19:02:16 #startmeeting storyboard 19:02:18 Meeting started Wed Mar 21 19:02:16 2018 UTC and is due to finish in 60 minutes. The chair is SotK. Information about MeetBot at http://wiki.debian.org/MeetBot. 19:02:19 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 19:02:21 The meeting name has been set to 'storyboard' 19:02:47 Hello :) 19:02:53 hi! 19:03:05 howdy 19:03:11 #link https://wiki.openstack.org/wiki/Meetings/StoryBoard#Agenda_for_next_meeting Agenda 19:04:20 Boom. Made some changes to the agenda. 19:04:26 \o/ 19:04:27 #topic Announcements 19:04:35 Sahara migrated! 19:04:39 YAY! 19:04:57 Gaining traction :) 19:05:21 excellent news 19:05:49 indeed 19:06:02 and the agenda is full of plans for more 19:06:03 next week we will have more exciting annoucements :) 19:06:13 #topic In Progress Work 19:06:53 I re-sent the patch to fix syntax highlighting 19:07:12 SotK, will review today/tomorrow. 19:07:29 thanks! 19:07:46 I have two patches out- one for private labels in the webclient, the other for a fix to the migration script to hopefyully stop making duplicate entries for users in the db. 19:07:57 Both already have +2's from fungi- thanks fungi :) 19:08:31 I will try to find some time to review them this week 19:08:47 we've noticed that the lp query api is feeding us "old" login.launchpad.net openids rather than login.ubuntu.com ones, and this has resulted in duplicate accounts. i'm working on correcting and deduplicating these, but the main challenge will be finding all the tables which have a column for user ids as the columns aren't consistently named for that 19:09:00 SotK, that would be awesome. They are both sure small so aside from testing the second one, it shouldn't take too long. 19:09:05 related to the seconf patch diablo_rojo mentioned 19:09:15 I can't promise anything as I've had approximately 0 free time so far this week 19:09:25 SotK, fair :) 19:09:50 this is also why I have failed to see if master is copying fonts properly despite saying I'd do it on monday :( 19:10:07 worst case i'll single-core approve the migrate script fix for rewriting openid urls before i do the next migration (probably friday) 19:10:43 fungi: I can try to make a list of the relevant fields at some point this week, that sounds like a short enough task I can fit it in somewhere 19:10:57 Looked like corvus is in need of help with the db model too. 19:11:26 luckily the test migrations on storyboard-dev have caused the same problem, so i have both an easy place to test the deduplication work _and_ a place to test that subsequent test migrations with diablo_rojo's patch work as advertised 19:11:52 yeah, I didn't have any obvious pointers after a quick glance at it, I will also try to fit looking into that in somewhere xD 19:12:04 Easy peasy- and some day I might have perms to do test migrations there too :) 19:12:14 SotK: if you think you can, that would be a huge help to me. otherwise i anticipate just reading through the entirety of the db schema picking out column names which "look like user ids" 19:12:31 * diablo_rojo cringes at the thought 19:12:34 (and hoping for the best) 19:13:25 diablo_rojo: I would be happy to +1 a patch to the place that patches go to grant that permission 19:13:32 if i at least know which columns are user ids, then i can work out which ones need rewriting vs which ones need their duplicate rows deleted 19:14:01 SotK, https://review.openstack.org/#/c/552667/ 19:14:05 diablo_rojo: has that system-config change to get you access to sb-dev.o.o not merged yet? 19:14:23 fungi, it hasn't was going to poke at clarkb today about it 19:14:24 i'll try to drum up another reviewer to approve it 19:14:33 fungi, that works too :) 19:14:45 clarkb may be hesitant to approve without another non-foundation-staff core reviewer giving a +2 19:15:00 fungi, oh yeah that makes sense. 19:15:05 ya I've tried to follow fungi's example there 19:15:14 Good practice to have. 19:15:49 especially where granting exceptional access to things comes into play 19:15:51 it has a non-foundation +2 now :) 19:15:58 thanks corvus!!! 19:16:15 Thank you corvus :) 19:16:58 \o/ nice 19:17:39 So hopefully that gets merged soon and I can carryon running all the test migrations there 19:17:48 that would be good 19:18:01 anything else in progress not being covered further down the agenda? 19:18:13 SotK, indeed, especially since there seems to be an increasing interest in what it the end result looks like. 19:18:30 SotK, don't think so. 19:18:37 oh, I did also for fun get a low priority patch for redesigning the dashboard a bit into a reviewable state over the weekend 19:19:16 #topic Migration Updates 19:19:34 this is an exciting looking section 19:19:42 :) 19:19:53 Ironic wants to migrate this week- Friday. 19:20:06 I also was notified that the UI squad in tripleO is ready to go. 19:20:21 So Friday will be a full day of migrations. 19:20:29 yep, i'll hopefully have the account deduplication tested and completed by then 19:21:09 sounds great :D 19:21:28 hopefully they go well 19:21:35 I also need to circle back to an email thread I have with Vitrage about migration- they asked a lot of the same questions- how do I differentiate bugs and new features? 19:21:36 either way i can do the imports friday 19:21:44 I didn't have the will power to answer the other way. 19:22:01 just the deduplication script (once i write it) will take longer the more db content we have 19:22:22 fungi, sounds like the sooner the better on that one then 19:23:01 yeah, but either way it's still just a fire-and-forget and odds are we're talking about a runtime difference on the order of seconds 19:23:19 i'm more concerned with additional new users of sb.o.o getting confused by the extra accounts 19:23:35 fungi, yeah that can be offputting. 19:23:40 so still, the sooner the better 19:24:44 Only other thing I have is that I'm currently running openstacksdk migrations and so far so good. I migrated three of the launchpad projects so far- I thiiiink there is one or two more left. 19:25:17 They've all been pretty quick too. 19:25:27 nice! 19:25:34 That's all from me for now :) 19:25:59 thanks for all your work on this diablo_rojo and fungi :) 19:26:15 #topic Open Discussion 19:26:37 SotK, you skipped the outreachy intern topic, but we can cover it here :) 19:26:50 ooops 19:26:52 #undo 19:26:53 Removing item from minutes: #topic Open Discussion 19:26:59 #topic Outreachy Intern! 19:27:04 Was chatting with Mridu this morning :) 19:27:10 i still haven't started on the mysql 4-byte unicode adjustment... which test migrations were failing due to extended codepoints? 19:27:20 She is working on applying to be an intern. 19:27:29 oh, i jumped the gun on open discussion, sorry 19:27:39 fungi, heat was struggling with that one... there was another. I will try to remember and let you know. 19:27:45 thanks 19:27:57 fungi, oh it was mistral I think 19:28:06 helps to have test cases 19:28:12 fungi, indeed it does :) 19:28:41 So, intern. 19:28:49 diablo_rojo: cool, I guess we should have some stories in mind to point people at during the application stage? 19:28:55 any news on sweksha who also popped into the channel? 19:29:12 SotK, yeah. I have been pointing to the make cli commands and the make tests not suck one 19:29:18 fungi, sadly not. 19:29:39 i think improving testing for the project would be at or near #1 for me, if they're interested in hacking on that 19:29:47 yeah I agree 19:29:49 I talked to vkmc and she didn't have sweksha's info so I don't have any way of contacting them unless they come back. 19:30:25 :( 19:30:40 fungi SotK agree. So I pointed Mridu at that this morning :) 19:30:42 Hi all. I am Mridu. I am looking forward to contribute to the project. Work more on Python. And learn to code better. Through this opportunity. :) 19:31:00 hi Mridu! and welcome 19:31:03 Mridu, we are thrilled to have you :) 19:31:39 welcome Mridu! 19:31:41 I pointed her at going through all the tasks tests and checking that they actually cover all the api functions for tasks. 19:31:47 Thank you so much :) 19:31:53 Unless there is somewhere else we want her to start on tests? 19:32:37 Still curious on knowing this project involves more of testing or more of development? 19:33:45 There are pretty equal opportunities of both I would say. 19:34:00 Our tests need a lot of help, but there are still a lot of things to be developed :) 19:34:05 particularly where it comes to developing tests 19:34:19 And anything you develop has to be tested to be merged. 19:34:37 fungi, lol :) 19:34:44 basically, we have a lot of features which got added in recent years with limited or sometimes even no test coverage at all 19:34:49 anything that levels up the coding skills. And helps me gain confidence. I am fine with it :) 19:35:08 Mridu, perfect :) I think you are in the right place then :) 19:35:13 so it's hard for us to know when subsequent changes break those features unless we have also developed regression tests to check that they continue to work 19:35:24 tests are definitely code, though 19:35:28 +1 19:35:31 based on what fungi, is saying it seems. Lots and lots of tests. :p 19:35:56 I think there is likely to be opportunity for both, but yeah, our tests are lagging behind everything else a lot atm 19:36:22 sometimes some of the most complicated kinds of software development, since it involves thinking about mocking out or similarly faking parts of existing software so that you can test other parts efficiently 19:36:41 er, test are sometimes some of the most complicated kinds of software development 19:36:45 tests 19:36:51 i should give up on typing today 19:37:44 but yeah, developing tests is still very much a kind of software development 19:37:58 fungi, I typed whole sentences that made no sense last night during the fc sig meeting- I wouldn't sweat a few typos. 19:38:01 hahaha. Also, I am a bit confused with the choices. There are 2-3 projects under Openstack which one to opt for. Also, is this the right place to ask such a question in first place. 19:38:46 Mridu 2-3 projects that have outreachy intern slots? 19:38:52 I think there are two. 19:38:53 yes 19:38:56 diablo_rojo: i read that meeting log. seemed like an excellent one 19:39:26 https://www.outreachy.org/2018-may-august/communities/openstack/ see. 19:40:15 Storyboard and one on Redfish it looks like. 19:40:48 Yes yes 19:40:58 One is there on Keynote - SAML as well 19:41:17 keystone, yes 19:41:33 also a (incomplete?) one for qinling, looks like 19:41:43 yes 19:41:50 yeah looks like it 19:42:07 complete or incomplete I am unaware about that. 19:43:00 It might be completed. 19:44:00 Mridu, well, it depends on what your interests are which project you want to jump into I suppose. Of the options I think we have the most flexbility in coming up with work to fit your interests, but I might be biased :) 19:44:27 * persia is biased 19:44:42 What is SAML project about? 19:44:58 I am also biased, but agree with diablo_rojo 19:45:07 These are the 2 projects where I can see skill level 5 that is needed :') 19:45:10 Storyboard is the best intern project for OpenStack because it lets one code in more languages, leverages more different testing components, and has active users all over the OpenStack developer community, allowing one to build strong relationships that will serve later. 19:45:57 Storyboard is also one of the few projects with full continuous deployment, so one can see the effects of one's work immediately (although the deployment was broken, last I checked) 19:46:03 Mridu: the #openstack-keystone channel can probably give you better answers about that one, but saml is "security assertion markup language" (a standard to serialize authn/authz data) 19:46:11 Mridu, so Keystone is actually the OpenStack project you would be working on for the SAML one. That project is focused on authetication, service discovery and authorization. 19:47:24 Basically it comes down to- do you like the idea of developing a task tracker tool or are you more interested in authentication and authorization? 19:48:13 Something that serves more end users I would go for that 19:49:18 it's hard to compare the end users of keystone (who are mostly interacting with it through a remote api) and the users of storyboard (which has a ui) 19:50:09 you could consider that the "end users" of keystone are other applications (which may be run by people, or may be run by yet other applications still) 19:50:10 I also don't know. I will give this a shot. Let's see what happens. :) 19:50:50 Wonderful :) 19:50:55 excellent :) 19:51:09 SotK, now you can go to open discussion :) 19:51:15 haha 19:51:20 #topic Open Discussion 19:51:37 yes! whether you want to work on storyboard or one of the other openstack projects, or on other free software completely, i think it's still awesome. free software always needs more people 19:51:48 fungi, +1 19:52:09 yeah. That's right 19:52:23 Thank you all for helping out :) 19:53:00 and thank you for wanting to help out too 19:53:02 Mridu, happy to help :) If you need anything else getting started- we all hang out in the #storyboard channel 19:53:29 Sure. 19:56:51 while we're here, did anyone happen to have ideas off the tops of their heads about the problem corvus encountered with session management in tests? 19:57:47 #link https://review.openstack.org/553102 WIP: test subscribers and permissions 19:57:53 not off the top of my head no 19:58:06 I will attempt to give it some thought when I have time 19:58:21 #link http://logs.openstack.org/02/553102/1/check/tox-py35/fee289f/testr_results.html.gz mysterious failure 19:58:40 thanks 19:59:02 Perhaps more mysterious is how the few are passing 19:59:07 Wheneverything else fails 20:00:13 well, that pep8 failure is just a style consistency nit "imports not in alphabetical order" 20:00:56 we're out of time 20:01:00 thanks for coming all 20:01:04 #endmeeting