[01:10:18] (PS2) Eileen: Additional NYSS funded fixes to contact layout editor [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/451534 [01:16:10] (PS4) Eileen: WIP drush to repair omnirecipient entries [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/451228 [01:25:33] (PS5) Eileen: WIP drush to repair omnirecipient entries [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/451228 [02:55:41] (PS6) Eileen: WIP drush to repair omnirecipient entries [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/451228 [03:01:23] (CR) jerkins-bot: [V: -1] WIP drush to repair omnirecipient entries [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/451228 (owner: Eileen) [03:36:42] (PS7) Eileen: drush to repair omnirecipient entries [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/451228 [03:40:03] (PS1) Eileen: Change backupGlobals to false. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/453078 [03:42:23] (CR) jerkins-bot: [V: -1] drush to repair omnirecipient entries [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/451228 (owner: Eileen) [03:42:48] (CR) jerkins-bot: [V: -1] Change backupGlobals to false. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/453078 (owner: Eileen) [03:44:11] (PS2) Eileen: Change backupGlobals to false. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/453078 [03:49:47] (CR) jerkins-bot: [V: -1] Change backupGlobals to false. [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/453078 (owner: Eileen) [03:54:58] (PS8) Eileen: drush to repair omnirecipient entries [wikimedia/fundraising/crm] - https://gerrit.wikimedia.org/r/451228 (https://phabricator.wikimedia.org/T200240) [04:31:35] (Abandoned) Eileen: Composer update [wikimedia/fundraising/crm/vendor] - https://gerrit.wikimedia.org/r/451563 (owner: Eileen) [04:43:07] Fundraising Sprint Owls, Fundraising Sprint Pluto is still a planet, Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM, Patch-For-Review: slow anonymous save - https://phabricator.wikimedia.org/T199753 (Eileenmcnaughton) @LeanneS I just tried to replicate this on staging & wasn't sure that I... [16:34:00] Fundraising-Backlog, fundraising-tech-ops, Epic: fundraising database improvements for 2018 - https://phabricator.wikimedia.org/T183138 (Jgreen) [16:34:02] Fundraising-Backlog, fundraising-tech-ops, Epic: work around and report bug in mariadb.org 10.0.36+maria-1~jessie /etc/init.d/mysql - https://phabricator.wikimedia.org/T201898 (Jgreen) Open>Resolved fixed the init script, did not bother to upstream a bug report/fix [17:04:15] ejegg, do we have an integration with stripe? [17:04:28] re; the email thread [17:11:03] jgleeson: not a full integration [17:11:23] I think major gifts' in-person events use stripe somehow [17:11:33] but we just get them into Civi via CSV imports [17:11:38] ah [17:11:53] you can take a look at the code in the offline2civicrm module [17:12:04] There's a base 'ChecksFile' class [17:12:17] and a bunch of other things that inherit from it [17:15:56] thanks! [17:20:54] if you have any questions I'd be happy to talk through any of those bits! [17:22:14] Fundraising-Backlog: Delay in GC transactions making it into Civi - https://phabricator.wikimedia.org/T202079 (DStrine) [18:19:32] have a good evening fr-tech! [18:19:53] o/ awight [18:22:09] http://saurabhbatra96.github.io/2018/08/15/catching-fraud-at-wikipedia/ if you're up for some reading! [18:28:24] saurabhbatra: oh hell yeah [18:28:31] :-) [18:30:34] saurabhbatra: This is great writing! [19:48:35] Fundraising-Backlog: Delay in GC transactions making it into Civi - https://phabricator.wikimedia.org/T202079 (Ejegg) Hi @krobinson! How are you looking for the donations in Civi? The Wednesday tests have been using our new integration, which codes the donations as gateway='ingenico' rather than gateway='glo... [19:51:39] Fundraising-Backlog: Delay in GC transactions making it into Civi - https://phabricator.wikimedia.org/T202079 (Ejegg) payments_fraud table numbers are a bit funny, though - the old integration only shows 757 rows from the 15th, while the new integration shows 9,044. Checking to see if the old integration do... [19:56:04] Fundraising-Backlog: Delay in GC transactions making it into Civi - https://phabricator.wikimedia.org/T202079 (Ejegg) OK, numbers from the old integration on the 15th check out too. 1,315 of them were recurring, so that's 709 new one-time donations, a bit under the number of payments_fraud rows. [19:56:46] dstrine: I think we're good on the payments infrastructure for the test on the 15th^^^ [19:56:54] what may have been lacking is my communication [19:57:07] maybe folks were still looking for 'globalcollect' rows [19:57:23] when the new integration is coded as 'ingenico'; [19:57:45] Fundraising-Backlog: Delay in GC transactions making it into Civi - https://phabricator.wikimedia.org/T202079 (krobinson) Ha! My bad completely - I was searching GLOBALCOLLECT %. I just tested and those numbers are what I see, so apologies. I was using the wrong gateway. Sorry for the confusion! [19:59:17] ejegg: yeah I saw. thanks for that! [20:02:40] Fundraising-Backlog: Delay in GC transactions making it into Civi - https://phabricator.wikimedia.org/T202079 (Ejegg) No worries, I probably should have communicated this more clearly! So, as long as both the integrations are active you'll have to complicate your queries with extra parentheses and an OR trx... [20:09:23] Fundraising-Backlog: Delay in GC transactions making it into Civi - https://phabricator.wikimedia.org/T202079 (krobinson) Thanks for the tips :) Now I know I'll check both. Ingenico fraud scans are more Michael's territory and I just hadn't realized the new integration had a different gateway. Cheers! [20:12:01] hehe I should probably just into the private channel if I’m gonna drop emails like that [20:38:25] Am just reading saurabhbatra blog - has anyone forwarded to comms yet - if not I will [20:38:40] eileen erm [20:38:48] hey [20:38:49] what would we do without you :-) [20:39:39] o/! [20:39:46] i just posted the link on the mail thread, that's all [20:43:27] saurabhbatra: yeah it’s for us not you to communicate with our internal comms dept [20:44:05] ah, okay [20:45:26] saurabhbatra: so the thing I notice missing in your blog is any mention of GSOC [20:45:53] or CiviCRM for that matter [20:46:06] Although I suspect there is no real Civi involvement in what you did [20:46:12] eileen: oops, i completely skimped on that :O [20:46:53] true about the Civi thing though [20:47:22] yeah & TBH maybe we should use a WMF blog rather than civi given that [20:47:35] it has a very very indirect impact on them - if we successfully de-couple it from wmf data schema some time in the future [20:48:28] GSoC should be fixed by adding it in the title and the acknowledgements maybe [20:50:31] yeah & perhaps in the start of ‘Research’ ie when saying ‘we’ elaborate there [20:51:12] IMO the potential for reuse in Civi is actually very high, our inputs are only coupled by a light preprocessing layer. [20:51:32] i guess that's true [20:52:13] but we're also picking up wmf specific data from fredge [20:59:14] Good point, feature engineering is nothing to sniff at [21:04:13] I guess we could provide with suggested features and a default integration with Civi fields [21:04:24] Then the extra inputs are pluggable [21:04:40] * awight hopes that saurabhbatra fell asleep ;-) [21:05:35] awight: nope, running queries rn :-) [21:05:47] /o\ [21:06:05] i'll sleep once i check out the new graph [21:09:25] awight: btw i tried running the tuning code, it doesn't work on sklearn 0.14 [21:09:38] model selection was introduced in 0.15 :-( [21:09:39] oh noes [21:10:08] Well I’m sure frdev will be upgraded soon enough, if that’s a blocker! [21:12:01] let's hope so! [21:12:16] What is it, btw? virtualenv doesn’t work at all? [21:12:44] alright, i'm off now. i'm scared that we might start a convo and i'll not make it to my day job tomm :-) [21:12:53] oof o/ [21:13:15] o/ [21:38:08] Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Thank You email for Stripe Import - https://phabricator.wikimedia.org/T202099 (jkim_wikimedia) [22:08:32] Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: FY17-18 total not reflected in WMF Donor section - https://phabricator.wikimedia.org/T202102 (LeanneS) [22:12:49] Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Thank You email for Stripe Import - https://phabricator.wikimedia.org/T202099 (jkim_wikimedia) Ooh, if that's the case, I can go ahead and do that for July's reconciliation. [22:29:52] Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Thank You email for Stripe Import - https://phabricator.wikimedia.org/T202099 (Eileenmcnaughton) @jkim_wikimedia maybe just try one or 2 first with the column 'no_thank_you' & check how it goes - if the thank you then goes out you should be able to see... [22:32:16] Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Thank You email for Stripe Import - https://phabricator.wikimedia.org/T202099 (Ejegg) Shoot, we're still using this in the query to get the batch of donations to send thank you letters for: ``` WHERE receive_date > :earliest AND thankyou_date IS NULL... [22:37:22] Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Thank You email for Stripe Import - https://phabricator.wikimedia.org/T202099 (jkim_wikimedia) Yeah, it didn't work. I can just import and do thank you emails manually this month since Leanne needs the totals in soon. [22:46:02] Fundraising-Backlog, Wikimedia-Fundraising-CiviCRM: Thank You email for Stripe Import - https://phabricator.wikimedia.org/T202099 (jkim_wikimedia) Hey all, the direct mail appeal from the import wasn't added either. [22:48:35] ejegg: I’m gonna deploy that latest civi update - it’s slightly higher risk so I don’t want it sitting undeployed & getting deployed at a weird time [22:49:03] oh - do we have any campaigns running? [22:50:42] eileen: I think there's an email out for Japan, but it should be fine to turn off the queue consumers for an hour [22:51:05] ejegg: I mean I don’t think anything needs to turn off [22:51:10] there is just a slight risk [22:51:25] no schema changes with this one? [22:51:33] No - but a cache change [22:51:38] ie it needs truncating [22:51:47] because it goes from one format to another [22:52:05] hmm, let's turn the qcs off for a little bit just to be safe - I'll do the switch flipping [22:52:38] thanks for the initiative on that, btw! [22:53:21] ejegg: ok - I’ll send out an email saying we’ll do it at 11 [22:53:34] & just let people know if there are any issues then holler [22:53:42] 11 your time, yah? [22:57:30] yeah [22:57:35] just doing the merging now [23:00:57] ok, I've got the process-control off-switch primed [23:06:00] just trying to figure out if a vendor update is needed [23:07:13] hmm... doesn't look like it [23:07:38] since I've been dragging my feet on reviewing your omnimail updates :S [23:07:57] anyway, i see nothing from git log deployment..master composer.lock [23:09:34] (PS1) Eileen: Merge branch 'master' of https://gerrit.wikimedia.org/r/wikimedia/fundraising/crm into deployment [wikimedia/fundraising/crm] (deployment) - https://gerrit.wikimedia.org/r/453293 [23:09:40] I found one commit in vendor [23:09:55] oh hang on - was that local only [23:10:01] (CR) jerkins-bot: [V: -1] Merge branch 'master' of https://gerrit.wikimedia.org/r/wikimedia/fundraising/crm into deployment [wikimedia/fundraising/crm] (deployment) - https://gerrit.wikimedia.org/r/453293 (owner: Eileen) [23:10:16] hrm [23:10:52] that rsync error always throws me off [23:11:07] missing remote ref 3c1bf07440ed786e04d3c49f97ffe103df9e60dd [23:11:19] ohh, that was your local-only vendor thing [23:11:27] got it [23:11:37] (PS2) Eileen: Merge branch 'master' of https://gerrit.wikimedia.org/r/wikimedia/fundraising/crm into deployment [wikimedia/fundraising/crm] (deployment) - https://gerrit.wikimedia.org/r/453293 [23:11:49] yeah - updated [23:11:55] :) [23:12:34] that looks better [23:13:20] (CR) Eileen: [C: 2] Merge branch 'master' of https://gerrit.wikimedia.org/r/wikimedia/fundraising/crm into deployment [wikimedia/fundraising/crm] (deployment) - https://gerrit.wikimedia.org/r/453293 (owner: Eileen) [23:14:29] (Merged) jenkins-bot: Merge branch 'master' of https://gerrit.wikimedia.org/r/wikimedia/fundraising/crm into deployment [wikimedia/fundraising/crm] (deployment) - https://gerrit.wikimedia.org/r/453293 (owner: Eileen) [23:14:43] ejegg: I’m ready to go [23:15:34] !log civicrm revision changed from b60ceb3d2f to 3f6dd6c9a6, config revision is 74ae910b0e [23:15:38] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log [23:15:41] tell me when to rsyncblaster it [23:18:14] !log disabled CiviCRM scheduled jobs for version upgrade [23:18:18] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log [23:19:29] should be fine to go now eileen [23:21:23] ejegg: ok - code out & cache cleared. Update script failed but prob cos I didn’t turn off logging & it tried on those tables - I think it’s fine [23:21:50] yeah domain value is 5.4.0 [23:21:54] ok, cool! [23:21:55] so it ran [23:22:28] guess I should still turn stuff on slowly [23:23:44] yep [23:24:07] OK, did a very limited run of the donations consumer, checking the resulting donations in the UI [23:24:45] looks sane [23:26:22] thank you mail seems to be working fine too [23:26:29] gonna turn all the rest back on [23:26:48] cool [23:27:02] thanks eileen! [23:27:21] so, to start using redis for the caching... [23:27:25] glad it’s out - I was worried it would ‘sneak out’ & cause chaos with unclered caches [23:27:35] I guess we want to coordinate that with ops? [23:27:44] Yep - you can try locally if you want [23:28:11] is it using that 'predis' php backend? [23:28:19] or talking directly? [23:28:46] * ejegg could just look [23:30:20] oh hey, seems like it's a PHP extension [23:30:31] what does predis get us, anyway? [23:30:48] so there is this page https://docs.civicrm.org/sysadmin/en/latest/setup/cache/ [23:33:15] (Tim asked if it was appropriate he do that documentation update from our funding & I said yes) [23:33:28] :) [23:33:28] https://github.com/civicrm/civicrm-sysadmin-guide/pull/109/files [23:33:42] yay documentation! [23:34:17] :-) [23:39:02] cool cool, supports the redis PW auth, lets you set a cache prefix [23:40:23] eileen: think we can try turning it on next week? [23:40:52] I'd love to see what the queue graph looks like if we try activating it for the Wednesday banner test [23:40:54] ejegg: so I did log a Phab - my main concern is to discuss how we will compare performance [23:41:29] oops, forgot to log turning the jobs back on [23:41:43] !log re-activated CiviCRM scheduled jobs [23:41:47] Logged the message at https://wikitech.wikimedia.org/wiki/Server_Admin_Log [23:42:33] eileen: do you think the queue backup during the Wednesday test will be enough data? [23:43:11] So 1) we should switch staging to redis - I think we have a drush command that will do an ‘in isolation’ performance test we can use there [23:43:31] ah yeah [23:43:40] 2) I guess it’s enough data - I just really want to have some sort of graph that shows speed over time in a usable way [23:43:58] e.g if we could ‘pin’ peak times & put just those in a graph [23:47:45] (PS6) XenoRyet: Fix mobile orientation of radio buttons on payments.wiki opt-in [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/451804 (https://phabricator.wikimedia.org/T201528) [23:49:05] (CR) XenoRyet: "Ok, this one works on safari and the rest. It does have the bigger text block wrapping the radio button, but that doesn't end up looking " [extensions/DonationInterface] - https://gerrit.wikimedia.org/r/451804 (https://phabricator.wikimedia.org/T201528) (owner: XenoRyet) [23:53:07] eileen: hmm, so like getting smoothed highlights of this graph: https://grafana.wikimedia.org/dashboard/db/fundraising-overview?orgId=1&panelId=22&fullscreen [23:53:46] yeah to see over time [23:54:16] hm https://grafana.wikimedia.org/dashboard/db/fundraising-overview?orgId=1&panelId=22&fullscreen&from=1519490824209&to=1533937624209 [23:54:50] Ok so I look at https://grafana.wikimedia.org/dashboard/db/fundraising-overview?orgId=1&panelId=22&fullscreen&from=1490460424209&to=1533937624209 [23:54:56] & that is over time & it looks bad [23:55:19] you mean it looks like perf has declined? [23:55:22] yeah [23:55:35] yeah, i'm afraid it might have a bit [23:55:59] there's some times with pretty scarce data [23:57:24] but in reality, we are getting queue size warnings an hour into each test these days, while we were able to consume at pretty much the rate things came in during November/December [23:59:02] XenoRyet: did you mean to make more changes in PS6? [23:59:14] all I see is the commit message and deleting a blank line in the css