[00:01:43] (03PS4) 10Cstone: DI part of switching getHostedPaymentStatus to return a PaymentDetailResponse [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/724233 (https://phabricator.wikimedia.org/T291223) [00:03:32] (03CR) 10jerkins-bot: [V: 04-1] DI part of switching getHostedPaymentStatus to return a PaymentDetailResponse [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/724233 (https://phabricator.wikimedia.org/T291223) (owner: 10Cstone) [00:05:01] (03PS5) 10Cstone: DI part of switching getHostedPaymentStatus to return a PaymentDetailResponse [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/724233 (https://phabricator.wikimedia.org/T291223) [00:07:10] (03CR) 10jerkins-bot: [V: 04-1] DI part of switching getHostedPaymentStatus to return a PaymentDetailResponse [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/724233 (https://phabricator.wikimedia.org/T291223) (owner: 10Cstone) [00:09:48] ugh they pass by themselves but not together?? [00:16:13] (03PS6) 10Cstone: DI part of switching getHostedPaymentStatus to return a PaymentDetailResponse [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/724233 (https://phabricator.wikimedia.org/T291223) [00:18:23] (03CR) 10jerkins-bot: [V: 04-1] DI part of switching getHostedPaymentStatus to return a PaymentDetailResponse [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/724233 (https://phabricator.wikimedia.org/T291223) (owner: 10Cstone) [00:20:14] okay i see [00:20:46] (03PS7) 10Cstone: DI part of switching getHostedPaymentStatus to return a PaymentDetailResponse [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/724233 (https://phabricator.wikimedia.org/T291223) [00:22:34] (03CR) 10jerkins-bot: [V: 04-1] DI part of switching getHostedPaymentStatus to return a PaymentDetailResponse [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/724233 (https://phabricator.wikimedia.org/T291223) (owner: 10Cstone) [00:26:26] (03PS8) 10Cstone: DI part of switching getHostedPaymentStatus to return a PaymentDetailResponse [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/724233 (https://phabricator.wikimedia.org/T291223) [00:28:15] (03CR) 10jerkins-bot: [V: 04-1] DI part of switching getHostedPaymentStatus to return a PaymentDetailResponse [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/724233 (https://phabricator.wikimedia.org/T291223) (owner: 10Cstone) [00:28:48] alright when you get back ejegg|food it's mad about the return declaration being array but I cannot find where that is being set [01:01:42] starting the maintenance for fran1001 and frdb1003 [01:07:17] looking cstone [01:07:40] i dont know if it just needs to be set to stop it or what [01:08:46] arr, maybe it needs the new SmashPig library? [01:09:18] so I think it's because we're setting up the mock based on the version of HostedCheckoutPaymentProvider that's in vendor [01:09:22] and locally you' [01:09:25] ve got that updated [01:09:33] ooooh [01:09:34] but in CI it's still the last tagged version [01:09:45] yeah it doesnt break locally [01:09:53] anyway, the mock would have the same return type as the thing it's mocking [01:10:31] feel like tagging a SmashPig version with whatever's merged right now? [01:12:02] sure [01:13:02] both hosts are back up. [01:15:24] ok 0.7.6 is there [01:18:24] and ok yes my local tests break with old smashpig [01:20:31] cool dwisehaupt nice to hear it was the 5-min version [01:21:00] yeah. $daughter credits using her sonic screwdriver to fix it. [01:31:52] (03PS4) 10Ejegg: Use risk scorer for Ingenico hosted checkout status [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/735447 (https://phabricator.wikimedia.org/T291230) [01:31:54] (03PS1) 10Ejegg: Map tokens to RecurringPaymentToken field [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/735464 (https://phabricator.wikimedia.org/T291230) [01:32:31] cstone: so... the stupid part of our test setup is that we have to update our mediawiki vendor for real to get it to test on that version [01:32:47] yeah I was just sanity checking that it wouldnt explode [01:32:58] ah cool [01:33:17] right, lemme take a look at the changes since current deployed version too [01:34:26] so looks like your ingenico patch is the only one that DonationInterface codepaths should touch [01:35:03] cause the smashpig update would change the rawresponse there but then the DI part is still missing [01:35:06] which I think would break things? [01:35:31] cstone it sure would, if anyone deployed without your patch [01:35:53] oh ugh, but if we're running DonationInterface test on vendor we've got a chicken/egg issue [01:35:59] yeah [01:36:25] ok, we need to fix that CI dumbness [02:12:49] I guess we would start bumping the SmashPig version in DonationInterface's composer.json any time we had an incompatibility like this, then find the composer option to not complain on missing .git dir [02:13:09] or figure out how to publish a zip for each version bump [02:18:51] (03PS2) 10AndyRussG: Adyen Checkout: map error code 905 to submethod [wikimedia/fundraising/SmashPig] - 10https://gerrit.wikimedia.org/r/732803 (https://phabricator.wikimedia.org/T294032) (owner: 10Ejegg) [02:29:19] 10Fundraising-Backlog: Get composer.json in crm deployable - https://phabricator.wikimedia.org/T117898 (10Ejegg) 05Open→03Resolved [02:29:21] 10Fundraising Sprint William Shatner, 10Fundraising Sprint X-Ray Spex, 10Fundraising Sprint Yo La Tengo, 10Fundraising Sprint Zapp, and 3 others: [EPIC] Build Frig: tools to help update our deployment branches - https://phabricator.wikimedia.org/T116820 (10Ejegg) [02:30:01] 10Fundraising-Backlog, 10Composer: Use composer-merge-plugin for paymentswiki - https://phabricator.wikimedia.org/T141434 (10Ejegg) 05Open→03Resolved [02:30:45] (03PS1) 10Ejegg: Attempt to make SmashPig install from dist [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/735468 [02:31:19] cstone maybe going via github it will install from zip and stop complaining about the missing .git dir ^^^ [02:32:21] (03CR) 10jerkins-bot: [V: 04-1] Attempt to make SmashPig install from dist [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/735468 (owner: 10Ejegg) [02:32:26] oh derp, needs a non-ssh URL [02:32:52] err, wait, that wasn't an ssh URL [02:33:58] 10Fundraising-Backlog, 10Spike: Investigate possible composer strategies to make hygiene easier - https://phabricator.wikimedia.org/T154792 (10Ejegg) 05Open→03Invalid [02:34:22] weird, why did zuul try to clone it using ssh? [02:35:37] (03PS2) 10Ejegg: Attempt to make SmashPig install from dist [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/735468 [02:35:51] is it because I used type: vcs instead of type: git? [02:36:35] k, looks like there's another possibility: type:github and no-api:true [02:36:49] (03CR) 10jerkins-bot: [V: 04-1] Attempt to make SmashPig install from dist [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/735468 (owner: 10Ejegg) [02:37:37] (03PS3) 10Ejegg: Attempt to make SmashPig install from dist [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/735468 [02:38:50] (03CR) 10jerkins-bot: [V: 04-1] Attempt to make SmashPig install from dist [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/735468 (owner: 10Ejegg) [02:39:04] 10Fundraising-Backlog, 10Patch-For-Review: We should not lock composer deps to dev-master - https://phabricator.wikimedia.org/T133556 (10Ejegg) 05Open→03Resolved a:03Ejegg [02:39:59] i see it adding it at the top but then does it get overwritten ? [02:41:32] 10Fundraising-Backlog, 10FR-Adyen: Adyen new integration 2021 - https://phabricator.wikimedia.org/T268072 (10Ejegg) We can use this ticket for the update to the NEW new version... https://docs.adyen.com/online-payments/release-notes#releaseNote=2021-10-06-web-componentsdrop-in-5.0.0 [02:43:20] cstone yeah it checks out the existing vendor, then it does a composer update [02:44:22] ok, let's see how mediawiki/mediawiki-codesniffer is publishing .zip files [02:44:27] Installing mediawiki/mediawiki-codesniffer (v34.0.0): Extracting archive [02:45:27] hmm, pointing to github on the packagist page [02:45:51] well, I guess I can try doing that for smashpig [02:46:00] and hope it doesn't blow away everything! [02:47:36] ok, seems to have maintained all the stuffs [02:47:43] let's try rechecking your patch [02:48:05] (03CR) 10Ejegg: "recheck" [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/724233 (https://phabricator.wikimedia.org/T291223) (owner: 10Cstone) [02:49:35] oh maybe the composer.lock needs an update [02:51:07] ok, maybe we can do one more bogus update to get the composer.lock pointing to the zip [02:54:08] dang, I gotta get to bed [02:54:48] yeah its late for you! [02:54:50] and me hah [02:54:57] (03PS1) 10Ejegg: Update SmashPig [core] (fundraising/REL1_35) - 10https://gerrit.wikimedia.org/r/735470 [02:55:00] see ya! [02:58:10] (03CR) 10jerkins-bot: [V: 04-1] Update SmashPig [core] (fundraising/REL1_35) - 10https://gerrit.wikimedia.org/r/735470 (owner: 10Ejegg) [06:24:57] (03CR) 10jerkins-bot: [V: 04-1] Localisation updates from https://translatewiki.net. [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/735528 (owner: 10L10n-bot) [06:31:28] (03CR) 10Raimond Spekking: [C: 03+2] "false positive" [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/735528 (owner: 10L10n-bot) [06:33:59] (03CR) 10jerkins-bot: [V: 04-1] Localisation updates from https://translatewiki.net. [extensions/DonationInterface] - 10https://gerrit.wikimedia.org/r/735528 (owner: 10L10n-bot) [14:59:29] fr-tech we are trying to add Apple Pay in the Adyen console and it is asking for our domain. What should I enter? https://www.wikimedia.org/? [15:03:47] emartin i think probably payments.wikimedia.org [15:04:08] that's the one that our certificate is attached to [15:05:52] noted, thanks ejegg [15:06:48] https://www.irccloud.com/pastebin/Hf7e3ytW/ [15:07:20] ejegg, could it be another URL? [15:09:46] Thanks ejegg|away for troubleshooting. Feels like its a more extensive, Apple specific version of URL they want :) [15:10:27] they claim to have more info here: https://docs.adyen.com/payment-methods/apple-pay/web-drop-in#going-live [15:42:59] ejegg and fr-tech I have been on the phone with Adyen around Apple Pay. The console setup steps for Apple pay asks for the URL attached to the certificate, however, we get an error message stating that "Validation file missing, url: https://payments.wikimedia.org/.well-known/apple-developer-merchantid-domain-association". Manavi is troubleshooting this error on her end. She wasn't expecting that we needed a [15:42:59] validation file given we are using our own certificate. She had to go check with her colleagues re what is going on. She is telling me we are not live yet with Apple Pay, but I assured her we are seeing transactions in the transaction list and we have definitely pushed the code into production. She mentioned that we haven't sent captures for the Apple Pay transaction reference numbers I shared with her as [15:42:59] evidence. (Does anyone know why this is?). She went away to review this and we may get on another call. I may need FRtech support as I definitely wouldn't know what to do with a validation reference file if I saw one. [15:44:09] ^^ good progress! Thanks Evelyn and fr-tech [15:49:22] hi emartin and Ppena jack and elliott know the most there and are both off today but I can look and see where we put the well known file [15:54:55] emartin: the well known link is here https://payments.wikimedia.org/.well-known/apple-developer-merchantid-domain-association.txt [16:28:37] fr-tech here's the apple pay doc for Adyen: https://docs.adyen.com/payment-methods/apple-pay/web-component [16:28:55] thanks AndyRussG I was failing so much at their search haha [16:29:22] ah I had just had the tab open [16:29:38] I don't see anything there saying that ApplePay requires a separate capture step? [16:29:49] yeah same [16:30:03] remember the doc changes based on whether you click "use your own certificate" or not [16:30:31] they have also released a new version since [16:37:57] emartin: we don't see anything in the Adyen doc requiring a separate capture step just for Apple Pay [17:31:15] fr-tech. It seems the configuration issue was on the Adyen side. They had another step that Manavi was not aware of but she has now enabled Apple Pay on our account and I am seeing settlements. Here's hoping we are good to go. We will keep Apple Pay in the France banners over the weekend and seek to add it to 6ENC banner tests next week. Thanks for talking with me about this and hope the play was a success. [17:31:42] thanks emartin ! [17:41:12] yay! [17:43:17] emartin: great news! Hehe also thanks, my daughter's play went fine, was super short, part of a Halloween/Day of the Dead event here at here school [17:45:13] (I do have IRC on my phone and also brought my laptop, if something urgent does come up.) [22:31:48] fr-tech. Regarding Apple Pay, I am watching the Adyen console to see if the transactions are settling. I am seeing transactions from yesterday that are unsettled and those from earlier today that are still unsettled. Does anyone know what the settlement capture logic is? How long after a transaction is authorized should we expect to send a capture? Can we confirm we are sending captures? As and example, here is [22:31:48] one from yesterday that isn't: PSP ref# 3736354875825546 [22:33:58] emartin: I don't know there, from our end it's just one call where we send the info to adyen there aren't separate authorize / capture steps like the previous integration [22:34:40] We have the call scheduled on Monday morning with their engineer we can ask there [22:34:44] When I spoke to Manavi this morning she indicated we send a separate capture on everything and we should do so on Apple Pay as well. [22:35:06] Hmmm maybe I am thinking about it differently then [22:35:30] I am only seeing authorized Apple Pay transactions. I manually settled a few to see if I can capture them. [22:37:00] emartin: one sec I'll send a relevant link in the Adyen doc and from our code [22:37:44] as cstone says, at least her and my understanding is that there's no separate step now, and nothing significantly different with ApplePay vs other Adyen methods in that specific regard [22:39:42] Thanks for the finding links AndyRussG I'm on my phone or I would hunt down the code one [22:40:34] yee no worries :) [22:42:20] Ok, I am scrolling through and found some that settled. This is interesting. Note the time lag from recieved to Settle. Perhaps I'm just being over zealous in expecting the captures too soon. It seem to take many hours to settle in this example: https://usercontent.irccloud-cdn.com/file/qwrykVP9/image.png [22:43:45] With Apple Pay we are going through Adyen>Apple Pay>issuing bank so perhaps it just takes longer? One to watch...and understand. [22:43:58] yeah sounds like a decent hypothesis [22:44:13] and if there are scaling issues with Adyen's console, maybe they have scaling issues elsewhere too [22:44:28] yes perhaps! [22:44:34] emartin: so, just to double-check, this is something that never happens with other Adyen payment methods? maybe it does happen but things go faster? [22:45:23] Yes! Everything goes much faster. Things are settling quickly. Let me look at the timing on a basic Visa and let you know specifics. [22:45:31] oki great thanks! [22:46:32] also unrelated note, my daughter just fell asleep while gaming on the iPad, after two days of staying up way later than she should, and I guess de-compressing from the stress of the plahy [22:48:09] emartin: also to confirm, so some of them seem to get settled at least eventually, but there's no discernible pattern there? [22:48:48] Here's a standard visa timing. I've looked at a few and this is indicative. https://usercontent.irccloud-cdn.com/file/L7pytBLc/image.png [22:49:11] Poor thing. It's not easy being a star! [22:49:35] How old is your daughter, Andy? [22:49:56] this is the younger one, she's 14 [22:50:27] Fun age! Blessings! [22:50:44] aww thanks! it is quite the age, indeed [22:51:26] I'll keep watching Apple Pay over the weekend as I live and learn. Thanks! [22:53:03] Thanks emartin ! [22:54:23] emartin: cool thx! yeah looking at stuff now... btw I do normally work hours some of my normal over the weekend, and this weekend will be no exception, so pls feel free to ping anytime (disclaimer: not encouraging anyone to not take their weekends :) ) [23:00:30] Thanks Andy. We did have a config issue today which Adyen addressed so perhaps we need a couple more days to make an assessment. Perhaps anything from yesterday won't settle because of the setup problem? We'll see. [23:00:46] yeah also sounds like a possibility [23:00:56] cstone: we're on API v67? or is it 66, do you recall? (also, if that's not documented somewhere, it should be.....) [23:06:23] I'm pretty sure 67 [23:06:37] oki thx [23:07:02] did you or someone say earlier today something about Adyen coming out with a new something-or-other? or was that not Adyen... sorry for the fuzzy recall [23:10:19] Adyen will deploy new console infrastructure capability on 8 November which is posing issues with transaction search. I am not sure if this got back to you perhaps? [23:11:23] hmmm maybe but I think it was something else hehe... [23:11:58] emartin: I guess that's why when I add a payment method filter for Apple Pay, it doesn't show any results, but when I just search for "Apple Pay" it does show stuff? [23:12:47] yes, they have issues. It's frustrating. We are all trying to cope with their lackluster console at the moment. Better days are coming. [23:15:24] just one more thought here AndyRussG, I manually captured 5 transactions and they settled near instantly so it feel like we aren't waiting on the issuers but something is delayed on the capture process. I'll see what tomorrow brings.... [23:17:24] emartin: oki cool, yeah I feel pretty confident that it isn't something we're missing on our end. Also, I am able to see both "Authorized" and "SettledBulk" transactions, so another thing I can do is look at the logs to see if any of the older unsettled ones say anything different [23:20:59] In case you're curious, here is another technical Adyen doc. If you look under "How it works", there are a series of steps, none of which indicate any extra action we need to do to capture or settle transactions: https://docs.adyen.com/online-payments/web-components#how-it-works [23:21:37] In the more complicated flow, the most we have to do is redirect the user to another site, for example for 3D Secure, but even then there's no "capture" step for us to perform [23:22:37] If Manavi thinks we're missing an action that we need to be doing, it'd be great to see the specific documentation for the action she's thinking of [23:38:16] mpphhh on my local dev setup now Apple Pay is borked [23:39:59] getting an error that says "SIP not enabled" in the logs, and then "Error Domain=PKPassKitErrorDomain Code=-2000" [23:40:19] aaand trying to update my Mac and it gets stuck looking for updates [23:48:50] oki now my mac is updating, only 12 GB to download... [23:49:26] emartin: I have a path to digging more deeply into this issue to be ready for the Adyen meeting on Monday. Looks like I have to wait for my Mac to update again to be able to proceed tho [23:50:47] In any case, what I'll do is check carefully the logs on the servers for payments that are sitting in Authorized and not getting captured, compare those to successful transactions in my local test environment, and successfully captured ones on prod [23:51:22] for now I think the recommendation remains to leave Apple Pay up, and I'll ping folks if I see anything that indicates otherwise [23:51:26] hope this is helpful!