[02:34:30] jdlrobson: I have a question regarding https://gerrit.wikimedia.org/r/#/c/226919/ , Let me know whenever you can answer. [14:52:01] Krinkle_: hey, you gotta sec to talk Gerrit->GitHub? [15:39:43] hashar: can you add me as an admin on GH? [15:39:45] plz :-) [15:40:11] also coreyfloyd [15:42:37] or bd808 if you wouldn't mind inviting me to the "owners" group [15:42:59] or bearND.. i'll just keep going down the list ;-) [15:43:44] bgerstle: do you need full owners or a group that owns select repos? [15:44:07] bd808: i need admin perms for the iOS repo, at least [15:44:11] so i can enable travis [15:44:34] so, not sure if you want to tweak the "mobile" group to do that or add me to "owners" [15:44:50] There is https://github.com/orgs/wikimedia/teams/ios too [15:44:59] bd808: 404 [15:45:09] try now [15:45:11] i only see 2 teams [15:45:19] haha that's an ooold [15:45:19] team [15:45:32] bd808: the repo it has isn't even maintained anymore [15:45:43] so we should just nuke it [15:46:18] i'll make a separate ticket for removing that repo [15:46:28] bgerstle: k. I'll grant you into admin and you can clean things up [15:46:38] bd808: much appreciated [15:46:49] bd808: would you be so kind as to add coreyfloyd too? [15:46:57] (and mhurd if it's not already) [15:47:05] might as well get the whole team ready to go [15:48:24] bgerstle: {{done}} for you and cory. Use your superpowers wisely :) [15:48:59] also if you don't have 2-factor enabled on your github account yet, now would be a good time to do that [15:49:15] bd808: http://24.media.tumblr.com/tumblr_mddzufjPer1ri1mxoo1_500.gif [15:49:19] bd808: already done [15:50:17] bd808: thank you sir! [16:00:03] hey mobile apps teams, please add anything you think is relevant to this afternoons reading showcase/demo: https://etherpad.wikimedia.org/ReadingShowcase [17:02:57] jdlrobson: there is a lot that needs to be done still [17:03:12] jdlrobson: but it is good to see there is interest in testing :-} [17:03:22] I am off [17:15:19] dbrant|brb: mdholloway coreyfloyd bgerstle standup [17:16:48] brt [17:28:29] mbinder: are we doing retro now or on friday? [17:28:49] bgerstle: my understanding is we cancelled the one that was moved [17:29:00] mdholloway: actually, do you want to hash out the search-order bug in batcave? [17:29:04] right, i see one for now on my cal though [17:29:10] dbrant: sure [17:29:12] bgerstle: sorta like taking a prescription :) [17:29:15] dbrant: mdholloway not sure if you saw my comment about the "index" field [17:29:23] bgerstle: should say "CANCELLED" on it [17:29:34] mbinder: ahhh [17:29:35] was cut-off [17:29:38] got it! [17:29:39] thanks [17:29:46] 👍 [17:35:57] joakino: jdlrobson rmoen bmansurov bearND mdholloway got any goodies for https://etherpad.wikimedia.org/p/ReadingShowcase ? rough demos and even non-user-y things welcome! [17:37:42] mbinder: are we having retro now? [17:39:04] coreyfloyd: see above :-) [17:39:14] oh right [17:49:00] dr0ptp4kt: whoops, i put my demo under last month's meeting... [17:54:18] mdholloway: language selection: the past, present, and future of wikimedia [18:07:00] agh kristenlans i am trying so hard to RSVP appropriately to the Showcase and can't seem to get gcal to cooperate. I'll be out for this one, but in for the future ones! [18:10:23] bgerstle: What up [18:10:31] Krinkle: hey! [18:10:33] niedzielski: since you are making Jenkins changes, I think it would be good to change the links from Gerrit apps-android-wikipedia-gradlew to go to the status page instead of the console page (basically remove the console part of the URL) [18:10:50] Krinkle: was hoping you could help me confirm that Gerrit => GitHub replication is disabled for the ios repo [18:11:05] do you know where that switch is? [18:11:11] bgerstle: I don't think that exists [18:11:17] I mean, in theory, it can be disabled [18:11:19] there is a config for it [18:11:20] bearND: oh like the job status? [18:11:25] but the repos shouldn't diverge [18:11:38] if GitHub is ahead, Gerrit will refuse to sync [18:11:59] niedzielski: yes, I think it's more convenient if you are looking for a checkstyle issue to use the link on the status page [18:12:08] bearND: sounds good. i'll take a note [18:12:10] Krinkle: ok, cool, i would like to change the config as well just to be sure [18:12:15] https://github.com/wikimedia/operations-puppet/blob/production/manifests/role/gerrit.pp [18:12:23] niedzielski: thank you [18:12:38] Krinkle: that's for the gerrit role itself.. what about the replication config? [18:12:47] quick question: if you cherry pick something preserved, is that a cherry pickle? [18:12:53] that's part of it [18:13:01] niedzielski: 😑 [18:13:11] bgerstle: couldn't be avoided [18:13:16] lol [18:13:48] Krinkle: i see that "github" is one of the slaves [18:14:04] but i don't see how we can disable replication for a specific repo [18:14:21] Ask ^demon / chad for the syntax there, we used to have it disabled for some in the past [18:14:23] w/o a separate manifest for iOS [18:14:33] we don't need another manifest.. do we? [18:14:58] However I think he'll recommend the same as me which is: Disable access in Gerrit ACL so that nobody can touch the repo, and once GitHub diverges, replication will refuse naturally in the event someone does do someting to that repo [18:17:00] Krinkle: ok, i guess it's fine then. it'll be up to us to make sure we don't push to or merge patches in gerrit [18:17:11] (instead creating a representative patch in GH once approved) [18:17:15] bgerstle: I can disable access to that gerrit repo if you want me to [18:17:28] Just say the word :) [18:17:29] Krinkle: that's OK, i've tweaked the ACLs before [18:17:30] thanks though [18:17:35] OK. So it's fine then [18:17:47] Krinkle: to be clear, you're saying i would remove push access [18:17:55] Krinkle: although, preventing even Jenkins from merging would be great [18:17:59] so if we +2 something, it doesn't get merged [18:18:01] Yes [18:18:07] Just remove it and make it exclusive [18:18:16] Admins can't push to a repo without access either [18:18:30] in theory an admin can grant themselves access, but it's not granted by default [18:19:18] k, jenkins too? [18:20:13] Jenkins has access by default, but if your acl is [x] Exclusive [18:20:15] then it won't [18:26:25] Krinkle: https://phabricator.wikimedia.org/T106966 < is that likely to be an issue in MobileFrontend or core? [18:26:38] i'm unaware of us knowingly doing any mangling :-/ [18:26:52] any theories? [18:26:59] jdlrobson: Neither, it's just wmf-config [18:27:15] ah ok. What's the variable? Want me to submit a change to fix it? [18:27:22] wgLoadScript [18:27:33] k i'll take a look now [18:28:03] ah it uses wgServer.. interesting [18:28:11] does it need to? [18:28:54] I'm not sure. It depends. [18:29:13] Krinkle: ok i'll investigate. Leave it with me. [18:29:42] jdlrobson: If we can guarantee that all host names we serve mobile page views out of also have a multiversion-supported wgScriptPath with load.php, then we don't. [18:55:10] atgo no worries [19:01:26] kristenlans: do you have a link to the reading showcase hangout? [19:01:36] bmansurov: in calendar invite [19:01:39] dr0ptp4kt: i don't see link for video for showcase [19:01:43] (in decsription) [19:01:55] mhurd: bmansurov in decription, since it's a hangout on air [19:01:56] kristenlans: thanks [19:02:45] kristenlans: the youtube link says not started yet [19:03:00] mhurd: that is correct we are not on air yet :-) [19:03:11] mhurd: if you will be participating, use the hangout lik [19:03:15] kristenlans: ah sry [19:03:29] mhurd: if you are a spectator, you tube s fine [19:04:03] kristenlans: i wasn't planning to speak, didn't want to crowd the hangout [19:06:23] kristenlans: i can drop off to youtube if needed [19:06:37] kristenlans: k thx [19:07:34] kristenlans: ECHOCHAMBERNIGHTMARE! [19:12:26] moushira youtube stream success? [19:12:59] yup kristenlans :) [19:21:00] "19:17 robh: etherpad was giving errors, apache restart fixed" [19:32:59] bmansurov: are you able to create new pages on mobile btw? For me the next button is permantely disabled [19:33:10] let me check [19:33:56] jdlrobson: yes, locally on master [19:34:06] oh dear [19:34:11] i'll raise a bug [19:34:17] cache maybe? [19:34:37] just to be clear, yes, I'm able to create new pages [19:36:11] kristenlans, so this current demo isn't coming on stream :) [19:36:32] bmansurov: how do you create them if next disabled? [19:36:36] Moushira audio, video...or everything? [19:36:42] jdlrobson: next is not disabled [19:36:50] once i start typing [19:36:53] kristenlans, audio yes,but screen isn't shared. [19:37:08] oh yeah she stopped sharing :-) [19:37:09] oh okay. mm.. i guess this could just be a jon issue then.. but it has me worried! [19:37:31] jdlrobson: yeah, i think the user has to enter a body to create a page [19:37:50] moushira she stopped sharing pretty quickly, only had demo up for 1 minute-ish [19:41:49] kristenlans, it seems the youtube stream is not sharing the screen [19:42:02] oh snap rmoen [19:43:10] rmoen: is it bahas demo right now? [19:43:30] kristenlans, maybe you just need to "share with all" whomever is presenting..on hangout [19:43:34] its a silhouette of a non-pictured user [19:43:42] dr0ptp4kt ^ ? [19:43:57] because it shows in the thumbnail, just needs to be on fullscreen [19:44:44] seems fixed now [19:45:56] rmoen: did *all* demos not show? we were generally using present to everyone [19:46:29] dr0ptp4kt: approximately 50% didn't show [19:46:43] dr0ptp4kt: its as if someone clicked on a user thumb and it locked on [19:47:15] rmoen: bummer [19:47:26] yeah no worries, just alerting [19:48:30] dr0ptp4kt: I think it started w/ Sherah's demo as reported by moushira [19:48:33] maybe [19:50:13] kristenlans: rmoen - thanks. we'll have to just pause in the midst of the showcase if needed in the future [19:56:01] Sorry rmoen moushira :-( [19:56:51] no worrie kristenlans ..we are all remoties :) [19:59:30] kristenlans: video, with part after 35:07 removed available here for posting https://www.youtube.com/watch?edit=vd&v=CEipt68-k2g [19:59:41] kristenlans: may take a while for the system to encode the video [20:00:01] dr0ptp4kt: kewl thanks [20:26:02] kristenlans: i'm going to actually create a shorter one. without video, it's not so useful. so i think the new one will be like 24 minutes long :( [20:30:22] kristenlans: final link - https://www.youtube.com/watch?edit=vd&v=UM9YwsAMC_o we'll have better luck next time. we'll create luck if we need to. https://www.youtube.com/watch?edit=vd&v=UM9YwsAMC_o [20:31:25] dbrant: interestingly, now the api sandbox is reliably returning "obama" as the first redirect for our testing query, even though "obama crush" is reliably the first result when searching in the app [20:32:05] dbrant: even though we both saw the first result fluctuate rapidly with our very eyes this morning. [20:37:02] mdholloway: in the sandbox, make sure to set the number of results to 20 (gpslimit), which is what we have in the app. It's reliably returning "obama crush" for me. [20:37:50] dbrant: aha. i had it at 16. what did we have as the limit earlier this afternoon? [20:38:06] mdholloway: 16, but I just picked that arbitrarily. [20:38:38] mdholloway: it seems it's dependent on that, as well. [20:39:33] dbrant: looks like 17 is the where we reach the point of fluctuation now [21:25:53] mdholloway: can you try my patch again but clear data in the app first? [21:26:09] bearND: sure [21:29:10] Krinkle: https://github.com/wikimedia/wikipedia-ios/pull/16#issuecomment-125346223 [21:29:18] mdholloway: niedzielski dbrant|brb bearND ^ [21:29:27] dr0ptp4kt: ^ [21:29:45] bgerstle: very nice [21:29:53] nice! [21:29:59] someone mentioned codecov on our phab ticket, so i used that since it's all open source [21:30:02] even supports gitlab [21:30:05] bgerstle: congrats. What is "prs"? [21:30:08] [slow clap] [21:30:12] it's another setting for travis [21:30:24] it can build prs, or the merge commit from that pr [21:30:28] AFAICT [21:30:30] it's a bit confusing [21:30:44] ah, plural of PR [21:30:46] yeah [21:34:01] Kristen|break: https://www.youtube.com/watch?edit=vd&v=UM9YwsAMC_o is live now [21:40:36] Whoa haha dr0ptp4kt my giant head is front and center on that video ;-) [21:40:54] :) [21:40:58] Also I have resting happy face [21:41:29] ^ oops, forgot to switch my nick back. [21:41:34] dr0ptp4kt: mediawiki page is updated https://www.mediawiki.org/wiki/Reading#Reading_showcase.2Fstaff_meetings [21:41:47] kristenlans: cool. thanks [21:46:14] bearND: OK, no crashing after clearing data, except after selecting one of the load strategy options, which I think is expected for purposes of this patch [21:47:11] mdholloway: noooo, it should not crash [21:47:26] bearND: hmm. [21:48:09] bearND: it crashes with each selection. [21:48:37] mdholloway: hmm, that's weird. I'm going to check it out from Gerrit [21:53:50] mdholloway: hmm, weird. Not sure why I didn't notice the crash earlier. Thanks! I think I'm going to revert this back to before I introduced the ListPreference. If you want I can hide the old HTML boolean so there is only one boolean flag left. [21:56:01] bearND: yeah, I think that's fine, this has turned into a bigger change than I anticipated it being. [21:56:33] bearND: maybe, if there's space, note in the HTML entry description something like (overrides JSON selection)? [21:59:35] bearND: That's if you leave both boolean flags, that is. Or, you could just hide the HTML boolean as you said, which might make more sense if it's not going to be in active development in the near future in any case [22:00:06] mdholloway: yeah, I'm just commenting out the html pref in the dev settings [22:00:12] and leave the rest as is [22:00:30] bearND: sounds good to me [22:04:46] @dbrant|brb you around? [22:12:11] bmansurov do you have EventLogging installed? [22:12:26] jdlrobson: no, it [22:12:29] 's disabled [22:12:49] Tjat [22:12:55] that's what is breaking my editor [22:13:07] i see [22:13:07] i don't know why... i guess the default settings do not work [22:14:40] bmansurov: could you enable and see if you can replicate? [22:14:47] jdlrobson: yes, i'll do so shortly [22:14:54] thanks bmansurov i appreciate it [22:14:58] also marxarelli you around? [22:15:10] yep yep [22:15:13] marxarelli: https://phabricator.wikimedia.org/T106878#1486341 [22:15:17] any thoughts around this? [22:15:31] oh fun [22:15:35] seems logging in is sometimes taking too long on mobile - do you think we should replace this step with an api call? [22:15:54] and have explicit tests for login workflow [22:17:00] bearND: for ripple effect, you mean just for API 21+? [22:17:13] looks like the webdriver request is timing out [22:19:10] marxarelli: :( [22:19:18] jdlrobson: mw-selenium supports a `browser_timeout` setting to increase the http timeout for webdriver connections [22:19:25] mdholloway: yes [22:19:30] let me see if the jjb builder is using it [22:19:36] bearND: cool, just checking [22:22:42] marxarelli: thubms up [22:23:38] rmoen: can you replicate https://phabricator.wikimedia.org/T107075 ? [22:26:15] mbinder: hey, sorry. what's up? [22:27:11] jdlrobson: re logging in via the api, there's a wip for that [22:27:31] jdlrobson: https://gerrit.wikimedia.org/r/#/c/226950/ [22:28:27] dbrant Mind taking a look at this epic and seeing if it covers tickets that meet the android team phab needs? https://phabricator.wikimedia.org/T105404 [22:29:27] We're trying to reach critical mass and deliver some improvements [22:29:50] mbinder: sure, will do! [22:30:10] Feel free to make new tickets (or find existing ones) and add them as blockers to the epic [22:37:20] marxarelli: cool. I'll keep an eye on that bug [22:37:37] marxarelli: just debugging another issue with browser tests [22:50:16] * rmoen just read the staff handbook word for word and is reminded how amazing it is to work for such an organization is. [22:50:31] (minus) the last is [23:13:47] coreyfloyd: in https://phabricator.wikimedia.org/T106106 iOS Release 4.1.7, there is: Release version: 4.1.7 and Release tag: 4.1.7.168 [23:14:11] coreyfloyd: I have beta 168 and alpha 170 - which of them will be released? [23:16:09] rmoen: bmansurov may want to fix https://phabricator.wikimedia.org/T107090 before the cut off [23:16:11] any takers? [23:21:09] also https://phabricator.wikimedia.org/T107104 [23:25:43] jdlrobson: i'll take the first one [23:25:54] bmansurov: awesome [23:26:21] jdlrobson: i was unable to reproduce https://phabricator.wikimedia.org/T107090 locally [23:26:47] rmoen i seem to be able to only reproduce it when running the search.feature browser test [23:26:48] it's super weird [23:27:02] jdlrobson: ok i'll look into it [23:27:06] further [23:27:22] i'm just running through beta labs on my phone [23:27:27] seeing if i can find anything broken [23:28:14] jdlrobson, rmoen: i see it on master locally [23:28:16] article page [23:33:35] jdlrobson: this patch (https://gerrit.wikimedia.org/r/#/c/224194/) caused the regression [23:33:51] jdlrobson: reverting it will bring back the chrome bug [23:33:53] bmansurov: really? mm that's weird [23:34:03] bmansurov: to last modified? [23:34:08] or to the clear in pages [23:34:12] jdlrobson: clear in pages [23:34:16] bmansurov: lol oh so any text is then shown ;/ [23:34:19] ah okay i thought you were looking at last modified [23:34:46] no, the last modified bug is the second one above ;) [23:36:09] it's logical. The text indent isn't big enough I guess. Damn browser bugs. [23:36:14] when will they end! [23:36:27] i'm not sure which is the more evil bug. [23:36:39] jdlrobson: i think we can limit that rule to the back button only [23:36:41] since i was only one who could reproduce the chrome issue - i could have a look and see if there's a better fix [23:36:50] bmansurov: yeh that might be a good compromise [23:36:55] bmansurov, my thoughts too [23:37:00] i'll look at the modifed bug [23:44:05] jdlrobson, rmoen: https://gerrit.wikimedia.org/r/#/c/227374/ [23:47:34] bmansurov: seems reasonable to apply the fix to the element affected. Hopefully we don't have another disappearing icon in the header in the future. [23:48:02] rmoen: agreed [23:48:36] jdlrobson: can you advise what page https://phabricator.wikimedia.org/T107104 ? i've tried using the html for the last edited bar locally and experienced no issue.