[03:00:46] <Negative24>	 Is the sudo policy users field the users that can run sudo?
[03:08:50] <jeremyb>	 Negative24, not necessarily users. could be groups. why?
[03:09:37] <Negative24>	 jeremyb: I'm just trying to figure out the interface. I think I broke it. Its not letting me sudo any more.
[03:10:22] <Negative24>	 I have everything set as default but it won't apply the changes.
[03:11:52] <jeremyb>	 ldap could be cached on server
[03:13:20] <Negative24>	 ?
[03:14:10] <Negative24>	 I thought running puppet would fix it but I can't sudo so I tried to reboot and it says Failed to reboot instance performance-testing.
[03:15:32] <jeremyb>	 idk about that
[03:15:52] <jeremyb>	 i don't remember exactly how it's set up
[03:16:01] <jeremyb>	 and anyway, it's at least 6 months since i looked
[03:16:02] <Negative24>	 What are the default fields for Sudo Policy.
[03:16:13] <jeremyb>	 i *thought* it's all just pulled from LDAP
[03:17:40] <Negative24>	 Well now my instance is hosed. I can't reboot or sudo so I'm powerless.
[03:18:01] <Negative24>	 Perhaps andrewbogott_afk can help when not afk.
[03:19:32] <jeremyb>	 isn't he supposed to be mostly offline atm?
[03:20:52] <Negative24>	 Yea. Everyone is AFK because of Wikimania. We should get more people on after Mon.
[03:27:42] <jeremyb>	 no...
[03:27:50] <jeremyb>	 he's afk unrelated to wikimania
[03:27:53] <jeremyb>	 but you're gone now
[03:35:38] <wikibugs>	 3Wikimedia Labs / 3tools: MediaWiki blank page after uploading LocalSettings.php - 10https://bugzilla.wikimedia.org/69154 (10Liangent)
[03:47:38] <wikibugs>	 3Wikimedia Labs / 3tools: MediaWiki blank page after uploading LocalSettings.php - 10https://bugzilla.wikimedia.org/69154#c3 (10jeremyb) Maybe you know this page already but since no one mentioned it yet:  https://www.mediawiki.org/wiki/Manual:How_to_debug
[09:55:20] <AndyRussG>	 Hi... I have an instance that was configured with labs-vagrant. Then somehow it borked: some stuff under the /vagrant/ directory is gone and labs-vagrant returns an error. I can't find any log about what happened... Any ideas? This is it: https://wikitech.wikimedia.org/w/index.php?title=Special:NovaInstance&action=configure&instanceid=26b0e67d-f1b2-4ab1-bdf0-ef088edc85bc&project=editor-engagement&region=eqiad
[14:09:08] <andrewbogott>	 yuvipanda: I can log in as root on diamond-collector but not as myself.  The system seems very slow… also, puppet is disabled.
[14:09:15] <andrewbogott>	 May I enable puppet and refresh?
[14:09:21] <yuvipanda>	 andrewbogott: sure
[14:32:44] <andrewbogott>	 yuvipanda: now puppet is hanging… I think that instance must be OOM.  Will it break things if I reboot it?
[14:37:40] <yuvipanda>	 andrewbogott: also merge https://gerrit.wikimedia.org/r/#/c/125241/? :)
[14:46:50] <Negative24>	 andrewbogott: It seems sudo policies aren't applying and I can't reboot performance-testing (instance).
[14:47:27] <andrewbogott>	 Negative24: can you give me an example of a sudo problem?
[14:48:22] <Negative24>	 negative24@performance-testing:~$ sudo su [sudo] password for negative24:  Sorry, user negative24 is not allowed to execute '/bin/su' as root on performance-testing.eqiad.wmflabs.
[14:48:42] <Negative24>	 Take a look at the sudo policy for performance
[14:49:09] <Negative24>	 I shouldn't have to authenticate
[14:50:41] <andrewbogott>	 Negative24: lamp::labs includes apache and php, so I'm going to uncheck those other packages.  Defining them twice is causing puppet to fail
[14:50:45] <andrewbogott>	 (unrelated to your other questions)
[14:51:10] <Negative24>	 I was trying to run puppet but I couldn't sudo.
[14:51:40] <andrewbogott>	 yeah, makes sense :)
[14:51:45] <Negative24>	 Btw, I checked those to investigate the problem with apache that I was getting. Thought that those had a different config.
[14:52:20] <Negative24>	 and wouldn't wipe my config out of /etc/apache2/sites-available.
[14:52:27] <andrewbogott>	 what is your shell name?  I'm going to su to your account to see what sudo is doing.
[14:52:41] <Negative24>	 negative24
[14:52:52] <Negative24>	 I just logged out.
[14:53:27] <Krenair>	 Is it possible to host Magus Manske's wikidata-game locally?
[14:53:34] <Krenair>	 Magnus*
[14:53:58] <Krenair>	 I wanted to make a contribution or two
[14:54:14] <Krenair>	 But it relies on widar and I have no idea what to do about that
[15:03:51] <andrewbogott>	 Negative24: it looks like there's something wrong with the 'allow running as' bit in the sudo policy GUI.  When I unset that, it works properly.
[15:03:59] <andrewbogott>	 I'll look into this later on, but in the meantime you should be able to proceed.
[15:05:09] <Negative24>	 andrewbogott: Can you give me the defaults for the sudo policy?
[15:05:23] <Negative24>	 I thought that I maybe set something wrong.
[15:05:29] <Negative24>	 *I
[15:05:49] <Danny_B>	 yuvipanda: many thanks for your assistance!
[15:05:54] <wikibugs>	 3Wikimedia Labs / 3tools: MediaWiki blank page after uploading LocalSettings.php - 10https://bugzilla.wikimedia.org/69154#c4 (10Liangent) Looks like some configuration error.  See https://wikitech.wikimedia.org/wiki/User_talk:Liangent#MediaWiki_hosting_on_tools
[15:05:55] <yuvipanda>	 Danny_B: \o/
[15:05:56] <andrewbogott>	 I think the current settings (as I just changed them) are now the default.
[15:06:53] <yuvipanda>	 andrewbogott: that was the bug me and Coren_WM2014 filed too
[15:07:19] <andrewbogott>	 yuvipanda: I thought that was slightly different, about allowing 'run as' for service members?
[15:08:05] <yuvipanda>	 andrewbogott: yeah, but even attempting to use that made sudo to anything impossible, including audo to root
[15:08:16] <andrewbogott>	 ok, good to know :(
[15:13:23] <wikibugs>	 3Wikimedia Labs / 3tools: MediaWiki blank page after uploading LocalSettings.php - 10https://bugzilla.wikimedia.org/69154#c5 (10Jamison Lofthouse) 5UNCO>3RESO/WOR Problem was the wrong port set on $wgServer.
[15:16:20] <Negative24>	 andrewbogott: I don't know if you fixed it but the failing reboot command resolved itself.
[15:16:50] <andrewbogott>	 Negative24: cool.  I told the nova backend to reset the state.  I don't know why it was stuck like that.
[15:17:24] <Negative24>	 andrewbogott: It also was saying that it failed to reboot the instance when I sent a reboot request.
[15:17:41] <andrewbogott>	 Yeah, that was because nova thought it was already rebooting and wouldn't try again
[15:18:13] <Negative24>	 Makes sense
[15:21:23] <Negative24>	 andrewbogott: Wait... Did puppet delete my public_html with the last run?
[15:21:47] <andrewbogott>	 I doubt it… where was it?
[15:21:50] <andrewbogott>	 in your homedir?
[15:22:31] <Negative24>	 it was in /srv and I have enabled role::labs::lvm::srv
[15:22:45] <andrewbogott>	 Ah, well -- in that case, sort of...
[15:22:53] <andrewbogott>	  /srv is now a different volume.   So the
[15:23:07] <andrewbogott>	 So the files that were in /srv before still exist, but are hidden because /srv is pointing someplace else.
[15:23:20] <Negative24>	 Well it was enabled even before I put the public_html there.
[15:23:35] <andrewbogott>	 enabled but maybe puppet hadn't run?
[15:23:56] <andrewbogott>	 I don't know -- most likely explanation is that it remounted after the files were there.  If they're now missing.
[15:24:05] <Negative24>	 Well...
[15:24:08] <andrewbogott>	 yuvipanda: can you remind me how to reveal files that are hidden by a mount
[15:24:22] <Negative24>	 They're replaceable but I don't want junk laying around.
[15:24:28] <yuvipanda>	 andrewbogott: unsure, bd808|BUFFER posted in labs-l last month
[15:25:24] <andrewbogott>	 yuvipanda: ok, I see it
[15:25:59] <yuvipanda>	 andrewbogott: cool
[15:26:13] <Negative24>	 I should have started puppet before touching /srv. I forget that it doesn't auto run.
[15:26:19] <andrewbogott>	 Negative24: here's an email that explains things (it's about a different but similar issue) http://lists.wikimedia.org/pipermail/labs-l/2014-July/002752.html
[15:26:29] <andrewbogott>	 But if you want to just recreate those files and leave the old ones stranded, that's fine :)
[15:27:55] <Negative24>	 I'll move them.
[15:29:22] <Negative24>	 Found it.
[15:29:51] <Negative24>	 So they're still on /dev/vda1 but srv is mounted on a different volume.
[15:30:18] <andrewbogott>	 yep.
[15:30:39] <andrewbogott>	 It's a klutzy issue with that mounting code.  I'm not sure what the right solution is
[15:31:43] <Negative24>	 This move is taking awhile. I'll be back in a few...
[16:24:20] <Negative24>	 andrewbogott: Puppet also deleted the root password when you unchecked the mysql config.
[16:24:40] <Negative24>	 the root password of mysql
[16:24:58] <andrewbogott>	 Negative24: that seems unlikely, since turning a class off doesn't cause puppet to do anything active, only stop doing what it was doing.
[16:25:12] <andrewbogott>	 It might be that that's another thing which the basic class refreshes someplace?
[16:25:27] <andrewbogott>	 Anyway, I hope that didn't scramble anything...
[16:25:45] <andrewbogott>	 You're free to switch things back on again as well, I just needed to get a clean puppet run in order to see what was happening.
[16:26:02] <andrewbogott>	 Or, rather, it turned out to not matter, but that's always the first thing I do :/
[16:26:18] <Negative24>	 I just ran puppet again with the mysql root password set again and it didn't clear it, so that's good.
[16:26:57] <Negative24>	 It only disconnected MW from the database. Not the end of the world.
[16:51:53] <wikibugs>	 3Wikimedia Labs / 3tools: MediaWiki blank page after uploading LocalSettings.php (due to wrong port set on $wgServer) - 10https://bugzilla.wikimedia.org/69154 (10Andre Klapper) 5RESO/WOR>3RESO/INV
[18:52:43] <spagewmf>	 bd808|BUFFER , yuvipanda : I just ran `labs-vagrant provision` forgetting to prepend `sudo `, and it seemed to work!  Did something change, or did I just get lucky, because it's always failed if I don't run `sudo labs-vagrant provision`
[19:08:08] <wikibugs>	 3Wikimedia Labs / 3Infrastructure: List of SVN users who were not migrated - 10https://bugzilla.wikimedia.org/58687#c6 (10Tim Landscheidt) Nemo, could you rephrase what list you are looking for, or close this bug?
[19:13:52] <spagewmf>	 bd808|BUFFER , yuvipanda : I just ran `labs-vagrant provision` forgetting to prepend `sudo `, and it seemed to work!  Did something change, or did I just get lucky, because it's always failed if I don't run `sudo labs-vagrant provision`
[19:16:04] <spagewmf>	 labs-vagrant instance http://flow-tests.wmflabs.org/wiki/Main_Page is giving "Service not available"  Assuming it's hhvm, how do I troubleshoot?
[19:16:39] <Negative24>	 Is the Parsoid repo down? http://parsoid.wmflabs.org:8080/deb doesn't seem to work.
[19:57:30] <spagewmf>	 FYI I did `sudo restart hhvm` and our labs instance is OK now.
[20:56:24] <wikibugs>	 3Wikimedia Labs / 3tools: Please install php5-imagick on Tool Labs - 10https://bugzilla.wikimedia.org/69078 (10Andrew Bogott) 5PATC>3RESO/FIX
[21:09:24] <wikibugs>	 3Wikimedia Labs / 3tools: install GD perl lib - 10https://bugzilla.wikimedia.org/67199 (10Andrew Bogott) 5PATC>3RESO/FIX
[22:10:35] <Dispenser>	 /usr/bin/sql is a fucking mess.  Here's Toolserver version http://lists.wikimedia.org/pipermail/toolserver-l/attachments/20131218/00e063f8/attachment.ksh
[22:50:57] <spagewmf>	 latest hhvm/labs-vagrant at http://flow-tests.wmflabs.org/wiki/User_talk:Spage latest labs-vagrant is reeeeally slow even after restarting apache and hhvm.  Load is low, no swap; I don't know what's going on.