[13:34:24] miriam_: hey, tell me when you're around [13:34:45] Hey Amir1 [13:34:48] I am here :) [13:36:26] okay, I want to self merge it as no one is around but I want you to help me make sure nothing breaks :D [13:37:28] ok :) [13:37:58] Also, is there a phabricator ticket for that? [13:38:57] no, I just discovered the issue [13:39:29] I wanted to check with you first if there was a way to get the data in some way :) [13:39:35] let me open it [13:41:12] miriam_: about that, I looked at the database, It's not there. It means the frontend didn't send anything [13:41:25] Oh no :( [13:44:30] yeah :/ I'm so sorry [13:44:36] I wish I could do something about it [13:45:57] https://phabricator.wikimedia.org/T202732 [13:46:13] oh no, it's fine :) our mistake [13:46:21] thanks a lot Amir1 for your help [13:47:39] miriam_: the update is in prod [13:47:41] let's test [13:47:50] yes :) thanks [13:48:10] you have no idea what mistakes I've made [13:48:23] The most prominent one brought down all of wikipedia for half an hour [13:52:59] <3 thanks Amir, and, the new test works for me :) [13:53:58] miriam_: the data is in the database now [13:54:07] it has labels of notes-false and notes-true [13:54:18] 553157 | 51551975 | 2018-08-24 13:48:52.333106 | {"notes-false": "false-1", "notes-true": "true-empty", "notes": "", "citation_needed": false} [13:54:18] 553838 | 51551975 | 2018-08-24 13:49:59.069975 | {"notes-false": "false-empty", "notes-true": "true-7", "notes": "", "citation_needed": true} [13:54:19] I see it also on my side [13:54:21] yes [13:54:23] :) thanks [13:58:49] yw :) [15:09:20] hm, the formatting of the projects list looks broken https://meta.wikimedia.org/wiki/Research:Index [19:52:41] 10Quarry: Do the big Quarry migration - https://phabricator.wikimedia.org/T202588 (10zhuyifei1999) The current NFS implementation of resultsets is flawed. The runners have user 'quarry' with UID of 998, but main has it as 997. Currently it's not too bad, just that puppet keeps changing the ownership of `/data/pr... [20:03:43] bmansurov_: will you be able to start the emails today? (just saw Ramtin's new matches) [20:28:42] leila: yeah that's the plan, I'll send some test emails first. [20:30:33] bmansurov_: great. I'll be around to circle back. [20:53:51] leila: while cleaning data and matching users, I came across a situation where a matched username doesn't exist in the list of usernames we sent emails to [20:54:17] ow boy. how different? [20:55:03] bmansurov_: and is the username for a newly registered account or you'd think someone was just curious and participated. [20:55:38] leila: not sure, but I'm wondering how that escaped our filtering. Double checking some things. [20:55:58] bmansurov_: yes. good point. not sure if you checked yesterday for this or not. [20:56:14] leila: yeah, it's in the list of discarded users [20:56:51] bmansurov_: hmmm. I'm confused. so then how were they included in the matched results? mistake on the matching end? [20:57:12] bmansurov_: and when you say the list of discarded usernames, which list are you referring to? the 10 that were tests, etc.? [20:57:40] leila: some data corruption on my end it seems. [20:57:51] leila: those 64 [20:57:59] bmansurov_: ah! I see. [20:58:58] bmansurov_: do we have to only drop all triangles that username was in? (I'm wondering if Ramtin should rerun the matching) [20:59:40] leila: he did [20:59:45] and I think he dropped the triangles [21:01:42] bmansurov_: ow okay. then it seems all good, and on your end. [21:02:36] leila: not yet, still trying to figure out why I'm seeing missing usernames in the list. I'll find the cause and then talk to you guys if that's something serious. [21:02:49] yup. sounds good. [21:24:46] 10Quarry, 10Patch-For-Review: Show desktop notification when a query is done - https://phabricator.wikimedia.org/T124625 (10Framawiki) 05Open>03Resolved [21:42:10] leila: guess we'll have to wait until Monday [21:53:57] * leila checks the email [21:55:57] bmansurov_: I see an updated list (just arrived), but it's already late for you. If you don't manage to send it out today, Monday is fine. It's in general better if we catch the people before the weekend, but we should definitely not send emails out over the weekend when we're not around full force to catch issues if they occure. [22:05:31] leila: oh cool, i'll send the first batch soon [22:08:10] great [22:46:32] lzia: sent! [22:46:38] 31 users [22:52:23] bmansurov_: \o/