[00:23:29] <TimStarling>	 bpirkle: https://gerrit.wikimedia.org/r/#/c/mediawiki/core/+/469354
[00:28:59] <TimStarling>	 anyone still around? it ideally needs to be cherry picked and deployed right now
[00:31:24] <TimStarling>	 it seems to work, although I have no idea why selectRowCount() is doing what it's doing
[00:32:25] <TimStarling>	 oh, the solution is to just not use it
[00:32:38] <TimStarling>	 the documentation says "This is useful when trying to do COUNT(*) but with a LIMIT for performance."
[00:32:45] <TimStarling>	 which we don't need to do
[00:35:15] <TimStarling>	 PS2 coming up
[00:36:56] <James_F>	 TimStarling: I can be around if needed.
[00:40:32] <legoktm>	 TimStarling: I'm here now
[00:40:42] <TimStarling>	 daniel is reviewing it
[00:42:09] <James_F>	 Kk.
[00:42:45] <James_F>	 I'll assume that between you, you can get it deployed, so I'll go home. :-)
[00:44:29] <legoktm>	 the patch looks fine to me
[00:48:36] <paladox>	 Should that be backported to 1.32 too?
[00:49:08] <legoktm>	 paladox: yeah, once it's merged in master
[00:49:13] <paladox>	 Ok
[00:52:33] <legoktm>	 paladox: and thanks for being so on top of that :) 
[00:53:41] <paladox>	 Your welcome :)
[01:33:07] <TimStarling>	 so slow, I guess it's only pizza though
[01:58:11] <TimStarling>	 ok, deployed twice, and tested in production, I'm going to get pizza now, will be out of comms
[01:59:24] <TimStarling>	 I'll take my laptop in case there's a problem
[02:22:25] <legoktm>	 Enjoy your pizza!
[17:21:01] <Krinkle>	 PHPUNit tests failing on all PHP/db versions:
[17:21:03] <Krinkle>	 > 
[17:21:03] <Krinkle>	 1) MediaWiki\Tests\Revision\MutableRevisionRecordTest::testSetGetPageId
[17:21:03] <Krinkle>	 Failed asserting that 2 is identical to 0.
[17:21:06] <Krinkle>	 on Travis-CI
[17:21:28] <James_F>	 Krinkle: https://gerrit.wikimedia.org/r/c/mediawiki/core/+/469468 See -dev.
[17:27:02] <Krinkle>	 James_F: Hm.. I'm not seeing the link between that and the 2 != 0
[17:27:27] <James_F>	 Oh, are there two sets of errors in Travis?
[17:29:31] <legoktm>	 Krinkle: in travis or jenkins?
[17:29:35] <Krinkle>	 Travis
[17:49:24] <Krinkle>	 cscott: Is https://gerrit.wikimedia.org/r/#/c/mediawiki/core/+/468589/ safe to backport?
[17:49:46] <Krinkle>	 also, are the fatals from https://phabricator.wikimedia.org/T207433 new since 33.wmf.1 or 32.x?
[17:50:21] <Krinkle>	 wmf.26, introduced it right?
[18:06:21] <James_F>	 Krinkle: wmf.26, but they're entirely synthetic (caused by Formafix writing bad URLs). I back-ported to REL1_32 but no need for prod.
[18:06:29] <Krinkle>	 k
[18:06:43] * Krinkle adds to the Logstash ignore list
[18:07:13] * Krinkle sees that the number of excludes is now more than 2 15" screens full, so he can't edit the labels without scrolling due to a bug in Kibana.
[18:07:28] * Krinkle considers getting a larger monitor.
[18:07:53] <Krinkle>	 https://logstash.wikimedia.org/app/kibana#/dashboard/0a9ecdc0-b6dc-11e8-9d8f-dbc23b470465 :)
[18:08:19] <James_F>	 Krinkle: https://www.samsung.com/us/computing/monitors/gaming/49--chg90-qled-gaming-monitor-lc49hg90dmnxza/
[18:08:34] <Krinkle>	 Thx, that'll go nicely on my expense report.
[18:08:47] <Krinkle>	 "Logstash monitor monitor": Motivation: Too many errors
[18:09:07] <Krinkle>	 "Technical debt" - it's expensive
[18:09:33] <Krinkle>	 James_F: That's actually cheaper than I thought it would be. Not bad!
[18:10:38] <James_F>	 The downside of using great external monitors is that you adjust your workflow and become dependent on them for productivity. I hate hate hate being away from the office and trying to get anything done on just my laptop without the other two screens.
[18:11:05] <Reedy>	 Monitor as handluggage
[18:11:07] <Reedy>	 $profit
[18:11:21] <Krenair>	 lol
[18:13:23] <legoktm>	 that's how I feel about my mechanical keyboard...
[18:13:33] <legoktm>	 I've been looking into buying a portable one to take with me everywhere
[18:16:35] <Reedy>	 legoktm: Hello Yuvi
[18:16:46] <legoktm>	 haha
[18:24:55] <Reedy>	 https://www.candykeys.com/product/vortex-core-rgb
[19:23:43] <James_F>	 legoktm: If it's portable, it's not a mechanical keyboard. ;-)
[19:30:43] <MatmaRex>	 tenkeyless keyboards are luggable ;)
[20:22:55] <greg-g>	 I really should bring my Kinesis on more trips. It'd be one thing if it was a team offsite, but this conference thing moving between rooms would be hard
[20:37:36] <legoktm>	 greg-g: Yuvi did it!
[20:40:01] <greg-g>	 legoktm: yeah, I'm not as cool as yuvi. Obviously. :)
[20:40:20] <legoktm>	 greg-g: well if you die your hair green.... ;-)
[20:40:59] <greg-g>	 I've been considering a quasi-mowhawk
[20:48:00] <greg-g>	 https://phabricator.wikimedia.org/T207881 can someone here look at this, please. cc anomie AaronSchulz 
[20:48:20] <greg-g>	 wondering if https://gerrit.wikimedia.org/r/#/c/mediawiki/core/+/467882/ is implicated or not
[20:49:13] <AaronSchulz>	 that change should only affect logging (and lessen it, since less things are seen as contention)
[20:50:19] <greg-g>	 AaronSchulz: hi btw! thoughts on what might be the cause of the spike?
[20:51:25] <AaronSchulz>	 I see a lot about 'Query: UPDATE  `change_tag_def` SET ctd_count = ctd_count + 1 WHERE ctd_name = 'OAuth CID: 699''
[21:01:03] <AaronSchulz>	 at the least, that sort of update should use onTransactionPreCommitOrIdle, which it doesn't
[21:02:07] <AaronSchulz>	 https://logstash.wikimedia.org/app/kibana#/doc/logstash-*/logstash-2018.10.24/mediawiki?id=AWanzDom00on8STvWRay&_g=h@97fe121
[21:20:16] <twentyafterfour>	 hmm
[21:20:42] <twentyafterfour>	 I also saw similar query that set edit_count +1 showing up a bunch 
[21:21:44] <AaronSchulz>	 that's a bit of an older problem, but also has a patch