[00:00:49] cwdent: other times I've had differences between Jenkins and locally it's had to do with lack of network connectivity [00:00:52] Fundraising-Backlog, Recurring-Donations: drush command does not handle some exceptions, and incorrectly returns success - https://phabricator.wikimedia.org/T107380#1493620 (atgo) NEW [00:00:53] on Jenkins [00:01:29] hrm. i redefined $.ajax in there but maybe i missed something [00:01:35] Fundraising Tech Backlog, Fundraising-Backlog, Recurring-Donations: drush command does not handle some exceptions, and incorrectly returns success - https://phabricator.wikimedia.org/T107380#1493620 (atgo) [00:01:52] cwdent: maybe something to do with different copies of $ because of scoping issues, so the test is actually making a real ajax call, which works locally but not on Jenkins? [00:02:14] maybe try running locally in Special:JavascriptTest and watching the dev tools network tab? [00:02:49] cwdent: also remember you can debug into the tests from that special page by passing the debug=true URL param, which de-minifies the scripts so you can use the browser's debugger on 'em [00:03:26] K family food time calls! Bye everyone, see you soooooon! [00:04:00] K4-713: ejegg|away: Doing a payments config deployment, hopefully minor... [00:04:52] woo [00:05:04] Fundraising Tech Backlog, Fundraising-Backlog, Recurring-Donations: Recurring Globalcollect gateway_txn_id is incorrect - https://phabricator.wikimedia.org/T107382#1493642 (atgo) NEW [00:05:39] cool AndyRussG|souper i will look into it. have a good night! [00:05:59] cwdent: thanks likewise! :) [00:06:57] Fundraising Tech Backlog, Fundraising-Backlog, Recurring-Donations: Redirect old PayPal IPN URL to active listener - https://phabricator.wikimedia.org/T107384#1493671 (atgo) NEW [00:07:29] fundraising-tech-ops: Temporary read access to all frack redis boxen - https://phabricator.wikimedia.org/T107386#1493688 (awight) NEW [00:08:09] !log deployed interesting gc-cc-limbo config [00:08:16] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log, Master [00:09:01] !log crazy previous message was all about: I pointed the DonationInterface frontends to mirror limbo messages to a Redis server on localhost. [00:09:07] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log, Master [00:10:04] Fundraising Tech Backlog, Fundraising-Backlog, Recurring-Donations: Don't HTML escape watchdog logs - https://phabricator.wikimedia.org/T107387#1493705 (atgo) NEW [00:10:20] !log rolled back config [00:10:26] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log, Master [00:10:28] oops [00:10:50] Fundraising Tech Backlog, Fundraising-Backlog, Recurring-Donations: RGC batch size config is being overridden in the Jenkins job config - https://phabricator.wikimedia.org/T107388#1493712 (atgo) NEW [00:10:50] whahappened? [00:12:37] Fundraising Tech Backlog, Recurring-Donations: DonationInterface / batch mode / drupal is repeatedly trying to do geoIP lookups on 127.0.0.1 - https://phabricator.wikimedia.org/T106199#1493731 (atgo) p:Triage>Normal [00:13:02] K4-713: hehe, I need to deploy libraries first. [00:13:06] Wanna cr... [00:13:15] https://gerrit.wikimedia.org/r/#/c/227908/1 [00:13:39] :p [00:13:51] awight: Sure, if you read this thing I'm gonna send you in a second. [00:14:04] consider it shaken upon [00:15:06] Just shared a thing. [00:15:39] Fundraising-Backlog, Recurring-Donations: Spike: Investigate GC subscriptions that just kind of end - https://phabricator.wikimedia.org/T107390#1493749 (atgo) NEW [00:16:13] awight: But you're, like, test-failing [00:16:23] I'd CR but it wouldn't help. :p [00:16:53] Fundraising-Backlog, Recurring-Donations, Epic: Amazon recurring - https://phabricator.wikimedia.org/T107391#1493759 (atgo) NEW [00:17:02] bah [00:17:08] not even my tests... [00:17:35] They shouldn't have to be. :D [00:18:53] Fundraising-Backlog, Recurring-Donations: Create alternate recurring schedules - https://phabricator.wikimedia.org/T107392#1493772 (atgo) NEW [00:19:41] Fundraising-Backlog, Recurring-Donations: Add a record in Civi when a recurring contribution fails on GC side - https://phabricator.wikimedia.org/T107393#1493779 (atgo) NEW [00:55:59] I... can no longer run MW tests [00:56:15] I think that ability is at home, sick [00:59:21] oh dear. [00:59:41] * K4-713 looks relieved [01:00:19] So... what's up with these tests? [01:00:43] Anyone remember dealing with a thing, where MW phpunit.php doesn't see any of the DonationInterface tests? [01:01:05] I'm hoping the tests are just fallout from some of the Astropay patches. There were some messages renamed... [01:01:11] --group DonationInterface damage? [01:01:15] yeah [01:01:18] empty [01:01:24] I've had this before... [01:01:25] hargh [01:01:34] Can't remember if the solution is to go home, or throw computer at wall [01:01:54] or make quiet sad noises in my kewb [01:02:01] Q-b [01:04:49] (CR) Katie Horn: "Just a couple inline comments..." (2 comments) [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/227908 (owner: Awight) [01:05:56] (CR) Awight: Clean up Redis and slayer stuff (2 comments) [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/227908 (owner: Awight) [01:14:01] awight: I think the last time I had that, was on the 6th floor. [01:14:06] Which makes no sense. [04:08:14] (CR) AndyRussG: "If you prefer, I guess we could get rid of them in this patch. I'd been thinking of keeping all config variables sent to the client unchan" [extensions/CentralNotice] (campaign_mixins) - https://gerrit.wikimedia.org/r/227647 (https://phabricator.wikimedia.org/T98747) (owner: AndyRussG) [16:04:21] hey, I have evidence that some of you are really good at git branches. Anyone know how to push two dependent feature branches to github, without the second branch's commits also showing up in the PR for the first branch? [16:09:13] awight: dang, I'm not very familiar with github. Last time I tried to do two pull requests, I had to force my github repo back to the older commit, make a PR, then push the 2nd up and make another PR [16:10:10] bah. I would totally do that, but the branches really are dependent [16:17:39] * AndyRussG waves [16:17:55] Hey awight ejegg! [16:18:55] hi AndyRussG [16:19:03] :) [16:20:28] awight: what do you think about the comment I put on the un-landingpage patch? If you still think it's best to remove those globals in this patch, I can do so right away... the other option would be just including them in a general unused-global-removal patch once we're sure a full rollback won't be needed (there are more newly-unused-but-left-in-but-marked-for-removal globals in the big patch, too) [16:22:39] AndyRussG: either way! My logic, fwiw, was that I'm very confident that nothing is using the feature, and adding TODOs is not really reducing the complexity, it's increasing it in a way. But that's just cos I have a phobia of keeping details in my head :) [16:23:15] Let's just merge what u have here, then fully remove later, that's fine too! [16:23:34] awight: cool! [16:23:54] I actually didn't mean to post a CR-1 but when it happened, I thought to myself it sort of did reflect the spirit of my comment ;) [16:23:57] Heh I hope it's all un-sitebreakage-ful [16:24:16] np! Totally valid of course :) [16:32:55] Hi ejegg. The Brazil banners just went up. Let me know if you see anything funny [16:33:06] looking good so far! [16:33:26] Plenty of donations started [16:35:07] hi [16:35:25] I'm trying lingo but I have no history on the chat :/ [16:35:30] good morning everyone [16:35:41] I see you! [16:35:55] morning Dstrine :) there are logs here - http://bit.ly/1eJqqCc [16:35:56] heyo [16:36:03] Dstrine: don't worry, i never have chat history :) [16:36:06] dang, getting a few API call timeouts even at 12 seconds [16:36:12] https://kmpunksays.files.wordpress.com/2012/12/tumblr_m7y8aex1td1qfy2kdo1_r2_500.gif?w=830 [16:37:03] Good morning! yes, donations seem to be coming in from Brazil [16:37:55] AndyRussG: i'm not having any luck running these qunit tests from bash, but per the gruntfile it looks like it's just going to do the same thing i've been doing manually, loading localhost jstest special page in chrome [16:38:04] where the tests pass for me, but they fail on jenkins [16:38:24] https://www.youtube.com/watch?v=WEWG6kSYqlY [16:38:55] i am looking into this https://github.com/axemclion/grunt-node-qunit [16:41:18] the-wub_: I'm seeing a bunch of form chooser errors that looks like they're getting a null currency [16:41:39] b1516_0716_ptbr_p1_lg_amt_10.no-lp.cc [16:42:40] ejegg: um, is there not a 'dsk/mob/ipd' in there? [16:43:03] nope, that's the whole utm_source [16:43:52] B1516_0716_enBR_p2_sm_amt_20.no-LP.bt shows up on another one of the form chooser errors [16:44:27] ah nevermind. looks like we missed that out. investigating... [16:45:43] those banners haven't changed since the last test though. so not sure why there would be a new issue [16:46:10] dang, hope I didn't screw something up in the code... [16:46:56] hrm, DonationInterface does some utm_source munging--should there be a fourth term? [16:47:13] We're definitely assuming only three terms, separated by dots [16:49:02] cwdent: hmmm... K I think eventually you should get 'em working on the command line, but since they work for my on the command line, getting that working locally probably won't help u too much [16:50:07] yeah, i will figure it out [16:50:58] the-wub_: sorry, disregard my utm_source comment--I see what you're talking about now, the device name should show up in the banner name. [16:51:29] cwdent: whoa funny! They work for me on the command line but not in Special:JavaScriptTest/qunit?module=ext.centralNotice.geoIP [16:51:40] Lemme debug 'n' pastebin [16:52:04] oh weird! yeah lemme get a look at that [16:52:10] awight: yeah it is usually in there. but jessica or I must have forgotten it for these desktop banners (we only started adding it for desktop recently) [16:52:33] cool, not a code issue then [16:52:46] cwdent: https://tools.wmflabs.org/paste/view/3cc5e20b [16:53:21] heh, definitely different errors than jenkins [16:53:30] but it's a good lead, thanks! [16:53:58] ejegg: I can't reproduce a null currency error with these banenrs. tried chrome/firefox/IE. how many are you seeing? [16:54:54] eek. 105 so far [16:55:19] oh dear. think we should pull the test? [16:55:46] oops, sorry, most of those are not Brazil [16:56:20] still 33 errors [16:56:32] Jeff_Green: Yubikey time? [16:58:40] the-wub_: looks like a bunch of different banners too, so the problem is probably not on that end [17:00:29] ejegg: weird. well we've had over 900 clicks from the Brazil banners, so 33 is a relatively small portion. I'm happy leaving this test up for the planned hour [17:00:42] k, i'll keep investigating [17:02:05] ejegg: I can't reproduce the null currency error, even w/ the B1516_0727_jaJP_dsk_p1_lg_txt_yuk banner [17:02:37] ??? so odd.. all of the 'Attempting to set a valid form' log lines in astropay have currency:null [17:03:17] something's not getting logged right, at the very least [17:03:29] Jeff_Green: ^ remind me if we can get to the payments apache access log? [17:04:45] awight|afk: not programmatically because apache still doesn't do access logs to syslog [17:05:02] awight|afk: it would be a post parameter, too [17:05:19] ejegg: ah, you are right [17:05:44] dang, another rash of API timeouts [17:06:44] ejegg: actually--weirdly enuf, we pass currency_code in as a GET and a POST param [17:06:58] o rly? [17:07:44] gotta look like I'm eating breakfast, brb [17:20:25] huh... 15 of the ppl who initially got those form chooser errors went on to get through the process [17:25:03] ejegg awight|afk: I think there has been some confusion over this in the past. we should be using POST for everything we send to the form chooser? think there's some legacy banner code using GET, which I'll have a go at cleaning up [17:25:34] the-wub_: I'm starting to think those log lines may be spurious [17:26:38] ejegg: okay that's good to know [17:26:51] trying to see if the proportion of people who get past the first click is the same with or without that error [17:27:15] test is over in 5 minutes btw. results look pretty good on my end [17:30:04] hmm, looks like getting that error makes you half as likely to get past the first click? methodology is totally ad-hoc [17:30:34] meganhernandez, the-wub_, ejegg, awight : test is down. Thanks guys! [17:30:59] thanks jessicarobell ! [17:31:24] thanks jessicarobell ejegg awight|afk! [17:32:48] ejegg: can you email me some of the error log lines? I'll see if I can spot anything common in the banners [17:32:51] got an entirely new error message from AP: 'Could not make the deposit' [17:33:11] the-wub_: sure, but it was 8 or so different banners [17:34:41] the-wub_: oh man, there are even a couple with utm_source donate and Waystogive. Banners are almost definitely not the cause [17:35:57] ah okay ejegg. thanks for investigating, lmk if there is anything I can do to help! [17:36:54] I don't see anything obvious in the code... [17:40:54] awight: the really odd thing is that it says ffname 'astropay' is invalid. Assigning ffname '' [17:42:03] yah! [17:43:39] ejegg: oh, when the currency is null, GatewayFormChooser::isValidForm( $ffname, $country, $currency, ... will return false [17:44:31] awight: I dunno if we can trust currency:null in those log lines though [17:44:50] look at all the 'Attempting to set valid form' lines [17:45:17] seems consistent [17:45:43] just about all of them have currency:null, and most don't get to the 'is invalid' error [17:46:53] Perhaps those are rescued by $this->session_getLastRapidHTMLForm(). that logs at debug level, unfortunately. [17:48:53] yeah, here's an example where that must have happened: 27139744 [17:48:54] hah, getLogDebugJSON has been wrong since forever [17:49:09] it's funky, but how is it wrong? [17:49:25] 'currency' should be 'currency_code' [17:49:54] that is a fact [17:51:10] (PS1) Ejegg: Fix currency code debug logging [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228028 [17:52:31] (PS2) Awight: Fix currency code debug logging [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228028 (owner: Ejegg) [17:52:48] (CR) Awight: [C: 2] "Thanks!" [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228028 (owner: Ejegg) [17:53:21] (CR) jenkins-bot: [V: -1] Fix currency code debug logging [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228028 (owner: Ejegg) [17:53:45] oh. right, I was in the middle of trying to diagnose the spurious test failures. [17:53:54] I think it's related to the msg renaming [17:54:15] But still at step zero, getting my phpunit to discover mw-extension test cases [17:56:30] speaking of spurious test failures... [17:56:40] maybe caffeine will help [18:14:15] (PS1) Awight: Fix name conflict with mw-core phpunit [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228038 [18:14:42] (CR) jenkins-bot: [V: -1] Fix name conflict with mw-core phpunit [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228038 (owner: Awight) [18:14:55] (PS2) Awight: Fix name conflict with mw-core phpunit [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228038 [18:15:17] (CR) jenkins-bot: [V: -1] Fix name conflict with mw-core phpunit [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228038 (owner: Awight) [18:15:34] ahh hell, fragile translation-dependent tests [18:15:52] I got this one... [18:16:00] oh? [18:16:19] or at least--just a moment ago I became able to run the tests at all :) [18:34:07] ejegg: Checkin? [18:34:10] ejegg: Wanna CR my queue patches? [18:34:16] I WON. [18:34:19] K4-713: o sorry, one sec! [18:34:36] K4-713: i o u a coke [18:34:46] heh [18:35:08] I'll just take it off of the list of a thousand cokes I owe you. [18:36:20] coke on credit... now that's a friend :p [18:42:50] Fundraising-Backlog: Change error messages on new GC red text to match field name - https://phabricator.wikimedia.org/T106172#1495558 (atgo) [18:56:26] (PS1) Awight: Set language globals from more tests [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228049 [18:56:44] ejegg|mtg: ^ [18:57:55] awight: huh, that's annoying to have to do [18:58:29] *really* annoying, but I'm not sure if we should be working around it [18:59:09] like, in the minimal case we'd be testing getErrorMessage, and it really does rely on global state [19:00:27] (CR) Ejegg: [C: 2] "Thanks for the fix!" [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228049 (owner: Awight) [19:00:43] awight: k, i'll take a look at that queue stuff in a sec [19:05:31] K4-713 dstrine-hiding lunch? [19:05:34] i'm hungrrrry [19:05:41] hi! [19:05:53] yeah sure [19:06:09] (Merged) jenkins-bot: Set language globals from more tests [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228049 (owner: Awight) [19:06:54] (PS3) Awight: Fix currency code debug logging [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228028 (owner: Ejegg) [19:07:25] (PS3) Awight: Clean up Redis and slayer stuff [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/227908 [19:07:33] (PS3) Awight: Fix name conflict with mw-core phpunit [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228038 [19:07:42] (PS8) Awight: Orphan slayer reads from frack Redis [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/226947 (https://phabricator.wikimedia.org/T99017) [19:08:12] (CR) jenkins-bot: [V: -1] Orphan slayer reads from frack Redis [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/226947 (https://phabricator.wikimedia.org/T99017) (owner: Awight) [19:17:59] ejegg: fyi, I want to deploy the cleanup patch first, which gives us the Predis library. Then I'll wait at least an hour before trying any crazy stuff with the real change [19:19:03] sounds like the way to go [19:19:45] gotta populate the queue before reading from it... [19:27:05] (PS4) Awight: Fix currency code debug logging [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228028 (owner: Ejegg) [19:28:19] (CR) Awight: "recheck" [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228028 (owner: Ejegg) [19:28:34] (CR) Awight: [C: 2] Fix currency code debug logging [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228028 (owner: Ejegg) [19:28:59] (Merged) jenkins-bot: Fix currency code debug logging [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228028 (owner: Ejegg) [19:43:18] MBeat: hey! I'll make sure to give you notice before I try deploying this beast. [19:43:34] thanks awight [19:53:04] (CR) Ejegg: [C: -1] "$this-warning isn't a thing, is it?" (1 comment) [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/227908 (owner: Awight) [19:53:33] lol thanks again [19:53:48] heh [19:54:16] (PS4) Awight: Clean up Redis and slayer stuff [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/227908 [19:57:33] AndyRussG: https://lists.wikimedia.org/pipermail/wikitech-l/2015-July/082547.html [19:58:05] (PS5) Ejegg: Clean up Redis and slayer stuff [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/227908 (owner: Awight) [19:58:26] awight: thx! [19:58:55] (CR) Ejegg: [C: 2] Clean up Redis and slayer stuff [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/227908 (owner: Awight) [19:59:19] (Merged) jenkins-bot: Clean up Redis and slayer stuff [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/227908 (owner: Awight) [20:00:14] ejegg: thx again. I'll deploy this right after standup, and wait for my weir to fill with fish [20:00:49] heh, nice [20:02:21] awight: so, writing to one redis server will replicate to all just via the redis configuration, nothing we have to do in code? [20:30:54] ejegg: It's actually incredibly fishy... https://phabricator.wikimedia.org/T103206 [20:31:05] The idea is that it's robust against single machine failures. [20:31:23] payments1001-3 (DI frontends) write to a redis queue on localhost [20:31:52] payments1004 (slayer) reads from each of those, round-robin, and if communication fails, that server is kicked out of the pool for the duration of the batch run. [20:31:53] Ahh, that works too! [20:32:12] So the read pool has all 6 in it? [20:32:41] The three masters are replicating to three slaves on payments1004, and to um "cold ready" or something failover cluster at payments2001-3 [20:32:57] ejegg: no, not yet. I wanted to KISS for this first iteration [20:33:19] but yeah, in the future we'll be fine reading from slaves and dealing with the duplicates [20:33:42] OK, so now in case of failure we still have the data, but we need to update config if we want to point to the backup [20:34:19] in the future? not sure, that would be a good way to go, though [20:34:21] hey awight, is the work you’re doing impacting geolocation on donate.wikimedia.org? [20:34:30] ccogdill: nope. Are you seeing anything funny? [20:34:34] just wondering… I’m getting country=XX and that was weird [20:34:36] or no so funny [20:34:38] ooh [20:35:17] ccogdill: consistently? [20:35:21] cwdent: hi! do let me know if you feel really stuck on the geoIP tests, if maybe there's something I can help with too [20:35:24] :) [20:35:29] ccogdill: can you open the debugger and email me your cookies? [20:35:42] consistently on chrome awight, but firefox just worked so maybe it is a cookie thing [20:36:04] AndyRussG: yeah i might do that, i've got a few more ideas [20:36:07] ejegg: one weakness is that anyone mid-session will lose their data if their frontend goes down [20:36:51] hmm [20:38:21] ejegg: I'm holding out for Redis Cluster, which is part of the next major release. It will do everything we want, transparently. [20:38:44] cleared the cookie awight. sorry I didn’t investigate that before! [20:38:48] It'll just look like we're hitting a single server, and can reprovision its own nodes [20:39:01] ccogdill: rats--next time, save that cookie for us! [20:39:07] sounds highly preferable! [20:39:27] ccogdill: the fact that you got XX at all was definitely caused by a bug [20:40:03] ohhh sorry…. it was called GeoIP [20:40:09] drat, sorry again! [20:42:13] no worries! I'm sure it'll happen again [20:42:22] fingers crossed ;) [20:42:25] hahaha [21:05:03] (PS1) Awight: Follow composer.json: include Predis [extensions/DonationInterface/vendor] - https://gerrit.wikimedia.org/r/228145 [21:09:26] (PS1) Awight: Merge master into deployment [extensions/DonationInterface] (deployment) - https://gerrit.wikimedia.org/r/228146 (https://phabricator.wikimedia.org/T99015) [21:10:30] ejegg: Am I backwards? https://gerrit.wikimedia.org/r/#/c/228146/ we should figure that out... [21:10:42] there can be only one ordering! [21:11:23] (CR) Awight: [C: 2] Follow composer.json: include Predis [extensions/DonationInterface/vendor] - https://gerrit.wikimedia.org/r/228145 (owner: Awight) [21:11:26] oops, i hadn't even been paying attention to the order! [21:11:34] what command do you use again? [21:11:58] git cherry -v deployment master [21:12:28] I usually delete the plus signs, cos the negative patches in theory weren't part of the merge [21:12:41] ah, cool. [21:12:56] my command is longer, but doesn't require you to edit the output [21:13:15] git log --no-merges --oneline deployment..master [21:14:08] but, so... they are reversed relative to one another [21:14:16] I think we're the confuse [21:14:30] yeah, I agree we should pick one [21:15:11] well, yours is like git log and mines is like a timeline [21:15:36] guess yours reads more intuitively [21:15:49] That's the part where I really can't say [21:16:09] feel free to cheat. --reverse [21:16:12] and sometimes the short hashes are all numeric and gerrit doesn't link em right [21:16:15] Mine can't do that :) [21:16:23] hahaha really? what a POS [21:17:08] (CR) Awight: [V: 2] Follow composer.json: include Predis [extensions/DonationInterface/vendor] - https://gerrit.wikimedia.org/r/228145 (owner: Awight) [21:17:27] (CR) Awight: [C: 2] Merge master into deployment [extensions/DonationInterface] (deployment) - https://gerrit.wikimedia.org/r/228146 (https://phabricator.wikimedia.org/T99015) (owner: Awight) [21:20:43] (Merged) jenkins-bot: Merge master into deployment [extensions/DonationInterface] (deployment) - https://gerrit.wikimedia.org/r/228146 (https://phabricator.wikimedia.org/T99015) (owner: Awight) [21:20:47] (PS1) Awight: update DonationInterface submodule [core] (fundraising/REL1_25) - https://gerrit.wikimedia.org/r/228148 [21:20:58] Fundraising-Backlog: Change error messages on new GC red text to match field name - https://phabricator.wikimedia.org/T106172#1496237 (atgo) a:atgo>None [21:21:18] Fundraising Sprint The Pogues, Fundraising-Backlog, Astropay Integration, Unplanned-Sprint-Work: Add 'loading' indicator to AstroPay form - https://phabricator.wikimedia.org/T107321#1496239 (atgo) p:Triage>Normal [21:21:45] (CR) Awight: [C: 2] update DonationInterface submodule [core] (fundraising/REL1_25) - https://gerrit.wikimedia.org/r/228148 (owner: Awight) [21:22:14] good luck awight :) [21:22:32] ha, thanks. This should be the safe part... [21:24:59] Fundraising research, Research-and-Data: Review Finland banner testing strategies - https://phabricator.wikimedia.org/T105561#1496289 (atgo) [21:25:43] AndyRussG: i think i figured out what i'm doing wrong [21:25:51] cwdent: ah cool! [21:26:12] you can call setWindowGeo() multiple times now, but the promise is resolved the first time and stays that way [21:27:05] i'm going to try messing with that [21:33:20] AndyRussG: of course this is sort of the point of using a promise vs. a callback... i hate thwarting good code for the test [21:34:13] Hmmm [21:35:09] but yeah since mw.geoIP doesn't get burned down between tests it seems like you can really only run one test [21:36:10] Looking at https://gerrit.wikimedia.org/r/#/c/227889/3/tests/qunit/subscribing/ext.centralNotice.geoIP.tests.js... [21:37:39] Fundraising Sprint The Pogues, Fundraising-Backlog, MediaWiki-extensions-CentralNotice, Performance: Test for performance regressions - https://phabricator.wikimedia.org/T107503#1496359 (awight) NEW [21:38:22] fundraising-tech-ops, Traffic, operations, Patch-For-Review: Decide what to do with *.donate.wikimedia.org subdomain + TLS - https://phabricator.wikimedia.org/T102827#1496380 (CCogdill_WMF) Update from IBM/Silverpop: it turns out there is a way for us to use https without losing all of our click da... [21:40:25] cwdent: yeah! you should be re-instantiating mw.geoIP.deferred for each test. That's why it's publicly available there :) [21:40:54] oh! ok let me try that [21:42:28] !log updated paymentswiki from fd0060bf86777ee6b7acd205d134066356da69e8 to 6854683083cabc730f37b6a79d559f23e7ff7b0f [21:42:36] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log, Master [21:42:47] MBeat: ^^ that should be a nothing, but fyi [21:42:57] Looks safe so far. [21:43:05] thx [21:43:17] cwdent: also maybe move madeAjaxCall to a local private variable within each test function? [21:45:22] cwdent: also, does the geoIP ajax call lookup set a cookie? I have the feeling it doesn't, could be wrong... [21:45:33] AndyRussG: yeah my local copy looks nothing like that one anymore :-\ [21:45:50] doesn't varnish set the cookie? [21:45:54] Ah heh OK [21:46:15] cwdent: varnish sets the cookie on the initial pageview, i.e. with the request for the main HTML [21:46:34] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Spike: Clear up Contribution Type usage - https://phabricator.wikimedia.org/T101321#1496431 (atgo) [21:46:39] Then the ajax call sets window.Geo (by returning a script that says, "Geo = {..." ) [21:46:49] But I don't think it sets a cookie [21:48:34] hmmm, AndyRussG at the bottom of the .always() i see a cookie with a todo [21:49:43] cwdent: right, but that's the code under test, not the the ajax function you should be mocking out [21:50:17] oh yeah you're right [21:50:19] sorry [21:50:23] np! :) [21:50:32] i am starting to get cross eyed [21:50:53] heh don't worry it happens ;) [21:52:39] cwdent: I would like to be AFK for about 1/2 an hour sometime over the next hour and a half, but I can move my absence about anytime in that space... When do you think it'd be most productive for you for me to be around? right now? or a little later? [21:53:38] AndyRussG: oh do whatever you need to do, i can figure this out! [21:53:59] Mmm it's more or less the same over here :) [21:54:10] i'll be heading out about 5 but maybe back on later [21:54:20] 6 your time right? you are central? [21:54:43] cwdent: no I'm on EST, it's almost 6 now [21:54:54] It's 4 there now? [21:54:58] yep [21:55:00] MST [21:55:10] we're missing the CST contingent on our team! [21:55:28] Heh, when I was in Mexico that was me... [21:55:38] One of these days I'll have to make a matrix for what time zone everyone is in. [21:55:57] Heh then we'll have to choose red or green pill [21:56:02] crossed with, whether the person is nocturnal or crepuscular [21:56:15] buh [21:56:44] * awight sweeps all our confetti off the balcony [21:57:37] You forgot to add a confidence measure. [21:57:53] Actually... [21:58:27] ...could you just... combine gerrit and IRC logs to make a nice picture of when the person is usually active? [21:58:30] !log paymentswiki config: jiggle the handle [21:58:37] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log, Master [21:58:52] K4-713: also consider mediawiki.org edits [21:59:12] Actually, scratch that, hack our computers and install keyloggers [21:59:21] O_O [21:59:30] I'm picturing an output not unlike the spectra you get in astronomy when you're trying to tell what the stuff out there is made of. [22:00:11] Well... the stuff on fire, anyway. [22:01:09] u read about the megaaurora? [22:01:14] I did. [22:01:23] Wooooo [22:01:25] New holiday destination... [22:04:00] Fundraising Sprint Lou Reed, Fundraising Tech Backlog, Fundraising-Backlog, Continuous-Integration-Config, Patch-For-Review: mediawiki/core fundraising/REL branches should use git submodule - https://phabricator.wikimedia.org/T100637#1496497 (atgo) [22:04:14] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Compile release notes and gotchas - https://phabricator.wikimedia.org/T99841#1496498 (atgo) [22:05:04] Fundraising-Backlog, Wikimedia-Fundraising, MediaWiki-extensions-CentralNotice: BUG: Campaign date fields cannot be edited as text - https://phabricator.wikimedia.org/T97159#1496500 (atgo) [22:05:15] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM, Documentation: Document wmf_contribution_extra table - https://phabricator.wikimedia.org/T97190#1496505 (atgo) [22:05:29] scuze me while i blow up IRC [22:06:00] Fundraising-Backlog, Wikimedia-Fundraising, MediaWiki-extensions-DonationInterface: Display a button or link for manually switching country - https://phabricator.wikimedia.org/T96047#1496507 (atgo) [22:06:12] Fundraising-Backlog, Wikimedia-Fundraising, MediaWiki-extensions-DonationInterface: Display a flag and/or country name on the donation form - https://phabricator.wikimedia.org/T96046#1496508 (atgo) [22:06:33] Fundraising Sprint House of Pain, Fundraising Tech Backlog, Fundraising-Backlog: Backfill donor language information from logs? - https://phabricator.wikimedia.org/T94670#1496509 (atgo) [22:06:45] Fundraising Sprint House of Pain, Fundraising Tech Backlog, Fundraising-Backlog: Backfill donor language information from logs? - https://phabricator.wikimedia.org/T94670#1169480 (atgo) [22:07:14] Fundraising Sprint House of Pain, Fundraising Tech Backlog, Fundraising-Backlog: Backfill donor language information from logs? - https://phabricator.wikimedia.org/T94670#1169480 (atgo) hey @awight did this get taken care of? It feels like we should take care of this before too long [22:08:10] Fundraising-Backlog, Wikimedia-Fundraising: Make it easier to cancel recurring giving - https://phabricator.wikimedia.org/T94530#1496517 (atgo) Adding to the fr-tech workboard. [22:08:41] Fundraising Tech Backlog, Fundraising-Backlog: Put payments-initial feed on wikikiosk - https://phabricator.wikimedia.org/T94127#1496519 (atgo) [22:09:19] Fundraising Tech Backlog, Fundraising-Backlog, MediaWiki-extensions-DonationInterface: Give error feedback when iDeal cannot connect to bank - https://phabricator.wikimedia.org/T92567#1496521 (atgo) [22:10:22] Wikimedia-Fundraising, WMF-Design, Easy: Polish : Credit card input form should reflect organization of credit card - https://phabricator.wikimedia.org/T1189#1496526 (atgo) [22:10:29] Fundraising-Backlog, WMF-Design, Easy: Polish : Credit card input form should reflect organization of credit card - https://phabricator.wikimedia.org/T1189#1496529 (atgo) [22:10:57] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Keep track of online vs major gifts contributions - https://phabricator.wikimedia.org/T93286#1496530 (atgo) [22:11:08] Fundraising Tech Backlog, Fundraising-Backlog, MediaWiki-extensions-DonationInterface: See if we can log acquirer authorization information - https://phabricator.wikimedia.org/T92467#1496532 (atgo) [22:12:08] Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Stray comma causes all kinds of failures - https://phabricator.wikimedia.org/T90232#1054175 (atgo) [22:13:09] Fundraising Tech Backlog, Fundraising-Backlog, MediaWiki-extensions-DonationInterface, Patch-For-Review: Move GET/POST/COOKIE/SESSION assumptions into the UI controller - https://phabricator.wikimedia.org/T91670#1496589 (atgo) [22:13:29] Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Allow mismatched refunds - https://phabricator.wikimedia.org/T89437#1496603 (atgo) [22:13:43] Fundraising Tech Backlog, Fundraising-Backlog, Epic: Write puppet modules to provision all fundraising components - https://phabricator.wikimedia.org/T88658#1496610 (atgo) [22:13:56] Fundraising Tech Backlog, Fundraising-Backlog: Tracking: Fundraising Tech Staging - https://phabricator.wikimedia.org/T88657#1496611 (atgo) [22:14:16] cwdent: so... netime u want to send a WIP... I have more thoughts about what's in Gerrit, not sure if I should bother commenting [22:14:30] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising, I18n: Set up out-of-sync alerts so that the tech team immediately discovers translations ready for publishing - https://phabricator.wikimedia.org/T89910#1496627 (atgo) [22:14:31] Fundraising-Backlog, Wikimedia-Fundraising, Analytics, Analytics-Backlog: Provide performant query access to banner show/hide numbers - https://phabricator.wikimedia.org/T90649#1496625 (atgo) [22:14:34] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising, MediaWiki-extensions-CentralNotice: Untranslated CentralNotice banners should be flagged or forbidden - https://phabricator.wikimedia.org/T90634#1496626 (atgo) [22:14:38] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising, MediaWiki-extensions-DonationInterface: Client-side validation breaks credit card button - https://phabricator.wikimedia.org/T89412#1496647 (atgo) [22:14:40] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising, Wikimedia-Fundraising-CiviCRM: Improve Civi error reporting - https://phabricator.wikimedia.org/T89868#1496646 (atgo) [22:14:42] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising, MediaWiki-extensions-DonationInterface: Expiration date error message is wrong when date < now - https://phabricator.wikimedia.org/T89388#1496648 (atgo) [22:14:52] AndyRussG: sure! just a couple minutes if that's ok... [22:14:59] cwdent: you bet! :) [22:15:06] i'll just todo the stuff that's busted right now [22:15:31] I'm so crept out. The frontend limbo code is making lots of connections to Redis, but the total_commands_processed doesn't increment. [22:15:33] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising, MediaWiki-extensions-DonationInterface: Error popup is not translated - https://phabricator.wikimedia.org/T89376#1496703 (atgo) [22:15:35] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising, Wikimedia-Fundraising-CiviCRM: Make CRM version stamp thing robust on dev boxes - https://phabricator.wikimedia.org/T88376#1496708 (atgo) [22:16:19] MBeat: Looks like 2 or 3 people are retrying repeatedly... No problem reports yet? [22:16:38] no, nothing in Zendesk [22:18:56] Argh, I get nothing but errors [22:18:59] Wikimedia-Fundraising: Credit card form should not be replaced if we're already on the correct type of credit card form - https://phabricator.wikimedia.org/T35422#1496866 (atgo) @danielfriesen - have you tried this again/heard anything else from lastpass? if there's nothing we can do about it, we should clos... [22:19:07] Nothing in the log, though [22:19:23] rolling back the config... [22:20:04] looks like they go from Status 20 to 400 at GC [22:20:14] awight: Gateway page errored out due to: ERR operation not permitted ?? [22:20:45] hehe well, there's the log thx [22:21:04] right after Clearing transaction from limbo store [22:21:40] (PS4) Cdentinger: WIP: adds geoIP tests [extensions/CentralNotice] (campaign_mixins) - https://gerrit.wikimedia.org/r/227889 [22:21:48] !log rolled back paymentswiki config [22:21:55] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log, Master [22:21:57] AndyRussG: ^ totally broken right now :-\ [22:22:24] ejegg: oh, sorry those are creepy but were about 40 min ago [22:22:50] (CR) jenkins-bot: [V: -1] WIP: adds geoIP tests [extensions/CentralNotice] (campaign_mixins) - https://gerrit.wikimedia.org/r/227889 (owner: Cdentinger) [22:23:45] Fundraising Tech Backlog, Fundraising-Backlog, Astropay Integration: Diagnose AstroPay ffname errors - https://phabricator.wikimedia.org/T106050#1496890 (Ejegg) [22:23:51] MBeat: okay, looks like CC functionality is restored again. [22:24:10] thanks awight [22:24:42] awight: looks like you deployed 2 min before them though, right? [22:25:26] ejegg: That ERR is totally a redis error, which I'm accidentally letting go to the top level of the stack. I realized this is a simple password issue, btw... [22:25:38] oho! [22:25:41] What I'm confused about is why the ERRs stopped, though everything was still borken [22:27:21] awight: I don't see any other attempts to delete from limbo store till after the rollback [22:27:42] and that's a frontend, not the slayer! [22:27:55] ok, I'm trying this again with the password, but gonna make a TODO about the exceptions [22:29:20] Wikimedia-Fundraising: Credit card form should not be replaced if we're already on the correct type of credit card form - https://phabricator.wikimedia.org/T35422#1496909 (DanielFriesen) Open>Resolved a:DanielFriesen The issue doesn't appear to exist anymore. Either the donation form changed or Last... [22:31:37] !log redeploying paymentswiki config: with password this time [22:31:40] MBeat: trying again... [22:31:43] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log, Master [22:32:50] gah! [22:33:08] awight: yeah, odd - looks like nobody else came back to the resultswitcher for like 40 minutes [22:34:05] thanks awight - when it’s working normally, ping me & i’ll follow up with the donors [22:34:08] cwdent: taking a look! [22:34:19] !log paymentswiki: rolled back again [22:34:25] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log, Master [22:34:50] Works again. [22:34:58] Fundraising Tech Backlog, Fundraising-Backlog, MediaWiki-extensions-CentralNotice, I18n: zh-..., pt, and other language variants are not supported by CentralNotice - https://phabricator.wikimedia.org/T55641#1496954 (atgo) [22:35:02] awight: what was 'gah' ? didn't see anything wrong in the logs [22:38:46] Fundraising-Backlog, MediaWiki-extensions-CentralNotice: BUG: Campaign date fields cannot be edited as text - https://phabricator.wikimedia.org/T97159#1496981 (atgo) [22:39:30] ejegg: The logs r lying... [22:39:42] Fundraising research, Analytics, Analytics-Backlog: Provide performant query access to banner show/hide numbers - https://phabricator.wikimedia.org/T90649#1064105 (atgo) [22:39:43] oh? [22:39:48] The exception must be swallowed somewhere [22:40:04] I should probably fix that before firing myself by deploying this wrong again. [22:40:14] But here goes anyway, wheee! [22:40:25] Fundraising-Backlog, Wikimedia-Fundraising, Analytics: Public dashboards for CentralNotice and Fundraising - https://phabricator.wikimedia.org/T88744#1497015 (atgo) p:Normal>Low [22:40:31] srsly though, grep for resultswitcher in the gc log - big gap from 21:44 to 22:22 [22:41:11] oh, that's not what you say they're lying about [22:41:36] Fundraising Tech Backlog, Wikimedia-Fundraising, MediaWiki-extensions-DonationInterface: Error popup is not translated - https://phabricator.wikimedia.org/T89376#1497022 (atgo) [22:41:46] lying about false negative [22:41:53] Fundraising Tech Backlog, Fundraising-Backlog, I18n: Set up out-of-sync alerts so that the tech team immediately discovers translations ready for publishing - https://phabricator.wikimedia.org/T89910#1497027 (atgo) [22:41:58] ooh [22:42:14] !log paymentswiki: config the IIIrd [22:42:16] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Improve Civi error reporting - https://phabricator.wikimedia.org/T89868#1497030 (atgo) [22:42:20] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log, Master [22:43:01] Fundraising Tech Backlog, Fundraising-Backlog, MediaWiki-extensions-DonationInterface: Client-side validation breaks credit card button - https://phabricator.wikimedia.org/T89412#1497033 (atgo) [22:43:49] !log paymentswiki config rolled back [22:43:56] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log, Master [22:44:20] Well that was different. Jul 30 22:43:06 payments1001 globalcollect_gateway: PHP Fatal error: Cannot access self:: when no class scope is active in /srv/www/org/wikimedia/payments/extensions/DonationInterface/vendor/coderkungfu/php-queue/src/PHPQueue/Backend/Predis.php on line 241 [22:44:34] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising, MediaWiki-extensions-DonationInterface: Expiration date error message is wrong when date < now - https://phabricator.wikimedia.org/T89388#1497035 (atgo) This seems fine on the GC form. [22:44:36] ack! [22:44:41] Fundraising Tech Backlog, Fundraising-Backlog, MediaWiki-extensions-DonationInterface: Expiration date error message is wrong when date < now - https://phabricator.wikimedia.org/T89388#1497036 (atgo) [22:45:43] Fundraising Tech Backlog, Fundraising-Backlog, MediaWiki-extensions-DonationInterface: Error popup is not translated - https://phabricator.wikimedia.org/T89376#1497041 (atgo) [22:45:56] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Skip tests when modules not enabled - https://phabricator.wikimedia.org/T88379#1497042 (atgo) [22:46:01] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Remove tests from CRM deployment - https://phabricator.wikimedia.org/T88378#1497045 (atgo) [22:46:10] awight: sorry for blowing up your phab notifications... doing some cleanup [22:46:18] (PS1) Awight: Fix self:: style PHP version incompatibility [wikimedia/fundraising/php-queue] - https://gerrit.wikimedia.org/r/228160 [22:46:31] awight: ah, 5.3 closures... [22:46:43] how did you know! [22:46:48] atgo: thanks! [22:46:49] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising, Wikimedia-Fundraising-CiviCRM: Thank You module should default to using a dummy mailer - https://phabricator.wikimedia.org/T88377#1497047 (atgo) @ccogdill_wmf any thoughts? [22:46:55] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Thank You module should default to using a dummy mailer - https://phabricator.wikimedia.org/T88377#1497049 (atgo) [22:47:03] haha a lot of them are yours :) [22:47:17] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: rMake CRM version stamp thing robust on dev boxes - https://phabricator.wikimedia.org/T88376#1497057 (atgo) [22:48:13] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising, Wikimedia-Fundraising-CiviCRM: Make credible dummy data set for fundraising CiviCRM testing - https://phabricator.wikimedia.org/T88012#1497061 (atgo) [22:48:15] Fundraising Dash, Fundraising Sprint Miles Davis, Fundraising Sprint N*E*R*D, Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Generate fake data for Dash/CiviCRM - https://phabricator.wikimedia.org/T101904#1497062 (atgo) [22:48:31] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Queue consumer should warn about unknown fields - https://phabricator.wikimedia.org/T88339#1497063 (atgo) [22:48:59] Fundraising Tech Backlog, Fundraising-Backlog: Look at using Lua for donatewiki - https://phabricator.wikimedia.org/T87669#1497067 (atgo) [22:49:10] (CR) Ejegg: [C: 2 V: 2] "Much neater than aliasing $this and passing it into the use block!" [wikimedia/fundraising/php-queue] - https://gerrit.wikimedia.org/r/228160 (owner: Awight) [22:49:59] (PS1) Awight: Bump vendor php-queue version [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228162 [22:50:00] Fundraising Tech Backlog, Fundraising-Backlog, MediaWiki-extensions-DonationInterface, Patch-For-Review: Tracking: General DonationInterface code cleanup - https://phabricator.wikimedia.org/T87043#1497072 (atgo) [22:50:07] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: CiviCRM queries become slower over time, require manual "flushing" - https://phabricator.wikimedia.org/T87032#1497073 (atgo) [22:50:15] ejegg: hehe, neater than blood all over the elevator [22:50:23] Fundraising Tech Backlog, Fundraising-Backlog: Scrap custom fundraising logging, integrate with PSR-3 core stuff - https://phabricator.wikimedia.org/T87065#1497074 (atgo) [22:50:47] conveninent that we only needed a const [22:50:50] Fundraising-Backlog, MediaWiki-extensions-DonationInterface: Provide a country switching form on payments.wmo - https://phabricator.wikimedia.org/T86942#1497078 (atgo) [22:51:10] yeah faux real [22:53:36] (PS1) Awight: Merge master into deployment [extensions/DonationInterface] (deployment) - https://gerrit.wikimedia.org/r/228163 [22:53:57] AndyRussG: i have a PS that is closer to working but i'm not having any luck getting that deferred to reset [22:54:05] (PS1) Awight: update DonationInterface submodule [core] (fundraising/REL1_25) - https://gerrit.wikimedia.org/r/228164 [22:54:21] Fundraising Tech Backlog, Fundraising-Backlog: Scrap custom fundraising logging, integrate with PSR-3 core stuff - https://phabricator.wikimedia.org/T87065#1497084 (Ejegg) We've migrated to Monolog with a custom processor to add our contribution tracking prefixes and a custom handler to log the right way... [22:54:43] Ah one comment is that you need to do that in every test, not the setup [22:54:46] (CR) Awight: [C: 2] Bump vendor php-queue version [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228162 (owner: Awight) [22:55:03] So slooow [22:55:11] (Merged) jenkins-bot: Bump vendor php-queue version [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/228162 (owner: Awight) [22:55:28] (CR) Awight: [C: 2 V: 2] Merge master into deployment [extensions/DonationInterface] (deployment) - https://gerrit.wikimedia.org/r/228163 (owner: Awight) [22:55:39] (CR) AndyRussG: [C: -1] "Great start!! :D" (3 comments) [extensions/CentralNotice] (campaign_mixins) - https://gerrit.wikimedia.org/r/227889 (owner: Cdentinger) [22:55:52] cwdent: [22:55:56] ^ [22:55:56] (CR) Awight: [C: 2 V: 2] update DonationInterface submodule [core] (fundraising/REL1_25) - https://gerrit.wikimedia.org/r/228164 (owner: Awight) [22:56:19] Oh, nice! Our deployment branches are all hotwired sufficiently that we can merge w/o waiting for silly tests. [22:56:27] Wikimedia-Fundraising: campaign-vs-amount.csv has fundraising campaigns started in 1970-01-01T00:00:00 - https://phabricator.wikimedia.org/T86620#1497091 (atgo) p:Triage>Low That is super weird. [22:56:59] Fundraising-Backlog, Epic: [epic] Integrate processing in Brazil via Astropay - https://phabricator.wikimedia.org/T77907#1497094 (atgo) [22:57:37] !log updating paymentswiki from 6854683083cabc730f37b6a79d559f23e7ff7b0f to 02db5f7f77b667da06b882b2f66de9c5546230bc [22:57:43] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log, Master [22:59:00] !log redeploying sketchy paymentswiki config [22:59:03] AndyRussG: i have been messing around with variants of that but i can't get anything to work. have you tried programatically recreating the deferred object like that? it just seems to never resolve again [22:59:06] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log, Master [22:59:35] cwdent: hm! No I haven't tried it yet 8p [22:59:56] !log rolling back. paymentswiki. [23:00:02] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log, Master [23:00:53] * awight looks for teddy bear [23:01:47] Wikimedia-Fundraising: Credit card form should not be replaced if we're already on the correct type of credit card form - https://phabricator.wikimedia.org/T35422#1497117 (atgo) Thanks for checking! [23:02:33] Fundraising-Backlog, MediaWiki-extensions-CentralNotice, Browser-Support-Google-Chrome, Patch-For-Review: Multiselects in Special:CentralNotice sometimes load incorrectly - https://phabricator.wikimedia.org/T74421#1497120 (atgo) [23:03:38] Wikimedia-Fundraising: Following link to donate fails to return user to previous page upon completion - https://phabricator.wikimedia.org/T60506#614152 (atgo) [23:03:40] Wikimedia-Fundraising: Redirect donators back to the page they were originally looking at before donation - https://phabricator.wikimedia.org/T57354#1497128 (atgo) [23:04:13] Somehow didn't deploy my changes. /me changes own diaper [23:04:25] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Thank You module should default to using a dummy mailer - https://phabricator.wikimedia.org/T88377#1497144 (CCogdill_WMF) As in a replacement to Silverpop? I'm open to the idea if it will be robust and reliable. [23:05:13] AndyRussG: now i'm thinking the second test only ever passed when it was expecting the same state as the first, since the .when callbacks all fire immediately [23:05:27] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Thank You module should default to using a dummy mailer - https://phabricator.wikimedia.org/T88377#1497147 (CCogdill_WMF) I'd like to be involved in any scoping for a project like this; to me, it isn't worth the time unless... [23:05:51] i don't have a good explanation for why the promise doesn't get recreating like i'd expect [23:06:21] (CR) AndyRussG: WIP: adds geoIP tests (4 comments) [extensions/CentralNotice] (campaign_mixins) - https://gerrit.wikimedia.org/r/227889 (owner: Cdentinger) [23:06:37] (PS1) Krinkle: Remove use of ResourceLoader::makeLoaderConditionalScript() [extensions/CentralNotice] - https://gerrit.wikimedia.org/r/228167 [23:07:43] cwdent: would it make sense to set up a Hangout to go through this sometime, maybe tomorrow? what do you think? [23:07:51] ergh yeah i think it makes sense, all these callbacks are registered at the same time [23:08:06] so they all resolve with the first promise and the added one never resolves [23:08:10] Wikimedia-Fundraising, I18n: MediaWiki:Donate interface-error-msg-field-correction not properly translatable due to several grammatical genders - https://phabricator.wikimedia.org/T48395#1497155 (atgo) Hey @tolino am I reading this right that you were able to resolve it? [23:08:15] sure AndyRussG that'd be great [23:09:19] (PS1) Awight: Actually update php-queue library [extensions/DonationInterface/vendor] - https://gerrit.wikimedia.org/r/228169 [23:09:38] cwdent: cool! My day tomorrow is basically open, so anytime you like is cool! [23:09:46] (PS1) Awight: update vendor submodule [extensions/DonationInterface] (deployment) - https://gerrit.wikimedia.org/r/228170 [23:10:12] (PS1) Awight: update DonationInterface submodule [core] (fundraising/REL1_25) - https://gerrit.wikimedia.org/r/228171 [23:10:27] (CR) Awight: [C: 2 V: 2] Actually update php-queue library [extensions/DonationInterface/vendor] - https://gerrit.wikimedia.org/r/228169 (owner: Awight) [23:10:30] yep me too [23:10:49] (CR) Awight: [C: 2 V: 2] update vendor submodule [extensions/DonationInterface] (deployment) - https://gerrit.wikimedia.org/r/228170 (owner: Awight) [23:11:10] (CR) Awight: [C: 2 V: 2] update DonationInterface submodule [core] (fundraising/REL1_25) - https://gerrit.wikimedia.org/r/228171 (owner: Awight) [23:11:12] cwdent: hmmm how about mid-morningish? Or whenever you think is good in terms of helping you be most productive w/ this :) [23:12:17] !log updating paymentswiki from 02db5f7f77b667da06b882b2f66de9c5546230bc to d4bdce1cae168448b116d75e3dcd3303b0f13dd2 [23:12:23] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log, Master [23:12:27] AndyRussG: sounds perfect, i am usually available by 10 EST [23:13:10] AndyRussG: it looks like to achieve the test behavior we want it'd be resetting the deferred in nested callbacks [23:14:13] cwdent: huh I dunno. Let's take it 1 test at a time, and first isolate fully... then see how to fix that? [23:14:36] Fundraising Tech Backlog, Wikimedia-Fundraising: Special:GlobalCollectGateway without any parameters throws an internal error - https://phabricator.wikimedia.org/T42961#1497187 (atgo) [23:15:18] How is 11 EST/9 MST [23:15:25] ? [23:15:30] sounds good! [23:16:35] Fundraising-Backlog: Remove Amazon from forms - https://phabricator.wikimedia.org/T107528#1497196 (atgo) NEW [23:16:40] cwdent: great! just sent an invite :) [23:17:48] thanks! [23:17:55] i gotta run, talk to you then! [23:19:30] cwdent:yw, cya! [23:23:59] awight: fyi: [23:24:00] 23:22 < ebernhard> ori: pushed them all out...but i've left DonationInterface un-updated since it was just lingering in the deploy branch but not deployed. [23:24:03] that was during swat [23:24:24] a few minutes previously: [23:24:24] 23:17 < ebernhard> ori: there look to be a few extra commits from adam as well [23:24:27] 23:18 < ebernhard> well, hmm not sure ordering is a bit wonky here, should be fine [23:25:41] greg-g: thx. FWIW, the DonationInterface submodule can always be deployed to the main cluster. We use the "deploy" branch for paymentswiki stuff, but on the cluster we go out with the train. [23:25:49] I hope that's reflected in the release configuration? [23:26:12] awight: /join -operations, I'll paste what you just said here [23:26:24] On the main cluster, we only use DonationInterface for its translations, on donatewiki. [23:26:27] k [23:38:39] Fundraising Sprint The Pogues, Fundraising Tech Backlog, MediaWiki-extensions-DonationInterface, Unplanned-Sprint-Work: Some errors are not being logged, and are not caught in the right context - https://phabricator.wikimedia.org/T107531#1497277 (awight) NEW [23:41:49] (CR) Awight: [C: 2] "Thanks! This is currently unused, good to catch it proactively." [extensions/CentralNotice] - https://gerrit.wikimedia.org/r/228167 (owner: Krinkle) [23:42:48] (Merged) jenkins-bot: Remove use of ResourceLoader::makeLoaderConditionalScript() [extensions/CentralNotice] - https://gerrit.wikimedia.org/r/228167 (owner: Krinkle) [23:43:54] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Thank You module should default to using a dummy mailer - https://phabricator.wikimedia.org/T88377#1497298 (awight) My card wan't clear enuf, but I'm talking about the other end of the spectrum here. This is only on develo... [23:45:46] Fundraising Tech Backlog, Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Thank You module should default to using a dummy mailer - https://phabricator.wikimedia.org/T88377#1497309 (CCogdill_WMF) Oh, okay. Well that sounds good to me! [23:46:58] ejegg: Thanks again for that currency patch! [23:47:04] hehe [23:47:14] deprecating currency_code is probably worth its own task... [23:47:24] It's just too confusing [23:47:29] nice to take the scales from our eyes [23:48:40] what year is it?? [23:49:08] K4-713: it's about time we invested in a modern payments system... [23:49:34] How about paypal mini express mobile-only embedded iframe javascript button? [23:50:03] * K4-713 flips over a desk [23:50:39] SOrry, that wasn't a question, it was a statement. I have some serious Mickey Mouse mercenaries on it already [23:50:43] (╯°□°)╯︵ ┻━┻ [23:51:31] c'mon, it's fun. Flip, flip, flip... [23:51:49] On a somber note, I'm going to have to play with my kid pretty soon... as soon as I can stop babysitting this botched semideployment, in fact [23:52:15] "semideployment" is not the name of my first band. But it will be the name of our second album. [23:53:05] WE... almost made it. [23:53:26] You could tour like Primus, just do the parking lots of famous venues [23:53:58] If I were to say "Primus Sucks"... [23:54:07] ino [23:54:14] The Melvins suck even harder [23:54:18] :) [23:54:53] Well, we have actually paymentswiki s*it writing to Redis, finally. [23:55:00] This is the "fill the goblet" phase. [23:55:03] yay! [23:55:27] ok, i'll /actually/ look at the next phase patch now :) [23:55:32] My friend joyously shouted "Primus Sucks" at one of their shows in the last... 10 years, and some ignorant kid got all up in her face about "THEN WHY DID YOU COME TO THE CONCERT." [23:55:38] * K4-713 sighs [23:55:39] All the objects have an instrinsic 1 hour expiration, so it shouldn't fill up over the weekend. [23:55:50] lol, a learning experience [23:56:02] big sister for life [23:56:13] and quit playing with awk to calculate exactly how much suckier AP's response time is than our other processors [23:56:22] awww [23:56:25] they're little. [23:56:42] bahaha [23:56:43] they need to send away for the charles atlas kit [23:56:46] You guys. [23:57:03] After a few months of our broken testing, they can buy a php 5.3 server for all the load, on the subpenny layaway [23:57:06] globalcollect used to kick sand in my face! [23:57:13] 8D [23:57:16] They're still doing it. [23:57:23] :P [23:57:43] WP switch from sand to decomposed granite, to large gravel, finally they want to be nice [23:57:46] it's pretty sweet [23:58:20] like how you make friends with someone after clinging to the back of their SUV's ski rack for a few miles, and finally they just accept your doggedness [23:58:32] you have to make them respect youl