[00:00:11] otherwise I can delete an instance from my project, and someone could re-create it in another one [00:00:14] with the same name [00:00:16] and I wouldn't know [00:00:24] ssh-keyscan may be a viable option [00:00:55] this likely would also work: http://www.php.net/manual/en/function.ssh2-fingerprint.php [00:01:28] I can add a job into mediawiki's jobqueue to keep trying until it gets a fingerprint, or until the instance is deleted [00:01:29] heh. but i actually want more than the fingerprint. i want the whole key ;) (i think) [00:01:59] what's the need for the full key? [00:02:12] well i'm still thinking about it [00:02:22] ssh-keyscan gives the whole key [00:02:25] yeah [00:10:50] Ryan_Lane, sorry I dropped off for a bit. I'd love to help with puppet cleanup and system administration if I can, let me know what I need to do to get started. [00:11:28] cool [00:11:35] !account-questions | lemurph [00:11:35] lemurph: I need the following info from you: 1. Your preferred wiki user name. This will also be your git username, so if you'd prefer this to be your real name, then provide your real name. 2. Your SVN account name, or your preferred shell account name, if you do not have SVN access. 3. Your preferred email address. [00:15:40] !initial-login | lemurph [00:15:40] lemurph: https://labsconsole.wikimedia.org/wiki/Access#Initial_log_in [00:15:41] back in a bit [00:50:11] Ryan_Lane: re your mailing list message, I heard that there are a lot of problems with LiquidThreads? [00:50:24] why will a standard discussion page not do? [00:50:54] because standard discussion system sucks [00:51:04] liquidthreads usually works fine [00:51:11] it's in use on mediawiki.org for years [00:51:40] apparently so does LQT (there was a discussion about getting it on meta and people were complaining about it) [00:51:57] something about it not being actively developed with bug fixes etc? [00:52:36] it's maintained [00:52:44] no new features until a re-write [00:53:28] labsconsole also doesn't update with the rest of the infrastructure [00:53:46] so, we can wait till it stableizes before upgrading [00:53:55] Granted, I've not used it myself so perhaps this could be seen as an opportunity to see what everyone else is complaining about ;) [00:54:02] heh [00:54:18] the UI isn't amazing, but it's so much better than discussion pages that it hurts [00:54:39] regular discussion pages make me want to slit my wrists [00:55:01] I guess I should make sure it's actually working :D [00:55:08] !project bastion [00:55:08] https://labsconsole.wikimedia.org/wiki/Nova_Resource:bastion [00:55:23] please don't go to enwiki too much then [00:55:36] https://labsconsole.wikimedia.org/wiki/Nova_Resource_Talk:Bastion [00:55:52] we don't want to lose you! [00:55:59] :D [00:56:50] !project testlabs [00:56:50] https://labsconsole.wikimedia.org/wiki/Nova_Resource:testlabs [00:57:06] !project openstack [00:57:06] https://labsconsole.wikimedia.org/wiki/Nova_Resource:openstack [00:58:31] seems to work fine [01:48:54] !project bots [01:48:54] https://labsconsole.wikimedia.org/wiki/Nova_Resource:bots [01:50:42] andrewbogott_afk: ugh. seems gluster is more of a PITA than I thought [01:51:34] it requires directories to preexist to use them in brick definitions [02:01:27] I guess the volume driver could stick something into the queue for all volume nodes that creates the directory, then sticks something into the queue when it's done [02:01:41] which would allow it to work [02:43:13] New patchset: Ottomata; "Added manifests/misc/analytics.pp - misc::analytics::packages: class to install packages analytics needs to process data" [operations/puppet] (test) - https://gerrit.wikimedia.org/r/2754 [02:43:34] New patchset: Ottomata; "manifests/misc/analytics.pp - some comments" [operations/puppet] (test) - https://gerrit.wikimedia.org/r/2755 [02:43:53] New review: gerrit2; "Lint check passed." [operations/puppet] (test); V: 1 - https://gerrit.wikimedia.org/r/2755 [02:46:49] New review: Dsc; "Damn fine." [operations/puppet] (test) C: 1; - https://gerrit.wikimedia.org/r/2754 [02:46:57] New review: Ottomata; "(no comment)" [operations/puppet] (test) C: 1; - https://gerrit.wikimedia.org/r/2754 [02:47:14] New review: Rich Smith; "(no comment)" [operations/puppet] (test) C: 1; - https://gerrit.wikimedia.org/r/2754 [02:50:11] New review: Dsc; "Very nice. Damn fine. Beautiful." [operations/puppet] (test) C: 1; - https://gerrit.wikimedia.org/r/2754 [08:44:29] aude: hi [08:44:37] is all on bots working ok [08:50:22] !log nagios fixed irc [08:50:28] ok [08:50:31] Logged the message, Master [08:51:08] houston, we made it yay [08:53:47] I promise to work properly now [08:53:58] :o [08:54:07] good bot [08:54:40] RECOVERY Disk Space is now: OK on deployment-web deployment-web output: DISK OK [08:54:50] RECOVERY host: deployment-web is UP address: deployment-web PING OK - Packet loss = 0%, RTA = 2.44 ms [08:55:10] RECOVERY Total Processes is now: OK on deployment-web deployment-web output: PROCS OK: 108 processes [08:55:30] RECOVERY dpkg-check is now: OK on deployment-web deployment-web output: All packages OK [08:55:50] RECOVERY Free ram is now: OK on deployment-web deployment-web output: OK: 92% free memory [08:56:30] RECOVERY Current Load is now: OK on deployment-web deployment-web output: OK - load average: 0.20, 0.16, 0.07 [08:59:10] RECOVERY SSH is now: OK on deployment-web deployment-web output: SSH OK - OpenSSH_5.3p1 Debian-3ubuntu7 (protocol 2.0) [08:59:20] RECOVERY Current Users is now: OK on deployment-web deployment-web output: USERS OK - 0 users currently logged in [10:50:09] quiet [11:54:35] petan|wk: RAWRRRRRRRRRRRRRRRRR better? [12:23:56] :o [12:51:11] 02/24/2012 - 12:51:11 - Updating keys for ashishd [13:06:15] PROBLEM Free ram is now: CRITICAL on bots-3 bots-3 output: Critical: 3% free memory [13:07:57] @search gang [13:07:57] Results (found 2): load, load-all, [13:10:07] @search access [13:10:07] Results (found 4): access, bastion, account-questions, account, [13:10:10] !access [13:10:10] https://labsconsole.wikimedia.org/wiki/Access#Accessing_public_and_private_instances [13:10:18] !instance [13:10:18] https://labsconsole.wikimedia.org/wiki/Help:Instances [13:10:23] !instances [13:10:23] https://labsconsole.wikimedia.org/wiki/Help:Instances [13:10:52] !terminology is help page https://labsconsole.wikimedia.org/wiki/Help:Terminology [13:10:52] Key exist! [13:11:10] !security [13:11:10] https://labsconsole.wikimedia.org/wiki/Security_Groups [13:11:15] RECOVERY Free ram is now: OK on bots-3 bots-3 output: OK: 59% free memory [13:45:09] someone around [13:45:20] I replaced the main page [13:45:29] Ryan is gonna kill me [13:48:44] * Barras watches who petan|wk is gonna be killed. [13:48:49] how* [14:32:16] YEAH NEW MAIN PAGE [14:32:41] magic! \o/ [14:36:13] it's nice! [14:36:16] thanks petan [14:37:20] it's basically english wikipedia heh [14:37:30] just changed the content ;) [14:38:57] <^demon> Have we voted on the slogan yet? [14:38:58] <^demon> :) [14:39:13] I think it's Ryan who made it [14:39:19] it's in wikitech somewhere heh [14:39:41] ah, but it was explanation why we have unicorn as logo, not a slogan [14:39:45] <^demon> I think we agreed on the jist, but we never bikeshedded on the exact words. [14:40:13] it's a wiki page :) [14:40:17] anyone can change it [14:40:43] <^demon> We should use the same for [[MediaWiki:Tagline]] :) [15:03:26] 02/24/2012 - 15:03:26 - Creating a project directory for fundraising [15:03:26] 02/24/2012 - 15:03:26 - Creating a home directory for jgreen at /export/home/fundraising/jgreen [15:04:27] 02/24/2012 - 15:04:27 - Updating keys for jgreen [15:15:45] PROBLEM Free ram is now: WARNING on bots-cb bots-cb output: Warning: 12% free memory [15:15:55] Damianz: ^ [15:20:50] PROBLEM Free ram is now: CRITICAL on bots-cb bots-cb output: Critical: 5% free memory [15:22:12] Damianz Damianz [15:22:25] Yeah 1min [15:22:29] :O [15:22:42] Could have guessed... [15:23:09] cb3 really needs some re-writing to stop it doing that. [15:23:17] 85% of ram usage isn't cool [15:23:39] pick a good language for that [15:23:45] like c instead of php [15:23:58] C would take a long time to port the bot into [15:24:00] php for executable programs suck afaik [15:24:00] PROBLEM Current Load is now: CRITICAL on fundraising-db fundraising-db output: CHECK_NRPE: Error - Could not complete SSL handshake. [15:24:01] Possibly python [15:24:11] python isn't better heh [15:24:14] Python with thread pools would do nicely [15:24:20] both php and python eat shitload of resources [15:24:29] c++ or asm ftw [15:24:32] :) [15:24:33] PHP more so. [15:24:38] C++ is harder to write [15:24:40] PROBLEM Current Users is now: CRITICAL on fundraising-db fundraising-db output: CHECK_NRPE: Error - Could not complete SSL handshake. [15:24:44] I like c++ [15:24:51] it seems to me easier than php a bit [15:25:03] * Damianz really needs to get around to learning C better as C is painful currently [15:25:15] but maybe because I spend more than 7 years programming in c languages and php only few months hehe [15:25:20] PROBLEM Disk Space is now: CRITICAL on fundraising-db fundraising-db output: CHECK_NRPE: Error - Could not complete SSL handshake. [15:25:41] I don't have a need for C in a lot of stuff but little things I have to write in C. [15:25:41] RECOVERY Free ram is now: OK on bots-cb bots-cb output: OK: 83% free memory [15:26:07] I think that whatever you want to do c++ is better than anything for a mediawiki bot [15:26:10] PROBLEM Free ram is now: CRITICAL on fundraising-db fundraising-db output: CHECK_NRPE: Error - Could not complete SSL handshake. [15:26:35] CBNGs core is C++ [15:26:43] it would be really faster [15:26:55] that's maybe only reason why it works :) [15:27:09] <^demon> Compile your PHP to C++. [15:27:13] <^demon> :) [15:27:15] :o [15:27:25] isn't php interpretor written in c too :P [15:27:32] <^demon> Yes. [15:27:33] I think so [15:27:40] PROBLEM Total Processes is now: CRITICAL on fundraising-db fundraising-db output: CHECK_NRPE: Error - Could not complete SSL handshake. [15:27:46] <^demon> But I wasn't talking about Zend PHP :p [15:27:58] ah [15:27:59] <^demon> Use HipHop so you can compile your php to c++ [15:28:20] PROBLEM dpkg-check is now: CRITICAL on fundraising-db fundraising-db output: CHECK_NRPE: Error - Could not complete SSL handshake. [15:28:21] compile or convert [15:28:36] HipHop is weird [15:28:37] <^demon> Well, converts then compiles it to binary. [15:28:44] <^demon> But there's also an interpreted mode. [15:28:51] hm... [15:29:14] <^demon> The hardest part is actually compiling hiphop :p [15:29:16] question is how it looks converted heh [15:29:44] <^demon> Dunno. But MediaWiki is fast :p [15:29:48] disassembled binaries created in msv looks crazy [15:30:01] I guess it's gonna be similar heh [15:30:13] msvs [15:30:23] IIRC the generated C++ is pretty neat [15:30:54] ^demon: how could you run it [15:31:01] it's a library for apache or what [15:31:21] when you compile a mediawiki you need to implement it to apache somhow [15:31:37] Run it as fcgi [15:31:40] you can't just upload a static binary to /var/www [15:31:42] <^demon> No apache, the compiled binary can run as a webserver on a specified port [15:31:44] It also has its own server [15:31:48] aha [15:32:12] <^demon> The code for this stuff is in phase3/maintenance/hiphop/ [15:32:44] at some point it would be faster if you just used a c++ library for a web server and use it instead [15:32:52] :) [15:33:41] btw how do you multithread php? [15:33:59] is there any memory handling too? [15:34:24] like delete in c++ [15:34:52] interpretor handle it but if it's a binary there needs to be a garbage collector or not? [15:37:19] <^demon> petan|wk: PHP itself is mostly threadsafe when running from the command line (although not all extensions are threadsafe). [15:37:36] <^demon> It is *not* ok to do multithreaded things when running from the web--that invariably fails. [15:38:10] <^demon> Closest thing to delete in c++ is unset() in php. [15:39:05] <^demon> But its behavior is different. [15:39:22] can you unset objects too? [15:40:21] probably yes [15:40:25] RECOVERY Disk Space is now: OK on fundraising-db fundraising-db output: DISK OK [15:40:54] <^demon> You could unset local instances of them, yeah. [15:41:02] <^demon> And unset() on an object's properties work. [15:41:05] RECOVERY Free ram is now: OK on fundraising-db fundraising-db output: OK: 94% free memory [15:41:18] <^demon> unset() on a global within function-level context doesn't behave exactly as you'd expect though. [15:41:37] <^demon> You can't attempt to unset($this) :p [15:42:35] RECOVERY Total Processes is now: OK on fundraising-db fundraising-db output: PROCS OK: 89 processes [15:43:06] but there is probably some automated delete because I never saw unset for each newly created object as in c++ [15:43:29] RECOVERY dpkg-check is now: OK on fundraising-db fundraising-db output: All packages OK [15:43:48] <^demon> Well yeah, most languages implicitly clean up when things go out of scope. [15:43:49] usually in c++ when you create a new instance of something you need to delete it or it's going to be stored in memory untill process exit [15:43:58] RECOVERY Current Load is now: OK on fundraising-db fundraising-db output: OK - load average: 0.08, 0.12, 0.09 [15:44:26] * Damianz tries compiling hiphop [15:44:27] it's done using garbage collectors but these can cause the very bad performance sometime [15:44:38] RECOVERY Current Users is now: OK on fundraising-db fundraising-db output: USERS OK - 0 users currently logged in [15:45:11] I would prefer if it was possible to use both, it seems that php can do that, while c# not... [15:45:32] like use either delete or leave the interpretor handle it [15:45:47] * let [15:45:53] <^demon> petan|wk: Well php does gc for you :) [15:46:01] <^demon> But you can manually invoke it, iirc. [15:46:19] <^demon> http://us2.php.net/manual/en/features.gc.php [16:53:55] PROBLEM Current Load is now: CRITICAL on fundraising-civicrm fundraising-civicrm output: CHECK_NRPE: Error - Could not complete SSL handshake. [16:54:35] PROBLEM Current Users is now: CRITICAL on fundraising-civicrm fundraising-civicrm output: CHECK_NRPE: Error - Could not complete SSL handshake. [16:55:15] PROBLEM Disk Space is now: CRITICAL on fundraising-civicrm fundraising-civicrm output: CHECK_NRPE: Error - Could not complete SSL handshake. [16:56:05] PROBLEM Free ram is now: CRITICAL on fundraising-civicrm fundraising-civicrm output: CHECK_NRPE: Error - Could not complete SSL handshake. [16:57:25] PROBLEM Total Processes is now: CRITICAL on fundraising-civicrm fundraising-civicrm output: CHECK_NRPE: Error - Could not complete SSL handshake. [16:58:15] PROBLEM dpkg-check is now: CRITICAL on fundraising-civicrm fundraising-civicrm output: CHECK_NRPE: Error - Could not complete SSL handshake. [17:05:10] 02/24/2012 - 17:05:09 - Creating a home directory for dsc at /export/home/reportcard/dsc [17:06:10] 02/24/2012 - 17:06:09 - Updating keys for dsc [17:07:10] 02/24/2012 - 17:07:09 - Creating a home directory for declerambaul at /export/home/reportcard/declerambaul [17:08:10] 02/24/2012 - 17:08:10 - Updating keys for declerambaul [17:08:17] 02/24/2012 - 17:08:16 - Creating a home directory for dsc at /export/home/mobile-stats/dsc [17:09:16] 02/24/2012 - 17:09:16 - Updating keys for dsc [18:15:13] 02/24/2012 - 18:15:13 - Creating a home directory for tparscal at /export/home/bastion/tparscal [18:16:12] 02/24/2012 - 18:16:12 - Updating keys for tparscal [18:30:12] 02/24/2012 - 18:30:11 - Updating keys for tparscal [18:47:12] hi labs ops! sumana sent me here -- may I get an account/password for gerrit/labconsole? [18:47:20] (my username is "au" on the old svn machine and "Au" is my mediawiki.org account name.) [18:49:05] I guess you want your username to be au ? [18:49:16] correct, if it's ok with you. [18:49:38] well, I can't create it, but I could ask you what Ryan needs so it's faster for him [18:49:55] ok. my primary contact email is audreyt@audreyt.org and fallback is audreyt.org@me.com [18:49:58] there was a bot command for that... [18:50:27] !account-questions | au [18:50:28] au: I need the following info from you: 1. Your preferred wiki user name. This will also be your git username, so if you'd prefer this to be your real name, then provide your real name. 2. Your SVN account name, or your preferred shell account name, if you do not have SVN access. 3. Your preferred email address. [18:51:07] hrm, so Ryan showed me how to do this once... [18:51:14] let's see if i remember ... [18:51:16] ["au", "au", "audreyt@audreyt.org"] [18:51:26] hehe :) [18:51:37] :p [18:52:18] if i were smarter i would have written this down.. [18:52:49] au: are you the audrey who is going to start working with us ? [18:52:58] why yes, one and the same [18:53:02] the perl guru ? [18:53:03] sweet! [18:53:17] I've committed a few bugfixes to Parsoid already, I think [18:53:21] thanks! [18:54:25] gah, found the page on disabling accounts... [18:58:20] LeslieCarr: I think you just volunteered to write some pages for the Help namespace :P [18:58:29] au: you are definitely possibly a maybe working labs user ;) [18:58:44] what would my password be? [18:58:47] try logging into labsconsole.wikimedia.org and resetting your password [18:58:50] Same as svn [18:58:53] got it [18:59:08] Bah wrong window [19:04:15] LeslieCarr++ ItWorked! [19:04:35] yay! [19:05:58] * Damianz gives up on making hiphop compile and goes back to writing this html transform in perl. [19:06:08] thanks to ssmollett for helping me to do it :) [19:07:09] ssmollett++ Platonides++ wm-bot++ all around :) [19:07:58] wm-bot doesn't deserve ++, she dies too much [19:09:19] wm-bot-- Damianz++ then :) #karmaTransfer [19:10:34] lol [19:11:53] hehehe [21:06:11] any opinions on alternatives to gitweb? [21:36:09] Ryan_Lane, I got my account set up. Can you let me know what else I need set up. Or who to ask? [22:00:49] lemurph: there's someone else working on puppet cleanup [22:00:57] we should add you to that project, and the two of you can work together [22:01:02] !project puppet-cleanup [22:01:02] https://labsconsole.wikimedia.org/wiki/Nova_Resource:puppet-cleanup [22:01:12] did you subscribe to the labs-l list? [22:01:36] danakim: lemurph would also like to work on the puppet repo [22:01:43] What's puppet-cleanup? [22:01:47] danakim: fine if I add him to your project? [22:02:09] RoanKattouw: turning our puppet repo into modules, and ensuring the puppet manifests run cleanly on initial run [22:02:35] brb [22:03:41] Ryan_Lane: HEART [22:07:30] Ryan_Lane: HEART^2 [22:08:40] heh [22:08:42] agreed [22:09:23] lemurph: I'll add you to the project. if you see danakim around, talk with him about what you guys are working on [22:09:41] lemurph: using the server admin log will make it much easier to coordinate [22:10:08] !log puppet-cleanup adding lemurph to the project. [22:10:11] ^^ like so [22:10:13] Logged the message, Master [22:10:31] it'll show up on the project page: https://labsconsole.wikimedia.org/wiki/Nova_Resource:Puppet-cleanup [22:11:13] if you log that you are working on specific manifests, it'll make it more likely that the two of you aren't duplicating efforts [22:12:07] * Damianz feeds Ryan_Lane cookie [22:12:35] argh. I *must* improve the project interface [22:12:40] it makes me want to punch kittens [22:13:15] bahahahahahahahahahaha [22:13:22] https://labsconsole.wikimedia.org/wiki/Main_Page [22:13:27] <3 [22:13:30] <3 so much [22:14:11] petan: that is the funniest main page ever [22:14:48] Excellent, <3 [22:15:01] Just the top third or so, <3 [22:15:16] It has a fade-in unicorn! [22:18:52] :D [22:18:53] That's awesome [22:19:02] Pretty colours too [22:19:37] super awesome petan [22:19:45] I especially love that it's a pink unicorn [22:19:48] * Damianz gives petan 1 bot [22:21:54] Thanks Ryan_Lane [22:22:01] yw [22:22:13] hopefully we'll have per-project puppet soon [22:22:21] it'll make your life a lot easier ;) [22:22:27] lemurph: have you cloned the git repo yet? [22:22:32] No [22:22:33] !git [22:22:33] for more information about git on labs see https://labsconsole.wikimedia.org/wiki/Git [22:23:39] ssmollett: were you done with the testing-ldap instance? I'm going to delete it [22:28:12] Ryan_Lane: you can delete it. [22:28:18] ok. cool [22:28:45] PROBLEM host: testing-ldap-build is DOWN address: testing-ldap-build check_ping: Invalid hostname/address - testing-ldap-build [22:29:15] PROBLEM host: testing-scheduler is DOWN address: testing-scheduler CRITICAL - Host Unreachable (testing-scheduler) [22:31:51] Got it, thanks, I'm looking over the manifests now [22:32:04] !log bots added a bots-labs instance for running bots meant for labs [22:32:05] Logged the message, Master [22:32:13] lemurph: cool [22:43:56] PROBLEM Current Load is now: CRITICAL on bots-labs bots-labs output: CHECK_NRPE: Error - Could not complete SSL handshake. [22:44:36] PROBLEM Current Users is now: CRITICAL on bots-labs bots-labs output: CHECK_NRPE: Error - Could not complete SSL handshake. [22:45:16] PROBLEM Disk Space is now: CRITICAL on bots-labs bots-labs output: CHECK_NRPE: Error - Could not complete SSL handshake. [22:46:06] PROBLEM Free ram is now: CRITICAL on bots-labs bots-labs output: CHECK_NRPE: Error - Could not complete SSL handshake. [22:46:37] That's really annoying [22:46:42] yeah [22:46:49] I restarted nrpe on the instance [22:46:53] it should stop soon [22:47:03] we need to figure out why that doesn't work properly on instance creation [22:47:15] Doesn't it sitll have that stupid on creation puppet restart issue? [22:47:20] lemurph: that would be a really awesome thing to fix :) [22:47:26] PROBLEM Total Processes is now: CRITICAL on bots-labs bots-labs output: CHECK_NRPE: Error - Could not complete SSL handshake. [22:47:34] I should fix my loose screw [22:47:36] labs-nagios-wm: wtf. I restarted nrpe! [22:47:48] maybe puppet needs to run again [22:48:06] I don't know why it requires two puppet runs, but it enrages me [22:48:16] PROBLEM dpkg-check is now: CRITICAL on bots-labs bots-labs output: CHECK_NRPE: Error - Could not complete SSL handshake. [22:48:46] sure enough, a second puppet run installed the config [22:48:52] weird [22:48:56] RECOVERY Current Load is now: OK on bots-labs bots-labs output: OK - load average: 0.47, 0.29, 0.29 [22:48:59] our repo is so fucked up [22:49:36] RECOVERY Current Users is now: OK on bots-labs bots-labs output: USERS OK - 1 users currently logged in [22:49:47] heh [22:49:59] Can we not just trash it and start again? [22:50:07] Or is there still the gerrit branch delete issue. [22:50:11] production is also screwed up [22:50:16] RECOVERY Disk Space is now: OK on bots-labs bots-labs output: DISK OK [22:50:18] Sounds annoying [22:50:20] the test branch needs to be deleted and recreated [22:50:25] it's beyond savign [22:50:30] *saving [22:50:37] lemurph: yeah, it is :( [22:50:49] because every new instance causes nagios alerts for 30 minutes [22:51:06] RECOVERY Free ram is now: OK on bots-labs bots-labs output: OK: 92% free memory [22:51:16] There [22:51:29] unless you run puppet again right after it's created? [22:51:32] yep [22:51:45] there's some bad dependency or something broken in the creation [22:52:00] you can see it in an instance's console log [22:52:11] since I redirect puppet messages there on creation [22:52:26] RECOVERY Total Processes is now: OK on bots-labs bots-labs output: PROCS OK: 85 processes [22:52:38] should also be in /var/log/messages [22:52:59] hm. not there [22:53:16] RECOVERY dpkg-check is now: OK on bots-labs bots-labs output: All packages OK [22:53:20] /var/log/daemon.log [22:54:46] puppet-agent[1431]: (/Stage[main]/Nrpe::Packages/File[/etc/nagios/nrpe_local.cfg]/ensure) change from absent to file failed: Could not set 'file on ensure: No such file or directory - /etc/nagios/nrpe_local.cfg.puppettmp_8137 at /etc/puppet/manifests/nrpe.pp:32 [22:54:54] o.0 [22:55:05] likely a bad dependency [22:57:56] I don't understand why that wouldn't work [22:58:21] the service requires the file, and is subscribed to it [22:59:36] hm. maybe the directory is missing? [22:59:43] the file requires the package, though [22:59:47] Shouldn't it ensure the dir exists? [22:59:52] the package should [23:00:25] though maybe it's saying the file didn't get downloaded [23:02:10] oh well. I give up for now [23:03:09] * Damianz gets Ryan_Lane to do the dusting [23:13:24] Looks like fun so far. I'll try connecting with danakim later on. [23:25:01] lemurph: great :) [23:55:41] damn it. I built a tiny rather than a small [23:57:48] * Damianz makes Ryan_Lane tiny [23:58:04] * Ryan_Lane shrinks [23:58:30] * Damianz sends tiny Ryan_Lane to make food