| *** sjmc7 has quit IRC | 00:02 | |
| *** gokrokve has quit IRC | 01:22 | |
| *** TravT has quit IRC | 01:41 | |
| *** gokrokve has joined #murano | 01:47 | |
| *** sab has joined #murano | 04:13 | |
| *** sab has quit IRC | 04:14 | |
| *** openstackgerrit has quit IRC | 04:57 | |
| *** chandan_kumar has joined #murano | 05:40 | |
| *** gokrokve_ has joined #murano | 06:58 | |
| *** gokrokve has quit IRC | 07:02 | |
| *** stan_lagun has joined #murano | 08:07 | |
| *** slagun has quit IRC | 08:10 | |
| *** chandan_kumar has quit IRC | 09:09 | |
| *** katyafervent_awa is now known as katyafervent | 09:53 | |
| *** katyafervent is now known as katyafervent_awa | 10:35 | |
| *** stanlagun has joined #murano | 10:52 | |
| *** openstackgerrit has joined #murano | 10:55 | |
| *** katyafervent_awa is now known as katyafervent | 10:58 | |
| openstackgerrit | Anastasia Kuznetsova proposed a change to stackforge/murano-dashboard: Add Test package, remove uploading of extra apps https://review.openstack.org/100846 | 11:01 |
|---|---|---|
| openstackgerrit | A change was merged to stackforge/murano: Change how actions are stored in Object Model https://review.openstack.org/101194 | 11:07 |
| openstackgerrit | A change was merged to stackforge/murano: Support specifying ``sort_dir`` key in packages.search call https://review.openstack.org/100283 | 11:16 |
| openstackgerrit | A change was merged to stackforge/murano-dashboard: Add 'Previous Page' button to the App Catalog page https://review.openstack.org/100284 | 11:22 |
| *** asalkeld has quit IRC | 11:40 | |
| *** gokrokve_ has quit IRC | 11:54 | |
| openstackgerrit | Sergey Kolekonov proposed a change to stackforge/murano-deployment: added new parameter for engine tests https://review.openstack.org/101506 | 12:09 |
| *** gokrokve has joined #murano | 12:32 | |
| *** gokrokve has quit IRC | 12:37 | |
| openstackgerrit | Ruslan Kamaldinov proposed a change to stackforge/murano: Added DB migrations on Alembic https://review.openstack.org/96471 | 13:30 |
| ruhe | patch-set #22. i hope this one will be merged :) | 13:31 |
| *** gokrokve has joined #murano | 13:34 | |
| *** gokrokve has quit IRC | 13:39 | |
| stanlagun | ruhe, you missed one place :) | 13:39 |
| stanlagun | status.entity_id | 13:39 |
| ruhe | ok. i will calmly update my patch and will not smash my keyboard | 13:40 |
| stanlagun | :)) | 13:40 |
| stanlagun | I don't know why this happens. I commented every single place in my first review | 13:41 |
| *** iyozhikov is now known as IgorYozhikov | 13:53 | |
| *** gokrokve has joined #murano | 13:55 | |
| *** gokrokve has quit IRC | 14:00 | |
| openstackgerrit | Ruslan Kamaldinov proposed a change to stackforge/murano: Added DB migrations on Alembic https://review.openstack.org/96471 | 14:01 |
| *** gokrokve has joined #murano | 14:02 | |
| *** sjmc7 has joined #murano | 14:09 | |
| openstackgerrit | Steve McLellan proposed a change to stackforge/murano: Add 'userdata_format' to Server heat template https://review.openstack.org/100344 | 14:17 |
| sjmc7 | can someone approve https://review.openstack.org/#/c/99757/ please? | 14:19 |
| *** gokrokve has quit IRC | 14:38 | |
| stanlagun | sjmc7, done | 14:49 |
| sjmc7 | thanks | 14:50 |
| *** stanlagun has quit IRC | 14:51 | |
| *** chandan_kumar has joined #murano | 14:55 | |
| sjmc7 | stanlagun, regarding the inheritance thing you commented on - how would an Application require a dependency on a class instance with the behavior of multiple subclasses? | 15:00 |
| sjmc7 | e.g. a HeatSWConfigInstance and LinuxInstance ? | 15:00 |
| sjmc7 | it feels like there should be a cleaner way to do this | 15:00 |
| sjmc7 | mix-ins are the way python tackled this; would the same approach work here? | 15:02 |
| sjmc7 | problem is, the subclasses are overriding behavior in most cases, not adding it | 15:05 |
| openstackgerrit | A change was merged to stackforge/murano-dashboard: Make individual flavor filters optional https://review.openstack.org/99757 | 15:15 |
| openstackgerrit | Ekaterina Fedorova proposed a change to stackforge/murano-dashboard: Sync developer settings https://review.openstack.org/101578 | 15:21 |
| openstackgerrit | Ekaterina Fedorova proposed a change to stackforge/python-muranoclient: Add package-create command https://review.openstack.org/99177 | 15:29 |
| openstackgerrit | Ekaterina Fedorova proposed a change to stackforge/python-muranoclient: Infrastructure update https://review.openstack.org/100954 | 15:29 |
| *** gokrokve has joined #murano | 15:41 | |
| *** TravT has joined #murano | 16:04 | |
| *** PierreRambaud has quit IRC | 16:05 | |
| tsufiev | sjmc7, he'll be in an hour or so | 16:10 |
| sjmc7 | ok, np | 16:10 |
| *** btully has joined #murano | 17:41 | |
| btully | tsufiev: i just responded to your comments re bug #1328512 and code review https://review.openstack.org/#/c/99050/ | 17:43 |
| *** gokrokve_ has joined #murano | 18:49 | |
| *** gokrokve has quit IRC | 18:52 | |
| tsufiev | btully, hi! i've seen your comment at https://review.openstack.org/#/c/99050/4/muranodashboard/templates/catalog/index.html | 19:44 |
| tsufiev | i've realized that I don't know either why buttons became non-responsive in that case - in first patchset I used special style provided in html | 19:45 |
| btully | k | 19:45 |
| btully | when you say unresponsive, do you mean they couldn't be clicked? | 19:45 |
| tsufiev | yes | 19:46 |
| tsufiev | unresponsive intentionally | 19:46 |
| tsufiev | perhaps some of the classes you've assigned to div had the same meaning | 19:46 |
| btully | i didn't change the button classes, no | 19:47 |
| btully | buttons are the same | 19:47 |
| btully | i didn't realize we had a function that intentionally makes the buttons unresponsive | 19:47 |
| tsufiev | weird... I'll take another look in a couple of minutes and then we can discuss that issue in hangout | 19:48 |
| btully | do you know where in the code the function is that enables the buttons? | 19:48 |
| tsufiev | or here | 19:48 |
| btully | i'm actually in a meeting right now unfortunately :( | 19:48 |
| tsufiev | np | 19:49 |
| tsufiev | irc is ok | 19:49 |
| tsufiev | https://review.openstack.org/#/c/99050/1/muranodashboard/templates/catalog/index.html so here the buttons became non-responsive because I'd covered them with transparent div | 19:51 |
| tsufiev | https://review.openstack.org/#/c/99050/4/muranodashboard/templates/catalog/index.html now I'll try to find out what magic did disable buttons here :) | 19:52 |
| btully | right you added the transparent div | 19:58 |
| btully | i thought you meant that there was some other function that was making the buttons non-responsive | 19:58 |
| btully | but just to take a step back | 19:58 |
| btully | i thought the purpose of your work was to purposefully make the buttons non-responsive | 19:59 |
| btully | i.e., don't allow a user to click before the page loads | 19:59 |
| btully | due to someone at one time clicking a button before the page was fully loaded, and some error occurred | 20:00 |
| btully | can you elaborate on that error? maybe i don't understand the purpose of what bug we are trying to fix | 20:00 |
| btully | it almost sounds like we fixed one bug but created another? | 20:01 |
| tsufiev | btully, yes, the purpose of that commit was indeed to prevent user from clicking a button before a page is fully loaded | 20:07 |
| tsufiev | i've did a little investigation... | 20:07 |
| tsufiev | 1. .modal-backdrop class is doing almost the same thing as my div with hard-coded style | 20:08 |
| tsufiev | 2. although some .modal-backdrop css-defined properties start working immediately, namely all mouse events interception, some others (like 0.8 alpha black color) are applied at the very end | 20:11 |
| btully | right | 20:11 |
| tsufiev | they are applied at roughly the same time when the spinner is shown | 20:11 |
| tsufiev | i suspect that this behaviour is not specified and depends on browser implementation | 20:12 |
| btully | so what is it we want to do, remove the spinner? | 20:13 |
| tsufiev | now it seems like a challenge to me: to save the spinner and show it in the right time :) | 20:14 |
| btully | or is the background that covers the buttons not loading fast enough? | 20:14 |
| tsufiev | removing the spinner is simply... | 20:15 |
| btully | the spinner is font-awesome package which i think you guys added, right? | 20:15 |
| btully | however it uses css3 animations to spin the icon | 20:16 |
| btully | so maybe that's part of the problem? | 20:16 |
| tsufiev | yes, it's ours | 20:16 |
| tsufiev | perhaps it's 'fade in' alone... | 20:16 |
| btully | right | 20:17 |
| btully | maybe we remove fade in | 20:17 |
| btully | and then the animation to fade in the background and spinner will hopefully not occur | 20:17 |
| tsufiev | no, doesn't work :( | 20:18 |
| tsufiev | opacity: 0.8 is applied both to .modal-backdrop alone and to .modal-backdrop.fade.in | 20:18 |
| btully | is the opacity 0.8 an issue? | 20:18 |
| tsufiev | i could only guess | 20:19 |
| tsufiev | maybe it's not | 20:19 |
| btully | for me, when i load the page, i barely see the background and spinner. it disappears almost immediately | 20:20 |
| tsufiev | are you using development server in debug mode? | 20:20 |
| btully | are you testing in some way that slows the browser down, and if so, how are you doing that? | 20:20 |
| tsufiev | as I said, django dev-server, debug mode | 20:21 |
| btully | the devstack server or my local running of murano and muranodashboard? | 20:21 |
| tsufiev | also, I'm using Linux inside VBox at home | 20:21 |
| tsufiev | local one | 20:21 |
| btully | let me check | 20:21 |
| btully | in my murano.conf, debug and verbose are set to true | 20:22 |
| btully | in local_settings.py for muranodashboard, debug is also set to true | 20:26 |
| tsufiev | so, you're using debug mode | 20:26 |
| *** gokrokve_ has quit IRC | 20:26 | |
| tsufiev | perhaps your hardware is more powerful, so the page is served faster than on my machine | 20:26 |
| tsufiev | it seems like an interesting puzzle, solving it could advance us in understanding how browsers load&render page content - but unfortunately it will be of little use for the Murano UI as a whole :( | 20:28 |
| sjmc7 | which page are you talking about? | 20:29 |
| sjmc7 | the catalog page? | 20:29 |
| tsufiev | i'll read some articles on the internet on how browsers do the rendering, but suspect the most practical solution here is to drop the spinner which appears to late | 20:29 |
| tsufiev | sjmc7, yes | 20:29 |
| sjmc7 | how many applications do you have? | 20:29 |
| tsufiev | i have six on first page | 20:30 |
| btully | i have 3 and only 1 page | 20:30 |
| sjmc7 | i'm on a remote server with a few packages, but it takes almost no time. a second or less | 20:31 |
| tsufiev | sjmc7, could you try this one https://review.openstack.org/#/c/99050/ - to see how late the spinner appears? | 20:32 |
| tsufiev | the idea was to make buttons not responsive (because pressing them until page is fully loaded causes an error) and show the spinner immediately | 20:33 |
| sjmc7 | have people pressed the button before it's loaded? | 20:40 |
| sjmc7 | what do other horizon dashboards do? | 20:41 |
| tsufiev | yes, they pressed | 20:42 |
| tsufiev | hm... you're asking the exact question i should had asked myself before doing the initial fix :-| | 20:43 |
| tsufiev | first have to find dashboards with page structure similar to AppCatalog | 20:44 |
| openstackgerrit | Brian Tully proposed a change to stackforge/murano-dashboard: Disable all buttons and links until AppCatalog page is loaded https://review.openstack.org/99050 | 20:50 |
| tsufiev | I've tried to reproduce initial bug on Project->Images, 'Lauch' button, but failed | 20:51 |
| btully | tsufiev: can you try the patch i just committed to see if the spinner loads faster now? | 20:51 |
| tsufiev | btully, yes | 20:52 |
| sjmc7 | which button(s) cause problems if they're hit too early? | 20:53 |
| tsufiev | sjmc7, Add to Env and Quick deploy | 20:55 |
| tsufiev | btully, it looks good when switching from one panel to another and almost the same when doing hard reload | 20:56 |
| *** gokrokve has joined #murano | 20:57 | |
| tsufiev | i think i should look into that issue with a clear head | 20:57 |
| btully | the only other options i can think of would be to explicitly disable the buttons | 20:57 |
| btully | actually i don't think they are buttons | 20:58 |
| tsufiev | sjmc7, i suspect other standard dashboards don't have such problem with buttons because they load faster (no pictures, no special fonts) | 20:58 |
| *** gokrokve_ has joined #murano | 20:59 | |
| btully | yeah they are not buttons, they are links | 20:59 |
| sjmc7 | what's not getting loaded fast enough? | 20:59 |
| tsufiev | so, it's virtually impossible to press the button before page is loaded - because they loaded just a nano-second before loading is complete | 20:59 |
| tsufiev | perhaps, it's App logos | 21:00 |
| btully | which is what we want, right? | 21:00 |
| sjmc7 | i don't understand why slow images would make buttons not work | 21:01 |
| btully | the problem is not that the buttons don't work | 21:01 |
| *** gokrokve has quit IRC | 21:02 | |
| *** gokrokve_ has quit IRC | 21:04 | |
| tsufiev | sjmc7, the bug happens for sure, ativelkov could confirm that. it occurs when the button (link rendered as button) is pressed, the required action is to draw modal form, but the java-script for handling modals is not yet loaded, so the wrong template is rendered | 21:16 |
| sjmc7 | nova must have the same problem | 21:17 |
| tsufiev | another way to solve it - is to provide another 'Add App' wizard template, for non-modal (non-ajax) requests | 21:18 |
| tsufiev | but it would look ugly | 21:18 |
| sjmc7 | what i am really getting at is - this seems like it should be something that has already been solved | 21:18 |
| sjmc7 | and also something that would very rarely affect real users | 21:19 |
| sjmc7 | since the javascript is tiny and cached | 21:19 |
| tsufiev | second point is definitely right | 21:19 |
| tsufiev | but because of this it may not be solved at all :) | 21:20 |
| sjmc7 | what's the error that happens? | 21:20 |
| tsufiev | anyways, I'll look search for horizon LP for that kind of bug | 21:20 |
| tsufiev | TemplateNotFound | 21:20 |
| sjmc7 | could we make the error nicer, if fixing it properly is very hard? | 21:21 |
| sjmc7 | "An error occurred because the page was slow to load - please try again" ? | 21:21 |
| tsufiev | simplest way is to prevent clicking the buttons for a little while | 21:21 |
| tsufiev | IMHO | 21:21 |
| tsufiev | it would be less irritating | 21:22 |
| sjmc7 | it'll be super irritating when i'm testing and can't click :) | 21:22 |
| tsufiev | hm... maybe we are irritated by different things :) | 21:23 |
| btully | right but if we prevent clicking, we should visually indicate that, which was what I was attempting to do | 21:25 |
| btully | otherwise yes, very annoying if i see a button and keep clicking on it while nothing happens | 21:25 |
| btully | tsufiev: on your dev server, do you not have js aggregated? | 21:26 |
| tsufiev | btully, no | 21:26 |
| tsufiev | I need to check whether js was packed on ativelkov machine who initially filed that bug on Safari | 21:27 |
| btully | so it sounds like a performance issue. maybe assets are being blocked due to non-aggregation | 21:28 |
| tsufiev | perhaps | 21:28 |
| btully | and because js is taking long to load, the modal system isn't fully initialized | 21:28 |
| btully | like you mentioned above | 21:29 |
| tsufiev | yep | 21:29 |
| btully | so even a JS solution wouldn't be helpful | 21:29 |
| tsufiev | will check on monday | 21:30 |
| * tsufiev leaves to sleep | 21:30 | |
| btully | so one way might be to explicity hide (or make invisible) the quick eploy and add to env buttons via css | 21:31 |
| btully | and then unhide them with js | 21:31 |
| btully | but kind of a wonky experience | 21:31 |
| btully | there's also a css "pointer-events" attribute that you can set to none, which prevents clicking, but it's not 100% supported in all browsers yet | 21:33 |
| btully | but given how much Horizon appears to use the modal system, i'm surprised this bug hasn't come up before | 21:34 |
| btully | i'll stop talking to myself now :) | 21:36 |
| btully | have a great weekend | 21:37 |
| *** btully has quit IRC | 21:37 | |
| sjmc7 | you too | 21:37 |
| *** gokrokve has joined #murano | 21:39 | |
| *** sjmc7 has quit IRC | 22:53 | |
| *** TravT has quit IRC | 23:36 | |
| *** gokrokve has quit IRC | 23:50 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!