenda for today 21:02:29 #link https://wiki.openstack.org/wiki/Network/Meetings 21:02:35 #topic Announcements 21:03:28 the design summit is one week away if you have not noticed 21:03:31 o/ 21:03:33 the summit schedule is up 21:03:41 https://www.openstack.org/summit/barcelona-2016/summit-schedule/global-search?t=Neutron%3A 21:04:25 this is the last summit in the well known format and as such it’s gonna be the last one where we’re gonna have a contributors meetup 21:04:27 #link https://www.openstack.org/summit/barcelona-2016/summit-schedule/events/17227 21:04:52 if any of you is staying the Friday, I’d strongly invite you to sign up on the related etherpad 21:07:02 because of the summit it’s safer to cancel the next couple of meetings 21:07:18 o/ 21:07:50 so the next Neutron team meeting is going to be on 21:08:06 Tue 8th 21:08:11 November 21:09:02 I'll be serving as an official in the US election, so I will miss it. 21:09:24 njohnston: you running for president? 21:10:00 Yes, haven't you seen the TV ads for the FWaaS party? 21:10:10 :) 21:10:42 any other announcement? 21:10:51 it’ll be difficult to top njohnston’ one 21:11:07 Social event in Summit? 21:11:16 mlavalle organizing one 21:11:19 yeah 21:11:31 http://lists.openstack.org/pipermail/openstack-dev/2016-October/105757.html 21:11:32 I’ll defer to him for any further details 21:11:34 here's the message I sent to the ML 21:11:55 #link http://lists.openstack.org/pipermail/openstack-dev/2016-October/105757.html 21:11:57 mlavalle: if you struggle with finding a good venue, we can always meet on the beach 21:12:09 we have a good venue 21:12:14 mlavalle: +1 :) 21:12:21 so far 30 people have confirmed 21:12:32 usually about 50-100 show up, iirc. 21:12:47 mlavalle: let’s hope they don’t cancel on us as it happened in Austin? 21:13:19 armax: I'll keep reminding people 21:13:41 mlavalle: ack 21:13:49 ok next topic 21:14:21 #topic blueprints 21:14:52 I still see people registering neutron blueprints in launchpad being completing oblivious to the RFE process 21:16:02 I’d like to remind here that blueprints are created at the very end of the RFE process and if and only if the drivers team gives a green light 21:16:18 is there any way to add a warning when creatin ga blueprint in LP? 21:16:27 unfortunately I can’t see a good one 21:17:32 but the landing page, if someone cared to read it, does provide instructions 21:17:40 anyhoo 21:19:45 we have not scrubbed the dashboard for the first Ocata milestone 21:19:47 #link https://launchpad.net/neutron/+milestone/ocata-1 21:19:59 but we’ll do it soon 21:20:24 if you are involved with any of the activities being tracked on that dashboard 21:20:43 please consider providing an update on the blueprint whiteboard 21:21:34 #topic Bugs and Gate failures 21:22:06 Last week’s deputy was john-davige, nothing scary to report 21:22:18 even our Grafana dashboards 21:22:20 #link http://grafana.openstack.org/dashboard/db/neutron-failure-rate 21:22:29 don’t seem to reflect anything alarming 21:22:37 yay 21:22:49 kevinbenton: you must have been on vacation? 21:22:56 ouch 21:23:01 yeah, i didn't get anything merged :) 21:23:07 kevinbenton: that explains it 21:23:08 :) 21:23:44 I am still licking the wounds post dhcp-turmoil 21:24:41 as far as bugs being reported since the start of Ocata I can already count 150 in the span of a month 21:25:10 we only fixed a third, if any of you has any idea on how to improve that ratio, I’d love to hear it 21:25:27 because even though we managed to keep the size of our backlog relative steady 21:25:39 we can’t seem the squash it 21:26:06 and we hover over the 800 bug count mark 21:26:10 which is pretty annoying 21:27:07 other than revoking kevinbenton 21:27:10 gerrit rights 21:27:16 I can’t think of anything else 21:27:33 and I must add 21:27:40 i think kevin is a net win, so that might hurt. 21:27:43 revoking my +2/+A rights too 21:28:06 One way to keep down the bug count would encourage developers to submit only fixes instead of a bug a nd a fix. Right? or do we prefer how it is now? 21:28:08 dougwig: ok, I would agree with that statement, actually 21:28:25 trevormc: you mean cheating? :) 21:28:33 trevormc: or lying? 21:28:38 the other thing would be to delete a code. 21:28:45 then we wouldn't have a bugs. am i rite? 21:28:51 because that doesn’t change the fact that hte codebse is buggy :) 21:29:03 dasm then there will be failures 21:29:07 I think dasm is onto something! 21:30:31 this probably deserves more discussion, but I think one of the issues I see is that developers hog the bug report for too long 21:30:44 +1 21:30:46 and the bug stays in progress for ever 21:30:59 I've encountered several times where it's already assigned so i won't dig into it 21:31:11 because from the number I can see that the number of new/unassigned bugs is relatively low 21:31:14 is there a way we can auto unassign after a week of no activity? 21:31:31 an auto-unassign after N days of idle is intriguing. 21:31:32 kevinbenton: I do that, even though not that frequently 21:31:59 and the reason I do that less frequently is that you can’t just rely on a script to auto unassign 21:32:35 you need some time in figuring out whether the bug report next status 21:32:37 armax: in that case, maybe would be good to add one more thing for bug deputy? to verify that? 21:32:43 it might be useful to listi such bugs though. 21:32:49 What about those patch sets where it has jenkins +1 just no one has reviewed it? Is it the owner's responsibility to bring it up in meetings or the IRC chat? 21:32:59 trevormc: ++ 21:33:11 dasm: right, but ihar and I were contemplating making the deputy duties doing some proactive backporting 21:33:12 trevormc, +++ 21:33:13 trevormc: agere 21:33:21 agree* i'm glad i can type 21:33:27 we don’t want to swamp the deputy too much 21:33:36 armax: true 21:33:48 ok, let me mull over on this a bit more 21:34:10 we can certainly monitor this area during the team meeting 21:34:11 armax: bug deputy and stable team are two different roles. instead of merging them, perhaps we have a stable nominee for each week as well? 21:34:17 that’s what this topic is for after all 21:34:36 dougwig: you mean estabilish a stable backport deputy? 21:34:44 armax: yes. 21:34:46 dougwig: intriguing idea 21:35:02 dougwig: we can talk about this next week and see where we go with it 21:35:19 armax: "Nov 8th" week 21:35:28 I think it was always ihar’s intention of making this a rotating effort 21:35:42 dasm: we are going to meet secretly at the summit 21:35:46 dasm: hence next week 21:35:54 NOOO!!!1111onoenoe :'( 21:36:01 dasm: and tell no-one about the outcome 21:36:10 and ignore any suggestion 21:36:18 armax: how other projects keeps their numbers low? 21:36:21 comment or feedback 21:36:28 that's where we adopt a complete NFV focus, right? 21:37:11 electrocucaracha: numbers are pretty high overall 21:37:34 electrocucaracha: In another project I work in, I observe developers uploading patch sets without a bug. They reserve bug reporting for users and operators. 21:37:48 electrocucaracha: but I must admit I have not talked to the respective PTL’s 21:37:54 so it’s something I can work on 21:38:39 yeah, maybe it's a common know issue, but it's better to keep a short active list instead of a longer wishlist 21:39:07 again, I am less concerned to have a 1000 bugs in the backlog at any given time 21:39:50 assumed we can close bugs as fast as they are openee 21:39:53 opened 21:39:59 or faster 21:40:19 but first and foremost 21:40:23 for some bugs discussion goes too long on just choosing a solution 21:40:34 not give reporters the chance to file them in the first place would be the way to go 21:40:50 but one can only hope for so muh 21:40:51 much 21:41:13 ok let’s move on for now 21:41:17 but before we do 21:41:26 dasm: was wondering about a deputy post-summit 21:41:34 if anyone is interested, please raise your virtual hand 21:41:41 if not, it’ll fall back to me 21:42:03 but please, don’t be shy/lazy and raise your virtual hand 21:42:11 I would be interested in shadowing, just so I know what to do when I volunteer for it sometime. 21:42:28 trevormc: there’s a very detailed how to guide 21:42:31 * manjeets wanted to do that but it would be his first time 21:42:57 #link http://docs.openstack.org/developer/neutron/policies/bugs.html#neutron-bug-deputy 21:42:58 armax, post-summit means... the first week of november right? 21:43:19 the week after the summit? i can do it. 21:43:35 electrocucaracha: correct 21:43:37 electrocucaracha: yes, until next meeting when someone will be picked 21:44:22 ok between manjeets, trevormc and dougwig who wins? 21:44:31 dougwig: ++ thanks 21:45:00 ok 21:45:03 I vote for dougwig and trevormc before me would be right choice 21:45:13 #action dougwig deputy for the week of Oct 31st 21:45:24 i win. i win all the things. i will be the best bug deputy you have ever seen, you don't even know what you're missing yet . make launchpad great again! 21:45:43 dougwig: you’ll get some tricks and treats in the post 21:45:44 we will have the smallest bug counts 21:45:45 * electrocucaracha lol 21:46:29 ok next one up 21:46:32 #topic Docs 21:46:47 I am not sure we have an official Doc dude or gal 21:46:59 all I know is that our docs still shows emagana 21:47:03 and we should change that 21:47:19 sam-I-Am doesn’t seem to be online either 21:47:44 He is flying today 21:48:22 we have 33 docs bug open 21:48:24 #link https://bugs.launchpad.net/neutron/+bugs?field.tag=doc 21:48:26 we suck 21:48:32 we should have 0 21:48:59 or close to zero 21:49:13 HenryG: where to? 21:49:14 isn't this that, person who implemented the change, should also update the docs? 21:49:32 dasm: ideally yes, and that’s the operating assumption I am under 21:49:47 I like to tag the person as a reviewer at least. 21:49:50 armax: around in circles, probably 21:50:14 Just my two cents: but being able to fix doc bugs is helpful as a newbie because it requires me to read up on Neutron while still being able to look into some of the code 21:50:24 I’ll attempt to clean up the list between now and the time after the summit 21:50:59 we need a new official doc liasion regardless 21:51:33 this table 21:51:35 #link http://docs.openstack.org/developer/neutron/policies/neutron-teams.html#core-review-hierarchy 21:51:39 is a little outdated 21:52:35 #action armax to refresh page http://docs.openstack.org/developer/neutron/policies/neutron-teams.html 21:53:17 next topic 21:53:21 #topic OSC 21:53:31 amotoki anything newsworthy you’d like to share? 21:53:45 I'd like to have a review for https://review.openstack.org/#/c/348097/ 21:53:59 neutron OSC plugins can use this as a utility. 21:54:13 amotoki: consider it done 21:54:34 amotoki: anything else? 21:54:52 For the summit, I am preparing a list of a gap and/or missing points. i'll share next week or some. that's all 21:55:00 amotoki: cool, thanks 21:55:07 #topic Neutron-lib 21:55:13 HenryG/dougwig 21:55:32 Nothing of great interest is happening 21:55:39 HenryG: nothing? 21:55:49 Just mundane things 21:55:52 didn’t you see my amazing patches? 21:56:08 That's api-ref 21:56:11 no 21:56:15 that one 21:56:36 the one that introduced the plugin registry, I’ll be working on bringing the context and policy engine to neutron lib later today 21:57:14 or directory as I have actually called it 21:57:20 OK I didn't see you took it out of WIP 21:57:24 but I am open to suggestion 21:57:26 s 21:57:42 what is the reason for brining the policy engine into neutron-lib 21:57:54 kevinbenton: because the context depends on it 21:58:05 but maybe I can get away with it 21:58:19 I haven’t actually finished everything up 21:58:23 ack 21:58:32 ok 21:58:47 kevinbenton: neutron-policy-oslo-thing-wrapper-lib, of course. 21:58:51 2 minutes left, I am no longer running these meeting as effeciently as I used to 21:58:55 I am getting sloppy 21:59:04 there are a few other interesting patches for review in lib.. such as the callback api updates and hacking checks rollout using incubation 21:59:08 if you get time 21:59:27 boden: aye, I’ll try to be more involved in neutron-lib this cycle 21:59:32 or at least I want you to think they’re interesting 21:59:40 And this one is lib-related: https://review.openstack.org/337731 21:59:55 HenryG: ack 21:59:57 ok 22:00:01 we’re almost at time 22:00:08 boden: are you the author? 22:00:13 thanks everyone for joining 22:00:18 bye 22:00:20 bye 22:00:21 and safe travels if you’re going to the summity 22:00:24 #endmeeting