| *** jtomasek has joined #storyboard | 07:46 | |
| *** fay_ is now known as faybrocklebank | 08:29 | |
| SotK | pedroalvarez: thanks! | 08:30 |
|---|---|---|
| pedroalvarez | :) | 08:59 |
| *** mrmartin has joined #storyboard | 09:53 | |
| Zara | \o/ | 10:11 |
| pedroalvarez | it was an easy one :) | 10:13 |
| openstackgerrit | Merged openstack-infra/storyboard-webclient: Upgrade 'marked' and 'highlightjs' libraries https://review.openstack.org/282871 | 11:08 |
| SotK | \o/ | 11:08 |
| Zara | :D | 11:09 |
| Zara | I can look at the 401 some more today, or look some more at search | 11:10 |
| Zara | I already have some search stuff waiting in review, so inclined to wait for that before pressing on with more search | 11:11 |
| persia | +1 on solving the 401 part | 11:13 |
| *** betherly has quit IRC | 11:40 | |
| *** betherly_ has joined #storyboard | 11:40 | |
| Zara | hm, seems the trying to resolve an abstract state was a different problem that happened to occur at the same time (at any rate, I just went through and hackily deleted every 'abstract: true', and the same number of errors persist) | 12:19 |
| *** mrmartin has quit IRC | 12:34 | |
| *** mrmartin has joined #storyboard | 12:36 | |
| Zara | and, with only one thing resulting in a 401, the posting the refresh token still fails, so I don't think it's failing because of lots of auth errors (unless some are still hidden, I guess...) | 12:37 |
| Zara | in conclusion, I still have no idea what's wrong, and I can only tell you that modules that the more things a module pre-resolves, the more 401s. | 12:38 |
| Zara | s/modules that// | 12:38 |
| Zara | which is what I noticed last week | 12:39 |
| * persia wonders if pre-resolution fails to carry the user's credentials | 12:40 | |
| SotK | it'd probably fail all the time in that case, rather than just when the token expires though I'd expect | 12:42 |
| SotK | the token is (normally) refreshed once the redirect has happened, I suspect we just need to find a way to retry the failed requests after a successful token refresh, rather than redirecting to the dashboard on a 401 | 12:49 |
| SotK | (if the refresh fails, we should still redirect to the about page, as currently happens) | 12:49 |
| Zara | yeah | 12:52 |
| Zara | the most annoying thing is the redirect, rather than the errors themselves | 12:52 |
| Zara | (I'm on the king of the north branch atm, which doesn't redirect afaict (though it would from the dashboard), but the user stays logged out until manually logging back in) | 12:53 |
| Zara | I should've mentioned I was on that branch | 12:54 |
| persia | Is there any potential to do a token refresh before the preresolution, or is that too much of a performance penalty? | 12:54 |
| Zara | (oh, huh, it still redirects) | 12:55 |
| Zara | hm, so if you click on something and you've been logged out, you do go to the right place (just stay logged out). but if you're not on sb, and you click on a link, and you've been logged out, you get redirected. | 12:56 |
| Zara | persia: honestly, I don't know yet (either how to do it or what the implications would be. I think theoretically it should happen first (or, at least, things should resolve, then the token refresh, then things resolve again), but that's just a guess | 12:57 |
| * persia also doesn't know | 12:58 | |
| Zara | it's in my box marked 'mysterious angularjs magic, dial krotscheck in emergency' | 12:58 |
| persia | heh | 13:01 |
| Zara | :) I've got as far as suspecting it's not happening for the reason people thought it was happening, anyway, so I've updated the story. | 13:02 |
| anteaya | morning | 13:10 |
| pedroalvarez | o/ | 13:11 |
| anteaya | so I got to wondering if there is a storyboard sandbox or test instance up anywhere | 13:11 |
| anteaya | morning pedroalvarez | 13:11 |
| anteaya | as getting folks to play around in a sandbox is always so much fun | 13:11 |
| anteaya | as opposed to cleaning up the db afterwards on a production instance | 13:11 |
| anteaya | has a sandbox ever been discussed? | 13:12 |
| pedroalvarez | this reminds me we never managed to get a dev-storyboard.openstack.org instance running | 13:12 |
| anteaya | interesting | 13:13 |
| Zara | there was one for a while, but someone rebooted it | 13:13 |
| anteaya | ;( | 13:13 |
| Zara | and ssh stopped working to the instance after that, for reasons unknown | 13:13 |
| anteaya | sorry that was :( | 13:14 |
| Zara | heh | 13:14 |
| anteaya | hmmmmm | 13:14 |
| anteaya | so the storyboard team would like a dev instance, is what I am hearing | 13:14 |
| Zara | yup | 13:14 |
| Zara | it's on the todo list and keeps being put back :S | 13:14 |
| anteaya | and there may even be puppety things available for said dev instance to be created | 13:14 |
| anteaya | well let's see if we can give this a bump | 13:15 |
| anteaya | Zara: you don't happen to know where the puppety bits are kept for the dev instance do you? | 13:15 |
| anteaya | to confirm they still exist in master? | 13:15 |
| Zara | hm, I think SotK set it up manually. tbh if we wanted to deploy something quickly, I think we'd probably use pedro's ansible playbooks https://galaxy.ansible.com/detail#/role/6187 , because it took SotK a while to set up one with puppet (that was a different instance) | 13:18 |
| Zara | unless I've got things mixed up, anyway | 13:18 |
| Zara | (puppety bits for storyboard.openstack.org are at https://github.com/openstack-infra/puppet-storyboard , I don't think we have any other puppety thigns) | 13:19 |
| Zara | *things | 13:19 |
| anteaya | awesome | 13:20 |
| anteaya | thank you | 13:20 |
| anteaya | Zara: do you mind putting that info in the infra channel | 13:20 |
| anteaya | with a request to set up a dev server? | 13:20 |
| Zara | that'd be fine by me, SotK, pedroalvarez? | 13:20 |
| anteaya | and can you use git.openstack.org instead of github when you post the link in infra, please? | 13:20 |
| Zara | ah, sure, it was just the one in my history | 13:21 |
| * SotK is fine with that | 13:21 | |
| pedroalvarez | Zara: I'm ok with whatever you are ok :) | 13:22 |
| anteaya | Zara: thanks | 13:22 |
| anteaya | SotK: morning :) | 13:22 |
| SotK | morning anteaya :) | 13:22 |
| anteaya | :) | 13:22 |
| *** mrmartin has quit IRC | 13:24 | |
| anteaya | Zara: so thanks for posting that | 13:25 |
| anteaya | things may or may not happen today or even this week on the request, but you posted it | 13:26 |
| anteaya | so yay! \o/ | 13:26 |
| anteaya | thank you | 13:26 |
| Zara | thanks :) I hope things happen! | 13:26 |
| *** mrmartin has joined #storyboard | 13:26 | |
| *** mrmartin has quit IRC | 13:27 | |
| pedroalvarez | thanks anteaya for helping :) | 13:27 |
| anteaya | doing my best | 13:28 |
| anteaya | :) | 13:28 |
| anteaya | keep being awesome! | 13:29 |
| *** mrmartin has joined #storyboard | 13:29 | |
| Zara | you too! :P | 13:32 |
| krotscheck | Eh, wha? | 13:35 |
| krotscheck | Zara: Mysterious magic? | 13:35 |
| Zara | Krotscheck: I tried looking at the events causing the 401 again... concluded that all angularjs was mysterious magic. (updated the story here with what I've found so far: https://storyboard.openstack.org/#!/story/2000407) | 13:37 |
| Zara | (we thought it was to do with sb trying to reload an abstract state; I tried hackily deleting every 'abstract: true' to see what happened; behaviour was exactly the same as before. :/ ) | 13:38 |
| krotscheck | Ok, I'll dig into it more once I'm done arguing with Google about Angular ;) | 13:38 |
| Zara | heehee | 13:38 |
| Zara | what's the argument about? | 13:39 |
| krotscheck | https://github.com/angular/angular.js/issues/9334 | 13:39 |
| krotscheck | Not an argument, just them asking me to update a patch | 13:39 |
| krotscheck | Oh wait | 13:39 |
| krotscheck | That's the argument | 13:39 |
| krotscheck | https://github.com/angular/angular.js/pull/13273#issuecomment-186313362 | 13:39 |
| krotscheck | that's my issue | 13:39 |
| Zara | ahhh | 13:41 |
| Zara | can someone give https://review.openstack.org/#/c/281898/ a +1? the current docs tell the user to install a video storyboard generator completely unrelated to StoryBoard, that happens to be called 'storyboard' | 13:52 |
| Zara | I have a few fixes waiting in review, but that's the funniest one. | 13:52 |
| betherly_ | LOL! thats brilliant | 13:54 |
| betherly_ | +1 done | 13:55 |
| Zara | thanks :) | 13:55 |
| Zara | I think someone generated those docs right at the start of the project and nobody noticed because we all used the developer ones. | 13:56 |
| persia | I used those docs once, a very long time ago, and they worked, although my instance only lasted a couple weeks before some patches broke it. | 13:57 |
| * persia doesn't remember the details, precisely, other than a conversation in which it was concluded that running separate VMs for the DB, the API service, and the web server was considered overkill for a development instance | 13:58 | |
| Zara | the user docs, rather than the developer ones? I think they've directed to the wrong package since Jan 2014... | 13:59 |
| * Zara checks diffs | 13:59 | |
| openstackgerrit | Merged openstack-infra/storyboard: Update User Docs https://review.openstack.org/281898 | 13:59 |
| persia | I would believe them being wrong since Jan 2014 :) | 14:00 |
| persia | I think they worked immediately before the meeting in Brussels, or something like that. | 14:00 |
| Zara | yeah, that's the earliest commit I can find for that file. it's possible it had a different name before that. | 14:03 |
| Zara | so I think it was a case of 'write placeholder text, replace later with real content', only nobody realised it was a placeholder | 14:05 |
| persia | hmmm. Perhaps | 14:05 |
| * SotK goes to look | 14:07 | |
| *** mrmartin has quit IRC | 14:09 | |
| SotK | looks to me like the `pip install storyboard` instruction originated in this patch: https://review.openstack.org/#/c/62239/ | 14:12 |
| *** openstackgerrit has quit IRC | 14:17 | |
| *** openstackgerrit has joined #storyboard | 14:17 | |
| Zara | yeah, that matches my findings. | 14:21 |
| pedroalvarez | hah, "Customizable video storyboard generator with metadata report" | 14:29 |
| Zara | hm, does https://storyboard.openstack.org/#!/story/2000163 count as merged, with the changes to default number of things returned (I believe SotK reduced it from 500 to 50) | 15:10 |
| Zara | or is it asking for something a bit different? | 15:10 |
| * SotK thinks that is something different | 15:11 | |
| Zara | fair enough | 15:11 |
| SotK | (the 500 to 50 reduction was just for the search when adding cards) | 15:11 |
| Zara | ohh, okay | 15:12 |
| Zara | It seemed to be faster all-round, but that might have been due ot other speedups | 15:12 |
| Zara | OR my imagination | 15:12 |
| Zara | huh https://storyboard.openstack.org/#!/story/24 | 15:12 |
| SotK | :D | 15:19 |
| *** b3rnard0_away is now known as b3rnard0 | 15:26 | |
| *** mrmartin has joined #storyboard | 15:43 | |
| Zara | hm, there's a 'figure out how to publish javascript libraries to npm' story from 2014 | 15:51 |
| Zara | I *suspect* I can mark that merged... | 15:51 |
| Zara | (I'm going through old stories atm; at some point we need to find a way to reduce duplicates. some advise doing a search on the title field, with suggested results, when people make new stories.) | 15:54 |
| persia | I've seen that both work very well and fail disasterously in the past. | 15:56 |
| persia | On the positive side, this results in many fewer bugs being filed, and concentrates discussion. | 15:57 |
| persia | On the negative side, it turns out that many people have difficulty differentiating "it crashes when I press the red button" from "it crashes when I press the red button", even with different stacktraces/backtraces. | 15:57 |
| persia | In other context, the recommended solution was to have project-specific configuration that allowed developers to enable or disable such a search. I'm not sure how that could work for Storyboard, really. | 15:59 |
| Zara | yeah, in that situation I think it'd be best to just put up with the negatives. it seems like more of a social problem anyway. it's not urgent, either way, just on my 'nice to have' list. tbh the main problem is when people treat stories like tasks, so that you end up with 5 different 'stories', all with the same goal, but a different task in each | 16:05 |
| Zara | that is annoying to garden | 16:05 |
| SotK | it is equally hard to garden monolithic stories with mountains of tasks I think | 16:08 |
| Zara | yeah, I'd like tasks to be ordered by status by default, I think that'd help with that a bit | 16:09 |
| persia | It really depends on the project. I worked on an operating system installer for a while, and nearly every reported issue was "cannot install", but it was very rare that any two of them had anything in common. | 16:09 |
| Zara | other than that... it's mainly on people to scope stories well, which can be tricky | 16:09 |
| SotK | (eg there is both https://storyboard.openstack.org/#!/story/2000322 and https://storyboard.openstack.org/#!/story/2000464, since I didn't want a single story for the entire boards/worklists stuff to avoid a giant task list) | 16:10 |
| persia | But for most projects, there is less magic and more user interaction, making it easier for users to know if they have the same or a different story they wish enabled. | 16:10 |
| persia | +1 on ordering tasks by status by default | 16:10 |
| * SotK wants to be able to filter the task list, but +1 on that in the short term from me too | 16:11 | |
| Zara | SotK: yeah, I think your separating those makes sense. I was thinking more of the extremes (that I'm guilty of as well). | 16:12 |
| Zara | partly because at first I thought stories were tasks... | 16:12 |
| persia | Unfortunately ,the viewpoint from which the nomenclature derives is not widespread, so few people actually write stories as stories. | 16:13 |
| persia | That said, the more I think about stories, the more I think that some scenarios and/or tasks extracted from the stories should be shared between stories (as one may need to achieve something in order to enable either of two different described behaviours). | 16:14 |
| persia | The conclusion I drew from this realisation was that it didn't really matter: the key exciting bit was that one could have a many-to-one relationship, and define semantics as needed. | 16:14 |
| persia | (although I would prefer if it was possible to assign tasks to branches, rather than just repos, but this may just be me) | 16:15 |
| * SotK believes that is possible in the API | 16:15 | |
| persia | Excellent. That gives leisure to think about presentation. | 16:15 |
| *** mrmartin has quit IRC | 17:26 | |
| *** mrmartin has joined #storyboard | 17:29 | |
| *** mrmartin has quit IRC | 17:33 | |
| openstackgerrit | Zara proposed openstack-infra/storyboard-webclient: Group Tasks By Status https://review.openstack.org/283183 | 17:33 |
| Zara | luckily, the alphabet goes M, then R, then T. perhaps the terms were chosen with this in mind | 17:34 |
| Zara | 'progress' is the sad one. | 17:35 |
| * Zara advocates renaming it to 'Ssssssprogress' to get round that | 17:36 | |
| * pedroalvarez suggests Zara to look at the alphabet | 17:37 | |
| pedroalvarez | oh no, I understand what you mean | 17:37 |
| * pedroalvarez pats himself | 17:38 | |
| Zara | :P | 17:38 |
| Zara | I KNOW MY ALPHABET VERY WELL THANK YOU | 17:38 |
| Zara | software, not so much | 17:38 |
| pedroalvarez | :) | 17:39 |
| pedroalvarez | so with that pach.. they are going to be grouped by status? | 17:39 |
| Zara | yeah | 17:39 |
| Zara | since people can't reorder them anyway | 17:39 |
| pedroalvarez | hahah, you can't reordem them, I'll set the order I want!!! HA! | 17:40 |
| pedroalvarez | I was just kidding, but I wonder if people will get upset because of this | 17:40 |
| Zara | we'll see if anyone complains about the patch, I guess. I thought it'd make it a bit easier to see how a story is progressing (we had some feedback asking for this), and I'm probably not going to get round to making them reorderable any time soon. | 17:41 |
| Zara | (jeblair had the same problem with big stories as SotK, and also suggested ordering by status) | 17:41 |
| Zara | so.... it's on people who like it the current way to protest, I guess | 17:41 |
| Zara | you can order things however you want in a worklist, anyway | 17:42 |
| Zara | huh... in storyboard's alphabet, P comes before M | 17:49 |
| Zara | I'm guessing 'progress' is short for 'inprogress' or something. | 17:49 |
| Zara | yup, argh. | 17:50 |
| Zara | welp, that's something to tidy up in the api at some point, I guess | 17:52 |
| Zara | ssssssssprogress gets my vote there, too | 17:52 |
| Zara | 'night, storyboard! I have a million things in review at this point! there's a very small patch that's 3 weeks old! | 17:58 |
| Zara | https://review.openstack.org/#/q/owner:zara.zaimeche%2540codethink.co.uk+status:open | 17:59 |
| Zara | I will leave that there :P | 17:59 |
| Zara | so yeah, night! | 17:59 |
| * pedroalvarez review that tiny patch | 18:02 | |
| pedroalvarez | I need to review things, I know, you have been coding really fast :) | 18:03 |
| pedroalvarez | </excuse> | 18:03 |
| * SotK also apologises for not reviewing for weeks | 18:34 | |
| *** mrmartin has joined #storyboard | 18:40 | |
| openstackgerrit | Merged openstack-infra/storyboard-webclient: Use 'Tasks' Icon for Tasks https://review.openstack.org/275130 | 18:42 |
| *** mrmartin has quit IRC | 19:08 | |
| *** mrmartin has joined #storyboard | 19:18 | |
| *** mrmartin has quit IRC | 19:25 | |
| *** mrmartin has joined #storyboard | 19:49 | |
| *** gothicmindfood has quit IRC | 20:19 | |
| *** persia has quit IRC | 20:19 | |
| *** persia has joined #storyboard | 20:19 | |
| *** persia has quit IRC | 20:20 | |
| persia_ | Everyone else managed to reconnect to a sane server. I try, and end up back on the doomed server. | 20:21 |
| *** persia has joined #storyboard | 20:21 | |
| *** gothicmindfood has joined #storyboard | 20:22 | |
| *** krotscheck has quit IRC | 20:37 | |
| *** anteaya has quit IRC | 20:38 | |
| *** krotscheck has joined #storyboard | 20:46 | |
| *** openstackgerrit has quit IRC | 20:47 | |
| *** openstackgerrit has joined #storyboard | 20:48 | |
| *** anteaya has joined #storyboard | 20:48 | |
| *** mrmartin has quit IRC | 21:54 | |
| *** jtomasek has quit IRC | 22:03 | |
| *** b3rnard0 is now known as b3rnard0_away | 23:29 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!