[00:16:19] <Krinkle>	 csteipp: FYI: The jenkins failures you might see are false, I aborted various jobs since you already merged the changes anyway.
[00:17:04] <csteipp>	 Krinkle: Thanks! I was wondering about them. It didn't seem right, but good to know for sure :)
[00:17:06] <Krinkle>	 There were about a dozen changes that each pushed 3 pipelines into halt for submission test, gate-and-submit and postmerge.
[00:17:24] <Krinkle>	 csteipp: In the future it may be best to just CR+2 instead of CR+/V+2/Submit
[00:17:40] <Krinkle>	 that way everybody waits a minute instead of you not waiting and everybody else waiting 15 minutes.
[00:17:59] <Krinkle>	 jenkins merges once the test complete.
[00:18:37] <Krinkle>	 Not long from now the V+2 and Submit button will be removed, just don't get used to them :)
[00:19:02] <csteipp>	 Yeah, I'll try and be more patient next time :)
[00:19:17] <Krinkle>	 https://integration.wikimedia.org/ci/ (sidebar)  https://integration.wikimedia.org/zuul/
[00:19:22] <Krenair>	 When that happens how will we be expected to merge into branches where the tests are broken?
[00:19:31] <Krinkle>	 complete halted, things have been stuck for 30 minutes. Still trying to clear stuff.
[00:19:58] <Krinkle>	 Krenair: Well, firstly, a failure is usually easy to fix. It'll take an extra commit, but it'll have to happen at some point anyway.
[00:20:20] <Krinkle>	 If it is of the utmost importance a gerrit admin (most people in ops) will be able to override still
[00:20:33] <Krinkle>	 though in such emergencies the code is probably already live and not in gerrit anyway.
[00:20:57] <Krinkle>	 in most if not all cases it'll be best for everyone to 'just pick a number'
[00:20:57] <Krenair>	 I don't really think we should be adding extra test fixes to the release branches, but okay
[00:21:15] <Krinkle>	 Krenair: If tests are failing, you're likely to ship broken code
[00:21:19] <Krinkle>	 ESPECIALLY to release branches.
[00:21:45] <Krinkle>	 backporting a commit that passes in master doesn't mean it'll pass in a release branch
[00:21:56] <Krinkle>	 It can interact with all sorts of different code that is not up to date in a release branch
[00:22:03] <Krinkle>	 thus cause all kinds of bugs that are had to for see.
[00:22:05] <Krinkle>	 hard*
[00:22:31] <Krenair>	 This was changes that were already released, just not put into the release branches
[00:22:45] <Krinkle>	 I assume that since the code was already released that csteipp or whomever has properly tested the code and installer locally manually.
[00:23:12] <Krinkle>	 since otherwise they wouldn't be bypassing jenkins (there shouldnt be a rush to get pass gerrit since it was already released)
[00:24:44] <csteipp>	 Yep. I think the issue is we've lived with broken unit tests (REL1_18) before, so when things start failing, people (or maybe just me..) assume that the testing is just broken.
[00:25:21] <csteipp>	 But good to know that it shouldn't be breaking. I aggree, backporting is more likely to break stuff.
[00:25:58] <Krinkle>	 csteipp: also, I believe that the person who merged these did not even wait for a failure to pop up
[00:26:01] <Krinkle>	 it was merged instantly
[00:26:07] <Krinkle>	 even before the submission tests completed.
[00:26:37] <Krinkle>	 If it failed it can be assessed and bypassed if need be
[00:26:40] <Krinkle>	 1.18 was a long time ago.
[00:28:17] <Krenair>	 This was code that was already officially released by csteipp. Some obviously erroneous failures I bypassed.
[00:29:11] <csteipp>	 Have to run for the train, but Krinkle, I do appreciate your input. Thanks.
[00:29:34] <Krinkle>	 csteipp: Alright
[00:29:42] <Krenair>	 If it's been properly released it should be in the branch, regardless of what Jenkins thinks.
[00:30:14] <Krinkle>	 Krenair: I understand the thoughts, but you're justifying it now with false reasons. There was no failure to be erroneous for you merged it before there was any response from jenkins.
[00:30:31] <Krinkle>	 Krenair: Yes, so it will be merged. But if there is a failure, it should be known.
[00:30:35] <Krinkle>	 You can bypass it, but it should be known
[00:30:37] <Krenair>	 Some of them already did fail:
[00:30:38] <Krenair>	 https://gerrit.wikimedia.org/r/#/c/59340/
[00:30:40] <Krenair>	 https://gerrit.wikimedia.org/r/#/c/59339/
[00:30:49] <Krenair>	 https://gerrit.wikimedia.org/r/#/c/59342/
[00:30:52] <Krinkle>	 Now it will not be known because if jenkins reports after you merged it, the score is ignored
[00:31:02] <Krinkle>	 Krenair: nonsense
[00:31:17] <Krinkle>	 Krenair: As I said, I aborted csteipp's stack of doom because the whole queue went clusterfuck
[00:31:44] <Krinkle>	 over a 100 jobs occupying the queue for over an hour now, nobody's been getting anything.
[00:32:15] <Krenair>	 I shouldn't have done the Verified+1 and merge without Jenkins on the REL1_19 branch which jenkins seemed happy with.
[00:32:36] <Krinkle>	 ?
[00:32:43] <Krinkle>	 I dont understand that sentence/
[00:33:33] <Krenair>	 Some of these backports had already been failed by Jenkins for reasons that seem to be completely unrelated to the changes themselves
[00:34:11] <Krinkle>	 I'm not discussing any of the failures. If there are failures, fine, bypass them. But you wait and see. Not assume it'll fail and assume it'll be a false positive and merge instantly.
[00:34:21] <Krinkle>	 That's what happened and that's what you need not do again. That's all I'm saying.
[00:34:52] <Krenair>	 The others had not and would not be, I was wrong to verified+2 (I said +1 before, oops) and merge those ones. I understand that.
[00:35:12] <Krinkle>	 alrighty then
[00:42:36] <Krenair>	 I'm sorry that it caused Jenkins to get a huge queue of tests Krinkle. I should've guessed that would happen
[00:46:21] <Krinkle>	 Krenair: Don't worry about it. The queue is getting near normal wait time (down to 12 minutes from 40 minutes, should be 2 ~ 5 ).
[00:47:02] <Krinkle>	 Krenair: Though the queue is annoying, it's nothing permanent. I'm more worried about not knowing whether our branches pass tests and/or introduced issues in recent back ports.
[00:51:16] <Krenair>	 I looked through those failed tests, nothing in them seems to be related to the changes themselves.
[00:55:34] <Krinkle>	 The ones you see now (the ones generated after they were already merged) fail because I aborted them.
[08:37:12] <Krinkle>	 hashar: Can you look into -qunit jobs failing? I added a hack into the shell script to do a curl fetch before the phantomjs run.
[08:37:16] <Krinkle>	 Looks like there are db errors
[08:37:20] <Krinkle>	 unable to read the sqlite file
[08:37:21] <Krinkle>	 readonly
[08:38:32] <Krinkle>	 hashar: sample: https://integration.wikimedia.org/ci/job/mediawiki-core-qunit/2091/console
[08:38:45] <Krinkle>	 HTTP/1.1 500 MediaWiki exception  Error: 8 attempt to write a readonly database
[08:38:49] <Krinkle>	 Doesn't happen all the time though
[08:39:08] <Krinkle>	 That's what I found yesterday, perhaps you can get investigate further?
[08:39:17] <Krinkle>	 I'll check in in about 10 hours
[08:39:29] * Krinkle  emails as well

[08:41:17] <Aaron|home>	 you can get those errors with excessive concurrency with sqlite
[09:31:10] <hashar>	 buuuug report
[12:04:44] <Danny_B>	 Dereckson: ping?
[12:41:08] <hashar>	 anyone around has any idea how we generate texvc  nowadays?
[13:57:38] <gerrit-wm>	 Change merged: Krinkle; [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/59358
[14:57:33] <^demon>	 qchris: How does gerrit calculate container.heapLimit if it's undefined?
[14:57:51] <^demon>	 I know docs say platform/JVM specific, but that's not all that clear to me.
[15:15:23] <qchris>	 ^demon: if heapLimit is undefined, gerrit.sh just does not pass a limit to java
[15:15:32] <qchris>	 So it's up to the jdk that you use
[15:15:37] <^demon>	 Ah, gotcha.
[15:17:34] <^demon>	 oracle site says that's 25% (so still 2G), but I can't find openjdk-specific docs on it.
[15:18:01] <qchris>	 the show-cache command should tell you
[15:18:21] <qchris>	 show-caches (s at the end)
[15:18:28] <qchris>	 There you'll find a "max" value
[15:19:09] <qchris>	 and that is computed via Runtime.getRuntime().maxMemory()
[15:21:36] <^demon>	 http://p.defau.lt/?tmYTpNgM2e10K5klM7nv_A
[15:22:06] <qchris>	 So 2G again :-)
[15:23:13] <qchris>	 But since the total used memory and the maximum available memory are close, maybe we could try going with 3G?
[15:23:38] <^demon>	 3G would be a 50% increase, but still far short of exhausting the total system memory.
[15:23:40] <^demon>	 Could try that.
[15:24:11] <qchris>	 I think it was on the repo-discuss mailing list that someone said they use 100G for gerrit :-)
[15:26:10] <^demon>	 Yeah, some installs have absurd amounts of memory and cpus.
[15:26:34] <^demon>	 I tried to convince ma rk to give me one of those, but he said I didn't need it ;-)
[15:27:06] <qchris>	 *gg*
[15:29:42] <Aayush|Away>	 c
[15:48:38] <zeljkof>	 guillom: do you have a minute now?
[15:49:18] <guillom>	 zeljkof: I need to get my blood results; can I ping you in ~20min?
[15:49:35] <zeljkof>	 guillom: sure, there is no rush
[15:49:45] <guillom>	 ok :)
[16:28:55] <guillom>	 zeljkof: I'm back
[16:29:24] <zeljkof>	 guillom: great
[16:29:37] <zeljkof>	 just a minute to finish something
[16:30:10] <guillom>	 sure
[16:36:55] <zeljkof>	 guillom: I am ready
[16:36:59] <guillom>	 ok
[16:37:04] <guillom>	 What can I do to help?
[16:37:30] <zeljkof>	 I have created this page
[16:37:31] <zeljkof>	 http://www.mediawiki.org/wiki/Bugzilla_Weekly_Report
[16:37:36] <zeljkof>	 it is simple
[16:37:40] <zeljkof>	 just embeds a few files
[16:37:51] <zeljkof>	 problem #1
[16:38:18] <zeljkof>	 for example for http://www.mediawiki.org/wiki/File:Status_changes_this_week.png
[16:38:52] <zeljkof>	 I have copy/pasted the page format from commons
[16:39:05] <guillom>	 Ah, I see
[16:39:19] <guillom>	 and you're getting red links because the templates don't exist on mediawiki.org
[16:39:27] <zeljkof>	 probably
[16:39:30] <guillom>	 Is there a reason you're not uploading the files to Commons?
[16:39:35] <zeljkof>	 no
[16:39:52] <guillom>	 Then that's a problem that's easily fixable :)
[16:39:53] <zeljkof>	 I just did not think those charts are good files for commons
[16:40:08] <guillom>	 I think they're fine.
[16:40:29] <zeljkof>	 so I should just upload the file to commons and change this links at the page to point to files at commons?
[16:41:02] <bawolff>	 I imagine such charts might be used in signpost, so commons would probably be good for that reason alone
[16:41:22] <guillom>	 Upload the files to Commons with the same name and delete the files on mediawiki.org; if there isn't a copy on mediawiki.org, the ones from Commons will be loaded automatically
[16:41:34] <zeljkof>	 did not know that
[16:41:37] <zeljkof>	 will try
[16:41:40] <guillom>	 The more you know :)
[16:41:48] <zeljkof>	 let me check if I can delete the files
[16:42:15] <zeljkof>	 no, I do not think I can delete the files
[16:42:21] <guillom>	 If not, I'm happy to do it once you're ready
[16:43:15] <zeljkof>	 ok, I will let you know
[16:43:31] <zeljkof>	 question/problem #2 (the last one)
[16:43:53] <zeljkof>	 if I upload new version of a file
[16:44:01] <zeljkof>	 will it automatically be updated at http://www.mediawiki.org/wiki/Bugzilla_Weekly_Report ?
[16:44:08] <guillom>	 yes
[16:44:11] <zeljkof>	 ok
[16:44:30] <zeljkof>	 since I am uploading new versions of the files every week
[16:44:55] <zeljkof>	 I will upload the files to commons and let you know when you can delete them from mediawiki.org
[16:45:10] <guillom>	 ok
[16:45:59] <zeljkof>	 should I upload the files with my personal or wmf account? :)
[16:46:14] <zeljkof>	 I am uploading files to commons usually with my personal account
[16:47:05] <guillom>	 honestly, I don't think it matters much :)
[16:47:11] <zeljkof>	 ok
[16:47:27] <zeljkof>	 I know there are conventions, it is sometimes better to ask :)
[16:48:13] <guillom>	 It may be better if you use your wmf account, but I don't think anyone will give you trouble for using your personal account
[16:48:41] <guillom>	 They're more guidelines than conventions, really. Many employees don't have a WMF-specific account.
[16:48:57] <zeljkof>	 I was told I should create wmf account
[16:49:46] <anomie>	 greg-g- Should we schedule a window to deploy the l10nupdate fix sometime today?
[16:50:30] <greg-g>	 anomie: sure thing, but ugh, 3-4pm Pacific is late your time...
[16:51:13] <greg-g>	 anomie: how long do you think it'll take you? not very?
[16:51:28] <marktraceur>	 Also you're talking to RoanKattouw in another channel about giving him 15-16
[16:51:48] <greg-g>	 on Thursday
[16:51:49] <anomie>	 greg-g- It's not too bad, and I have nothing else to do tonight besides cooking dinner. The deploy itself will be quick, but if we want to run a test instead of waiting to see if things blow up at 2:00 UTC, that will take longer
[16:52:17] <greg-g>	 right right....
[16:52:26] <greg-g>	 longer as in....?
[16:53:34] <anomie>	 No idea. Ideally no one will even notice the thing is running until a new message gets logged by logmsgbot.
[16:55:06] <greg-g>	 anomie: well, running an l10n update at ~22:30 seems.... really close to the 02:00 normal update
[16:55:52] <greg-g>	 I just have no idea how long those things take on the backend (like, is it an hour of grinding before the update?)
[16:56:57] <anomie>	 The update itself seems to take about 30 minutes. Then the new code I have no idea, it's too new.
[16:58:04] <greg-g>	 well, the only thing I'm worrying about is if anything bad happens if a new run if l10n starts when a previous one is still going
[16:58:17] <greg-g>	 if that won't be an issue, then it should be fine
[16:59:25] <anomie>	 If something does blow up between 02:00-03:00 UTC, I'll probably still be up. Or else Tim should be around to revert it. I just hate to push more to him with everything else he does.
[17:01:49] * greg-g  nods

[17:02:43] <greg-g>	 anomie: so, are you comfortable with the 3-4pm window? It sounds like your iffy, and if you're iffy, I'm more than iffy ;)
[17:05:05] <anomie>	 Just talked to robla about it, we have a plan. Let's do it.
[17:05:16] * robla  confirms

[17:05:29] <greg-g>	 alrighty
[17:06:13] <robla>	 anomie pointed out that we can always disable the cron job if it looks like it's going to run long
[17:07:01] <greg-g>	 true that
[17:07:28] <zeljkof>	 guillom: all files uploaded to commons http://commons.wikimedia.org/wiki/Category:Bugzilla_Weekly_Report
[17:07:53] <zeljkof>	 feel free to delete them from mediawiki http://www.mediawiki.org/wiki/Category:Bugzilla_Weekly_Report
[17:09:08] <guillom>	 will do now
[17:09:54] <zeljkof>	 guillom: thanks
[17:09:59] <guillom>	 sure
[17:11:53] <guillom>	 done :)
[17:12:57] <zeljkof>	 guillom: thanks :)
[17:14:00] <zeljkof>	 and the charts are really displayed :)
[18:14:58] <gerrit-wm>	 New review: Zfilipin; "I think we should report a bug that test2 and wmflabs do not behave the same. There is no reason for..." [qa/browsertests] (master) C: -1;  - https://gerrit.wikimedia.org/r/59194
[18:16:01] <gerrit-wm>	 New review: Zfilipin; "I think we should report a bug that test2 and wmflabs do not behave the same. There is no reason for..." [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/59194
[18:19:15] <gerrit-wm>	 New review: Zfilipin; "I do not think we should change "anonymous user creates page" test to "logged in user creates page" ..." [qa/browsertests] (master) C: -1;  - https://gerrit.wikimedia.org/r/59335
[18:33:00] <bsitu>	 Reedy: ping
[18:47:35] <gerrit-wm>	 New patchset: Cmcmahon; "login properly on beta, failed attempt to wait for upload to complete, not sure why it does not complete" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/59446
[18:56:22] <gerrit-wm>	 New review: Zfilipin; "(3 comments)" [qa/browsertests] (master) C: -1;  - https://gerrit.wikimedia.org/r/59446
[19:03:40] <Aayush251>	 yurik_ are you there?
[19:23:45] <mwalker>	 greg-g: may I get part of today's lightning deploy to push https://gerrit.wikimedia.org/r/#/c/59457/1
[19:57:56] <Aayush251>	 Hey yurik :)
[19:58:13] <yurik>	 Aayush251, hi
[19:58:30] <Aayush251>	 Are you free at the moment?
[19:58:35] <yurik>	 somewhat :)
[19:59:18] <Aayush251>	 I was studying about the API roadmap would be good if you sort sout some of my doubts .
[19:59:35] <Aayush251>	 Oh ok tell me when you are completely free .
[19:59:39] <yurik>	 sure, what are the questions :)
[20:01:29] <greg-g>	 mwalker: sure, it doesn't look intense, it isn't, right?
[20:01:34] <greg-g>	 :)
[20:02:08] <mwalker>	 greg-g: nope; not intense
[20:02:18] <mwalker>	 just pushing that extension to master
[20:03:24] <greg-g>	 mwalker: cool, adding on the calendar now, will ping you if anyone else requests any LD time
[20:03:35] <mwalker>	 awesome; thanks :)
[20:04:05] <greg-g>	 mwalker: thank you :)
[20:04:23] <Aayush251>	 ><
[20:08:51] <Aayush251>	 This project
[20:09:05] <marktraceur>	 Aayush251: Anything we can help with?
[20:09:31] <Aayush251>	 mark I had some query with API rodamap project
[20:09:41] <Aayush251>	 so asking yurik :)
[20:10:07] <marktraceur>	 !tabcompletion
[20:10:07] <wm-bot>	 Instead of manually typing another person's nickname in IRC, you can type the beginning of their name and hit TAB on your keyboard to get it autocompleted, like on the command line.  This avoids misspellings.  Just make sure it's the right person!
[20:10:25] <marktraceur>	 Cool!
[20:10:49] <yurik>	 Aayush251, i'm still not sure what you are asking :)
[20:11:01] <Aayush251>	 my irc lagging ><
[20:14:13] <Aayush251>	 Except php is there any other language needed to be have knowledge on for this project?
[20:19:08] <yurik>	 Aayush251, not really
[20:19:20] <yurik>	 you should know things like SQL
[20:19:42] <yurik>	 which is another language technically ;)
[20:23:53] <Aayush251>	 I've worked with SQL in past with php
[20:26:23] <yurik>	 Aayush251, than you should be all set :)
[20:27:10] <Aayush251>	 I'm worried haha
[20:27:30] <yurik>	 sorry, i'm passing out (still jetlagged a bit). Write to me yurik at wikimedia dot org
[20:27:40] <Aayush251>	 There are too many things I didn't get on the API roadmap page
[20:27:41] <yurik>	 will try to answer any questions
[20:27:48] <Aayush251>	 Oh ok seeya
[20:28:01] <marktraceur>	 Aayush251: I submit that you should apply.
[20:28:01] <yurik>	 i will try to answer that too :)
[20:28:27] <Aayush251>	 I will apply marktraceur
[20:28:37] <Aayush251>	 whats the fun on working things you already know about :D
[20:28:44] <yurik>	 Aayush251, and if you would have understood everything, there would be no point in applying ;)
[20:28:46] <marktraceur>	 That's the spirit! :)
[20:30:53] <Aayush251>	 Thanks
[20:31:52] <Aayush251>	 I wanted to work on Browser Test Automation
[20:32:17] <marktraceur>	 Aayush251: Oh, that's a cool project too. Sadly our QA folks seem to be out at the moment.
[20:32:18] <Aayush251>	 But sadly its for Outreach For Women program
[20:32:26] <marktraceur>	 Oh, hm.
[20:32:45] <marktraceur>	 Aayush251: You could probably still poke the QA people and get them to be GSoC mentors
[20:33:15] <Aayush251>	 I did but they said this project doesn't meet the GSoC criteria
[22:32:14] <mutante>	 welcome gerrit-wm
[22:32:18] <mutante>	 moved pre merged changes
[22:37:25] <gerrit-wm>	 New review: Anomie; "(1 comment)" [mediawiki/extensions/CentralAuth] (master) - https://gerrit.wikimedia.org/r/58924
[22:38:01] <gerrit-wm>	 New patchset: Brian Wolff; "RefreshLinks on imagelinks on new uploads to kill tracking cat." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/59359
[22:38:39] * bawolff  laughs about the whole #wikimedia-dev channel coming full circle

[22:40:51] <gerrit-wm>	 New patchset: Brian Wolff; "Use a combined const, instead of string concat in Exif.php" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/58903
[22:41:01] <James_F>	 bawolff: Finally. :-)
[22:41:08] <awjr>	 greg-g: we've got a couple of small js changes to fix some issues we discovered post deployment today - can i try to get those out during lightning deploy?
[22:41:35] <gerrit-wm>	 New review: Brian Wolff; "Jenkins said error, but looks unrelated at first glance, so seeing if trying again fixes things." [mediawiki/core] (master) C: 2;  - https://gerrit.wikimedia.org/r/58903
[22:41:49] <greg-g>	 awjr: mwalker is using the LD right now, but he's starting shortly so there might be time for you as well. If there is time, yeah, go for it
[22:42:02] <gerrit-wm>	 New patchset: Bsitu; "(bug 47095)  Email digest header tweaks" [mediawiki/extensions/Echo] (master) - https://gerrit.wikimedia.org/r/59537
[22:42:08] <awjr>	 sweet thanks greg-g - mwalker can you ping me when you're done?
[22:42:38] <mwalker>	 awjr: if you're ready go ahead -- I'm still doing some last minute prep
[22:42:54] <greg-g>	 awjr: do you have a gerrit change for this, btw?
[22:43:00] <awjr>	 ok, mwalker i'll start prepping and we'll see who gets there first :)
[22:43:01] <mutante>	 i think i would like it if gerrit-wm would talk even more now, like say when a human adds another human as reviewer to a change
[22:43:04] <awjr>	 yeah greg-g
[22:43:13] <awjr>	 https://gerrit.wikimedia.org/r/#/c/59532/ and https://gerrit.wikimedia.org/r/#/c/59535/
[22:43:21] <greg-g>	 cool, thanks
[22:43:22] <awjr>	 oh, still waiting for jenkins to merge :0/
[22:43:27] <greg-g>	 oh, right....
[22:43:36] <greg-g>	 effing jenkins today
[22:43:49] <gerrit-wm>	 New patchset: Bsitu; "(bug 47095)  Email digest header tweaks" [mediawiki/extensions/Echo] (master) - https://gerrit.wikimedia.org/r/59537
[22:44:09] <awjr>	 is brad still doing the l10n updates fixes? dunno his irc nick...
[22:44:16] <gerrit-wm>	 New review: Brian Wolff; "PS3: Restructure per Aaron to reduce duplicated code. Also fixes the issues Reedy pointed out." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/59359
[22:45:21] <greg-g>	 awjr: anomie is done, TimStarling did the deploy and ran a l10n update to test, but you're good to go
[22:45:39] <awjr>	 ok, then im just waiting on jenkins i guess
[22:45:40] <anomie>	 greg-g: s/ran/is running/
[22:45:49] <TimStarling>	 l10nupdate is still in progress
[22:46:09] <greg-g>	 right, meant in "started" sense, my bad
[22:46:52] <TimStarling>	 are you doing a git pull or updating submodules?
[22:47:19] <TimStarling>	 if so, the LC part may have to be run again
[22:47:42] <gerrit-wm>	 New patchset: Mwalker; "Update ContributionReporting for 2013 fundraiser" [mediawiki/core] (wmf/1.22wmf1) - https://gerrit.wikimedia.org/r/59539
[22:49:19] <gerrit-wm>	 New patchset: Mwalker; "Update ContributionReporting for 2013 fundraiser" [mediawiki/core] (wmf/1.22wmf2) - https://gerrit.wikimedia.org/r/59540
[22:49:33] <awjr>	 i havent done anything at all yet TimStarling; i can wait til you're done
[22:49:55] <gerrit-wm>	 Change merged: Mwalker; [mediawiki/core] (wmf/1.22wmf1) - https://gerrit.wikimedia.org/r/59539
[22:50:10] <gerrit-wm>	 Change merged: Mwalker; [mediawiki/core] (wmf/1.22wmf2) - https://gerrit.wikimedia.org/r/59540
[22:52:28] <gerrit-wm>	 New patchset: Aaron Schulz; "Reduced DB contention in User::saveOptions()." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/59490
[22:55:38] <gerrit-wm>	 New review: Aaron Schulz; "Triggering jenkins again maybe" [mediawiki/core] (master) C: 2;  - https://gerrit.wikimedia.org/r/59490
[22:57:05] * awjr  shakes jenkins

[22:59:49] <gerrit-wm>	 New review: CSteipp; "(1 comment)" [mediawiki/extensions/CentralAuth] (master) - https://gerrit.wikimedia.org/r/58924
[23:05:05] <gerrit-wm>	 New patchset: Anomie; "Add Javascript login check against the central wiki" [mediawiki/extensions/CentralAuth] (master) - https://gerrit.wikimedia.org/r/58924
[23:07:17] <gerrit-wm>	 New review: Anomie; "PS4: Rename stuff to not conflict so much with Ie3d6523d" [mediawiki/extensions/CentralAuth] (master) - https://gerrit.wikimedia.org/r/58924
[23:09:46] <gerrit-wm>	 New patchset: Mwalker; "ContributionReporting Requires Colour" [mediawiki/extensions/ContributionReporting] (master) - https://gerrit.wikimedia.org/r/59544
[23:10:08] <mwalker>	 awjr: I think that's the problem -- it already has shaken baby syndrome
[23:10:15] <gerrit-wm>	 New patchset: Demon; "Support HTTP timeout detection in curl backend" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/59545
[23:10:24] <awjr>	 lol mwalker
[23:10:26] * awjr  slaps jenkins

[23:10:36] <gerrit-wm>	 New patchset: Jdlrobson; "Add dependency to mobile module" [mediawiki/extensions/CentralNotice] (master) - https://gerrit.wikimedia.org/r/59546
[23:10:36] <gerrit-wm>	 New patchset: Jdlrobson; "Remove unnecessary global" [mediawiki/extensions/CentralNotice] (master) - https://gerrit.wikimedia.org/r/59547
[23:10:36] <gerrit-wm>	 New patchset: Demon; "Make logging of search failures more useful" [mediawiki/extensions/MWSearch] (master) - https://gerrit.wikimedia.org/r/59548
[23:10:46] <mwalker>	 also; I totally blame you for ContributionReporting being so aggravating
[23:10:55] <awjr>	 what!
[23:10:59] <awjr>	 i inherited that mess :p
[23:11:34] <awjr>	 iirc kaldari had a hand in its evolution as well...
[23:11:35] <gerrit-wm>	 New review: Dzahn; "no jenkins on this (yet)" [wikimedia/communications/WP-Victor] (master); V: 2 - https://gerrit.wikimedia.org/r/53701
[23:11:36] <gerrit-wm>	 Change merged: Dzahn; [wikimedia/communications/WP-Victor] (master) - https://gerrit.wikimedia.org/r/53701
[23:11:51] <mwalker>	 actually neither of you are listed in the contributors
[23:11:55] <gerrit-wm>	 New patchset: Demon; "Support HTTP timeout detection in curl backend" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/59545
[23:12:00] <mwalker>	 tfinc: !!! it's all your fault!
[23:12:03] <awjr>	 hehehe trying to escape blame
[23:12:20] <kaldari>	 you think I would sign that shit ;)
[23:12:26] <ori-l>	 mutante: it should go hand-in-hand w/ https://gerrit.wikimedia.org/r/#/c/53702/
[23:12:36] <gerrit-wm>	 Change merged: Dzahn; [wikimedia/communications/WMBlog] (master) - https://gerrit.wikimedia.org/r/53702
[23:12:52] <tfinc>	 mwalker: huh ?
[23:13:07] <mwalker>	 ContributionReporting -- you signed it!
[23:13:18] <tfinc>	 signed it ?
[23:13:22] <gerrit-wm>	 Change merged: Adamw; [mediawiki/extensions/ContributionReporting] (master) - https://gerrit.wikimedia.org/r/59544
[23:13:29] <mwalker>	 you're listed in the credits
[23:13:39] <kaldari>	 mwalker: Actually I think Trevor's to blame
[23:13:46] <awjr>	 oh yeah
[23:13:51] <awjr>	 totes
[23:13:52] <tfinc>	 ahh wrote it. yeah possibly. that would have been years ago though so you'll need a blame map
[23:14:06] <tfinc>	 bring it on !
[23:14:16] <mutante>	 !log deploying new EventLogging on blog server
[23:14:18] <mwalker>	 hehe -- I'm trying just to nuke this extension!
[23:14:29] <awjr>	 tfinc: i think we're all trying to pass the buck on that one...
[23:14:29] <mwalker>	 I think it's best for all if it just goes away
[23:14:38] <awjr>	 but… pretty graphs!
[23:14:44] <kaldari>	 mwalker: ha, you must be running into our anti-nuke booby traps
[23:14:54] <tfinc>	 is that the table layout or the fancy canvas graphs ?
[23:14:56] <mwalker>	 http://test-reportcard.wmflabs.org/dashboards/donations
[23:15:09] <mwalker>	 this is the table layout horridness
[23:15:17] <tfinc>	 kill it!
[23:15:35] <mwalker>	 I am! but limn isn't quite ready yet
[23:15:45] <mwalker>	 so I had to update it one last time :'(
[23:15:45] <tfinc>	 what's it missing ?
[23:15:58] <mwalker>	 dunno; pizzaz or something
[23:16:02] <mwalker>	 see the link above
[23:16:13] <mwalker>	 it's getting there; but... I dunno
[23:16:56] <ori-l>	 mutante: hey, can you hold off on deploying that change for a minute, if you haven't done it already?
[23:17:09] <gerrit-wm>	 New patchset: awjrichards; "Update MobileFrontend to tip of production branch" [mediawiki/core] (wmf/1.22wmf1) - https://gerrit.wikimedia.org/r/59549
[23:17:12] <gerrit-wm>	 New patchset: awjrichards; "Update MobileFrontend to tip of production branch" [mediawiki/core] (wmf/1.22wmf2) - https://gerrit.wikimedia.org/r/59550
[23:17:13] <mutante>	 ori-l: i was between adding the new stuff and removing the old stuf
[23:18:05] <mutante>	 ori-l: the new stuff is in it already, but removing the old.. there are SVN remnants, and diff between what is actually deployed and what is in git :/
[23:18:25] <mutante>	 i see changes to $wp_roles ..
[23:18:42] <gerrit-wm>	 Change merged: awjrichards; [mediawiki/core] (wmf/1.22wmf1) - https://gerrit.wikimedia.org/r/59549
[23:18:47] <gerrit-wm>	 Change merged: awjrichards; [mediawiki/core] (wmf/1.22wmf2) - https://gerrit.wikimedia.org/r/59550
[23:20:41] <gerrit-wm>	 New patchset: Ori.livneh; "Don't filter out searches + jshint fix" [wikimedia/communications/WP-Victor] (master) - https://gerrit.wikimedia.org/r/59551
[23:21:15] <mutante>	 ah i see
[23:21:18] <ori-l>	 mutante: do you think https://gerrit.wikimedia.org/r/#/c/59551/ could piggyback? it touches a file that was introduced in the commit you just merged, so it should be properly versioned
[23:21:31] <ori-l>	 yeah, just read your message, sorry
[23:22:09] <mutante>	 yea, merging that
[23:22:45] <gerrit-wm>	 New review: Dzahn; "piggyback 59551" [wikimedia/communications/WP-Victor] (master); V: 2 C: 2;  - https://gerrit.wikimedia.org/r/59551
[23:22:45] <gerrit-wm>	 Change merged: Dzahn; [wikimedia/communications/WP-Victor] (master) - https://gerrit.wikimedia.org/r/59551
[23:22:59] <mutante>	 i like that term,heh
[23:23:12] <ori-l>	 heh
[23:23:47] <mutante>	 it's live
[23:23:53] * ori-l  tests

[23:24:10] <mutante>	 as to removing the old stuff, i hope it doesnt have to be right the second
[23:25:05] <mutante>	 different repo
[23:25:27] <mutante>	 maybe having just a single one would be easier?
[23:26:02] <mutante>	 hmm, oh well, one is a theme and one is a plugin
[23:26:02] <ori-l>	 don't ask me, i just added eventlogging and tried to leave things in a better state than i found them by tidying up some of the JS
[23:26:22] <mutante>	 that was half just thinking out loud
[23:26:48] <mutante>	 thanks:) yea, the spring cleanup one is also merged
[23:26:51] <ori-l>	 mutante: well, since you're there, we might as well take notes about what it would take to puppetize
[23:27:00] <ori-l>	 do you notice anything out of the ordinary?
[23:27:08] <ori-l>	 is it just mysql, wordpress + those two repos?
[23:28:16] <mutante>	 our customizations, one in themes and one in plugins, are those 2 repos
[23:28:27] <mutante>	 but wordpress itself itsnt in puppet
[23:29:05] <mutante>	 and afaik wordpress admins upgrade it via web ui as well
[23:30:27] <mutante>	 http://packages.ubuntu.com/precise/wordpress  but that is not what is running on marmontel
[23:31:07] <ori-l>	 mutante: maybe tar/gzip the entire directory? that way we can diff it against a clean download of the same version from wordpress.org
[23:31:18] <ori-l>	 i mean, tar/gzip and put it somewhere where i can reach it :P
[23:32:01] <Reedy>	 Then run screaming because it's wordpress code
[23:33:03] * ori-l  has never looked

[23:33:21] <mutante>	 ok, can do in a little..need to move work space
[23:33:57] <mwalker>	 jdlrobson: so were we pushing banners to all mobile clients on alpha and beta?
[23:34:18] <jdlrobson>	 mwalker: the code is there yes but currently desktop is always the device - i sent you a patchset
[23:34:34] <mwalker>	 yep; I just saw it; that's what triggered my question :)
[23:39:28] <gerrit-wm>	 New review: MZMcBride; "Be nice." [mediawiki/extensions/ContributionReporting] (master) - https://gerrit.wikimedia.org/r/59544
[23:42:09] <gerrit-wm>	 New patchset: Demon; "Replicated repositories are at /var/lib/git/*" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/59556
[23:42:47] <Aaron|home>	 \o/
[23:42:56] <gerrit-wm>	 New review: Demon; "I'm not sure if this is correct, or if we really want the repos at the new location (and need to upd..." [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/59556