[15:05:11] <wikibugs>	 (03PS1) 10Andrew Bogott: passwords: remove dummy passwords for openstack ceilometer [labs/private] - 10https://gerrit.wikimedia.org/r/446037
[15:07:14] <wikibugs>	 (03CR) 10Andrew Bogott: [V: 032 C: 032] passwords: remove dummy passwords for openstack ceilometer [labs/private] - 10https://gerrit.wikimedia.org/r/446037 (owner: 10Andrew Bogott)
[16:46:45] <addshore>	 !log general-k8s create initial blank medium instances
[16:46:46] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:General-k8s/SAL
[21:10:49] <awight>	 !log deployment-prep ran namespaceDupes.php on beta enwiki
[21:10:53] <stashbot>	 Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Deployment-prep/SAL
[23:18:00] <yurb>	 Anyone using pipenv on tool forge?
[23:39:39] <yurb>	 I'm getting an error "/bin/sh: 1: mysql_config: not found" when trying to install mysqlclient with pipenv
[23:40:13] <yurb>	 inside a "webservice --backend=kubernetes python shell"
[23:47:39] <Cyberpower678>	 Any PHP experts about?
[23:48:05] <Cyberpower678>	 I could use one.  I've got the weirdest bug I could use some insight on.
[23:48:44] <Reedy>	 !ask
[23:48:44] <wm-bot>	 Hi, how can we help you? Just ask your question.
[23:48:51] <Cyberpower678>	 Essentially I have my IABot code on my dev machine and an exact copy of it on my production machine, IE toolforge
[23:49:21] <Cyberpower678>	 I'm processing a page in my ruwiki sandbox, but on production the output is wrong, but on my dev, the output is right.
[23:50:24] <Cyberpower678>	 Reedy: when I run IABot on my local machine on my sandbox, I get https://ru.wikipedia.org/w/index.php?title=Участник:Cyberpower678/Черновик&diff=94008025&oldid=94008019 which is correct
[23:51:00] <Cyberpower678>	 Reedy: when I run the exact same code on Toolforge or the cloud VPS, I get https://ru.wikipedia.org/w/index.php?title=Участник:Cyberpower678/Черновик&diff=94008084&oldid=94008072
[23:51:05] <Cyberpower678>	 which is incorrect.
[23:51:22] <Cyberpower678>	 No idea why.
[23:51:44] <Cyberpower678>	 At first I thought it's a difference in the PHP version except Toolforge and my dev machine are both on 5.5
[23:52:41] <Cyberpower678>	 Fundamentally, what could cause a difference of processing like that, if the code is identical, as well as the PHP version, and the input.  The only variable left is the machine.
[23:55:20] <Reedy>	 How much can you narrow a test case of your code?
[23:55:41] <Cyberpower678>	 Reedy: can you clarify?
[23:55:56] <Reedy>	 Minimise your reproduction case
[23:56:11] <Reedy>	 Rather than saying "it's somewhere in this codebase" to "it's in this function"
[23:56:43] <Cyberpower678>	 Reedy: oh that's easy.  I had it print out some stuff to see what it's thinking.  I've already narrowed it down to a function
[23:56:53] <Reedy>	 Link? Pastebin?
[23:57:45] <Cyberpower678>	 Reedy: https://github.com/cyberpower678/Cyberbot_II/blob/test/IABot/Parser/parse.php#L2516 is where the decision making happens.
[23:57:58] <Cyberpower678>	 For how to format the template.
[23:58:26] <Cyberpower678>	 As well as here ->https://github.com/cyberpower678/Cyberbot_II/blob/test/IABot/Parser/parse.php#L2585
[23:59:30] <Cyberpower678>	 To sum it up, for the first template that was collapsed, the final format decision came out to be "{key}       = {value} " on production which is incorrect.