[07:20:27] joakino: phuedx \o [07:32:32] o/ [08:01:20] bmansurov: you still up or in uzbekistan? [08:01:38] jdlrobson: up, my schedule has changed recently [08:01:49] owch :) early no? [08:02:08] jdlrobson: yeah, but haven't stayed up late, just started the day ;) [08:02:15] hah [08:02:17] smart [08:02:33] how's the vacation going? [08:02:38] anything i can help with? I was thinking of starting the task in todo but given im off tomorrow didnt want to step on any toes :) [08:03:12] yeh it's been great. very chill. I've been good at keeping away from the inbox and relaxing :) [08:03:17] jdlrobson: i'm preoccupied with the references patch, so feel free to pick up the task. I dont think others are done with their tasks yet either. [08:03:26] jdlrobson: great! [08:03:43] bmansurov: can also help out wiht the other tasks but not sure whether that's helpful or not [08:04:09] jdlrobson: joakino was having a hard time understanding what nrz wanted in the lazy loading of images task [08:04:18] jdlrobson: maybe you can help him clarify [08:04:41] sure [08:04:42] jdlrobson: i also replied to the references patch if you want to take a look [08:14:18] bmansurov: sure [08:26:20] yo [08:32:11] yo yo yo phuedx [08:36:24] hey jdlrobson [08:36:32] reading-web-staging-3 is in a funny state again [08:36:34] just rebooting [08:36:53] bmansurov: i think you're right about the error on staging [08:37:29] just updating everything to get rid of that error and i'll try again [08:37:29] (re: ie9 icons) [08:38:00] phuedx: i have the same problem locally, but couldn't figure out how to fix yet. spent quite a bit time playing with imagemagic and what not [08:40:23] bmansurov: possible authmanager updates? [08:41:38] phuedx: no, i don't have it installed [08:41:56] bmansurov: it's in core right? [08:42:45] phuedx: is it? anyhow, i haven't updated core recently I think, and I have had this problem for a long time, since the days when I started testing on opera mini [08:46:07] provisioning staging [08:48:11] updating staging [08:57:59] bmansurov: ok, i think i meed to ping tgr about this [08:58:14] en/en.m work fine but vagrant/labs-vagrant fail [08:58:29] something's up with the local sessionmanager config /maybe/ [08:58:45] phuedx: ok, i'd like to know the solution too, i'll ask you later once you talk to tgr [08:59:33] it looks as if the error is tracked here: https://phabricator.wikimedia.org/T127233 [09:00:08] ah tgr! speak of the devil and he doth appear :D [09:00:11] ;) [09:00:33] the power of phone notifs [09:00:53] phuedx: i think we have the same problem but different causes, my error message doesn't talk about auth manager [09:00:53] TURN THEM OFF [09:00:58] THEY ARE EVIL [09:01:08] give me a sec while I figure out how to scroll on this thing [09:01:15] oh good lord [09:01:25] it's not that urgent, i'll ping you in the task! [09:01:31] go back to whatever you were doing [09:01:31] ok, till later [09:02:00] tgr: bmansurov is seeing an error locally that i can reproduce on our staging server [09:02:06] the staging server has just been updated [09:02:11] sec [09:02:19] http://reading-web-staging.wmflabs.org/w/load.php?modules=skins.minerva.icons.images&image=mainmenu&format=rasterized&lang=en [09:02:21] ok, I'll look at it in an hour [09:02:32] "sessions are disabled for this entry point" [09:02:44] but the error doesn't occur in prod [09:02:50] kk [09:02:58] * phuedx copypastas the scrollback [09:20:56] hi! [09:21:08] a lot of activity today! [09:23:10] jdlrobson: i'm lost with https://phabricator.wikimedia.org/T135430#2378825 , animations block on scroll? [09:23:40] joakino: hey [09:23:43] it seems that way [09:23:56] i've not found any evidence of it [09:24:00] but transition appears much smoother [09:24:20] medium are also using visibility 0s linear 0s,opacity .4s 0s :) [09:24:38] might be an idea to reach out to a dev there and seeing if they can help us out [09:24:56] jdlrobson: i haven't seen any problems, is that what the original but report was about/ [09:24:57] ? [09:25:19] It's an ios specific problem so unfortunately i cant verify [09:25:24] you need real device [09:25:30] jdlrobson: i was just trying to confirm that the last change (which changes the UX of the lazy loaded images) would get signoff from nzr [09:25:34] actually i lie.. you can see it on chrome android [09:25:49] just scroll but dont release your finger [09:25:55] and the render will not happen until you lift it [09:26:16] aha [09:26:31] may that be because of the debouncing instead of throttling? [09:26:45] possibly... [09:30:31] joakino: so what seems to be the problem to be is that we trigger a keyframe animation when the loaded class is present [09:30:44] only according to http://developer.telerik.com/featured/scroll-event-change-ios-8-big-deal/ painting (any DOM changes) are paused via scroll - but this seems to include so this would include adding a class to the element [09:31:14] if the keyframe is on the element without the class then it will trigger straight away before the image hasn't loaded [09:31:32] so i'm thinking that's why they went with transition http://codepen.io/jmperez/pen/yYjPER [09:32:21] but yes debouncing is maybe playing some part in this and worth investigating [09:32:56] but the goal is that we should be showing the same animation as we currently are but it should also kick in during a scroll if that makes sense? [09:33:22] jdlrobson: oki, I get all this, but it seems like a thing for a followup bug or investigation, not what this bug says needs to be done [09:33:39] maybe something got lost in translation but this was the bug i was trying to get fixed [09:33:51] feel free to update description if necessary [09:34:06] i folded several bugs in there [09:34:10] not sure if you read the original bug reports [09:34:46] aha [09:35:05] i'll re-read and decide something about the task [09:37:37] thanks joakino [09:37:43] im wrapping up for day [09:37:46] anything needed before i go? [09:37:53] nopes, have fun [09:38:03] :D [09:41:49] jdlrobson, joakino: hopefully the commit i linked to explains why we originally switched to animations [09:42:06] i gave bman some more feedback [09:42:10] thunderstorm starting :D :D :D [09:42:53] tgr|away: vs Volker_E today in the Euros! :D [09:43:18] not sure if tgr|away cares about football though :) [09:44:24] joakino: any ideas what happened wth enabling lazy loaded images on Farsi wiki? [09:44:48] jdlrobson: sprint planning and product owner priorities I guess [09:45:01] don't remember anything special being mentioned [09:45:15] i sent a mail to try and get it back on the agenda as that was the one thing I wanted to happen while I was out :) [09:46:15] I just saw it, I know dr0ptp4kt is aware and I'm sure he'll fit it in if there's bandwidth [09:46:28] 👍 [09:46:28] sweet that makes me happier :) [09:56:50] 1 [15:06:20] joakino: my audio is broken. [15:06:29] joakino: i'll restart my browser [15:06:32] o noo [15:06:34] haha [15:06:38] i'm there waiting :) [15:56:05] joakino: uh oh, went black! [15:56:15] dr0ptp4kt: yeah here too [16:00:25] nzr: will we see you at the web prioritization ceremony? Web misses you... [16:00:34] maxbinder: yes :) [16:00:41] huzzah [16:58:37] phuedx: we will remain in the prioritization hangout [17:02:21] we will remain! [17:28:17] mdholloway: was there any crucial reason for the ZeroRatedMobileAccess varnish cleanup (that may have caused the bug) or just hygiene? [17:29:01] jhobs: that was just hygiene, the real reason i was poking around was to try to get rid of the need to call the mobile domain in order to get carrier tagging [17:29:16] jhobs: which it sounds like is going to need more discussion as well [17:29:22] mdholloway: it may be best to lean on the safe side and revert that then [17:29:36] jhobs: but yeah, the change seems like the most likely suspect so i'll get a patch up to revert it. sorry about that. [17:29:39] mdholloway: I commented on the task, but I do remember ZRMA special page doing way more than what it should [17:29:56] mdholloway: no worries; your reasoning WAS solid after all. It's the code that's not ;) [18:40:29] jhobs: dr0ptp4kt: sure enough, reverting fixed it [19:50:18] bearND: hey there, have you by any chance been running into 403s lately (i.e., today) when making content service requests? [19:50:58] bearND: "Error: 403, Insecure Request Forbidden - use HTTPS" [19:51:24] bearND: not every time, just sometimes (even for different tries of the same request) [19:51:41] bearND: i think prod is fine, just a dev/beta issue [20:21:51] mdholloway: thx for the update on the vcl [20:22:18] tgr, mdholloway bearND would you please update https://etherpad.wikimedia.org/p/Scrum-of-Scrums (even if it's "no update")? [20:22:47] dr0ptp4kt: sure, sorry about that! [20:22:55] mdholloway: np, thx [20:22:58] dr0ptp4kt: (gotta click that link before i close the hangout...) [20:24:49] jhobs: did you have a chance to look at the conversation in my patch? [20:27:48] dr0ptp4kt: boom, done [20:28:03] mdholloway: thx [20:34:22] dr0ptp4kt: updated sos etherpad [20:34:38] bearND: thx [20:34:55] mdholloway: I see 403 when running npm test [20:52:05] bmansurov: sorry, let me do that right now [20:52:11] bmansurov: got distracted [20:52:19] jhobs: thanks [20:52:29] jhobs: any tips to get wikidata working locally? [20:52:43] jhobs: i'm trying to test your patch, but have to use vagrant ;( [20:53:10] bmansurov: unfortunately, no. My plan of action thus far has just been to manually hard code in a description just to test [20:53:27] jhobs: ok [20:53:29] bmansurov: since none of the changes actually change how the code grabs from wikibase [20:54:46] mdholloway: looks like an intermittent backend issue. On different test runs I had different tests fail. To reproduce I had to refresh on lead section request 5 times. [20:55:17] bearND: yeah, i got similar results [20:57:36] mdholloway: sounds like varnish to me: "via cp1052 frontend, Varnish XID 1317010381
Error: 403, Insecure Request Forbidden - use HTTPS" [20:58:23] bearND: the error message seems to come from this file, and (consistent with the error message) it looks like there were some changes recently to tighten up around non-https requests (particularly from beta labs): https://github.com/wikimedia/operations-puppet/blob/production/modules/varnish/templates/vcl/wikimedia-frontend.vcl.erb [21:00:53] reets: o/ hello! would it be possible to get that fancy lightning bolt in the new zeplin "type C" mocks as an SVG? i can't seem to find it in the google icons. i can trace it if that's not possible [21:08:36] bmansurov: I chimed in, although I'm not sure I helped too much :/ [21:08:52] jhobs: thanks, i'll wait for others to express themselves too [21:09:19] bmansurov: yeah, I'd specifically love to hear phuedx's opinions on it [21:09:37] we'll get him to speak, he promised [21:10:04] mdholloway: hahah I just read your comment on the task :D [21:10:17] er... the zero task [21:22:23] hi niedzielski: I just updated the zeplin with slices for the varC screen. If you go to the "Assets" tab now on the the righthand-side menu you can now download the 4 icons as svg... also you'll notice they were updated to the white on circular background style after a consult with Nirzar this morning :D [21:22:30] jhobs: lol [21:23:14] mdholloway: with most backend stuff, adam & yuri are your guys, although I certainly had to read enough of it myself to be able to help out sometimes. Frontend I've got you covered. [21:23:17] reets: nice! thank you!! [21:23:34] jhobs: gotcha [21:32:19] joakino: hey... robson mentioned there is a way to do the animation while the page is scrolling. [21:32:42] i talked to jkatz also. i think its worth trying to fix and go with option A of yours. [22:50:34] bearND: what are you thinking about the service patches? [22:52:07] mdholloway: let's go with yours [22:52:55] bearND: ok, sounds good