Wednesday, 2018-05-30

*** kumarmn has quit IRC00:01
*** kumarmn has joined #openstack-tc00:03
*** aprice_ has joined #openstack-tc00:04
*** ildikov_ has joined #openstack-tc00:05
*** csatari_ has joined #openstack-tc00:05
*** csatari has quit IRC00:05
*** ildikov has quit IRC00:05
*** csatari_ is now known as csatari00:05
*** ildikov_ is now known as ildikov00:05
*** aprice has quit IRC00:06
*** mordred has quit IRC00:06
*** aprice_ is now known as aprice00:06
*** dhellmann has quit IRC00:06
*** melwitt has quit IRC00:06
*** melwitt has joined #openstack-tc00:11
*** dhellmann has joined #openstack-tc00:11
*** ChanServ has quit IRC00:12
*** mordred has joined #openstack-tc00:16
*** kumarmn has quit IRC00:18
*** ChanServ has joined #openstack-tc00:33
*** sets mode: +o ChanServ00:33
*** melwitt is now known as Guest4815300:39
*** ricolin has joined #openstack-tc00:40
*** kumarmn has joined #openstack-tc00:48
fungitc-members (and anyone else): since you've had a few hours to fully recover from the summit, i'm sure you now have all manner of things you want to discuss in the 01:00z office hour00:59
fungi#startmeeting tc01:00
openstackMeeting started Wed May 30 01:00:00 2018 UTC and is due to finish in 60 minutes.  The chair is fungi. Information about MeetBot at
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.01:00
*** openstack changes topic to " (Meeting topic: tc)"01:00
openstackThe meeting name has been set to 'tc'01:00
*** harlowja has quit IRC01:00
fungi#topic office hour01:00
*** openstack changes topic to "office hour (Meeting topic: tc)"01:00
clarkbI'd be curioua to hear how the goals session went as I had a conflict01:00
* fungi digs up the pads for that01:01
mnaserAs fungi mentioned, there was a lot of discussion about prior goals and their success01:02
mnaserI think one of the ideas brought up was that there was some operators mentioning that the goals were not oriented to them but rather to the developers / tech / code, if I understood01:02
fungipart of that was my misunderstanding01:03
fungimriedem had left comments in the previous session's pad that the cold upgrades goal had gotten backburnered in favor of the mox goal01:04
mnaser^ yep01:06
mnaseri think that was brought up01:07
fungiaround line 99 i think01:07
*** kumarmn has quit IRC01:07
fungii mistakenly thought it was input from someone representing operator concerns in the discussion, but turns out to have been mreidem (or i think he said that was him at any rate)01:07
mnaseryep, it was him who brought it up (on behalf of operator feedback)01:09
fungianyway, i did my best (in both sessions) to explain that we intentionally picked a rocky goal we thought would please operators (dynamic reconfiguration) and one which reduced our tech debt (mox removal)01:09
fungiso it was dynamic reconfig which was picked over cold upgrades, really01:10
mnaseri think the dynamic reconfig was far more popular because it was far easier to implement01:12
fungibut regardless, that's all water under the bridge now01:12
fungiwe did pitch some potential stein goals01:12
fungii did my best to represent the storyboard migration goal as diablo_rojo needed to be elsewhere for the latter half of the session01:13
fungianyway, the upshot of the session is that the champions for the pitched goals from the backlog would write up specs proposals for review01:16
fungi#info Champions for goals pitched at the Forum session on Stein Release Goals will write up formal goal proposal changes for review01:18
mnaseri guess there isn't much more to that ;x01:23
fungi#chair mnaser01:27
openstackCurrent chairs: fungi mnaser01:27
fungimay as well add other tc members present as chairs when we use meetbot for minutes so we can collaborate on chair-only actions01:28
clarkbthank you for the goals summary01:28
mnaserhopefully thursday we can talk about the whole idea of bringing back an actual agenda for meetings01:29
fungi#idea Add other TC members present as chairs when we use meetbot for minutes so we can collaborate on chair-only actions01:29
*** mordred has quit IRC01:31
fungi#info In the Forum session for a TC Retrospective some consensus was expressed to try using meetbot to record relevant minutes during office hours, hence... this!01:32
*** kumarmn has joined #openstack-tc01:37
*** kumarmn has quit IRC01:42
*** mordred has joined #openstack-tc01:45
*** kumarmn has joined #openstack-tc01:59
*** openstack changes topic to "OpenStack Technical Committee office hours: Tuesdays at 09:00 UTC, Wednesdays at 01:00 UTC, and Thursdays at 15:00 UTC | | channel logs"02:00
fungithanks for a far more productive office hour than i anticipated!02:00
openstackMeeting ended Wed May 30 02:00:05 2018 UTC.  Information about MeetBot at . (v 0.1.4)02:00
openstackMinutes (text):
*** kumarmn has quit IRC02:27
*** dklyle has joined #openstack-tc02:40
*** dklyle has quit IRC02:54
*** rosmaita has quit IRC02:57
*** kumarmn has joined #openstack-tc02:57
*** kumarmn has quit IRC03:02
*** masayuki- is now known as masayukig_03:08
*** masayukig_ is now known as masayukig03:12
*** kumarmn has joined #openstack-tc03:28
*** harlowja has joined #openstack-tc03:29
*** kumarmn has quit IRC03:36
*** kumarmn has joined #openstack-tc04:06
*** kumarmn has quit IRC04:10
*** harlowja has quit IRC04:14
*** kumarmn has joined #openstack-tc04:37
*** kumarmn has quit IRC04:42
*** kumarmn has joined #openstack-tc04:48
*** kumarmn has quit IRC04:52
*** ricolin has quit IRC05:05
*** kumarmn has joined #openstack-tc05:09
*** eandersson has joined #openstack-tc05:10
*** kumarmn has quit IRC05:14
*** kumarmn has joined #openstack-tc05:46
*** eandersson has quit IRC05:48
*** eandersson has joined #openstack-tc05:49
*** kumarmn has quit IRC05:51
*** kumarmn has joined #openstack-tc06:07
*** kumarmn has quit IRC06:12
*** kumarmn has joined #openstack-tc06:22
*** kumarmn has quit IRC06:26
*** kumarmn has joined #openstack-tc07:00
*** kumarmn has quit IRC07:04
*** jaosorior has joined #openstack-tc07:15
*** ricolin has joined #openstack-tc07:15
*** kumarmn has joined #openstack-tc07:31
*** kumarmn has quit IRC07:36
*** jpich has joined #openstack-tc07:53
*** kumarmn has joined #openstack-tc08:06
*** kumarmn has quit IRC08:11
*** kumarmn has joined #openstack-tc09:13
*** kumarmn has quit IRC09:18
*** gcb has joined #openstack-tc09:19
*** dtantsur|afk is now known as dtantsur09:38
*** jaosorior has quit IRC10:14
*** kumarmn has joined #openstack-tc10:28
*** jaosorior has joined #openstack-tc10:31
*** kumarmn has quit IRC10:33
*** kumarmn has joined #openstack-tc11:41
*** kumarmn has quit IRC11:47
*** rosmaita has joined #openstack-tc11:49
*** kumarmn has joined #openstack-tc12:18
*** kumarmn has quit IRC12:23
*** edmondsw has joined #openstack-tc12:50
*** kumarmn has joined #openstack-tc12:51
*** kumarmn has quit IRC12:56
*** mriedem has joined #openstack-tc12:59
*** kumarmn has joined #openstack-tc13:10
mnasermorning everyone o/13:11
dhellmannfungi : I like the office hour logs; thanks!13:48
fungiyou bet13:50
smcginnisNow we just need to remember to do that each time. :)13:51
openstackgerritDoug Hellmann proposed openstack/governance master: add python 2 deprecation timeline
*** EmilienM_PTO is now known as EmilienM14:08
*** Guest62139 has quit IRC14:13
*** flaper87 has joined #openstack-tc14:20
dimsbig-tent-like-problems at CNCF (and k8s) fyi -
dimsclick bait :) "This is important because we are already reaching the point where the sheer number of CNCF projects is becoming off-putting."14:23
dimsthere is some really good thoughts on current status and governance of k8s too :)14:28
*** hongbin has joined #openstack-tc14:33
*** dklyle has joined #openstack-tc14:33
dhellmannthanks, dims14:49
ttxdims: thanks! Looks like they are struggling to represent the various deliverables produced by the Kubernetes community (and whether to represent them as sub-components or full-fledged top-level projects)14:55
ttxand how the CNCF "project" structure is not a perfect fit14:56
dimsright ttx15:06
aspiershi all, before I start politely pestering poor Jimmy about this, please can you confirm I'm barking up the right tree?
*** Guest48153 is now known as melwitt15:25
zanebTheJulia: if summit is in Vancouver again I think we should lobby to hold the PTG on one of those cruise ships that dock alongside15:27
TheJuliazaneb: only if we undock the ship... :)15:29
smcginnisThe SS Croke Park15:29
zanebTheJulia: obviously :)15:30
TheJuliadeva tossed out the idea at one point way back when of seeing if we could do an ironic midcycle up in the mountains where we all fly in via seaplane. Seemed like a great idea collaboration wise.15:37
smcginnisI've seen too many horror movies for that to sound good to me. ;)15:38
dhellmannas far as I can tell those cruise ships are just vectors for norovirus15:47
*** mriedem has left #openstack-tc15:56
*** mriedem has joined #openstack-tc15:56
*** ricolin has quit IRC16:08
openstackgerritBrian Rosmaita proposed openstack/governance master: Add Glance completion artifacts for Pike py35 goal
zanebthere was an outbreak of norovirus (or the other one, not sure) in my house during the week of the PTG in Dublin16:38
fungiaspiers: it's something we've fought for going back years now. the foundation contracted with a web development company to do most of the and summit properties, and just convincing them to make the source code public has been a hard-fought win. i get the impression they want to be more open in their development processes but have been told that they feel web (and presumably mobile16:38
* zaneb was never so glad to be stranded on an island in ~4in of snow16:38
fungidevice application) development and event deadlines don't mesh well with free software ideals so even this much is a stretch for them apparently16:38
zanebfungi: I especially like how all the commit message headlines are just the name of the person who committed the change :D16:40
fungiaspiers: in short, the primary feedback loops to that development team involve marketing analysts and event coordinators16:40
fungizaneb: yes, presumably revision control systems are a bit of a novelty still as well16:41
* zaneb searching for a Denny Crane GIF16:42
aspiersfungi: it's really disappointing to me that we didn't use an open source solution from the beginning and throw money at that instead16:42
fungiaspiers: yep, preaching to the choir i'm afraid16:42
*** pabelanger has quit IRC16:42
*** pabelanger has joined #openstack-tc16:42
aspiersI don't understand why we didn't use something like
aspiersI did point the foundation at that ~3 years ago16:43
fungiit's one of the up-sides to the infra team switching to a different domain name... at least it'll be more clear what infrastructure is being run in an open fashion and what's being done in a more proprietary/corporate manner (though a few of our whitelabel systems will still likely have vhosts when they're for openstack-specific content)16:44
zanebI assume they're using a bridge to git from something hideous (svn? vss?), and not just completely new to version control16:44
fungii don't know for sure, but i gather they're using github as their primary vcs now. previously i'm not sure if they used local svn or something directly on the servers though much more of the content used to just be stored in some cms databases16:46
aspiersgit-svn does a better job of commit messages than that, IIRC16:46
fungiheh, yeah i think you're right about that ;)16:46
aspiersfungi: so when the mobile app has the same problems for several summits in a row, do we have any likely mechanism for getting them fixed?16:47
aspiersI feel inclined to submit issues to that github repo, if there is no better place to track them16:47
aspierseven if they are ignored by the developers, at least the rest of us can collaborate on them16:48
aspiersdoes that make sense?16:48
fungithey've been taking feedback in the feedback sessions and prioritizing fixes/wishlist additions in with whatever new demands they get from marketing and event coordination16:48
fungibut the biggest hurdle with the mobile app dev i think is getting releases certified by the phone os vendors16:49
fungiapple especially16:49
aspiers10 mins of feedback twice a year is not ideal :-/16:49
dhellmannit sounds like there's a bit of a split between the primary customers and primary users for the app16:49
zanebaspiers: I think what it comes down to is that dealing with issues opened by users and actual bugs require two different systems16:51
fungii do get the impression that they're doing some herculean work given it's software used for events given twice a year by a single organization (at least in terms of the summit schedule app), so not necessarily the same dynamic as developing openstack16:51
zanebit's actually correct to collect user feedback in a private system like zendesk, because it may contain private information about the user16:52
zanebbut that needs to then be scrubbed and put into a public bug tracker, which is the part that's not happening16:52
fungii used to be a lot more hard-nosed about the situation, but have come to realize that the most i can hope to do is be available to provide guidance if they ask for it and try to set good examples myself16:52
aspierszaneb: that should be up to the user to decide. 99% of feedback I give as a user is not confidential and I would *far* prefer that 99% to be done in the open16:52
aspierszaneb: using zendesk for that reason is optimising for the 1% case16:53
fungii and others do still regularly put a bug in the ear of the people setting priorities and paying the contracts on that stuff, but it's not something i really have control over16:54
* aspiers considers talking to the board members he knows16:55
fungithough they have picked up some ideas from how the community infrastructure team operates, and one of their projects (openstackid) is not only developed through gerrit (admittedly with the primary contributor self-approving most of their own patches) but also continuously-deployed to infra-managed servers16:55
zanebaspiers: as a dev, I am with you. as a 'contact' link in a mobile app used by thousands of people who may not have the same background... better to have one place and the developers creating public bugs when appropriate16:55
zaneb(I have opened several tickets about the app, and most of them were about authentication stuff and my foundation account)16:56
fungiyeah, i get the impression that the feedback they receive on the app is (by descending quantity, separated by an order of magnitude) 1. help requests involving account issues, 2. wishlist new feature requests, 3. reports of actual bugs16:57
zanebthat doesn't surprise me16:58
funginor i, but does lend some understanding as to whether they might prioritize feedback16:59
fungier, _where_ they might prioritize handling feedback16:59
*** dtantsur is now known as dtantsur|afk17:00
fungias for it's unfortunate that "The apps will be moving to Gerrit shortly after the Barcelona Summit" never happened17:00
fungino doubt priorities shifted17:00
openstackgerritMerged openstack/governance master: Clean up remaining docs links
openstackgerritMerged openstack/governance master: Import ansible-role-tripleo-modify-image
openstackgerritMerged openstack/governance master: Note that the old incubation/graduation process is obsolete
openstackgerritMerged openstack/governance master: Include a rationale for tracking base services
*** jpich has quit IRC17:31
*** mriedem1 has joined #openstack-tc18:21
*** mriedem has quit IRC18:23
*** mriedem1 is now known as mriedem18:24
*** diablo_rojo has joined #openstack-tc18:48
dhellmanntc-members: please review the current contents of the tracker when you have time
zaneb3rd & 4th item are the same thing18:50
dhellmannah, thanks18:51
dhellmannzaneb : do you want to fix that?18:51
zanebcan do18:51
dhellmannI realized there were 2 other items that were the same and merged those, too18:51
dhellmannabout python 2/318:51
EmilienMdhellmann: excellent update, I haven't seen missing items18:52
dhellmann23 different things -- we're doing more than I think a lot of you realized18:52
dhellmanncertainly more than I realized :-)18:53
zanebdhellmann: re 'Trove Project Health'... that may have been me who mentioned in the Forum session that it was unmaintained. and by 'unmaintained' I of course would have meant that it is in maintenance mode with no new development18:57
dhellmannhmm, I thought there was a question from an operator, but maybe it was you18:57
dhellmannmy notes on that were thin, and the etherpad doesn't mention it at all18:57
zanebit may also have come from someone else18:57
dhellmannI figured we could at least look at the recent review activity18:58
zanebthere's probably no reason to think that it's abandoned18:58
zanebbut the fact that we have public cloud operators wanting to use it but nobody to continue to develop it suggests a failure in the marketplace of supply and demand18:59
dhellmannprobably not. part of doing this is the exercise of following up on those sorts of questions in a more visible way so that (a) people realize we do that at all and (b) we get into the habit of doing it18:59
dhellmannyes, that's a good point18:59
zanebironically I tried to get the Trove and Zaqar developers together in Vancouver in 2015 to talk about that exact feature, and the Trove people made clear in no uncertain terms that they didn't want it19:01
dhellmannthe feature of publishing events to users?19:09
zanebdhellmann: the feature of not using a shared rabbitmq that the users' DB servers have access to for messaging between the servers and the control plane20:09
dhellmannthat's related but different20:10
zaneb <- the one that is near the top of the public cloud WG's todo list20:11
openstackLaunchpad bug 1752103 in OpenStack Public Cloud WG "Multi tenant message bus for Trove" [High,New]20:11
openstackgerritMerged openstack/governance master: Replace deprecated library function os.popen() with subprocess
mnasersent the starlingx email and updated the tracker repo20:33
dhellmannmnaser : thanks!20:34
EmilienMmnaser: thanks, I was about to sync with you (back from pto)20:35
mnaserEmilienM: no problem, i guess at this stage we're pending some of the normal infra review here
mnaserthat patch imports the non-openstack non-forked stuff, but i had a few things like empty repos and some "this is confidental" contents that should be stripped off20:36
EmilienMmnaser: your email is excellent, nice plan20:37
mnaseri had some help from other tc members :)20:37
dhellmannI'm starting to wonder if that wiki page is the best way to track these things, given the number20:37
EmilienMI thought we used storybard20:39
dhellmannwe do have storyboard, too, yes20:39
EmilienMmaybe we could convert each point into a story20:40
dhellmannhmm, storyboard is currently showing me a blank page20:40
EmilienMit works for me20:41
EmilienMmnaser: I'm going to add a story in SB20:41
dhellmannsafari caching issue again20:41
mnaserEmilienM: i'm okay with that, it would probably be nice that we can post comments that are tracked by time20:42
dhellmannI guess the issue with storyboard is it is hard to see who the drivers are if there are more than 120:42
dhellmannmaybe we just put that in the description20:42
mnaserit's not the prettiest20:42
dhellmanntags are global, so if you want your name to be a tag we could do that, but it seems not ideal20:42
mnaseroh, i thought tags were per project20:43
* mnaser has not used storyboard that muc20:43
dhellmannI don't think so? I certainly see tags that I didn't create20:43
mnaserthanks EmilienM20:45
EmilienMthanks for kicking this off :D20:46
mnaser:) anyways, i gotta get moving, bbl20:50
dtroyermnaser: thanks for the review on 569562… I am re-syncing things that changed while I was ignoring the world over the weekend.21:14
dhellmanndtroyer : now that we have that storyboard story (^^), maybe you want to mention that in your commit message to have zuul update storyboard automatically?21:25
dhellmanntc-members: please check out this ML thread from corvus about a new organization for infra and its relationship to openstack and other OSF projects, and of course please provide any input you have:
dimsdhellmann : y i read that thread, is there anything specific the TC has to vote as a resolution for that?21:44
fungilikely not at this stage21:44
fungithere will ultimately be a proposed change to remove a bunch of infra repos from governance21:44
dhellmannnot voting, but the more folks we have thinking about this the better our solution will be21:45
dimsack dhellmann fungi21:45
dhellmannconsider it a precursor to other similar sets of relationships with new OSF projects21:45
fungithough the proposed governance body of the new infra is likely to be a bit unique across osf projects in that it exists mostly to serve the rest so seeks to get more direct involvement/input from them21:48
*** harlowja has joined #openstack-tc21:52
*** kumarmn has quit IRC22:14
dtroyerdhellmann: will do, thanks.   That's the Story: header, right?22:16
*** kumarmn has joined #openstack-tc22:32
*** kumarmn has quit IRC22:35
*** kumarmn has joined #openstack-tc22:36
*** hongbin has quit IRC22:38
*** mriedem is now known as mriedem_afk22:48
*** edmondsw has quit IRC22:52
*** edmondsw has joined #openstack-tc22:53
*** edmondsw has quit IRC22:57
dhellmanndtroyer : yes22:58
*** kumarmn has quit IRC23:10
*** kumarmn has joined #openstack-tc23:35
*** kumarmn has quit IRC23:57

Generated by 2.15.3 by Marius Gedminas - find it at!