19:01:16 <SotK> #startmeeting storyboard
19:01:17 <openstack> Meeting started Wed Mar 20 19:01:16 2019 UTC and is due to finish in 60 minutes.  The chair is SotK. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:01:18 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
19:01:18 <_pewp_> [ MeetBot. - Debian Wiki ] - wiki.debian.org
19:01:21 <openstack> The meeting name has been set to 'storyboard'
19:01:34 <diablo_rojo> o/
19:01:36 <SotK> #link https://wiki.openstack.org/wiki/Meetings/StoryBoard#Agenda_for_next_meeting Agenda
19:01:36 <_pewp_> [ Meetings/StoryBoard - OpenStack ] - wiki.openstack.org
19:02:18 <SotK> #topic Announcements
19:02:33 <SotK> I don't think there are any things to announce this week
19:02:56 <diablo_rojo> None that I know of
19:03:10 <SotK> #topic Migration Updates
19:03:23 <SotK> going off the agenda, looks like the same story here
19:03:28 <diablo_rojo> YEah
19:03:39 <diablo_rojo> and no patches from the NDSU students that Ive seen
19:04:32 <SotK> #topic Outreachy
19:05:05 <SotK> there are many patches that we should review, since I think the submission deadline for applications is soon
19:06:01 <mkarray> Just out of curiousity, were all the outreachy contributions lowhanging stories?
19:06:16 <diablo_rojo> mkarray, I think basically all of them were
19:06:20 <diablo_rojo> They cleaned out out
19:06:28 <diablo_rojo> I had to do a saturday triage to tag some more
19:06:36 <diablo_rojo> But I would guess those are now claimed too
19:06:57 <diablo_rojo> I also think we might have some that are assigned but wont see patches for so we can wait a bit before un assigning them.
19:07:28 <SotK> I've already given a +2 to a number of the patches, but still have a heap to review
19:08:29 <diablo_rojo> I will do my due diligence today
19:08:41 <diablo_rojo> fungi, if you have some time to help us out it would be greatly appreciated
19:08:48 <SotK> thanks :)
19:09:16 <SotK> #topic Forum + PTG Planning
19:09:45 <SotK> #link https://www.openstack.org/summit/denver-2019/summit-schedule/events/23658/ibuprofen-for-your-storyboard-pain-points Forum Session
19:09:45 <_pewp_> [ Open Infrastructure Summit | Ibuprofen for Your StoryBoard Pain Points ] - www.openstack.org
19:09:46 <fungi> oh, yes indeed. reviewing
19:10:17 <SotK> I should probably add that to my schedule
19:10:30 <fungi> does anybody claim this _pewp_ bot?
19:10:37 <fungi> if not, i'll kick it
19:10:54 <diablo_rojo> fungi, a long time ago it was for doing table flips in the cinder meetings
19:11:08 <diablo_rojo> but it seems to have evolved into something more obnoxious
19:11:19 <diablo_rojo> hemna_, ? ^^
19:11:40 <diablo_rojo> But yes SotK please add it to your schedule
19:11:46 <fungi> yeah, if nobody's around to shut it up, i'll take care of it in a few
19:12:20 <smcginnis> hemna_: ^^
19:12:57 <SotK> does the Thursday bug triage idea still sound good to folks?
19:13:26 <diablo_rojo> Thursday works for me
19:13:31 <diablo_rojo> SotK, you will still be around right?
19:13:52 <SotK> yeah, I fly back on Saturday morning so just miss the last day
19:14:38 <fungi> sure, thursday at ptg sounds like a good time for that. i'll probably be splitting my time with other infra/opendev stuff but can pitch in
19:16:18 <diablo_rojo> Sweet.
19:16:45 <diablo_rojo> As soon as the ptgbot is setup we can make a storyboard track and say hiding in the infra room or whatever
19:17:24 <fungi> off-topic for this meeting, but lmk if you want to take a stab at pushing up the denver config for ptgbot and i'm happy to review
19:18:25 <SotK> sounds good to me
19:18:38 <SotK> #topic In Progress Work
19:18:57 <SotK> mkarray: I believe you wanted to discuss some query structure stuff
19:19:02 <mkarray> yup
19:19:57 <mkarray> So for some context, which items show up based on a query is defined by these objects which are usually scattered across the code https://github.com/openstack-infra/storyboard-webclient/blob/master/src/app/services/resource/story.js#L58-#L63
19:19:58 <_pewp_> [ storyboard-webclient/story.js at master · openstack-infra/storyboard-webclient · GitHub ] - github.com
19:20:02 <diablo_rojo> fungi, once we have the room names I think ttx or myself can handle that. Dont have all the details yet.
19:21:23 <mkarray> Originally I was going to remove the limitation of one value per key, but after looking into it, it seems like the only key which has multiple values is User
19:21:52 <mkarray> these consist of assignee_id, creator_id, subscriber_id, and user_id
19:23:23 <mkarray> it seemed sort of silly to remove the limitation just to fix it for that single key, so instead I think we should just make the values associated with the 'User' key, into their own keys. Like so: http://paste.openstack.org/show/748137/
19:24:17 <mkarray> Now if we want to add or remove what can be searched on a storyboard component, we either just add or remove a key/value pair
19:24:27 <fungi> seems fine to me. i guess we need to come up with new icons? that'll be the "hard part" ;)
19:25:13 <fungi> i honestly don't know why "user" got the amalgamation treatment there to begin with
19:25:19 <mkarray> I thought this was worth bringing up in the meeting because 1. What fungi said ^ and 2. It will require a file (I'll take care of this) for each key
19:25:48 <fungi> yeah, that was the biggest unknown for me... is anybody aware of why that choice was made in the first place?
19:25:56 <mkarray> And yes fungi, I haven't seemed to find a purpose for merging them into one type either
19:25:58 <fungi> it's entirely possible i'm just not aware
19:26:13 <fungi> and that there was a good reason
19:26:39 <mkarray> I bugged Sotk about it back when I first hopped on the project, but he wasn't sure either at the time
19:27:12 <fungi> i think if none of the present maintainers are aware of what problem that was solving, it's probably enough reason to assume it was just happenstance/convenience
19:28:09 <fungi> i suppose there might be some benefit to preserving the old "user" key semantics in the webui so that existing query urls continue to work the same, and just no longer expose that in autocompletion?
19:28:33 <SotK> I suspect it was just convenience
19:29:25 <SotK> the User there is the User $resource which maps to the /v1/users endpoint, so it would probably have to continue to exist to avoid lots of churn elsewhere
19:29:31 <fungi> so effectively add the new assignee, creator and subscriber keys and have them do normal autocomplete; leave user as a viable url parameter which maps to assignee+creator but no longer gets included in autocompletion
19:29:51 <mkarray> What Sotk said. It would be useful just to avoid refactoring everything else in the process
19:30:01 <fungi> sure
19:30:14 <SotK> it gets mapped to assignee_id in urls at the moment anyway iirc
19:30:19 <mkarray> yes^
19:30:32 <mkarray> the links wont be affected,
19:30:33 <SotK> its just needed for places we make general user-related queries in the code
19:30:48 <fungi> cool
19:31:37 <fungi> sounds like a solid plan to me in that case
19:31:40 <SotK> but yeah that paste looks nice to me, +1 to that approach
19:31:40 <diablo_rojo> I think that all sounds like a good plan
19:32:04 <fungi> mkarray: thanks for working on this! i think it'll be useful to a lot of folks
19:32:15 <mkarray> great, ill crack away at a solution and commit it when I have a reasonable one
19:32:29 <mkarray> My pleasure :)
19:32:33 <diablo_rojo> Sounds good.
19:32:35 <diablo_rojo> Thanks!
19:33:00 <SotK> yep, its much appreciated, thank you
19:33:13 <mkarray> That's all for me on this subject
19:33:43 <SotK> in other in progress work, I have many patches that need review :)
19:33:51 <diablo_rojo> And for once we did you topic during the meeting with adequate time mkarray :)
19:34:16 <mkarray> progress  is progress! diablo_rojo
19:34:18 <diablo_rojo> I'm so on it.
19:35:01 <SotK> a good portion of those are attachments-related, and the security team stuff is dependent on at least part of the attachment stuff at the moment, because that was easier than repeatedly up/downgrading my database
19:35:04 <fungi> and trying to nail down the apparent performance regressions on the storyboard.openstack.org deployment
19:35:46 <SotK> I'd especially appreciate review on https://review.openstack.org/#/q/status:open+project:openstack-infra/storyboard+topic:test-improvements
19:36:05 <fungi> which i thought might have been cache memory pressure, but even after temporarily releiving that with a restart of the offending memory hogs SotK reports similar poor performance on loading stories
19:36:48 <diablo_rojo> I'll start with those then.
19:37:26 <fungi> are people successfully running storyboard on python 3 at this point? if so, it might make sense to try switching storyboard-dev over to using python 3.5 instead of 2.7 and see if it's still working, then look at a similar switch for storyboard.o.o
19:37:48 <SotK> yeah, I run my development instance on 3.6
19:38:18 <fungi> 3.6 for us will probably have to wait for container-based deployment, or at least ansible-driven puppetless deployment
19:38:18 <diablo_rojo> Same
19:38:47 <fungi> since switching to a puppet which will work on a platform providing 3.6 is a bit of a hassle still
19:39:36 <fungi> (we're still wrapping up transition to puppet 4, and we need puppet 5 on newer platforms)
19:40:03 <SotK> I believe I've had success running on 3.5 in the past, but I don't remember for sure
19:40:15 <SotK> I definitely think its worth trying out switching storyboard-dev
19:40:25 <fungi> i'll make a note on my todo list
19:40:50 <diablo_rojo> +2
19:40:51 <fungi> mostly curious if performance will be any better. but regardless we need to consider moving off python 2.7 by the end of this year
19:41:18 <fungi> i don't want storyboard devs stuck trying to maintain compatibility with a python which is past eol
19:41:57 <mkarray> python 2 loses support beginning of 2020 iirc, probably a good idea
19:42:13 <diablo_rojo> Agreed. Not enough of us and not enough time
19:42:17 <fungi> exactly
19:42:18 <SotK> +1
19:44:58 <SotK> anything else in-progress that folk want to raise?
19:45:24 <fungi> nothing i'm aware of
19:45:26 <mkarray> I pushed my first patch earlier this week, would be nice if any of you could review when there's some time
19:45:35 <fungi> congrats!
19:45:51 <fungi> and thanks!
19:46:12 <mkarray> :)
19:46:23 <diablo_rojo> mkarray, definitely will take a look today
19:46:30 <diablo_rojo> Congratulations :)
19:46:37 <SotK> I plan to catch up on my review backlog in the morning, so hopefully I'll get to it then :)
19:46:48 <fungi> i'd like to say i'll take a look today, but i'll have to see how the rest of my day goes
19:46:48 <mkarray> Sounds good, thanks all
19:47:38 <fungi> sun will be down here soon and i still have too many irons in the fire i think
19:48:47 <diablo_rojo> fungi, theres always another day :)
19:49:01 <diablo_rojo> SotK, thanks for running the meeting :)
19:49:09 <fungi> there are no guarantees, but that's certainly what i hope for every nigvt when i go to sleep!
19:49:20 <SotK> np
19:49:48 <diablo_rojo> fungi, lol thats dark
19:52:10 <SotK> thanks for coming folks!
19:52:12 <SotK> #endmeeting