20:00:45 <david-lyle> #startmeeting Horizon
20:00:46 <openstack> Meeting started Wed Jan 27 20:00:45 2016 UTC and is due to finish in 60 minutes.  The chair is david-lyle. Information about MeetBot at http://wiki.debian.org/MeetBot.
20:00:47 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
20:00:49 <openstack> The meeting name has been set to 'horizon'
20:00:59 <robcresswell> o/
20:01:07 <matt-borland> hola
20:01:08 <hurgleburgler> ( ಠ‿ಠ)┘
20:01:17 <ying_zuo> :)
20:02:00 <piet> o/
20:02:02 <bpokorny_> o/
20:02:35 <david-lyle> Let's roll
20:02:51 <david-lyle> We tagged M-2 last week
20:03:03 <david-lyle> so we're officially in the race to the end
20:03:22 <david-lyle> target date for M-3 March 3rd
20:03:26 <hurgleburgler> \o/
20:04:00 <tsufiev> o/
20:05:14 <david-lyle> #link https://etherpad.openstack.org/p/mitaka-horizon-priorities
20:05:24 <david-lyle> is essentially the same as last week
20:06:09 <david-lyle> so I won't belabor it
20:06:37 <david-lyle> I don't have any other general items. There were a couple on the agenda
20:06:48 <r1chardj0n3s_afk> o/
20:06:51 <david-lyle> #link https://wiki.openstack.org/wiki/Meetings/Horizon#Agenda_for_2016-01-27_2000_UTC
20:07:07 <david-lyle> #topic Categorizing incoming bugs
20:07:32 <tsufiev> I thought now there should be much fewer opened themization patches :)
20:07:49 <hurgleburgler> :-O
20:08:04 <david-lyle> lhcheng added this item, but I think I can speak to it
20:08:26 <hurgleburgler> I can help with that
20:08:33 <david-lyle> the volume of bugs that comes in to Horizon via launchpad is quite large
20:09:10 <david-lyle> it would be worlds better to have someone actively watching them as they come in to triage/classify them
20:09:32 <david-lyle> otherwise we get into our current state with 400 uncategorized bugs
20:09:59 <david-lyle> when we were smaller, I could watch that too, but I cannot handle that any more
20:10:12 <david-lyle> hurgleburgler: did you volunteer?
20:10:16 <robcresswell> I think hurgleburgler volunteered
20:10:19 <hurgleburgler> I can help out
20:10:20 <robcresswell> lol
20:10:32 <david-lyle> I should have not said anything else
20:10:38 <robcresswell> haha
20:10:51 <hurgleburgler> I can help categorize, but I can't say I'd be the only one ;)
20:11:13 <david-lyle> it's a gloriously rewarding pleasure cruise
20:11:33 <matt-borland> I can help hurgleburgler
20:11:35 <hurgleburgler> david-lyle I'm not sure I believe you
20:11:43 <tqtran> so as we move away from a central horizon to plugin, we can encourage people to file bug their own respective repo instead
20:11:45 <robcresswell> I'm working on reducing the total too. We've knocked about 1/3 off the uncategorised since first bug day, and about 120 from total iirc.
20:12:03 <robcresswell> tqtran: We've already removed all the trove and sahara bugs.
20:12:06 <hurgleburgler> \o/
20:12:09 <robcresswell> These are mostly outs.
20:12:11 <robcresswell> ours*
20:12:37 <tqtran> ew... gl matt-borland and hurgleburgler!
20:12:41 <tsufiev> many bugs I�ve seen there are actually feature requests
20:12:48 <david-lyle> thank you matt-borland and hurgleburgler
20:12:53 <robcresswell> \o/
20:13:03 <david-lyle> tsufiev: we do have the wishlist category
20:13:19 <robcresswell> tsufiev: If they are small, wishlist, if big, mark invalid and ask for blueprint.
20:13:27 <david-lyle> I give up on trying to figure out how OpenStack thinks bugs and features should be tracked
20:13:36 <tsufiev> david-lyle, robcresswell: got it
20:14:00 <david-lyle> matt-borland, hurgleburgler I will add you to the horizon-bug team so you can do the things
20:14:22 <hurgleburgler> ┏(＾0＾)┛┗(＾0＾) ┓
20:14:27 * matt-borland updating todo list
20:14:27 * tsufiev imagines whole Horizon community walks in one of intra task-tracking sessions and start demanding new task tracker
20:14:33 <r1chardj0n3s> "congrats" hurgleburgler an matt-borland :-)
20:15:01 <tsufiev> I join the congratulations :)
20:15:52 <tyr_> Would "bug review" fit as a natural, regular part of each release? Say...the first 2 weeks after a Summit?
20:16:06 <david-lyle> ok added
20:16:25 <tsufiev> tyr_, unfortunately bugs are being added all the time :(
20:16:31 <david-lyle> tyr_: I would prefer an ongoing process
20:16:38 <robcresswell> tyr_: No, its too constant to do it that way
20:16:51 <robcresswell> We'd be ignoring 5.5 months of bugs every time
20:16:54 <david-lyle> my biggest concern is missing High/Critical defects
20:16:59 <tsufiev> tyr_, besides that we had already 2 Bug Days
20:17:02 <robcresswell> Plus RC1 is already supposed to be bug cleanup time.
20:17:05 <david-lyle> that could have been fixed prior to release
20:17:31 <tyr_> I understand that, and am not suggesting there isn't regular maintenance...just curious about having a bug party phase of the release. Perhaps it is already RC1.
20:18:28 <david-lyle> tyr_: certainly not a bad endeavor for the weeks between release and summit
20:18:28 <robcresswell> I have enough difficulty recruiting people for a single bug day :)
20:18:30 <tyr_> e.g. please don't pester the cores about our pet features during this window because we are collectively focused on bugs. Just an idea.
20:19:04 <david-lyle> but that's mostly the RC cycle for cores
20:19:07 <david-lyle> or at least should be
20:20:10 <david-lyle> I'm for anything that raises the visibility of bugs and their solutions :)
20:20:26 <david-lyle> even to the detriment of landing features
20:20:45 <david-lyle> but I'm going to move on
20:20:45 <r1chardj0n3s> (aka "new bugs")
20:21:14 <david-lyle> #topic QA and Cross project liaison, need a volunteer
20:21:28 <david-lyle> #link https://wiki.openstack.org/wiki/CrossProjectLiaisons
20:21:46 <david-lyle> There are a ton of cross project liaison roles for each project in OpenStack
20:22:00 <david-lyle> these two are currently open
20:22:02 <tsufiev> what are QA liason responsibilities?
20:22:19 <david-lyle> tsufiev: quite honestly, for Horizon not much
20:22:33 <tsufiev> oh, Tempest/Grenade, I�m staying away from that
20:23:01 <david-lyle> tsufiev: well think about our tempest footprint, 1 test
20:23:19 <david-lyle> additionally, it's on my list to make that a tempest plugin
20:23:28 <tsufiev> but if that is related to integration testing of Horizon against the rest of Openstack, I may sign in
20:24:08 <tyr_> I'm interested in the Cross Project Spec Liaison.
20:25:07 <tsufiev> david-lyle, is there any place besides https://wiki.openstack.org/wiki/CrossProjectLiaisons#QA with some info about this role?
20:25:27 <david-lyle> tsufiev: not that I'm aware of
20:25:41 <tqtran> i will sign up for zaqar and trove
20:26:20 <tsufiev> ok, I�ll other QA liasons
20:26:35 <david-lyle> and since all liaison roles default to me if no one occupies it, I can tell you that I don't recall anything that was required, but I do write tempest and devstack patches for horizon
20:26:43 <david-lyle> as required
20:27:05 <hurgleburgler> david-lyle I need one of those devstack patches, btw
20:27:07 <tqtran> we have our own liason list too right?
20:27:16 <david-lyle> tyr_ you'll mainly just need to review the cross project specs
20:27:28 <david-lyle> tqtran: not that I'm aware of
20:27:36 <david-lyle> maybe in an etherpad somewhere
20:27:41 <robcresswell> There's an interproject liaison list too if anyone is working with specific projects.
20:27:53 <tqtran> i could have sworn i saw a liason list for horizon somewhere
20:28:15 <david-lyle> those who volunteered please enter your name
20:28:19 <robcresswell> tyr_: Yeah, just need to pop up in the weekly meetings and let us know if there are any cross-project plans we should be aware of.
20:28:21 <tqtran> as in, horizon people checking out other projects to make sure we are in sync with their api
20:28:43 <tsufiev> david-lyle, ha, hope they have some courses for new QA liaisons on how to write devstack patches :)))
20:29:04 <david-lyle> tyr_:  you will find that since our architecture is so different from the other services that little applies to us directly
20:29:24 <david-lyle> tsufiev: they are mostly horizon code change driven
20:29:33 <david-lyle> ping me
20:29:55 <tyr_> Ok. I'm assuming things like WebSSO where Horizon has a role to play in the overall experience.
20:30:33 <tyr_> I'll add my name.
20:30:36 <david-lyle> tyr_: nothing that topical
20:30:39 <david-lyle> :P
20:31:03 <tyr_> ha! Ok, maybe a background task, but seems like good breadth opportunity.
20:31:06 <david-lyle> tyr_: https://review.openstack.org/#/q/project:+openstack/openstack-specs+status:+open,n,z
20:31:26 <david-lyle> I'll keep watching too, because I can't help myself
20:31:36 <robcresswell> Are there any meetings relevant to the cross-project specs?
20:32:08 <david-lyle> robcresswell: I'm not entirely sure any more
20:32:40 <tyr_> wiki claims one on Feb 2nd
20:32:42 <david-lyle> with the shift roll of the cross-project meeting, I'm uncertain as to it's intent anymore
20:32:59 <david-lyle> at one point it was to discuss cross-project specs
20:33:39 <david-lyle> additionally there is #openstack-meeting-cp but it doesn't seem to be used
20:34:26 <david-lyle> here's my cynical take, we're replacing all the parts of the car while in motion and I'm not certain what pieces are back, what have changed or why
20:34:52 <david-lyle> anyway
20:35:06 <david-lyle> thanks to the volunteers
20:35:15 <david-lyle> #topic Open Discussion
20:36:03 <robcresswell> Just a reminder for mid-cycle attendees: Fill in your name etc, and put topic ideas on the etherpad
20:36:08 <robcresswell> #link https://wiki.openstack.org/wiki/Sprints/HorizonMitakaSprint
20:36:11 <hurgleburgler> Would it be possible to get some eyes and testing help with : https://review.openstack.org/#/c/252161/
20:36:31 <david-lyle> by filling in your name and email, I can get your internet access set up at Intel ahead of time
20:37:38 <robcresswell> btw, why is everyones email AT and DOT instead of @ . ? Did I miss a memo?
20:37:55 <david-lyle> robcresswell: not easily scraped
20:38:05 <tsufiev> robcresswell, spammers don�t sleep?
20:38:18 <robcresswell> Ah, makes sense.
20:38:29 <david-lyle> just like tsufiev
20:38:51 <david-lyle> doesn't sleep
20:39:15 <tsufiev> I�m not a spammer :D
20:39:30 <robcresswell> idk, interview doodle says otherwise tsufiev...
20:39:34 <robcresswell> :D
20:39:39 <tqtran> tsufiev: you spam the horizon reviews, stop it!
20:39:43 <hurgleburgler> LOL
20:39:47 <david-lyle> no don't !
20:39:52 <robcresswell> haha
20:40:19 <tsufiev> circling back to serious business :)...
20:40:22 <tsufiev> do most of people here look through mitaka-priorities etherpad? Would it make sense if I wrote down the list of new integration tests ready for review (on a regular basis)?
20:40:57 <david-lyle> or provide the link to the gerrit topic at least
20:40:57 <hurgleburgler> tsufiev are you adding them to the weekly bug report?
20:40:57 <robcresswell> tsufiev: I have a potentially better (and selfish) idea. Use the weekly bug report, put a couple on each week.
20:41:21 <david-lyle> that wasn't on the agenda!
20:41:21 <tqtran> ok, since we have some time, i also like to bring up the plugin doc again. its high on priority atm but not getting enough love https://review.openstack.org/#/c/233709/ can we agree to push it through and make amends, the information is useful enough
20:41:24 <robcresswell> I've gotten slow on updating it, because its quite time consuming to do it on my own each week
20:41:32 <robcresswell> david-lyle: Yeah, its 2-3 weeks out of date iirc.
20:41:41 <david-lyle> ah ok
20:41:42 <tsufiev> robcresswell, hurgleburgler: okay, works either for me
20:42:14 <robcresswell> I'd like it if people could sensibly drop a patch or two in, and I'll moderate it anyway. I'm going through the bug list in my spare time, and when thats properly in order we can drop the weekly report I hope.
20:42:24 <tqtran> there are a few follow on docs that i would really like to get started on, but that cant happen until this one goes through
20:42:46 <tsufiev> robcresswell, I�ll just put there new integration tests and bump the date every week. If anyone would like add their bugs/bps, they�re welcome
20:42:58 <robcresswell> tsufiev: That would be great, thanks!
20:43:01 <r1chardj0n3s> I think it'd be good if hurgleburgler and matt-borland could punt the hard-to-reproduce bugs (ie. require edge cases or obscure openstack components to reproduce) to the group
20:43:20 <hurgleburgler> r1chardj0n3s +100
20:43:24 <tsufiev> robcresswell, to me this is only a question of moderating - to ensure that every bug/bp mentioned somehow fits our priorities for current release cycle
20:43:35 <matt-borland> yes, that would be good
20:43:48 <r1chardj0n3s> should use the weekly bug report for that
20:43:53 <robcresswell> Yeah, I don't expect them to be able to reproduce every bug, r1chardj0n3s, its just useful to have someone to keep the moving target-ness down
20:43:54 <matt-borland> ah, ok
20:44:19 <r1chardj0n3s> robcresswell: yeah, but we need to flag the hard-to-reproduce ones so they don't just sit and pile up :-)
20:45:04 <robcresswell> May be worth going over the bug tags so that people with specific expertise can check. Like scale requirements, or specific component setups.
20:45:04 <tqtran> its no fun when you have to restack to test something though. i usually lose a day trying to reproduce and get no where.
20:45:24 <tsufiev> in an ideal world the scenario for bug reproduction would be a devstack config/script...
20:45:37 <r1chardj0n3s> yep, and I've lost a day to trying to set up some part of openstack that I've never used before
20:45:50 <r1chardj0n3s> tsufiev: heh, and a pony!
20:46:11 <tsufiev> r1chardj0n3s, if you find a pony who could write devstack scipts :)...
20:46:13 <david-lyle> as I stated above "it's a gloriously rewarding pleasure cruise"
20:46:31 * tsufiev looking for such pony for at least 1 year
20:46:42 <robcresswell> If we got stackalytics to track karma, our bugs would be gone in a day.
20:46:45 <hurgleburgler> maybe if we get enough people on the list, we can make it a rotating duty
20:46:47 <hurgleburgler> weekly
20:46:53 * ducttape_ wonders about noro-virus on dave's cruise
20:46:54 <david-lyle> if the bug is vague and provides no clues for its reproduction, mark it incomplete
20:46:57 <hurgleburgler> if we have 4 people, then we can take a week a month
20:47:16 <david-lyle> ducttape_: it's a mix of cruise ship/oregon trail
20:47:35 <robcresswell> hurgleburgler: Can you and Matt alternate?
20:47:54 <robcresswell> Might make it easier to maintain, keeps it less time consuming.
20:48:03 <hurgleburgler> robcresswell thought you were going to help too?
20:48:15 <robcresswell> I'm trying to work through the backlog.
20:48:27 <david-lyle> robcresswell: is keen on the past
20:48:33 <robcresswell> haha
20:48:45 <robcresswell> Hey I got about 50 categorised so far this week.
20:48:51 <robcresswell> Its time consuming :)
20:48:54 <hurgleburgler> can we work some backlogging into the current persons rotation?
20:48:56 <david-lyle> \o/
20:49:26 <robcresswell> hurgleburgler: No need to overload, I think. Just keep on top of the new reports. If you have spare time, then by all means have at it.
20:50:21 <robcresswell> tqtran: I'll look through the docs again. The last comment I left (patch set.. 15?) was a complete blocker to betherly the other day.
20:50:42 <robcresswell> We need to double check its complete otherwise it doesnt save anyone time.
20:51:12 <david-lyle> there is a point of contention around pbr in there
20:51:17 <david-lyle> I will comment in the review
20:51:27 <david-lyle> but mandating pbr is not great
20:51:31 <robcresswell> Ah, okay, maybe my misunderstanding then.
20:52:03 <david-lyle> ok for openstack/plugin-x, less great for company-y/plugin-x
20:52:20 <david-lyle> probably both should be documented
20:52:25 <david-lyle> eventually
20:53:06 <robcresswell> I'm happy with getting a minimum viable doc situation, but it has to definitely work the whole way through for a new plugin.
20:53:22 <robcresswell> Otherwise I lose half a day debugging plugins that I'm unfamiliar with
20:53:34 <robcresswell> Not ideal :)
20:54:09 <robcresswell> As I said, I'll check it again, it looks mostly all done. I can carry it too if you're short on time tqtran.
20:54:12 <tqtran> its basically the same steps that i took for zaqar ui, which is last week
20:54:19 <robcresswell> or just fed up of my reviews :p
20:54:42 <tqtran> and its proven to work, i might have missed a few things here or there, but the bulk of the info is valid and should work
20:55:07 <tqtran> i'll also verify again end of today
20:55:21 <robcresswell> Thanks
20:55:26 <david-lyle> let's get something in
20:55:31 <david-lyle> and improve
20:55:43 <david-lyle> I don't think something wildly inaccurate is acceptable
20:55:53 <david-lyle> but reasonably accurate
20:55:55 <robcresswell> Sure
20:56:18 <david-lyle> if we release without it :(
20:56:50 <robcresswell> It needs to be part of the release. I'm sure we can merge it within the next couple days.
20:57:20 <tqtran> great, i can start documenting plugin localization and workflow extensions once it lands
20:57:26 <david-lyle> 3 minutes
20:57:29 <robcresswell> Sounds good
20:58:41 <david-lyle> anyone else?
20:59:30 <david-lyle> Thanks everyone!
20:59:36 <robcresswell> \o
20:59:36 <david-lyle> #endmeeting