[00:09:44] 10Phabricator, 10DBA, 06Operations, 13Patch-For-Review: Upgrade m3 (phabricator) db servers - https://phabricator.wikimedia.org/T138460#2464778 (10greg) No worries, thanks Jaime. [02:19:48] Danny_B yay it works [02:20:09] yes [02:20:11] twentyafterfour you can deploy https://phabricator.wikimedia.org/rPHABcf12fdf248df82dc414d96bddd147c058bc3d636 to production now please. [02:20:40] We found the cause of the problem it was opcache [02:20:43] causing the problems [02:20:48] restarting apache fixed it [02:22:48] Danny_B ok ive turned off something to do with rev timestamps in opcache which means it made us have to restart apache everytime we editted php [02:23:03] Danny_B could you try a syntax error again please [02:23:49] yup, works now without restarting [02:25:06] Ok, :) [02:25:09] thanks for testing [02:25:32] Parent tasks works too https://phab-01.wmflabs.org/T63 [02:26:55] Oh it seems task graphs have stopped working tottaly [02:26:56] https://phab-01.wmflabs.org/T6 [02:27:10] But dosent matter [02:30:04] twentyafterfour ^^ im not sure but it fixes graph not showing but on tasks that have one sub task or parent task it dosent show graph now [02:31:22] paladox: that should be ok [02:31:26] Ok [02:31:43] Yep i think so too, we can always fix that at another time [02:32:04] Anyways i have to go now, bye, it is 03:31 am here [02:32:18] having no parents in graph is desired [02:32:23] Oh [02:32:29] but what about subtasks [02:32:51] so if any task doesn't have subtasks, but parents only, then table should be displayed instead of graph [02:32:59] Oh [02:34:02] Yep that works [02:34:09] parent tasks seems to break graphs [02:34:59] But anyways have to go :) Bye Danny_B and twentyafterfour [02:35:08] paladox: https://phab-01.wmflabs.org/T63 [02:35:18] :) [02:35:22] (i've set the limit to 100 for testing) [02:35:26] Oh [02:35:29] just showing you it works [02:35:29] Ok [02:35:34] yep thanks [02:35:36] :) [02:36:03] Bye [02:37:06] hmm, actually that's a weird bug [02:37:34] having limit 50 https://phab-01.wmflabs.org/T6 shows table [02:37:46] having limit 100, it shows graph [02:40:06] the edge is 64/65 64 is table, 65 is graph [02:40:49] wondering if it has something to do with parent task which is 63 or if it is just a coincidence [02:41:56] nope. changed parent 63 to 59 and still table [04:32:48] 10Phabricator: Make OAuth session persistent in Phabricator - https://phabricator.wikimedia.org/T140448#2465121 (10Jay8g) [07:38:37] Oh [08:43:15] twentyafterfour hi, it seems to shows commits twice [08:43:16] https://phab-01.wmflabs.org/T59 [08:43:58] But why dosent parent task allow task graphs to work. It should only disable graph if it reaches 50+ task [08:45:26] Danny_B ^^ [08:55:47] Plus it shows commits is the wrong place. [08:55:57] I thought there meant to be under the description [10:58:52] 10Phabricator: Make OAuth session persistent in Phabricator - https://phabricator.wikimedia.org/T140448#2465508 (10Aklapper) Could you elaborate which steps you perform that require you to log in every time? I'm not sure I understand what is requested here... :-/ [15:05:02] 10Phabricator: Same task is shown in task graph - https://phabricator.wikimedia.org/T140473#2466110 (10Paladox) [15:08:35] 10Phabricator: Same task is shown in task graph - https://phabricator.wikimedia.org/T140473#2466142 (10Danny_B) 05Open>03Invalid That's intended behavior. You always need a root of graph. Which is the current task (which is highlighted). [15:10:13] 10Phabricator: Same task is shown in task graph - https://phabricator.wikimedia.org/T140473#2466154 (10Paladox) Oh, thanks for explaning. [16:16:14] phab just blippd when I tried to save a task.. [16:17:27] twentyafterfour ^^ [16:18:37] what dose blipped mean? [16:18:39] didn't save? etc [16:24:57] Yeah [16:25:10] Possibly generic error page from varnish [16:25:49] yeah that's on our end arlight [16:25:54] but I have no guesses [16:25:55] atm [16:26:12] I just refreshed, and Chrome gave the "Do you want to resubmit?" type message [16:40:19] 10Phabricator: Like/dislike token image was changed retroactively - https://phabricator.wikimedia.org/T140481#2466367 (10Nemo_bis) [17:49:20] 10Phabricator, 10Analytics-Tech-community-metrics: Decide on wanted metrics for Maniphest in kibana - https://phabricator.wikimedia.org/T28#2466546 (10Aklapper) [17:50:32] 10Phabricator, 10Analytics-Tech-community-metrics: Decide on wanted metrics for Maniphest in kibana - https://phabricator.wikimedia.org/T28#2466550 (10Aklapper) [18:37:55] hahaha, "view raw" now has a knife and fork icon next to it (when clicking the comment dropdown for eg: quoting) [18:46:34] 10Phabricator: Like/dislike token image was changed retroactively - https://phabricator.wikimedia.org/T140481#2466804 (10Aklapper) 05Invalid>03declined This looks like a "I personally don't like some color change as it reminds me of items on another website which I personally don't like" report without any f... [18:58:55] are phab notifications disabled? [18:59:23] No [18:59:46] Well not that i am aware of, ive been getting notifications all day. [18:59:54] You may want to check your preference. [18:59:57] mutante ^^ [19:00:22] mutante: You should be happy if noone pings you about work :P [19:02:21] ok, thanks. no, another user told me he didnt get any notifiations about access requests. and then it seemed like i also didnt get them, but i guess their is just no activity [19:02:47] no, he did stuff on them, like adding subtasks. hmm [19:06:09] Oh, we did do some changes to subtasks and parent tasks to fix graph [19:06:15] so not sure if that is related [19:07:15] i'll keep an eye on it [19:08:01] last one I got was 25 or so minutes ago [19:08:04] fwiw [19:08:34] about the GPG keys in Gerrit.. i dunno .. i mean it _could_ _maybe_ .. be nice for access requests if people want to sign that it's actually them requesting to replace that SSH key .. [19:09:04] chasemp: ok, thx [19:12:48] mutante: yeah, I'm like "neat, maybe we'll use it" but then also "I dont' have a use case in mind yet" [19:14:51] sometimes i was tempted to ask for GPG signing when people upload requests to replace their SSH keys [19:15:07] and it's like we are just trusting the gerrit user at that point.. [19:15:23] then most of the times we end up using another form of verification [19:15:35] i wouldnt mind it [19:21:36] greg-g it is optional [19:21:43] it wont cause any problems for anyone [19:22:09] except from those that actually use git push --signed and doint have a valid certificate and didnt upload there gpg key to gerrit [19:22:17] mutante ^^ [19:23:22] paladox: I know it's optional [19:23:27] i dont know about any details yet [19:23:29] Ok [21:00:45] btw paladox and Danny_B: thanks for helping with all the testing/code review last night to fix the task graph issue. [21:01:04] greg-g your welcome :) [21:16:57] Yeah for sure, super helpful I literally couldn't have gotten that done without the help [21:17:38] :) [21:21:06] * andre__ also thanks [21:21:25] Your welcome :) [22:31:32] it's always funny to see a task with only one parent and one subtask that doesn't get the task graph treatment. (must be a small part of a tangled web of work ;) ) [22:32:45] greg-g: ? [22:33:28] Yeh i think that may be a bug [22:33:33] i found that strange too [22:45:04] the one in question: https://phabricator.wikimedia.org/T138778 [22:45:21] two parents up to you get to: https://phabricator.wikimedia.org/T88445 [22:50:28] greg-g: earlier today i have reported some notes about it here [22:50:56] ah, I thought I read all of the scrollback, I was tired yesterday/this morning... [22:52:36] 04:37 UTC+2 was the start [22:53:09] utc +1 was for me [23:31:23] 05Gitblit-Deprecate, 06Project-Admins: Archive #Gitblit-Deprecate - https://phabricator.wikimedia.org/T138986#2467568 (10mmodell) [23:31:26] 05Gitblit-Deprecate, 06Operations: gitblit blobs not redirecting to the correct moved resource unless .git is part of repo in url - https://phabricator.wikimedia.org/T139027#2467566 (10mmodell) 05Open>03declined I don't think we need to care about redirecting ancient / malformed urls.