[00:09:02] eileen: Can you think of a reason that uniqid() would return empty string? [00:09:21] I'm testing your merge patch, but all my log_conn_ids (also from before your patch) are four-digit ints [00:10:21] uniqid() works fine from CLI scripts.... [00:10:47] ooh--log_conn_id is an int on my box. That would be it, eh? [00:12:20] Also, is there an easy way to turn safe mode off to allow me to do crazy batch merges? [00:12:58] awight: yeah log_conn_id was converted [00:13:01] to 17 char [00:13:13] ok that does it then. [00:13:14] there is a drush command to do that - I need to put up a blog explaining [00:13:48] drush cvapi System.updatelogtables [00:14:23] re turn off safe mode - what do you mean by that? [00:16:52] (CR) Awight: [C: 2 V: 1] "Works as advertised!" [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/294428 (https://phabricator.wikimedia.org/T135063) (owner: Eileen) [00:17:43] To force batch merging to ignore conflicts--I was able to test without that though, so nvm! [00:20:46] awight: I think there is a drupal permission to add that [00:21:49] (CR) Awight: [C: 2] Fix merge hook to remove calculated fields from merge_conflicts [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/285322 (https://phabricator.wikimedia.org/T133625) (owner: Eileen) [00:23:55] eileen: Fatal error: Uncaught CiviCRM_API3_Exception: [undefined: A fatal error was triggered: Dedupe rule not found for selected contacts [00:23:59] (CR) jenkins-bot: [V: -1] Per merge connection id [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/294428 (https://phabricator.wikimedia.org/T135063) (owner: Eileen) [00:24:47] what is this btw? 'rgid' => 9, [00:27:50] (CR) Awight: "I really like this implementation. A few questions inline, and I got this error when running locally:" (2 comments) [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/290828 (owner: Eileen) [00:27:59] gotta run. I'll review more tomorrow! [00:28:16] (PS6) Awight: Fix merge hook to remove calculated fields from merge_conflicts [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/285322 (https://phabricator.wikimedia.org/T133625) (owner: Eileen) [00:28:41] eileen: Yikes! This patch has a circular dependency on itself??? https://gerrit.wikimedia.org/r/#/c/285322/5 [00:28:47] That just appeared when I tried to rebase from the GUI [00:28:56] I'll.. leave that alone, sorry for the mess. [00:29:43] oh dear :-) [00:33:24] (PS7) Eileen: Fix merge hook to remove calculated fields from merge_conflicts [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/285322 (https://phabricator.wikimedia.org/T133625) [00:36:36] (CR) Eileen: [C: 2] "I'm transferring Adam's +2 to this version - which is the same as the last one he +2'd but there were rebasing issues in between" [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/285322 (https://phabricator.wikimedia.org/T133625) (owner: Eileen) [00:37:51] (CR) Eileen: "yeah - it is actually hard coded to the rule I chose on prod / staging - the rules are a bit of a mess actually. I probably need to tidy t" [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/290828 (owner: Eileen) [00:38:07] (PS2) Eileen: Per merge connection id [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/294428 (https://phabricator.wikimedia.org/T135063) [00:47:20] Fundraising Sprint Hermit Crab Husbandry, Fundraising Sprint Internet Exploring, Fundraising Sprint Jabberwock Slaying, Fundraising Sprint Killing Time, and 4 others: Write a hook to skip the calculated tables from conflict consideration - https://phabricator.wikimedia.org/T133625#2237657 (Eileenm... [00:57:27] (PS2) Eileen: Add mysql to drop triggers to git [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/293785 [00:57:29] (PS4) Eileen: Add prospecting custom fields [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/293793 (https://phabricator.wikimedia.org/T128818) [01:40:16] Fundraising Sprint Hermit Crab Husbandry, Fundraising Sprint Internet Exploring, Fundraising Sprint Jabberwock Slaying, Fundraising Sprint Killing Time, and 5 others: Clicktracking data not matching up with donation totals - https://phabricator.wikimedia.org/T132500#2384435 (Jgreen) Innnteresting... [03:38:46] (PS1) Eileen: Remove unused merge function. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/294659 (https://phabricator.wikimedia.org/T133625) [03:38:48] (PS1) Eileen: Remove code to include our custom tables. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/294660 (https://phabricator.wikimedia.org/T133625) [03:38:50] (PS1) Eileen: Alter logic for calculating wmf_donor fields [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/294661 (https://phabricator.wikimedia.org/T133625) [14:55:07] Fundraising Sprint Licking Cookies, Fundraising-Backlog, FR-PayPal-ExpressCheckout: Come up with ID scheme for "merchant reference" numbers - https://phabricator.wikimedia.org/T137837#2385467 (cwdent) Open>Resolved We decided on OID-PPEC- [15:50:39] MBeat, PPena, awight : are you all still happy to run a Ukraine adyen test this morning? [15:51:11] definitely, thanks the-wub [15:51:12] the-wub absolutely [15:51:18] the-wub: +1 Thank you! [15:51:42] cool! we'll start in 10 mins then :) [15:54:11] awight: are we already deploy frozen? or shall we get another out today? [15:55:13] cwd: We're frozen, unless the deployer wants to get all cowabunga and then babysit for 24h... [15:55:28] * cwd just might [15:55:30] U mean, for the PPEC tweaks? [15:55:31] hehe [15:55:57] yeah that, and just whatever we kill now won't add to the heap [15:56:06] FWIW, I don't think of those as minor patches--they have the potential to really choke our queue consumers, for one thing [15:56:27] tell me more [15:56:36] Actually, I was gonna ask if you want to come up with an outline of the phased deployment for the OID bits [15:56:54] The catch is that we currently expect a N.N format in various places [15:57:43] yeah. do you expect this value to hit any of those places? [15:58:04] So when we start spraying A-A-N.N stuff around, it will come in through DI, the listener, and the audit processor, which means there's at least a 24h window in which we should expect heavy showers of instability [15:58:04] it seemed to me like not much was looking at it but i was probably being pretty narrow [15:58:22] I hadn't looked yet... [15:58:38] hmm ok [15:58:57] I guess all I'm trying to say about the freeze is that technically it's begun earlier this week, but we're always open to caprice ;) [15:59:50] Here's an example of something that will kick our asses: [15:59:51] ./legacy-paypal/PaypalIPNListener.php: $contribution['contribution_tracking_id'] = $post_data['custom']; [15:59:59] That probably needs fixin [16:00:29] i thought only order_id was affected by this [16:00:46] * awight peeks at UA logs [16:01:05] well, on #master we're pushing order_id into all of PayPal's custom fields [16:01:06] MBeat PPena awight : banners are up! [16:01:17] so they'll squirt back at us through the IPN and audit [16:01:32] including "custom" [16:01:40] the-wub: thanks! [16:01:41] awesome, will keep an eye for any errors the-wub [16:01:44] I'll be pasting my results here: https://docs.google.com/spreadsheets/d/17arGLTPupnrCmo7O0He38js-gmOoF3OY53sOwiAbAp0/edit#gid=1606946668 [16:04:37] There's the first UA donation. Uh, oh [16:04:47] FormChooserNot able to find a valid form for country 'UA', currency 'UAH', method 'paypal', submethod '', recurring: 'false', gateway 'adyen' for utm source 'B1516_0616_enruukUA_dsk_p2_sm_pym_adyen.no-LP.paypal [16:04:59] method paypal but gateway adyen [16:05:26] I believe that donor was successfully sent to the PayPal flow, however [16:06:18] Failed a UA mastercard [16:06:41] the-wub: Only one of those form errors, fyi. [16:07:02] hmm... choosing paypal worked fine for me, no sign of an error [16:07:23] dang, one more form error [16:08:32] PayPal seems to be working fine, but yeah no CCs yet [16:09:51] the-wub: okay, a CC just went through successfully [16:10:22] awight: neat! I see it too :) [16:10:22] Through Ingenico, however. Are we testing both Ingenico vs PP and Adyen vs PP? [16:10:36] that's right [16:10:53] ok great [16:11:25] People are so generous... Sometimes I just can't believe that donations come in like this [16:12:45] noting showing at adyen's side yet [16:12:47] I still can't get over the speed of donations when we did a US test a few weeks ago. crazy. [16:12:50] takes foreeeeverrrr [16:13:17] PPena: yeah, I think it was about 15-20 minutes before anything showed up from them in the last tests [16:13:44] ok [16:13:52] is there any way we can make that faster, or push them to make it faster? [16:14:11] PPena: I'm only seen 4 pending mc's go through Adyen yet, and one failed visa. [16:14:26] 2 failed visas, now. [16:14:59] hey, there goes a pending mc. [16:15:23] I see 2 adyen donations in civi now [16:15:27] fr-tech: So... remind me, Adyen donations are settled at the listener eh? [16:15:34] yeah [16:15:38] ty [16:15:43] as i understand it... [16:18:50] seems weirdly slow today, even ingenico and paypal [16:18:57] awight: seen any more form errors? [16:19:05] no, just the two [16:23:43] Fundraising-Backlog, MediaWiki-extensions-DonationInterface: Tunnel and test Jessie paymentswiki - https://phabricator.wikimedia.org/T137975#2385725 (awight) [16:24:50] Jeff_Green: ^ mebbe put a note there when the system is ready to test and we'll get it into whatever sprint that may be. [16:25:32] awight: it's already ready [16:25:44] who whow whoa [16:26:07] Fundraising Sprint Licking Cookies, Fundraising-Backlog, fundraising-tech-ops, MediaWiki-extensions-DonationInterface, Unplanned-Sprint-Work: Tunnel and test Jessie paymentswiki - https://phabricator.wikimedia.org/T137975#2385739 (awight) [16:26:11] i mean, i won't be surprised if something funky turns up but puppet runs clean and I've worked out all the significant kinks I've found so far [16:26:11] k, ensprinted. [16:26:25] windfall! [16:26:43] yeah that is exciting [16:26:51] I've got some minor cleanup for stuff like pretty motd, and monitoring for package updates, but stuff should be generally ready to go otherwise [16:27:18] * awight is tempted to drop everything else and play in the new sandbox [16:27:21] mysql & redis are replicating like 1004 [16:27:46] oh beware one thing, if your testing populates the local redis master, that isn't being replicated to 1004, and can't be [16:27:52] no prob [16:27:59] That's just for orphan rectification [16:28:03] k [16:28:22] Which I'm practically happy to see break cos it will force us to parachute it out of there [16:28:36] I'm going to cut out a little early today (3PM my time) because we're roadtripping this evening, but I'll be on normal time tomorrow [16:33:59] the-wub: The form errors could be coming from people backing out and switching payment methods, but it seems too consistent... utm sources are B1516_0616_enruukUA_dsk_p2_sm_pym_adyen.no-LP.rpaypal and B1516_0616_enruukUA_dsk_p2_sm_pym_adyen.no-LP.paypal [16:34:07] Still only 2 though [16:38:45] this is so weird [16:39:03] sad we have nothing to comapre to [16:39:06] PPena: what's that--the Adyen console latency? [16:39:09] should we leave a little longer? [16:39:18] no the low low low traffic [16:40:17] about 2 attempted donations per minute... [16:40:49] Maybe the-wub can say whether the donations are reasonably proportional to banner impressions? [16:40:59] and the fact that I see 0 tukraininan trnx in teh console [16:42:14] PPena: agreed that the empty console is weird. Lemme squint at the listener log and see if I can get you the earliest successful authorization ID [16:42:33] hard to say, because banner impression numbers are on at least 15 min delay too [16:43:23] seeing a reasonable number of clicks though - about 230 in total (all methods) [16:44:42] might it be a broader Adyen console lag? most recent new payment I see there is 08:07:56 PDT [16:45:27] PPena: re something to compare it to, I actually just realised we've been running English low-level in Ukraine. Ingenico cc is about 44% conversion there, and paypal about 29% (since 2016-01-01) [16:48:08] oh cool!! that heps [16:48:44] the-wub can we keep it up for +1 hr? I can ask Adyen whats up with the console... [16:49:16] PPena: I can extend the campaign, but can't stay around to monitor. you okay with that? [16:49:23] the-wub: PPena: In case it helps to have a PSP id, here's the first Adyen transaction to apparently go through successfully: 4814660931758510 [16:51:02] cwd: oh wow, TIL the SmashPig job runner is on the Civi server and pulls jobs written by the listener box SmashPig [16:52:38] the-wub yes sounds fine, just wanted more data and reach out to Adyen. If you feel confortable that works for me. [16:52:52] awight thanks!! [16:53:04] the-wub: I'm happy to monitor for another hour if that's helpful [16:54:13] awight: is that...poorly compliant? [16:54:22] cool, thanks awight. PPena MBeat I've extended the test for another hour [16:54:37] cwd: it's great, and possibly even legal ;) I was just surprised [16:55:05] ty the-wub [16:55:13] oh...great! [16:56:14] the-wub: no big deal, but 9 of those form errors total. All the same banner [16:56:30] 10 now :) [16:57:31] awight: ah nuts. [16:58:24] are the donors all still getting through to paypal though? [16:59:41] the-wub: I donno, making a task to track this... [16:59:53] Unfortunately, that logline is missing any information that I could correlate with other logs [17:00:14] thanks. I do need to head out in a few minutes, so will look at it again later today. [17:00:27] Fundraising-Backlog, MediaWiki-extensions-DonationInterface: Form chooser error from Ukranian Adyen-PayPal form - https://phabricator.wikimedia.org/T137978#2385840 (awight) [17:00:44] Thanks, cul8r! [18:18:36] AndyRussG: d'oh! oops I had the mouse hovering over the hang up button :( [18:26:38] (CR) Awight: Kill staged_vars (1 comment) [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/293463 (https://phabricator.wikimedia.org/T137387) (owner: Ejegg) [18:30:58] (PS5) Awight: Drush merge command [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/290828 (owner: Eileen) [18:35:44] (CR) Awight: "Let me know if you plan to refine this patch, or if you'd rather just get it out there. It looks like it does what you intended, I'm happ" (1 comment) [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/290828 (owner: Eileen) [18:38:12] (CR) Awight: [C: 2] "Redundant!" (1 comment) [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/294660 (https://phabricator.wikimedia.org/T133625) (owner: Eileen) [18:43:57] (PS11) Awight: Encapsulated amount validation for better messages [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/286261 (https://phabricator.wikimedia.org/T105618) (owner: Ejegg) [18:44:16] (CR) Awight: "PS 11: manual rebase" [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/286261 (https://phabricator.wikimedia.org/T105618) (owner: Ejegg) [18:52:33] (CR) Awight: [C: -1] "We could push this through as an incremental improvement, but I wanted to check whether it's doing what you expected here... If I submit " [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/286261 (https://phabricator.wikimedia.org/T105618) (owner: Ejegg) [18:53:01] (PS5) Awight: Log an error when banner translation retrieval fails [extensions/CentralNotice] - https://gerrit.wikimedia.org/r/293147 (https://phabricator.wikimedia.org/T137247) (owner: Ejegg) [18:55:29] (CR) Awight: Log an error when banner translation retrieval fails (1 comment) [extensions/CentralNotice] - https://gerrit.wikimedia.org/r/293147 (https://phabricator.wikimedia.org/T137247) (owner: Ejegg) [18:56:41] Whew! Just these two still open for review ;) T133964, T108721 [18:56:41] T108721: Convert CentralNotice to use extension registration - https://phabricator.wikimedia.org/T108721 [18:56:41] T133964: Implement AtomicReadBuffer for choice of PHP-Queue backend - https://phabricator.wikimedia.org/T133964 [19:12:20] Fundraising-Backlog, FR-Smashpig: SmashPig throwing errors on fraud notification - https://phabricator.wikimedia.org/T137994#2386288 (awight) [19:47:15] dstrine sorry gonna be late 4 standup, stuck in traffic [19:47:28] AndyRussG: ok [20:01:49] hangouts glitches! [20:02:39] Oops, cos I upgraded FF. [20:07:49] Fundraising-Backlog, FR-Smashpig: SmashPig throwing errors on fraud notification - https://phabricator.wikimedia.org/T137994#2386563 (DStrine) p:Triage>Low [20:08:40] Fundraising-Backlog, MediaWiki-extensions-DonationInterface: Form chooser error from Ukranian Adyen-PayPal form - https://phabricator.wikimedia.org/T137978#2386568 (DStrine) a:Pcoombe [20:08:57] Fundraising-Backlog, FR-Adyen, MediaWiki-extensions-DonationInterface: Form chooser error from Ukranian Adyen-PayPal form - https://phabricator.wikimedia.org/T137978#2385840 (DStrine) [20:09:52] Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Provide ability to enter foreign currency into Source and have the USD amount populated by Civi - https://phabricator.wikimedia.org/T137931#2386573 (DStrine) p:Triage>Normal [20:10:02] Fundraising-Backlog, FR-Adyen, MediaWiki-extensions-DonationInterface: Form chooser error from Ukranian Adyen-PayPal form - https://phabricator.wikimedia.org/T137978#2386575 (awight) My first guess would be a ECMAScript glitch in the banner. [20:14:19] Fundraising-Backlog, FR-Ingenico, MediaWiki-extensions-DonationInterface: Ingenico orphan slayer incorrectly cancelling pending transactions - https://phabricator.wikimedia.org/T137905#2386612 (awight) This is probably fine. But we should double-check that the cancellation is actually going through,... [20:14:29] Fundraising-Backlog, FR-Ingenico, MediaWiki-extensions-DonationInterface: Ingenico orphan slayer incorrectly cancelling pending transactions - https://phabricator.wikimedia.org/T137905#2386613 (DStrine) p:Triage>Normal [20:15:06] Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Adding Gift Source option for Engage users - https://phabricator.wikimedia.org/T137569#2386617 (DStrine) p:Triage>Normal [20:17:40] Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Make summary fields in contact records more useful - https://phabricator.wikimedia.org/T137568#2386633 (DStrine) p:Triage>Normal [20:18:06] Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: A tool to evaluate and fix donor-entered data - https://phabricator.wikimedia.org/T137498#2386635 (DStrine) p:Triage>Normal [20:25:55] Fundraising-Backlog, FR-PayPal-ExpressCheckout, FR-Paypal: Renew PayPal API certificate. Due date: Aug 1st - https://phabricator.wikimedia.org/T136742#2386680 (DStrine) p:Triage>High [20:26:22] Fundraising Sprint Killing Time, Fundraising Sprint Licking Cookies, Fundraising-Backlog, Patch-For-Review: Cut fundraising/REL1_26 branch - https://phabricator.wikimedia.org/T137213#2386684 (cwdent) Open>Invalid 1.26 is EOL in November so let's not even bother moving to that and just go... [20:29:12] Fundraising-Backlog, FR-PayPal-ExpressCheckout, MediaWiki-extensions-DonationInterface: Merchant reference generation should have options for attempt number or API sequence - https://phabricator.wikimedia.org/T136733#2345971 (DStrine) [20:30:42] Fundraising-Backlog, MediaWiki-extensions-DonationInterface, Technical-Debt: Move controller code out of GatewayType#processResponse - https://phabricator.wikimedia.org/T136665#2386700 (DStrine) p:Triage>Low [20:37:14] Fundraising-Backlog, FR-PayPal-ExpressCheckout: Epic: Add semantic prefix to order_id - https://phabricator.wikimedia.org/T138013#2386753 (cwdent) [20:49:21] Fundraising-Backlog: Move payments to MediaWiki 1.27 - https://phabricator.wikimedia.org/T138016#2386861 (cwdent) [20:50:32] Fundraising-Backlog: Move payments to MediaWiki 1.27 - https://phabricator.wikimedia.org/T138016#2386898 (cwdent) [20:50:34] Fundraising-Backlog, fundraising-tech-ops, Epic: [Epic] Upgrade paymentswiki to MediaWiki REL_* release branch or follow the WMF train. Upgrage PHP to HHVM or PHP 5.5 - https://phabricator.wikimedia.org/T130130#2386899 (cwdent) [21:08:51] awight: if you agree that we should not deploy the order_id patches should i push a revert? or if that is not your take on it let me know [21:29:19] (PS1) Cdentinger: Revert "Add express checkout identifier to order_id" [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/294842 [21:31:10] (Abandoned) Cdentinger: Differentiate express checkout for the back end. [wikimedia/fundraising/PaymentsListeners] - https://gerrit.wikimedia.org/r/294639 (owner: Cdentinger) [22:14:23] cwd: I thought they were a great start--but not ready to be lightning deployed without having a plan in place [22:14:58] They're especially appropriate to merge now cos we're planning to do this bloated release in a few weeks, with the expectation that everything will be destabilized [22:15:26] If you want to rework them, then revert by all means, or just push forward incrementally. Up 2 u [22:16:33] I guess I'm confused about what the revert would accomplish [22:16:46] awight: i pushed a revert for one, and the other never merged cause listeners has no CI so i just abandoned [22:17:01] ok, sure! [22:17:29] Is the alternative approach still TBD, or did you have thoughts? [22:17:29] awight: my intent was not deploying them but keeping fast forward history [22:17:38] hrm? [22:18:33] IMO this is the perfect time to put edgy stuff like that on #master of all our repos, then we can just sit around and hammer on the whole thing in vagrant until it looks reasonable. [22:19:52] awight: do you think there's a better way to do it? marrying magic strings doesn't feel like the best thing. [22:20:03] maybe we could use another param [22:20:23] some other identifier, so we don't break stuff [22:22:03] It seems that the "custom" field is the only thing that comes back from PP. We could put different data in there, but that's sort of what we're doing. Agreed that semantic magic is not fun. [22:23:04] if paypal is the only case where we need this behavior -- routing from the listener -- maybe generalizing is not necessary [22:23:06] The only other approach I can think of is sort of reasonable, which is to be agnostic of what type of transaction it is until it gets back to somewhere that we can marry it to front-end logging. But what if that ship never comes in? [22:25:43] It's possible that we could make some API calls to determine whether it's a legacy charge or not. I won't rule that out. [22:26:19] awight: correct me if i'm wrong but we are shoving a parameter through a bunch of request stuff even though all we actually want is for it to be tacked on the returnurl. [22:26:40] and it can get trashed after hitting the listener and altering the message [22:26:52] nah, the returnurl has everything we need--that has the exact gateway and the order id [22:27:31] but aren't we also sending the value to paypal in their format? [22:27:45] which value? "custom"? [22:28:37] well i mean the piece of information that says "this is EC not legacy" [22:28:50] sure, we're using entirely different APIs [22:28:54] does that need to be comingled with the other request data? [22:30:36] paypal doesn't need to know that we are differentiating on the back end. so if we just tacked a param on to the return url that the listener could key off to change the message, that seems to cover it, no? [22:31:02] the message would have the new gateway in civi [22:31:10] and we wouldn't have to trash any other code [22:31:40] and there is already a helper for returnurl [22:31:55] The returnurl doesn't help us, unfortunately. [22:32:16] That looks like PaypalExpressGatewayResult&order_id=NNN, so we have all the info we need there. [22:32:22] ah shit that's right [22:32:23] but the listener doesn't see the return url [22:32:30] hrmrph [22:33:16] I haven't even entirely convinced myself that we need to know that charges come through pp_ec [22:33:25] Only mostly convinced [22:34:07] There are practical things about: is one gateway or the other systematically broken. Then there are theoretical things like, * maybe we need to know in order to refund via API, which we don't do yet. [22:34:40] well there is probably some field in the spec we could hijack for this... my question is, is this a behavior we need to generalize (as in changing the format of order_id) or is this an annoying thing about paypal [22:34:50] For recurring it's more obviously a big deal, cos we might have to use reference transactions, in which case subscriptions from the two gateways couldn't be more different. Actually, they're completely different to begin with. [22:35:00] I think it's gonna be a one-off need [22:35:25] however... It might be nice to have OID-STAGING-nnnn [22:35:29] por ejemplo [22:37:38] i can see use cases for it, but if this is the only current one it might be a big decision to make based on not enough data [22:38:30] maybe we can suture this on in such a way that it can get ripped off when a better method is needed elsewhere [22:39:51] definitely. I gotta run for now, but let's talk more next week. Thank for thinking about this! [22:39:54] *s [22:40:42] thank you! [22:40:49] have a good weekend, see you in italy [22:41:01] chow! [22:51:45] cwd: AndyRussG can either of you get into civi? CV is having trouble. I'm not on my work computer so I can't confirm on my end. [22:52:33] dstrine: the certificate I used to use went with the hard drive... haven't gotten a new one :( [22:52:42] yeah works for me [22:53:56] errything looks normal [22:54:38] ok thanks AndyRussG and cwd looks like a connection issue on her end. She eventually got in [22:55:02] word [22:55:31] apparently it runs much slower in the afternoon? I wasn't aware of this. [22:56:07] ¯\_(ツ)_/¯ [22:56:47] yeah [23:31:00] Fundraising-Backlog, fundraising-tech-ops: access request for users at fundraising analytics consultant CPS Data Consulting - https://phabricator.wikimedia.org/T135392#2387664 (DStrine) @Jgreen All three now have their Yubi keys. They do not have access to phab, collab or any common wikis yet. Do they n...