[00:17:34] RECOVERY dpkg-check is now: OK on wikidata-dev-3 i-0000021d output: All packages OK [00:18:44] RECOVERY Current Load is now: OK on wikidata-dev-3 i-0000021d output: OK - load average: 0.11, 0.09, 0.03 [00:19:24] RECOVERY Current Users is now: OK on wikidata-dev-3 i-0000021d output: USERS OK - 0 users currently logged in [00:20:44] RECOVERY Free ram is now: OK on wikidata-dev-3 i-0000021d output: OK: 74% free memory [00:21:54] RECOVERY Total Processes is now: OK on wikidata-dev-3 i-0000021d output: PROCS OK: 90 processes [01:04:14] PROBLEM Puppet freshness is now: CRITICAL on nova-production1 i-0000007b output: Puppet has not run in last 20 hours [01:08:14] PROBLEM Puppet freshness is now: CRITICAL on nova-gsoc1 i-000001de output: Puppet has not run in last 20 hours [01:16:56] Ryan_Lane: could you merge my bugzilla changes for puppet? [01:17:06] link? [01:17:14] 1s [01:17:20] or add me as a reviewer [01:18:25] done [01:23:22] oh, this is in the production branch? [01:23:27] you need to use the test branch for labs [01:23:36] hexmode: I reviewed it [01:23:37] :) [01:23:49] tyvm [01:24:34] Ryan_Lane: abandoning, switching to test branch [01:30:29] Ryan_Lane: ok, obviously I'm a git n00b. Where is all this documented? [01:31:02] on mediawiki's site somewhere [01:31:06] git checkout test [01:31:40] Then to reuse your existing change, git cherry-pick 09bd35a9ac49c8d48fdee1e9620e0ae0ecd9e4b5 [02:49:24] RECOVERY Current Users is now: OK on upload-wizard i-0000021c output: USERS OK - 0 users currently logged in [02:50:34] RECOVERY Free ram is now: OK on upload-wizard i-0000021c output: OK: 93% free memory [02:50:54] RECOVERY Disk Space is now: OK on upload-wizard i-0000021c output: DISK OK [02:51:54] RECOVERY Total Processes is now: OK on upload-wizard i-0000021c output: PROCS OK: 75 processes [02:53:44] RECOVERY Current Load is now: OK on upload-wizard i-0000021c output: OK - load average: 0.03, 0.36, 0.27 [03:44:39] PROBLEM Free ram is now: WARNING on test-oneiric i-00000187 output: Warning: 16% free memory [03:48:59] PROBLEM Free ram is now: WARNING on utils-abogott i-00000131 output: Warning: 15% free memory [03:57:10] PROBLEM Free ram is now: WARNING on nova-daas-1 i-000000e7 output: Warning: 14% free memory [04:02:34] PROBLEM Free ram is now: WARNING on orgcharts-dev i-0000018f output: Warning: 15% free memory [04:04:39] PROBLEM Free ram is now: CRITICAL on test-oneiric i-00000187 output: Critical: 4% free memory [04:08:59] PROBLEM Free ram is now: CRITICAL on utils-abogott i-00000131 output: Critical: 3% free memory [04:09:39] RECOVERY Free ram is now: OK on test-oneiric i-00000187 output: OK: 97% free memory [04:13:59] RECOVERY Free ram is now: OK on utils-abogott i-00000131 output: OK: 96% free memory [04:17:19] PROBLEM Free ram is now: CRITICAL on nova-daas-1 i-000000e7 output: Critical: 5% free memory [04:22:30] PROBLEM Free ram is now: CRITICAL on orgcharts-dev i-0000018f output: Critical: 3% free memory [04:27:10] RECOVERY Free ram is now: OK on nova-daas-1 i-000000e7 output: OK: 93% free memory [04:27:30] RECOVERY Free ram is now: OK on orgcharts-dev i-0000018f output: OK: 95% free memory [05:30:10] PROBLEM Puppet freshness is now: CRITICAL on wikidata-dev-2 i-0000020a output: Puppet has not run in last 20 hours [09:52:57] petan|wk: hello :) [09:53:10] petan|wk: is there any project name for the pre deployment cluster ? [09:53:27] wmflabs being too generic [09:53:30] 'beta' same [09:53:38] deployement-prep is too long :-] [10:00:41] PROBLEM dpkg-check is now: CRITICAL on wikidata-dev-3 i-0000021d output: DPKG CRITICAL dpkg reports broken packages [10:05:20] mutante: ping pong :-D [10:05:41] mutante: any idea about a name for the preproduction cluster / deployment tester / beta etc ? [10:23:44] PROBLEM Current Load is now: CRITICAL on wikidata-dev-4 i-0000021f output: CHECK_NRPE: Error - Could not complete SSL handshake. [10:24:24] PROBLEM Current Users is now: CRITICAL on wikidata-dev-4 i-0000021f output: CHECK_NRPE: Error - Could not complete SSL handshake. [10:25:04] PROBLEM Disk Space is now: CRITICAL on wikidata-dev-4 i-0000021f output: CHECK_NRPE: Error - Could not complete SSL handshake. [10:25:44] PROBLEM Free ram is now: CRITICAL on wikidata-dev-4 i-0000021f output: CHECK_NRPE: Error - Could not complete SSL handshake. [10:26:54] PROBLEM Total Processes is now: CRITICAL on wikidata-dev-4 i-0000021f output: CHECK_NRPE: Error - Could not complete SSL handshake. [10:27:34] PROBLEM dpkg-check is now: CRITICAL on wikidata-dev-4 i-0000021f output: CHECK_NRPE: Error - Could not complete SSL handshake. [10:34:20] hey there [10:35:19] i'd like to use the vcsrepo type in my manifest. [10:37:07] normally i'd install it with "puppet module install vcsrepo" [10:37:09] unfortunately, the puppet that is included with the labs instances is too old and doesn't have the module sub-command. [10:37:09] hints? [10:58:04] PROBLEM host: wikidata-dev-4 is DOWN address: i-0000021f check_ping: Invalid hostname/address - i-0000021f [11:01:50] hashar: yes it's deployment cluster or deployment test site [11:01:59] or just deployment-prep [11:02:07] deployment-prep is name of project in labs [11:02:24] beta.wmflabs.org and deployment.wmflabs.org are dns [11:02:34] ;) [11:03:35] that is confusing :-D [11:03:44] PROBLEM Current Load is now: CRITICAL on wikidata-dev-3 i-00000222 output: CHECK_NRPE: Error - Could not complete SSL handshake. [11:03:49] how is that confusing the purpose is to test [11:04:02] it's test of sw before it's deployed to prod :-) [11:04:07] so deployment site [11:04:24] PROBLEM Current Users is now: CRITICAL on wikidata-dev-3 i-00000222 output: CHECK_NRPE: Error - Could not complete SSL handshake. [11:04:42] like a site where we install the trunk mediawiki so that community of wikimedia projects can test if it's going to work with the project [11:05:04] PROBLEM Disk Space is now: CRITICAL on wikidata-dev-3 i-00000222 output: CHECK_NRPE: Error - Could not complete SSL handshake. [11:05:07] I can't think if any better name [11:05:14] PROBLEM Puppet freshness is now: CRITICAL on nova-production1 i-0000007b output: Puppet has not run in last 20 hours [11:05:44] PROBLEM Free ram is now: CRITICAL on wikidata-dev-3 i-00000222 output: CHECK_NRPE: Error - Could not complete SSL handshake. [11:06:54] PROBLEM Total Processes is now: CRITICAL on wikidata-dev-3 i-00000222 output: CHECK_NRPE: Error - Could not complete SSL handshake. [11:07:34] PROBLEM dpkg-check is now: CRITICAL on wikidata-dev-3 i-00000222 output: CHECK_NRPE: Error - Could not complete SSL handshake. [11:07:47] I've created an instance on my project Upload-wizard, I can see it in "Manage Instances", but why can't I see it in my project page? [11:08:33] Why the "Instances for this project" section of the page "Nova Resource:Upload-wizard" is still empty? [11:09:14] PROBLEM Puppet freshness is now: CRITICAL on nova-gsoc1 i-000001de output: Puppet has not run in last 20 hours [11:10:52] Shujenchang: you better use Special:NovaInstance [11:11:07] !instance [11:11:07] https://labsconsole.wikimedia.org/wiki/Help:Instances [11:11:52] !project deployment [11:11:52] https://labsconsole.wikimedia.org/wiki/Nova_Resource:deployment [11:11:54] yay [11:12:16] !project deployment-prep | hashar [11:12:16] hashar: https://labsconsole.wikimedia.org/wiki/Nova_Resource:deployment-prep [11:50:15] !project deployment-prep | hashar [11:50:16] hashar: https://labsconsole.wikimedia.org/wiki/Nova_Resource:deployment-prep [11:50:24] documentation :P [11:52:41] sorry went out for lunch [11:59:44] I've created an instance on my project Upload-wizard, I can see it in "Manage Instances", but why can't I see it in my project page? Why the "Instances for this project" section of the page "Nova Resource:Upload-wizard" is still empty? (My computer was crashed just then, and I rebooted it, so I may not recieve replies about the question I asked just then) [12:04:51] Shujenchang: https://labsconsole.wikimedia.org/wiki/Nova Resource:Upload-wizard?action=purge [12:05:00] Shujenchang: https://labsconsole.wikimedia.org/wiki/Nova_Resource:Upload-wizard?action=purge [12:05:53] thx for your help~ [12:14:43] and how to config the Puppet if I want to install a mediawiki and upload-wizard extension, what puppets should be checked? [12:18:15] !log deployment-prep petrb: updated git and commited all changes [12:18:16] Logged the message, Master [12:18:27] !log deployment-prep petrb: moved the log file storage to gluster [12:18:29] Logged the message, Master [12:19:03] Shujenchang: apache-php5-mysql [12:19:11] mediawiki is not puppetized so far [12:19:49] hello mutante [12:20:08] webserver::apache2 and webserver::php5-mysql? [12:20:26] Shujenchang: only one [12:20:33] mutante, any update on wikistats? I'd love to add a bunch of wikis to the tables :) [12:20:36] I think that php5 mysql works [12:21:22] should I checked something under "database"? [12:21:43] hashar: is there any way to debug why it loads so slow? [12:21:51] it ? [12:21:53] Shujenchang: I don't think so [12:21:56] hashar: mediawiki [12:22:18] need to debug it to find out the root cause [12:22:19] every page on deployment site loads slow [12:22:19] Nemo_bis: yea, there have been updates, actually i was on it almost the whole weekend, trying to make the unified update script to work for all. there are issues with letting a system user run cron jobs right now..opened a bug and talked to ryan [12:22:25] I see [12:22:41] hashar: in fact I believe that mediawiki itself is pretty slow, because I never managed to install it anywhere for it to be fast [12:22:48] Nemo_bis: in which form do you have the new wikis right now? [12:22:57] all servers I ever installed it on even on fast ones it was running slow [12:23:49] while sw like word press runs 200 times faster :) [12:23:51] I will look at it [12:24:02] meanwhile [12:24:06] you mean you will fix the mediawiki so that it's faster :P [12:24:10] nice [12:24:11] :D [12:24:12] do you guys use puppet ? [12:24:17] for? [12:24:20] sometimes we do [12:24:29] to install packages on the tests servers [12:24:37] we use it for apache [12:24:39] that's all [12:24:51] against the `test` branch? [12:24:59] everything else is installed by hand [12:25:05] on deployment site [12:25:16] yes it's test branch [12:25:20] whole labs [12:25:21] which is like … a very bad practice :-] [12:25:22] hashar: yea, test branch if you want to install packages on instances [12:25:32] hashar: it's only working practice [12:25:45] I can't wait 2 weeks for someone from ops to merge every little change [12:26:05] if I did, deployment site wouldn't even exist, we would be probably at very beginnig [12:26:24] mutante, a list of 7400 api.php [12:26:40] I can push in ops/puppet test branch [12:26:44] for example I wanted Ryan to puppetize squid from prod, I was waiting few months, then I decided to do it myself [12:26:44] well I could last time I checked [12:26:53] mutante, https://code.google.com/p/wikiteam/source/browse/trunk#trunk%2Fbatchdownload%2Flists [12:27:22] and yes, mutante, I saw something moving, that's why I asked :) [12:27:27] hashar: that doesn't change the fact that we can't and we definitely wanted the deployment site to work sooner than in few years :-) [12:27:45] mutante, any news from Wikia? I see they have some sweet central APIs with lists of active wikis and such [12:27:49] that explains why things were done by hand [12:28:00] also in beginning of labs it wasn't even possible to use puppet [12:28:03] unless you were ops [12:28:06] Nemo_bis: wow! nice . well for the "wikistats admin" thing that would be too many anyways.. it would just allow adding one by one.. so just mysql import [12:28:18] mutante, I see [12:28:38] Nemo_bis: no, i've been meaning to contact somebody at wikia for the data [12:28:39] those are all verified APIs, although some give DB errors on particular requests (!) [12:28:42] ok [12:29:02] mutante, do you know who to contact? [12:29:56] Nemo_bis: not a specific "best person to ask" yet.. but there should be good contacts between wmf and wikia [12:30:09] yep, more or less [12:31:26] interesting how people run wikis on IP addresses without DNS names [12:32:16] heh [12:32:31] there's even a mirror of Wikipedia on a test wiki now full of spam, for what it's worth [12:32:56] (I removed it from the lists, though) [12:33:20] Nemo_bis: i'll just import these to db with a manual import for now [12:33:29] mutante, sweet [12:33:34] Nemo_bis: and then run an update and we can look at the HTTP response codes [12:33:48] mutante, I didn't check if there are duplicates though [12:33:55] PROBLEM HTTP is now: CRITICAL on upload-wizard i-0000021c output: CRITICAL - Socket timeout after 10 seconds [12:34:09] Nemo_bis: don't worry, database won't allow duplicates in the URL to the api [12:34:28] mutante, yes, but you might still have some of those in the old Special:Statistics raw form [12:34:32] or not? [12:34:58] well, i can't guarantee but i have an "conversion" script [12:35:04] that tries to use new API URLs [12:35:10] and if that works it changes them [12:35:13] wow [12:35:33] Nemo_bis: what list you talk about? [12:35:44] petan|wk, https://code.google.com/p/wikiteam/source/browse/trunk#trunk%2Fbatchdownload%2Flists [12:36:06] I'm archiving all those wikis (well, not alone, but I did 1000 of them so far) [12:36:10] right [12:36:15] whatever the purpose is [12:36:50] petan|wk, the purpose is http://archive.org/details/wikiteam [12:37:19] it's trying to find all mediawiki installations on the internet [12:37:37] hm.. [12:38:10] https://code.google.com/p/wikiteam/wiki/TaskForce [12:38:45] http://www.cs.brown.edu/~pavlo/mediawiki/ already did it in 2009, we're now archiving all the survived wikis and then we'll try to copy him and run a crawler to find all the new ones and archive them too [12:38:52] petan|wk: I only checked webserver::php5-mysql, and run the command 'sudo puppetd -tv', but when the command was finished, it seemed php and mysql is not installed. [12:39:10] highlight + bold seems a very nasty thing [12:39:31] Shujenchang: that sounds weird [12:39:35] it should work [12:39:39] Nemo_bis: ah, well, for these, we don't have names yet, but i figured anyways we should ask the API itself for the Wiki name anyways [12:39:43] can you insert me to project? [12:39:43] It is said "The program 'php' is currently not installed. To run 'php' please ask your administrator to install the package 'php5-cli'" and "The program 'mysql' can be found in the following packages: [12:39:44] * mysql-client-core-5.1 [12:39:44] * mysql-cluster-client-5.1" [12:39:55] Nemo_bis: need to make one API call to get wiki name for all ... [12:40:07] Shujenchang: I will look in that, but I don't have access to that project [12:40:27] mutante, ? [12:40:29] * Nemo_bis afk now [12:40:37] petan|wk: sure,wait a moment~ [12:40:46] What's your username on labs? [12:40:50] petan|wk: I think we should add puppet class apaches::service on the Apaches [12:40:58] Nemo_bis: all we have in the lists are API URLs, the db wants an URL and a wiki name currently. we can retrieve the name from the API though [12:40:58] really? [12:41:03] that should resolve lot of issue [12:41:07] hashar: feel free to do that [12:41:09] Nemo_bis: ok, i'll show you later .. will import [12:41:13] such as including php5-apc a opcode accelerator [12:41:14] mutante, yeah, sounds silly to require both [12:41:20] hashar: don't forget to log it [12:41:20] mutante, thanksù [12:41:20] of course [12:41:21] hashar: log everything :-) [12:41:32] petan|wk: can you make me a sysadmin on the project ? [12:41:35] sure [12:41:58] petan|wk: What's your username on labs? [12:42:15] done [12:42:19] Shujenchang: Petrb [12:42:19] petan or petan|wk? [12:43:00] petan|wk: danke [12:43:26] 04/24/2012 - 12:43:25 - Creating a home directory for petrb at /export/home/upload-wizard/petrb [12:44:05] petan|wk: I've added you [12:44:26] 04/24/2012 - 12:44:25 - Updating keys for petrb [12:44:47] ok [12:45:25] petan|wk: thx~ [12:49:05] Reedy: could you comment on this: https://gerrit.wikimedia.org/r/#patch,sidebyside,5553,1,ProofreadPage_body.php [12:49:28] Reedy: or could you point me to someone who knows about extending the parser [12:49:52] Shujenchang: it seems to work [12:50:14] probably some problem in puppet [12:50:26] but apache server is surely running there [12:51:17] but php's not, I run the command "php", It is said "The program 'php' is currently not installed. To run 'php' please ask your administrator to install the package 'php5-cli'" [12:51:26] yes that's cli [12:51:38] php-cli is not a part of apache:php5:mysql [12:52:10] and mysql seems not be installed too [12:52:12] in fact I don't know if there is any puppet class for that [12:52:30] yes mysql server isn't puppetized, it installed only php modules for mysql [12:52:49] mysql server needs to be installed separately [12:52:58] Can command "apt-get" be used on instance? [12:53:02] yes [12:53:36] I see, and I'll use "apt-get" to install them? [12:53:45] ok [12:55:45] I've created an instance on my project Upload-wizard, I can see it in "Manage Instances", but why can't I see it in my project page? Why the "Instances for this project" section of the page "Nova Resource:Upload-wizard" is still empty? (My computer was crashed just then, and I rebooted it, so I may not recieve replies about the question I asked just then) [12:56:15] sudo apt-get install php5 [12:56:16] [sudo] password for shujenchang: [12:56:16] Reading package lists... Done [12:56:16] Building dependency tree [12:56:16] Reading state information... Done [12:56:16] The following NEW packages will be installed: [12:56:16] php5 [12:56:17] 0 upgraded, 1 newly installed, 0 to remove and 44 not upgraded. [12:56:17] 1 not fully installed or removed. [12:56:18] Need to get 1,108 B of archives. [12:56:18] After this operation, 20.5 kB of additional disk space will be used. [12:56:19] Get:1 http://ubuntu.wikimedia.org/ubuntu/ oneiric-updates/main php5 all 5.3.6-13ubuntu3.6 [1,108 B] [12:56:19] Fetched 1,108 B in 0s (53.0 kB/s) [12:56:20] Selecting previously deselected package php5. [12:56:45] It seemed install errored? What's wrong? [12:56:46] :o [12:56:58] it seems ok to me [12:57:33] php5 is already the newest version. [12:59:14] But why it is said "Errors were encountered while processing: ganglia-monitor"? [13:00:10] no idea [13:00:18] it must be something else broken [13:00:25] but apache works [13:02:18] and is php works on my instance? [13:02:24] mutante: automount on deployment-web2 is broken [13:02:25] petan|wk: Shujenchang : Ryan just added the generic mysql server class to be ready for use [13:02:40] ok [13:02:43] that's new to me [13:02:50] and besides that there is mariadb as alternative [13:03:10] I would recommend mysql for development since that's what most of people use [13:03:19] mutante: what is the name of class? [13:03:43] petan|wk: how to see the info of php? [13:04:06] Shujenchang: I think there is some function built in php [13:04:28] Shujenchang: generic::mysql::server [13:04:40] mutante: do you have any idea how we could add new puppet class in NovaInstance configuration screen? [13:04:51] mutante: can you check why deployment-web2 sucks so much :O [13:05:16] petan|wk: it is binary compatible though and the lead developer is former guy from mysql. it is free as opposed to Oracle licensing being unclear [13:05:36] wait a moment oracle licensing isn't clear? [13:05:41] how can we use it then [13:05:54] I thought it was GPL [13:06:08] http://en.wikipedia.org/wiki/MariaDB [13:06:33] see on labs-l [Labs-l] Problems when setting up an instance of Wikidata [13:08:05] in the beginning there was a "rather this than mysql on labs", but it isn't finally decided [13:08:34] ok [13:08:45] anyway I still would rather use same as we have on prod [13:08:59] are we switching to Maria there? :P [13:09:01] hashar: which one were you looking for? [13:09:11] sudo puppetd -tv [13:09:11] info: Loading facts in projectgid [13:09:11] info: Loading facts in default_gateway [13:09:11] info: Loading facts in projectgid [13:09:11] info: Loading facts in default_gateway [13:09:11] Could not retrieve project_gid: private method `chomp' called for nil:NilClass [13:09:12] Could not retrieve project_gid: private method `chomp' called for nil:NilClass [13:09:12] Could not retrieve project_gid: private method `chomp' called for nil:NilClass [13:09:13] Could not retrieve project_gid: private method `chomp' called for nil:NilClass [13:09:13] Could not retrieve project_gid: private method `chomp' called for nil:NilClass [13:09:14] Could not retrieve project_gid: private method `chomp' called for nil:NilClass [13:09:14] err: Could not retrieve catalog from remote server: Error 400 on SERVER: Could not find class generic::mysql::server for i-0000021c.pmtpa.wmflabs on node i-0000021c.pmtpa.wmflabs [13:09:15] warning: Not using cache on failed catalog [13:09:15] err: Could not retrieve catalog; skipping run [13:09:20] What's wrong? [13:09:34] hm... [13:09:47] Could not find class generic::mysql::server for i-0000021c.pmtpa.wmflabs on node i-0000021c.pmtpa.wmflabs? [13:09:50] yes [13:10:06] I think that it wasn't merged to puppet [13:10:16] "could not retrieve project_gid" sounds like a bug [13:10:17] Ryan probably forgot to do that [13:10:37] yes that's another problem [13:10:41] meh [13:10:51] please open bugs in BZ :p [13:10:52] mutante: apaches::service [13:11:03] !bz [13:11:03] https://bugzilla.wikimedia.org/$1 [13:11:06] mutante: looks like that class is the perfect one to setup the various deployement hosts [13:11:07] meh [13:11:13] err the various web deployement hosts [13:11:33] maybe I have to install mysql manually via apt-get too [13:13:11] Shujenchang: the wikidata guys just had that installed. are you on the labs-l mailing list ? would probably be best to reply there as well [13:13:27] hashar: looking ...brb [13:13:41] looks like I can add group per project [13:13:54] mutante: yes, I'm on the mailing list [13:13:56] [Manage Puppet Groups] [13:15:14] Shujenchang: reply here, see last sentence by Ryan http://lists.wikimedia.org/pipermail/labs-l/2012-April/000176.html [13:15:14] mutante: yeah I can add new puppet classes :-D [13:15:54] !report is to report new bug open: https://bugzilla.wikimedia.org/enter_bug.cgi [13:15:54] Key was added! [13:16:34] hashar: ah, so also classes that have not been in there at all? and done? [13:16:59] well I have managed to add a project specific puppet class [13:17:01] trying right now [13:17:05] should be in Apache group [13:17:43] !log added apache::service on deployment-web host [13:17:43] added is not a valid project. [13:17:48] grrr [13:17:54] well, project specific sounds good enough, right [13:18:32] hashar: yeah, if in doubt i would log it to bastion project, abusing that as "meta" or something :) [13:18:51] !log deployment-prep added apache::service on deployment-web host [13:18:53] Logged the message, Master [13:18:56] uhm.. in this case, probably to deployment-prep [13:18:57] yep [13:18:58] we need a shorter project name [13:19:10] something like `tentacle` `jailbreak` or whatever [13:19:38] so how can I run puppet update now ? :-D is there a nice sudo rule ? [13:19:44] sounds good for project names, but i think hostnames should be project-001 or something and not just random names [13:20:02] I was referring to the project name [13:20:39] hashar: sudo puppetd -tv ? [13:20:50] ask for password [13:20:54] hashar: given the brand new changes to sudo i did not try yet [13:21:01] but there is always "sudo su -" :p [13:21:21] I must not be allowed to use sudo on sep-prep [13:21:22] hashar: hmm, Shujenchang was just able to do that [13:21:38] ? [13:21:54] Shujenchang: you just ran "sudo puppetd -tv" and did not have permission issues [13:22:25] permission issues? [13:23:22] hashar: [Labs-l] Per-project sudo, managed via labsconsole, you saw that yet? pretty new [13:23:47] yup I see that [13:24:19] deployment-prep has ALL ALL , and the default [13:24:26] though sudo ask me for a password [13:24:41] isn't it because I am not allowed to "puppetd -tv" ? [13:24:53] oh.. [13:25:00] no,just type your labs password [13:25:24] oh of course [13:25:32] mutante: you mean I shoud change "ALL ALL"? [13:25:50] works! [13:25:52] Shujenchang: no, you are fine, i was just giving hashar an example it worked for you. dont worry [13:25:59] hashar: k:) [13:26:06] :D [13:27:08] next question, when login on deployment-web PS1 show the instance name i-0000000217 [13:27:22] can't I make it to show the deployement-web instead ? [13:27:44] no :( not that easy [13:28:05] there has been discussion on that as well [13:28:17] like when we wanted the "nice" hostnames in nagios [13:28:34] I will just skip that for now so [13:28:46] I'm going to have my dinner, see u later~ [13:29:39] hashar: well, you can change your prompt by setting PS1= in login script, but this was about the real hostname :) [13:30:23] maybe LDAP could be used to lookup the instance-name given the instance-ID [13:30:38] Ryan wants us to file bugs for all of that ;) [13:33:11] it is not that much important anyway [13:33:44] another more serious issue is the devops group [13:34:19] it has a local gid of 1075 [13:34:35] and members seems to be manually managed in /etc/group directly [13:34:52] also, it is not the devops primary group which cause all kind of funny issues for me [13:35:12] namely, I can not write a file owned by someone:devops -rw-rw---- [13:35:30] vim think I can not write to it and open it in read only mode [13:35:37] hashar: similar, just filed this https://bugzilla.wikimedia.org/show_bug.cgi?id=36206 [13:35:41] then when writing it, it ends up as : hashar:svn -rw-rw---- [13:37:45] how are the account created on deployment-prep project anyway? [13:39:26] hashar: human users via labsconsole, putting them into the project- primary group, system users should be created via puppet but i think they aren't (and would NOT be in the project- group and you can't add them via puppet).. and i don't think puppet is used for this ..but should [13:40:25] ohh I did not know labsconsole could let us add / create users [13:40:27] checking [13:41:50] hashar: well if you add a user to a project it creates the homedir and puts the key there on next puppet run [13:43:07] bbiaw, importing wikis [14:14:47] I'm back [14:20:30] bah [14:30:47] mutante: here is my blocking bug https://bugzilla.wikimedia.org/show_bug.cgi?id=36218 :D [14:34:55] petan|wk: did you ever get help for https://bugzilla.wikimedia.org/show_bug.cgi?id=36135 localization on labs commons? [14:36:11] I've already installed php and mysql, how can I install Mediawiki? [14:37:34] hashar: good, best way to go i think since the whole gid/LDAP thing is changing and being worked on [14:37:49]