February 18, 2011  
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28

[00:00:48] <lacostej> questions: does anyone know why the builds are so unstable on jenkins on jenkins ? I have the same problem with my fork (http://coffeebreaks.virtual.vps-host.net:8082/job/jenkins-mvn3-lacostej-master/) and I haven't had much the time to look into it the last days.
[00:01:21] *** avandendorpe has joined #jenkins
[00:04:41] *** girasquid has quit IRC
[00:05:36] <kohsuke> Hmm, the error is that there are too many threads
[00:05:57] <kohsuke> I remember seeing excessive number of HtmlUnit related threads left behind --- maybe related
[00:13:43] *** BrianF has joined #jenkins
[00:14:29] *** _marc` has quit IRC
[00:16:58] *** BrianFox_ has quit IRC
[00:21:48] <webbananaboy> im having trouble getting jenkins to run my ant script on windows
[00:22:05] <webbananaboy> I keep getting ant.bat is not recognized as an internal or extrnal command
[00:22:18] <webbananaboy> anybody know how to fix this
[00:22:24] <webbananaboy> apart from using linux instead
[00:22:42] <gmcdonald> install ant?
[00:23:11] <webbananaboy> yeah its installed
[00:23:17] <webbananaboy> i can run ant from my command line
[00:23:28] <webbananaboy> I also put it in the system %Path%
[00:23:28] *** jasedit has quit IRC
[00:24:04] <gmcdonald> so ANT_HOME is set and c:\path\to\ant.exe is in your %PATH%
[00:24:45] <webbananaboy> looks like ANT_HOME is not set
[00:24:47] <webbananaboy> let me try that
[00:25:12] <webbananaboy> should it be set to the ant.exe or just to the base ant home directory
[00:25:17] <gmcdonald> base
[00:25:46] *** drulli has quit IRC
[00:30:08] <gmcdonald> looks like for ant in our config we specify the path to an install in the tools directory
[00:30:11] <gmcdonald> c:\hudson\tools\ant\apache-ant-1.8.1
[00:30:23] <gmcdonald> etc..
[00:37:10] <mindless> abayer: reminder of stale changes in clone-workspace-scm plugin.. please run a release when you get a chance, thx
[00:37:27] <webbananaboy> hmm
[00:37:38] <lacostej> kohsuke: thanks for the answer
[00:37:40] <webbananaboy> so maybe if i move it to the tools directory it would work
[00:37:56] <webbananaboy> setting ANT_HOME didnt work
[00:38:02] <webbananaboy> still getting
[00:38:03] <webbananaboy> cmd.exe /C '"ant.bat && exit %%ERRORLEVEL%%"' 'ant.bat' is not recognized as an internal or external command, operable program or batch file.
[00:38:39] *** msm has quit IRC
[00:39:28] *** selckin has quit IRC
[00:40:19] *** lacostej has quit IRC
[00:40:46] *** kenneth_reitz has quit IRC
[00:41:41] <hornswoggles> hey all. anyone have an idea why hudson.model.UpdateCenter.xml is sometimes a binary file and sometimes not?
[00:42:35] *** BrianF has quit IRC
[00:43:03] *** lacostej has joined #jenkins
[00:43:03] *** ChanServ sets mode: +v lacostej
[00:45:17] <hornswoggles> trying to manage the file through bcfg2 and it needs to know if a file is binary or not.
[00:47:03] *** kenneth_reitz has joined #jenkins
[00:47:20] *** awb has quit IRC
[00:50:48] *** harryv has joined #jenkins
[00:50:58] <harryv> hey, anybody had succes with github post receive hooks and jenkins?
[00:53:11] <kohsuke> I have just that plugin in the works
[00:53:24] <kohsuke> I need to push it and release it...
[00:54:35] *** esteele has joined #jenkins
[00:56:08] <harryv> nevermind. actually got it working :)
[00:56:35] <kenneth_reitz> I've been thinking of adding a GET hook instead of a POST hook
[00:56:38] <kenneth_reitz> not a rubyist though
[00:57:12] <kenneth_reitz> to github-services that is
[01:06:29] *** mando has quit IRC
[01:15:58] *** lacostej has quit IRC
[01:15:58] *** noahcampbell has quit IRC
[01:26:54] *** eric_n_dfw has quit IRC
[01:30:00] <mindless> jobConfigHistory can generate some funky diffs
[01:30:43] <mindless> the diff sections have overlapping line numbers/content
[01:34:33] *** tieTYT has joined #jenkins
[01:34:37] <tieTYT> yo
[01:34:57] <tieTYT> hudson is constantly giving me this exception when a build is triggered: org.tmatesoft.svn.core.SVNCancelException: svn: No credential to try. Authentication failed
[01:35:10] <tieTYT> when I ssh into the box and manually do an svn update, the problem goes away temporarily
[01:35:23] <tieTYT> it used to occur ~once a month
[01:35:30] <tieTYT> now it's happened the last 3 days in a row
[01:36:55] *** nej has quit IRC
[01:37:23] <tieTYT> google to the rescue
[01:40:51] *** Ragnor has quit IRC
[01:43:10] *** Ragnor has joined #jenkins
[02:05:33] *** d2m has quit IRC
[02:08:18] *** sshaw has quit IRC
[02:13:49] *** mindless has quit IRC
[02:20:13] *** d2m has joined #jenkins
[02:20:56] *** SnagJJV has left #jenkins
[02:28:59] *** joewilliams is now known as joewilliams_away
[02:37:46] <evilchili> abayer: how does one debug the git plugin?  I've got a windows slave that is hanging when the plugin does a clone, but i can run the clone (as the jenkins user) manually
[02:38:18] <abayer> That is a good question. I have never actually used the thing on Windows. =)
[02:38:52] <rtyler> I hear it involves magic
[02:38:54] <rtyler> and cygwin
[02:39:27] <nairb774> for windows, thankfully not
[02:39:29] <evilchili> magic? crap. I've already killed 7 unicorns for this project
[02:39:40] <nairb774> evilchili, are you cloning over the git protocol or ssh?
[02:39:46] <evilchili> ssh
[02:41:43] <nairb774> how far in the clone does it get?
[02:41:49] <evilchili> no idea
[02:41:58] <evilchili> it appears to not be doing anything at all
[02:42:02] <nairb774> does it create the .git folder?
[02:43:06] <evilchili> yeah; seems fine, although refs/heads is empty
[02:43:18] <nairb774> any objects/packs showing up?
[02:43:22] *** ahhughes_ has joined #jenkins
[02:43:30] <evilchili> nossir
[02:43:51] <nairb774> is the server it is cloning from a machine you have access to?
[02:44:00] <evilchili> yep
[02:44:22] <evilchili> i was just about to put ssh into debug mode and watch from that end
[02:44:29] <evilchili> but was hoping I wouldn't have to
[02:44:36] <nairb774> git when sending, creates a pack file from all the data on the server
[02:44:38] <nairb774> then sends that
[02:44:45] <nairb774> I wonder if it is hanging doing that
[02:45:17] <evilchili> i wouldn't expect the problem to be on the remote end; if I run the clone from the command-line it succeeds
[02:45:18] <nairb774> git streams the pack file as it creates it, so it would be in the very early stages
[02:45:24] <evilchili> but yeah, something is whacked
[02:45:43] <nairb774> true, forgot about you mentioning that
[02:45:52] <evilchili> also, abayer, aborting a build during a git operation leaves hung ssh.exe and git.exe on the slave :/
[02:46:10] <evilchili> i'll open an issue for it if there isn't one already
[02:47:07] <nairb774> have you tried using a free style build to invoke git to see if that works?
[02:47:17] *** wyrdvans has joined #jenkins
[02:47:27] <nairb774> (where the build does not have any official scm attached to it)
[02:47:55] <abayer> Feel free - I'll try to get a Windows intance up to play with it at some point. Sorry. =(
[02:48:06] <evilchili> i can provide you with one if need be
[02:48:33] <nairb774> abayer, I can tell you that I use the git plugin on windows and it works great :)
[02:48:35] *** wyrdvans has quit IRC
[02:48:43] <evilchili> it *has* been working fine, until this week
[02:48:49] <evilchili> builds just started hanging
[02:49:15] <nairb774> I hate that
[02:49:31] <nairb774> any jenkins upgrades that might have triggered it?
[02:49:37] *** wyrdvans has joined #jenkins
[02:50:01] <nairb774> I can stop asking the easy questions if they are annoying ;)
[02:50:36] <evilchili> none that I'm aware of.  and no, not annoying.  that's how troubleshooting works. :)
[02:51:56] <evilchili> it doesn't appear tied to any particular git operation; if I do the clone and then start a new build, the fetch hangs
[02:52:03] <evilchili> but I can run the fetch from command-line
[02:52:18] <nairb774> so odd
[02:52:28] <nairb774> what version of jenkins?
[02:52:34] <evilchili> 1.397
[02:53:00] <evilchili> ah!
[02:53:14] <evilchili> when i went from hudson => jenkins it installed a new service for the jenkins slave
[02:53:24] <evilchili> that service isn't running as the proper user
[02:53:34] <nairb774> oops
[02:53:55] <evilchili> this would have been *much* easier to figure out if git's output showed up in the console log :(
[02:54:36] <nairb774> it doesn't? huh, I thought it did
[02:54:50] <evilchili> nope
[02:54:58] <evilchili> i imagine it's hanging waiting for credentials
[02:55:03] <nairb774> ah
[02:55:04] <evilchili> but there's no output in the console log
[02:55:17] <nairb774> what jvm are you using? hotspot?
[02:56:09] *** elpargo has joined #jenkins
[02:57:17] <evilchili> nairb774: i'm sure I don't know :x
[02:57:38] <evilchili> 1.6.0_17-b04 from sun
[02:57:42] <nairb774> ok
[02:58:07] <nairb774> I don't know the internals of the git plugin well, but there was this: https://github.com/jenkinsci/jenkins/commit/3c5c2c
[02:58:24] <nairb774> which made logs disappear on j9/ibm
[02:58:41] * nairb774 digs in the git-plugin code
[03:01:56] <nairb774> it seems it points the stdout to a ByteArrayOutputStream
[03:12:44] *** ahhughes_ has quit IRC
[03:12:57] *** ahhughes has joined #jenkins
[03:14:16] <ahhughes> I think I remember seeing a number of ${VARIABLE}'s that I can use in my hudson job config? I need to pass in some properties to my maven job... e.g. -Dfoo=bar
[03:14:20] <wyrdvans> Can someone tell me how to save a property in a groovy script?  I'm trying to set the TracProjectProperty.
[03:16:21] <wyrdvans> here's what I have so far but it doesn't work:  http://jenkins.pastebin.com/rf3fD8Q7
[03:20:14] *** bap2000 has quit IRC
[03:25:57] <wyrdvans> wow...I really know how to quiet a room
[03:27:43] <jieryn-w> evilchili: hey.. i think i updated console-column-plugin to your liking
[03:27:59] <jieryn-w> evilchili: unfortunately there's no easy way to do what you requested, but i think i got to the spirit of the issue
[03:28:26] <evilchili> jieryn-w: sweet! I saw the update but haven't had a chance to investigate yet
[03:29:13] <jieryn-w> so column names will be  L LC, LFC, LStC, LSuc, LUtc, LUcc
[03:29:22] <jieryn-w> sorry, no "L"
[03:29:31] <jieryn-w> but the other L* will be there
[03:29:42] <jieryn-w> the only other way to do this will be to create a custom, fake, locale
[03:29:52] <jieryn-w> and i18n it yourself
[03:30:04] *** kenneth_reitz has quit IRC
[03:30:05] <jieryn-w> well, i wanted to check with you before rolling out a release
[03:32:13] *** cowboyd has joined #jenkins
[03:32:26] *** cowboyd has joined #jenkins
[03:32:28] *** hornswoggles has quit IRC
[03:33:50] <ahhughes> is there a ways I can get the jenkins workspace dir to use in my maven -DsomeParam=${JENKINS_WORKSPACE} ?? goals and options...
[03:34:06] <abayer> ${WORKSPACE}. =)
[03:34:12] <evilchili> jieryn-w: I'll grab it tomorrow and have a look
[03:34:18] <evilchili> but it sounds good to me
[03:34:38] <jieryn-w> ok, i'll roll a release now so you can pick it up w/out hassle
[03:35:40] <jieryn-w> that is, except for the hassle of restarting jenkins  *coughs loudly*
[03:35:45] <jieryn-w> :-D
[03:38:17] <ahhughes> abayer, thaks... I can;t see where they are documented :'( and can I use them anywhere.. I think I need them in my maven "goals and options"
[03:39:08] <abayer> Yeah, you can - and there's a link below build steps in freestyle jobs, at least, to (your server)/env_vars.html. =)
[03:41:02] <ahhughes> make me wonder why that link is not under the maven "options and goals" section too if I can use it there :/
[03:43:48] * ahhughes tests this theory...
[03:47:52] <ahhughes> works!
[03:47:54] *** elpargo has quit IRC
[03:51:11] *** abayer has quit IRC
[03:51:44] <jieryn-w> kohsuke sounds like walking dead
[03:54:27] <ahhughes> submitted issue/request :) very simple one... but I think its worth having http://issues.jenkins-ci.org/browse/JENKINS-8820
[03:54:29] <kohsuke> you are in the call, too?
[03:54:31] *** rcampbell_ has quit IRC
[03:54:34] <jenkins-admin> JENKINS-8820:Maven Job Configure should have links to env-vars.html (Open) http://jenkins-ci.org/issue/8820
[03:55:24] <jieryn-w> i am listening, goto meeting hates kittens and linux
[03:57:52] <kohsuke> I guess the system doesn't show other participants because I only see me and Charles
[03:58:38] <jieryn-w> i can't join because i am a linux user
[03:58:54] <jieryn-w> but i can listen to call, i suppose i should drop off since i'm a perl person, not ruby :-)
[04:00:18] <jieryn-w> evilchili: ok, 1.2 released
[04:19:12] *** elpargo has joined #jenkins
[04:19:52] *** atmos has quit IRC
[04:21:16] *** esteele has quit IRC
[04:25:46] *** elpargo_ has joined #jenkins
[04:27:11] *** elpargo has quit IRC
[04:27:11] *** elpargo_ is now known as elpargo
[04:27:15] *** Lewisham has joined #jenkins
[04:27:15] *** ChanServ sets mode: +v Lewisham
[04:28:27] *** BrianF has joined #jenkins
[04:28:53] *** elpargo has quit IRC
[04:31:43] *** girasquid has joined #jenkins
[04:32:19] <cowboyd> I need to find a screen sharing soln that works on linux
[04:32:42] <cowboyd> you know I used to run Vbox with windows and screen share that way
[04:33:20] <lifeless> vnc, screen (with ssh),
[04:33:50] <mwalling> webex is java now, isnt it?
[04:38:07] *** rromanchuk_ is now known as rromanchuk
[04:41:04] <kohsuke> cowboyd: descriptors[FogCloud] = d
[04:52:40] *** Teo1 has quit IRC
[04:52:57] *** abayer has joined #jenkins
[04:52:58] *** ChanServ sets mode: +o abayer
[04:53:27] <kohsuke> cowboyd: org.apache.commons.beanutils.DynaBean
[04:56:29] *** girasquid has quit IRC
[04:56:57] <kohsuke> cowboyd: Object get(String)
[04:59:11] <kohsuke> cowboyd: *.iml, *.ipr, *.iws
[05:00:36] <kohsuke> cowboyd: git rm --cached
[05:02:14] *** kenneth_reitz has joined #jenkins
[05:04:09] *** Deesl has joined #jenkins
[05:06:23] *** BrianF has quit IRC
[05:07:27] *** kenneth_reitz has quit IRC
[05:08:33] *** noahcampbell has joined #jenkins
[05:10:37] *** ahhughes has quit IRC
[05:10:58] *** ahhughes has joined #jenkins
[05:10:59] *** wyrdvans has quit IRC
[05:13:02] <rtyler> cowboyd: got jruby support hacked in yet? :D
[05:13:44] <cowboyd> rtyler: working on it :)
[05:13:57] <rtyler> are there more people on the meeting? I don't have a headset
[05:14:03] <rtyler> and I also have nothing to contribute
[05:14:08] <rtyler> but I can make soothing ocean sounds
[05:15:34] *** rcampbell_ has joined #jenkins
[05:19:55] <rtyler> kohsuke: the dude that hacks on minecraft once did a hackathon while ustreaming his IDE
[05:20:04] <rtyler> I wonder if that'd be interesting to watch for you ;)
[05:20:23] <kohsuke> what's minecraft?
[05:20:35] <rtyler> WHAT
[05:20:37] <rtyler> THE
[05:20:41] <rtyler> WHAT
[05:20:47] <rtyler> how do you not know
[05:20:55] <rtyler> it's probably for the best
[05:23:04] *** kenneth_reitz has joined #jenkins
[05:25:41] <kohsuke> cowboyd: http://jenkins.pastebin.com/3QieGCh2
[05:25:45] <cowboyd> rtyler: it's just me and kohsuke
[05:25:51] <rtyler> bummer
[05:26:40] <rtyler> cowboyd: I'm bummed you don't like Jython instead of JRuby
[05:26:44] <rtyler> that'd make me so much happier
[05:26:55] <cowboyd> rtyler: jython was my first jvm language
[05:27:09] <cowboyd> I reckon if I can make it work for jruby, I can get jython to work too
[05:27:12] <rtyler> YUS
[05:27:16] <rtyler> THANK JEEBUS
[05:27:23] <cowboyd> it's mainly just figuring out where to hook in.
[05:34:19] *** kenneth_reitz has joined #jenkins
[05:34:53] <kohsuke> cowboyd: getWriter().println("Hello")
[05:35:42] <kohsuke> cowboyd: http://localhost:8080/cloud/Bob/
[05:39:51] <kohsuke> cowboyd: http://support.webex.com/support/system-requirements.html
[05:44:04] <mwalling> kohsuke: do me a favor. *DO NOT DISCOVER WHAT MINECRAFT IS*
[05:44:55] <rtyler> I'm sure Hudson would appreciated it :P
[05:45:42] <cowboyd> i'll have a look at doing it over webex next week
[05:45:48] <cowboyd> that way perhaps more people can join in.
[05:57:10] *** Deesl has quit IRC
[06:05:04] *** girasquid has joined #jenkins
[06:05:47] <girasquid> If I was going to keep track of whether jenkins was running or not (using something like, say, monit), what would be the best way to do that? Just pinging the port jenkins runs on, or something else?
[06:06:15] *** awb has joined #jenkins
[06:12:20] *** cowboyd has quit IRC
[06:14:31] *** cowboyd has joined #jenkins
[06:17:41] *** rcampbell_ has quit IRC
[06:20:40] *** cowboyd has quit IRC
[06:21:52] *** Deesl has joined #jenkins
[06:24:18] <mwalling> girasquid: i get /hudson/ for service up status (looking for the jenkins header), then check /manage/ to see if theres an upgrade notification
[06:24:54] <girasquid> mwalling: /hudson/ being the frontpage of your jenkins?
[06:25:00] <mwalling> yeah
[06:25:06] <mwalling> too lazy to rename it
[06:25:32] <mwalling> sue me... oracle probably will
[06:25:38] <girasquid> hah
[06:32:38] *** girasquid has quit IRC
[06:42:38] *** akostadinov has joined #jenkins
[06:43:05] *** akostadinov has quit IRC
[07:02:03] *** Deesl has joined #jenkins
[07:02:08] *** oeuftete has quit IRC
[07:13:13] *** ahhughes has quit IRC
[07:14:10] *** zaphX has joined #jenkins
[07:14:31] *** d2m has joined #jenkins
[07:17:59] *** noahcampbell has quit IRC
[07:31:04] *** Haloperidol has joined #jenkins
[07:35:02] *** akostadinov has joined #jenkins
[07:41:59] *** wolfs has joined #jenkins
[07:41:59] *** ChanServ sets mode: +v wolfs
[07:47:51] *** vila has joined #jenkins
[07:51:18] *** akostadinov has quit IRC
[07:51:28] *** akostadinov has joined #jenkins
[07:51:51] *** dotsev has joined #jenkins
[07:52:27] *** akostadinov has quit IRC
[07:52:38] *** akostadinov has joined #jenkins
[08:01:41] *** wolfs has quit IRC
[08:02:29] *** nej has joined #jenkins
[08:08:50] *** Weltraumschaf has joined #jenkins
[08:10:10] *** skataria has joined #jenkins
[08:11:37] *** Weltraumschaf has quit IRC
[08:16:53] *** nej has joined #jenkins
[08:17:34] *** Sebastian has joined #jenkins
[08:33:44] *** Haloperidol has quit IRC
[08:36:28] *** mconigliaro has quit IRC
[08:37:02] *** mconigliaro has joined #jenkins
[08:44:15] *** drulli has joined #jenkins
[08:44:17] *** ChanServ sets mode: +v drulli
[08:49:48] *** Deesl has quit IRC
[08:52:49] *** awb has quit IRC
[08:53:17] <Juri_Agarin> Has anyone else had problems with Jenkins+git plugin 1.1.5?
[08:53:45] <Juri_Agarin> We have build-jobs that work with 1.1.4 but not with 1.1.5. Git fetch seems to not work.
[08:54:16] *** emanuelez has joined #jenkins
[08:59:22] *** emanuelez has quit IRC
[08:59:50] *** emanuelez has joined #jenkins
[09:04:03] *** endless has joined #jenkins
[09:08:05] *** aheritier has quit IRC
[09:08:13] *** aheritier has joined #jenkins
[09:08:13] *** ChanServ sets mode: +v aheritier
[09:15:43] *** Stubbs has joined #jenkins
[09:24:12] *** emanuelez_ has joined #jenkins
[09:24:13] *** emanuelez has quit IRC
[09:25:02] *** endless is now known as patryk
[09:25:13] *** wolfs has joined #jenkins
[09:25:13] *** ChanServ sets mode: +v wolfs
[09:25:14] *** emanuelez_ has quit IRC
[09:25:26] *** emanuelez_ has joined #jenkins
[09:27:33] *** Stubbs has quit IRC
[09:27:38] *** cristiano has joined #jenkins
[09:27:56] *** Stubbs has joined #jenkins
[09:30:12] *** emanuelez_ has quit IRC
[09:30:41] *** emanuelez_ has joined #jenkins
[09:32:17] *** Lewisham has quit IRC
[09:36:09] *** d2m has quit IRC
[09:36:47] *** Heimidal has joined #jenkins
[09:37:47] *** Heimidal_ has joined #jenkins
[09:41:45] *** Heimidal has quit IRC
[09:42:24] *** Lewisham has joined #jenkins
[09:42:24] *** ChanServ sets mode: +v Lewisham
[09:43:20] *** cristiano has quit IRC
[09:45:00] *** imeikas has joined #jenkins
[09:51:18] *** calavera has joined #jenkins
[09:51:18] *** emanuelez_ has quit IRC
[09:52:08] *** emanuelez has joined #jenkins
[09:52:32] *** Sebastian has quit IRC
[09:56:01] *** cristiano has joined #jenkins
[09:57:51] *** slaboure has joined #jenkins
[09:58:34] *** Weltraumschaf has joined #jenkins
[10:03:30] *** _marc` has joined #jenkins
[10:04:18] *** d2m has joined #jenkins
[10:09:51] *** stigkj has joined #jenkins
[10:13:16] *** Lewisham has quit IRC
[10:13:18] *** emanuelez_ has joined #jenkins
[10:13:27] *** emanuelez has quit IRC
[10:14:36] *** markh_ has joined #jenkins
[10:18:18] *** emanuelez_ has quit IRC
[10:18:43] *** emanuelez_ has joined #jenkins
[10:18:57] *** cristiano has quit IRC
[10:20:35] *** Heimidal_ has quit IRC
[10:20:55] *** Heimidal has joined #jenkins
[10:24:07] *** d2m1 has joined #jenkins
[10:26:35] *** d2m has quit IRC
[10:28:27] *** Aetzel has joined #jenkins
[10:29:59] *** JHogarth has joined #jenkins
[10:32:29] *** DaveH has joined #jenkins
[10:33:35] *** voorth has joined #jenkins
[10:37:26] *** d2m has joined #jenkins
[10:39:28] *** d2m1 has quit IRC
[10:39:53] *** voorth has quit IRC
[10:44:47] *** Heimidal has quit IRC
[10:48:51] *** d2m has quit IRC
[10:51:38] *** d2m has joined #jenkins
[10:57:02] *** dienomight has joined #jenkins
[10:58:18] *** gmcdonald has quit IRC
[10:58:25] *** gmcdonald has joined #jenkins
[11:08:46] *** d2m1 has joined #jenkins
[11:09:38] *** gmcdonald has quit IRC
[11:09:38] *** gmcdonald has joined #jenkins
[11:10:47] *** ExtraSpice has joined #jenkins
[11:11:28] *** d2m has quit IRC
[11:26:15] *** Weltraumschaf has quit IRC
[11:26:15] *** emanuelez_ has quit IRC
[11:26:19] *** emanuelez has joined #jenkins
[11:29:26] *** d2m has joined #jenkins
[11:30:18] *** Weltraumschaf has joined #jenkins
[11:30:59] *** slaboure has quit IRC
[11:31:24] *** octe has joined #jenkins
[11:32:23] *** d2m1 has quit IRC
[11:34:45] *** atmos has joined #jenkins
[11:41:46] *** d2m has quit IRC
[11:46:17] *** atmos has quit IRC
[11:46:40] *** d2m has joined #jenkins
[11:48:48] *** d2m has quit IRC
[11:53:40] *** emanuelez_ has joined #jenkins
[11:53:40] *** emanuelez has quit IRC
[12:01:05] *** cld2403 has quit IRC
[12:18:39] *** emanuelez_ has quit IRC
[12:33:02] *** d2m has joined #jenkins
[12:35:17] *** aheritier_ has joined #jenkins
[12:35:18] *** ChanServ sets mode: +v aheritier_
[12:35:27] *** oeuftete has joined #jenkins
[12:36:12] *** slaboure has joined #jenkins
[12:37:36] *** edorian has quit IRC
[12:38:16] *** aheritier has quit IRC
[12:38:17] *** aheritier_ is now known as aheritier
[12:39:21] *** aheritier_ has joined #jenkins
[12:39:21] *** ChanServ sets mode: +v aheritier_
[12:40:18] *** aheritier_ has quit IRC
[12:40:25] *** aheritier_ has joined #jenkins
[12:40:25] *** ChanServ sets mode: +v aheritier_
[12:40:34] *** aheritier has quit IRC
[12:40:35] *** aheritier_ is now known as aheritier
[13:01:50] *** aheritier has quit IRC
[13:06:15] *** skataria has quit IRC
[13:09:31] *** aheritier has joined #jenkins
[13:09:31] *** ChanServ sets mode: +v aheritier
[13:16:39] *** aheritier_ has joined #jenkins
[13:16:40] *** aheritier has quit IRC
[13:16:40] *** aheritier_ is now known as aheritier
[13:16:40] *** ChanServ sets mode: +v aheritier
[13:19:16] *** larrys has quit IRC
[13:28:07] *** slaboure has quit IRC
[13:31:16] *** kstreith has joined #jenkins
[13:44:50] *** Lewisham has joined #jenkins
[13:44:50] *** ChanServ sets mode: +v Lewisham
[13:50:58] *** DamZzzz is now known as DamZ
[13:53:17] *** slaboure has joined #jenkins
[14:17:20] *** DamZ is now known as DamZzzz
[14:19:40] *** elpargo has joined #jenkins
[14:20:45] *** wyrdvans has joined #jenkins
[14:24:42] *** calavera has quit IRC
[14:29:57] *** Lewisham has quit IRC
[14:36:28] *** edorian has joined #jenkins
[14:36:38] *** rromanchuk has quit IRC
[14:37:21] *** rromanchuk_ has joined #jenkins
[14:43:28] *** ojacobson has joined #jenkins
[14:44:03] *** wolfs has quit IRC
[14:44:24] *** wolfs has joined #jenkins
[14:44:24] *** ChanServ sets mode: +v wolfs
[14:54:31] *** webbananaboy has quit IRC
[15:00:27] *** dogmatic69_ has joined #jenkins
[15:02:38] *** cowboyd has joined #jenkins
[15:02:41] *** dogmatic69 has quit IRC
[15:02:43] *** dogmatic69_ is now known as dogmatic69
[15:03:34] *** amitev has quit IRC
[15:03:48] *** cowboyd has quit IRC
[15:03:54] *** webbananaboy has joined #jenkins
[15:14:18] *** DamZzzz is now known as DamZ
[15:25:54] *** Weltraumschaf has quit IRC
[15:27:49] *** dotsev has quit IRC
[15:32:35] *** elliot has joined #jenkins
[15:32:57] *** stigkj has quit IRC
[15:34:41] *** BrianFox has joined #jenkins
[15:34:41] *** BrianFox has joined #jenkins
[15:38:05] *** larrys has joined #jenkins
[15:38:05] *** ChanServ sets mode: +v larrys
[15:41:36] *** kenneth_reitz has quit IRC
[15:41:42] *** msm has joined #jenkins
[15:47:50] *** elpargo_ has joined #jenkins
[15:49:21] *** nej2222 has joined #jenkins
[15:49:36] *** elpargo has quit IRC
[15:49:36] *** elpargo_ is now known as elpargo
[15:51:56] *** Stubbs has quit IRC
[15:58:17] *** dogmatic69 has quit IRC
[16:00:51] *** Stubbs has joined #jenkins
[16:00:56] *** rcampbell_ has joined #jenkins
[16:01:17] *** elpargo_ has joined #jenkins
[16:01:36] *** elpargo has quit IRC
[16:01:36] *** elpargo_ is now known as elpargo
[16:02:29] *** abayer has quit IRC
[16:02:53] *** Aetzel has quit IRC
[16:03:16] *** kenneth_reitz has joined #jenkins
[16:11:40] *** amitev has joined #jenkins
[16:16:21] *** akostadinov has quit IRC
[16:16:36] *** akostadinov has joined #jenkins
[16:17:12] *** nej2222 has quit IRC
[16:20:48] *** elpargo has quit IRC
[16:21:35] *** elpargo has joined #jenkins
[16:23:00] *** mando has joined #jenkins
[16:31:19] *** Stubbs has quit IRC
[16:35:05] *** SnagJJV has joined #jenkins
[16:35:40] *** rcampbell_ has quit IRC
[16:37:21] *** dienomight has quit IRC
[16:38:55] *** recampbell has joined #jenkins
[16:42:41] *** joewilliams_away is now known as joewilliams
[16:47:14] *** SnagJJV has quit IRC
[16:48:40] *** jasedit has joined #jenkins
[16:50:19] <jasedit> I enabled the security settings so only those who log in with a local Unix account can manage the server, and it isn't accepting my Unix account credentials. How do I go about getting back access?
[16:52:35] <mwalling> ~disablesecurity
[16:52:35] <javabot> mwalling, disablesecurity is http://wiki.jenkins-ci.org/display/JENKINS/Disable+security
[16:54:06] *** dogmatic69 has joined #jenkins
[16:54:19] <jasedit> Thanks :)
[16:55:28] *** kenneth_reitz has quit IRC
[16:55:45] *** girasquid has joined #jenkins
[16:56:26] *** sshaw has joined #jenkins
[16:56:35] *** nd__ has joined #jenkins
[16:57:45] *** abayer has joined #jenkins
[16:57:45] *** ChanServ sets mode: +o abayer
[16:58:28] *** drulli has quit IRC
[16:58:54] *** girasquid has quit IRC
[17:00:09] *** nd___ has quit IRC
[17:04:23] *** dogmatic69 has quit IRC
[17:09:04] *** wolfs has quit IRC
[17:09:45] *** amitev has quit IRC
[17:14:07] <Tartarus> Hmm
[17:14:17] <Tartarus> What's the biggest matrix anyone has ever built?
[17:15:06] <Tartarus> I'm looking at ~1053 valid combos on something I'm going to kick off over the weekend
[17:15:17] <ccutrer> whoa
[17:16:13] <ccutrer> my largest matrix is a 3-dimension, 2x3x5, with 11 valid combinations
[17:22:55] *** elpargo_ has joined #jenkins
[17:25:36] *** elpargo has quit IRC
[17:25:36] *** elpargo_ is now known as elpargo
[17:25:40] *** Ourson has joined #jenkins
[17:26:05] <Ourson> Hi
[17:26:55] <Ourson> I want to change the PATH variable for a slave, how can I do ?
[17:26:55] *** wolfs has joined #jenkins
[17:26:55] *** ChanServ sets mode: +v wolfs
[17:27:44] *** _marc` has quit IRC
[17:30:04] *** vjuranek has joined #jenkins
[17:31:58] <ccutrer> Ourson: click on the slave, click Configure, and check Environment Variables, and Add one for PATH
[17:34:18] <Ourson> but that replace the previous one, no ?
[17:35:03] <Ourson> what I want to do is  PATH=PATH:/my/path
[17:36:55] <Tartarus> That's fine
[17:37:03] <Tartarus> export PATH=${PATH}:/my/path
[17:37:08] <Tartarus> assuming your shell is bash or sh
[17:37:31] <Ourson> ok, thanks
[17:39:05] <ccutrer> yup, exactly (though I get the odd behavior that if PATH was originall /bin, and I set it to $PATH:~/bin; it ends up being /bin:~/bin:~/bin
[17:51:31] *** drulli has joined #jenkins
[17:51:31] *** ChanServ sets mode: +v drulli
[17:55:19] <Ourson> in a matrix project how specify more than one filter rule ?
[17:56:17] <Tartarus> Ourson, you don't, you use groovy regex
[17:56:23] <Tartarus> at least that's what I've had to do
[17:56:32] <Ourson> the following produces an exception : !(platform == "x86-windows" && arch=="x86_64")  (platform=="mac").implies(mac_version!="None")   (platform!="mac").implies(mac_version=="None")
[17:57:55] *** noahcampbell has joined #jenkins
[17:58:44] *** jfelchner has joined #jenkins
[18:00:46] *** Haloperidol has joined #jenkins
[18:00:50] *** patryk has quit IRC
[18:05:40] *** d2m has quit IRC
[18:06:31] *** dotsev has joined #jenkins
[18:09:41] *** Weltraumschaf has joined #jenkins
[18:14:19] *** steph021 has quit IRC
[18:17:15] *** Tartarus_ has joined #jenkins
[18:17:42] *** Tartarus has quit IRC
[18:17:42] *** vjuranek has quit IRC
[18:17:43] *** _W_ has quit IRC
[18:17:49] *** vjuranek has joined #jenkins
[18:18:16] *** Tartarus_ is now known as Tartarus
[18:18:40] *** _W has joined #jenkins
[18:18:40] *** d2m has joined #jenkins
[18:19:00] <ccutrer> Ourson: there are not operators between "!(platform == "x86-windows" && arch=="x86_64")", "(platform=="mac").implies(mac_version!="None")", and "(platform!="mac").implies(mac_version=="None")", which is why you're getting an exception
[18:19:51] <ccutrer> I have a hard time wrapping my head around what I want and how it needs to be processed, but try puttig || or && between each of those pieces, and see how it comes out
[18:20:32] <Ourson> ccutrer > Yeah, I finally suceed to do what I would
[18:20:58] <Ourson> thank you :)
[18:22:00] *** DamZ is now known as DamZzzz
[18:22:23] <Ourson> someone brave should make a nice UI with checkboxes to filter the matrix :)
[18:22:44] *** markh_ has quit IRC
[18:26:44] *** _marc` has joined #jenkins
[18:28:53] *** BrianFox has quit IRC
[18:30:09] *** sanitycheck has joined #jenkins
[18:30:31] <sanitycheck> why is #hudson even on freenode anymore?
[18:32:00] <ccutrer> sanitycheck: I think it's there simply as a place holder so people don't accidentally go into it (I can't join it, saying it's invite only)
[18:32:49] *** Weltraumschaf has quit IRC
[18:32:53] <sanitycheck> thx
[18:33:19] *** harryv has left #jenkins
[18:34:23] *** JHogarth has quit IRC
[18:35:13] *** mando has quit IRC
[18:35:59] *** wolfs has quit IRC
[18:36:33] *** Haloperidol has quit IRC
[18:37:25] *** Haloperidol has joined #jenkins
[18:45:56] *** cowboyd has joined #jenkins
[18:46:52] *** ian_mac has joined #jenkins
[18:49:21] <abayer> aheritier: did you just create "Sonacle" as a new word? =)
[18:52:15] *** SnagJJV has joined #jenkins
[18:59:21] *** elpargo has quit IRC
[19:06:10] *** SnagJJV has quit IRC
[19:07:00] *** BrianFox has joined #jenkins
[19:07:00] *** BrianFox has joined #jenkins
[19:07:20] *** SnagJJV has joined #jenkins
[19:08:03] *** DamZzzz is now known as DamZ
[19:10:46] *** Lewisham has joined #jenkins
[19:10:47] *** ChanServ sets mode: +v Lewisham
[19:14:42] *** kutzi has joined #jenkins
[19:14:42] *** ChanServ sets mode: +v kutzi
[19:16:19] *** DaveH has quit IRC
[19:17:40] *** slaboure has quit IRC
[19:19:56] *** dotsev has quit IRC
[19:27:50] *** dhackner has joined #jenkins
[19:29:34] *** DamZ is now known as DamZzzz
[19:32:18] *** rromanchuk_ is now known as rromanchuk
[19:33:51] *** elpargo has joined #jenkins
[19:37:20] <jenkinsci_builds> Starting build 520 for job jenkins_main_trunk (previous build: STILL UNSTABLE -- last SUCCESS #510 4 days 5 hr ago)
[19:38:31] *** BrianFox has quit IRC
[19:41:04] *** sanitycheck has left #jenkins
[19:41:50] *** wolfs has joined #jenkins
[19:41:51] *** ChanServ sets mode: +v wolfs
[19:43:47] <Creeture> What did the update-center.json URL change to?
[19:45:38] <evilchili> fun fact: if jenkins is doing a build against a git repository, and you replace the entire git repository, the build will succeed, but the entire repo history will be displayed under 'changes'
[19:45:39] <evilchili> :)
[19:46:40] *** recampbell has quit IRC
[19:47:15] *** recampbell has joined #jenkins
[19:48:46] *** bap2000 has joined #jenkins
[19:53:00] <bap2000> Does anyone know how the Maintainer(s) section of the plugin macro on the wiki is generated?
[19:53:03] <bap2000>  more specifically, how do I get something better than "n/a (id: bap)" in there?
[19:53:28] <drulli> It is generated from the pom.
[19:53:32] <abayer> bap2000: make sure you've got a committer section in your plugin's POM, and that'll get picked up int he next release.
[19:53:47] <bap2000> committer? I got developer
[19:53:53] <abayer> Sorry, developer.
[19:53:58] <abayer> Or it could be broken. =)
[19:54:09] <bap2000> .. and should I use my Jenkins bap, or my git bap2000?
[19:54:18] *** drulli has quit IRC
[19:54:33] <bap2000> guessing abayer is right ;-)
[19:54:57] <abayer> I'll take a look when I get a chance, but that probably won't be 'til sometime next week.
[19:55:27] <bap2000> no worries, just wondered why it didn't have a name in it like some of the others
[19:57:24] *** wolfs has quit IRC
[20:00:44] *** steph021 has joined #jenkins
[20:00:45] *** steph021 has joined #jenkins
[20:00:54] *** jasedit has quit IRC
[20:04:18] *** vjuranek has quit IRC
[20:06:05] *** DamZzzz is now known as DamZ
[20:10:36] *** sshaw has quit IRC
[20:13:08] <jieryn-w> http://daniel.gredler.net/2011/02/15/hudson-and-jenkins-two-weeks-later/
[20:17:51] <kohsuke> If you include commits to plugins, I bet the difference would be even bigger
[20:18:36] *** SnagJJV has quit IRC
[20:20:33] *** Weltraumschaf has joined #jenkins
[20:20:44] <jieryn-w> indeed
[20:22:21] *** kutzi has quit IRC
[20:26:22] <jieryn-w> now that they are using git, comparisons are going to be even easier ..
[20:32:29] *** awb has joined #jenkins
[20:33:07] *** akostadinov has quit IRC
[20:36:25] *** wolfs has joined #jenkins
[20:36:26] *** ChanServ sets mode: +v wolfs
[20:38:25] *** msm has quit IRC
[20:39:26] *** mando has joined #jenkins
[20:44:33] *** jonarcher has joined #jenkins
[20:44:52] *** msm has joined #jenkins
[20:48:32] *** tickez has quit IRC
[20:48:39] *** tickez has joined #jenkins
[20:49:48] *** wolfs has quit IRC
[20:50:37] *** SnagJJV has joined #jenkins
[20:50:54] *** SnagJJV has left #jenkins
[20:52:56] *** Aetzel has joined #jenkins
[20:58:31] *** kenneth_reitz has joined #jenkins
[21:02:05] <jenkinsci_builds> Project jenkins_main_trunk build #520: STILL UNSTABLE in 1 hr 24 min: http://ci.jenkins-ci.org/job/jenkins_main_trunk/520/
[21:02:06] <jenkinsci_builds> Kohsuke Kawaguchi: doc improvements
[21:02:08] <jenkinsci_builds> Starting build 521 for job jenkins_main_trunk (previous build: STILL UNSTABLE -- last SUCCESS #510 4 days 6 hr ago)
[21:14:39] *** ExtraSpice has quit IRC
[21:18:50] *** ka6sox has joined #jenkins
[21:20:17] <ka6sox> hi there.
[21:21:37] * rtyler runs and hides
[21:21:49] <ka6sox> I rtyler you can't hide
[21:23:06] <rtyler> so did you have specific questions for abayer, kohsuke, mindless and I?
[21:23:07] <ka6sox> I am doing research on an article about FOSS projects which had their core infrastructure tied to a "benevolent corporation" that changed direction and left them without support.
[21:23:12] <ka6sox> yes.
[21:23:14] <ka6sox> specific
[21:23:55] <ka6sox> was all your wiki/scm/feeds/etc all tied to the BC?
[21:24:05] <rtyler> BC?
[21:24:11] <rtyler> oh, benevolent corp
[21:24:12] *** drulli has joined #jenkins
[21:24:19] *** ChanServ sets mode: +v drulli
[21:25:25] <rtyler> wiki/issues were actually on Snoracle hardware
[21:25:31] <rtyler> scm was with java.net
[21:26:02] <ka6sox> okay, I don't know the relationship between java.net and snoracle.
[21:26:47] <rtyler> your guess is as good as mine :P
[21:27:07] *** harryv has joined #jenkins
[21:27:14] <harryv> when the irc bot crashes... how do I get him back?
[21:27:24] *** harryv has left #jenkins
[21:27:55] <rtyler> ka6sox: java.net started as a Sun + CollabNet collaboration to provide a home for java projects
[21:28:11] <rtyler> I think Oracle foots the bill, but there's a terms of service, etc tied to it
[21:28:14] *** DamZ is now known as DamZzzz
[21:29:18] <ka6sox> okay so basically you were tied directly or indirectly with "terms" to the BC.
[21:29:56] <rtyler> when kohsuke left Oracle, the actual "tie-in" to Oracle was not clearly enumerated or talked about
[21:30:00] *** robertjames has joined #jenkins
[21:30:19] <robertjames>   /msg NickServ VERIFY REGISTER robertjames rdblqskljkyf
[21:30:29] <ka6sox> mmmmm
[21:30:32] <rtyler> RUH ROH
[21:31:10] *** robertjames has left #jenkins
[21:31:18] <ka6sox> when someone with root@ in there signon comes on a channel I have the bot screams: rrrrrrrrrooooooot has landed!
[21:31:56] *** kenneth_reitz has quit IRC
[21:32:40] <ka6sox> second question: did you take anything (that wasn't restricted) in terms of sources/wikipages/anything else and how painful was that?
[21:33:12] <rtyler> dealing with importing JIRA and Confluence backups seemed painful, but kohsuke did that :P
[21:35:51] *** kenneth_reitz has joined #jenkins
[21:36:59] <ka6sox> I know in your case that OSUOSL is providing support, but not everyone can go there, unfortunately, how long did you search for alternatives till you landed there?
[21:38:42] <rtyler> our relationship with the OSUOSL extended back into early 2010 when I approached them about mirroring Hudson
[21:39:08] <ka6sox> okay so you didnt' search far.
[21:39:34] *** Teo1 has joined #jenkins
[21:41:10] *** kenneth_reitz has quit IRC
[21:41:11] *** msm has quit IRC
[21:44:48] *** drulli has quit IRC
[21:44:50] <ka6sox> how long from the time you broke off till you had a workin (and you define working) system?
[21:46:01] *** msm has joined #jenkins
[21:50:08] *** sshaw has joined #jenkins
[21:55:02] *** wolfs has joined #jenkins
[21:55:02] *** ChanServ sets mode: +v wolfs
[21:56:50] *** elliot has quit IRC
[21:57:52] *** elliot has joined #jenkins
[21:59:53] *** Aetzel has quit IRC
[22:16:09] *** wyrdvans has quit IRC
[22:17:19] <Creeture> If I wanted to provide a "Choice" in a parameterized build that is a dynamically generated list, where would I start?
[22:20:13] <Creeture> Run a script before the params prompt that generates a properties file, then use the Extended Choice Parameter plugin?
[22:22:02] <evilchili> Creeture: you could create a job that generates the choices, then trigger a parameterized build with that
[22:22:23] <evilchili> er, i guess that defeats the purpose of the list. nm
[22:26:42] <Creeture> Is there a plugin that will cause a build to execute something before prompting for parameters?
[22:27:29] *** msm has quit IRC
[22:29:01] *** ojacobson has quit IRC
[22:30:31] <jenkinsci_builds> Project jenkins_main_trunk build #521: STILL UNSTABLE in 1 hr 28 min: http://ci.jenkins-ci.org/job/jenkins_main_trunk/521/
[22:30:31] <jenkinsci_builds> Kohsuke Kawaguchi: fixed a possible race condition (INFRADNA-65)
[22:34:48] *** mando has quit IRC
[22:43:05] *** Stubbs has joined #jenkins
[22:58:21] *** stigkj has joined #jenkins
[22:58:29] *** Jay7 has joined #jenkins
[23:07:49] *** elliot has quit IRC
[23:16:54] <rtyler> ka6sox: we probably had a weekend or so, we already ran a lot of our infrastructure
[23:17:02] *** Weltraumschaf has quit IRC
[23:17:24] <ka6sox> rtyler, okay good...thats quite helpful.
[23:18:00] <ka6sox> switching gears:
[23:18:09] *** kstreith has quit IRC
[23:18:27] <rtyler> the stuff important to development we ran, ci.jenkins-ci.org, SCM (github) and the mailing lists (google groups)
[23:18:53] <ka6sox> a number of our "remote builders" are already running either hudson or jenkins, with Jenkins Peering can those remote builders just "report" back to the master jenkins?
[23:19:35] <ka6sox> okay...see you were further ahead than the others it seems.
[23:20:14] <ka6sox> (and they would switch to jenkins instead of hudson.
[23:20:53] <rtyler> huh?
[23:21:28] <ka6sox> a number of the folks who are providing build services for us already are either running hudson or jenkins
[23:21:43] <ka6sox> and they are behind corp FWs
[23:21:45] <Jay7> or may switch to it :)
[23:21:53] <ka6sox> right Jay7
[23:21:55] * Jay7 is one from #oe
[23:22:34] <rtyler> ah this is the issue we've talked previously about
[23:22:37] <rtyler> wtf is jenkins peering? :P
[23:23:24] <ka6sox> hell if I know...that was a term that somebody said..not me :D
[23:23:41] <Tartarus> i made it up :)
[23:24:18] <Tartarus> One use case, for example, is we have jenkins up, behind corp FW and bringing slaves up dynamically
[23:24:40] <Tartarus> So I said, it'd be nice if it could just tell the public jenkins "I did this job, here's the results and saved files"
[23:24:53] <Tartarus> And configure that to happen only on some jobs, and called it a peer :)
[23:24:59] <rtyler> Tartarus: that may fit under the external job use-case
[23:25:07] <Tartarus> OK
[23:25:14] <Jay7> my case is home buildserver on dynamic real ip which is may run builds only at specified time (e.g. over night)
[23:25:18] <Tartarus> I'm not partial to how it's solved, just the constraints we have :)
[23:25:29] * rtyler nods
[23:25:35] * ka6sox too
[23:27:21] <ka6sox> rtyler, in addition we expect some new hardware that is a few racks over in the DC  that we *will* need to control.
[23:27:56] <ka6sox> so we have 3 kinds
[23:28:04] <ka6sox> 1) controlled by Jenkins
[23:28:11] <ka6sox> 2) FT but only reporting
[23:28:21] <ka6sox> 3) PT but only reporting
[23:28:53] <Jay7> is FT/PT full and partial time?
[23:29:02] <ka6sox> yes
[23:29:14] <Jay7> 2 have no meaning imho then
[23:29:27] <Jay7> because it is subset of 3 :)
[23:30:00] <ka6sox> Jay7, okay that is correct
[23:30:03] <ka6sox> 2 use cases.
[23:30:51] *** kenneth_reitz has joined #jenkins
[23:31:26] <rtyler> the "reporting" usecase is confusing to me, it sounds like you more intend for batch jobs to be run and reported asynchronously
[23:31:35] <ka6sox> yes
[23:31:36] <Jay7> btw..
[23:31:46] <ka6sox> that is correct.
[23:31:52] <Jay7> ka6sox: we can't report state per package then
[23:32:03] <Jay7> or we should integrate bitbake and jenkins
[23:32:06] <ka6sox> we have to fix bitbake to do that.
[23:32:11] <ka6sox> right..integration.
[23:32:15] <rtyler> bitbake plugin!
[23:32:19] <rtyler> make it so Jay7
[23:32:20] * rtyler runs
[23:32:26] <Jay7> rtyler: we have one already..
[23:32:38] <Jay7> and second was removed about a week ago :)
[23:32:49] <Jay7> (tinderbox client)
[23:33:03] <ka6sox> hssssssssssss
[23:33:09] <ka6sox> bad client
[23:33:24] <Jay7> true
[23:34:11] <ka6sox> it reports every single step via a HTTP put...millions of them.
[23:34:17] <rtyler> heh
[23:34:36] <ka6sox> * however many builders are reporting.
[23:36:06] <Jay7> Tartarus: btw, we shouldn't forget about automated testbuilds reporting to replace that scary wiki table
[23:36:36] *** kenneth_reitz has joined #jenkins
[23:36:56] <ka6sox> Jay7, thats one of the nicest pics in Jenkins...status of everything
[23:37:10] <ka6sox> a whole table of status.
[23:37:58] <Jay7> ka6sox: but what about details of build env?
[23:38:02] <rtyler> the amount of builds OE is doing scares the bejesus out of me
[23:38:14] <rtyler> please tell me you have at least one person paid to deal with it full time
[23:38:20] *** edorian has quit IRC
[23:38:46] <Jay7> I mean e.g. buildhost arch, distro, build type, etc
[23:38:58] <ka6sox> rtyler, I'm a volunteer just trying to keep the infra runing...I was up @ 3:30am last night trying to keep this ....thing from killing off the server.
[23:40:01] *** _marc` has quit IRC
[23:40:11] <rtyler> I think you can solve the PT solution with a slave-side daemon that checks some URL for some work-spec (packages to build) and then submits the results to some defined external job in Jenkins
[23:40:34] *** wolfs has quit IRC
[23:40:35] <rtyler> of course, if you can SSH directly into slaves, that's far easier
[23:40:43] <ka6sox> not possible
[23:40:50] <ka6sox> has to be a PULL model for them.
[23:41:01] <rtyler> I meant for the machines connected directly to Jenkins
[23:41:03] <ka6sox> I'm ready...got work for me?
[23:41:14] * Tartarus scrolls up
[23:41:15] <Jay7> heh.. seems another oestats-client/server but under jenkins control..
[23:41:28] <ka6sox> for the "local ones" that are a few racks away thats possible
[23:41:55] <rtyler> well, you could use Jenkins to manage queueing work with jobs to "insert" work into a DB or something else for said clients to get
[23:41:58] <ka6sox> but the ones that are really remote they are behind FW's
[23:42:23] <ka6sox> that I like better...because it ties a job# to the reporting.
[23:42:30] <rtyler> Jay7: I'm sure you guys have already done some kind of this, I'm trying to help come up with solutions that play to jenkins' strengths
[23:42:43] <rtyler> queueing and reporting, Jenkins excels at
[23:42:55] <Jay7> well..
[23:43:11] <ka6sox> rtyler, the mechanism of transfer of the queue information is not "corp FW friendly"
[23:43:25] <ka6sox> or PT builder friendlyh
[23:43:34] <rtyler> what would be corporate FW friendly?
[23:43:50] <ka6sox> a place to pick up work with a job number
[23:43:51] <Tartarus> Well, I can post things out
[23:44:00] <rtyler> right, like I said :P
[23:44:00] <Jay7> so every slave should have jenkins on the buildhost to run builds and queue/report results to upstream (master jenkins somewhere aroung ka6sox)
[23:44:03] <ka6sox> a mechanism that allows posts back with the job number.
[23:44:06] <Tartarus> And I have lots of resources, but they're shared (Which is why I wrote that LM Slave plugin :))
[23:44:48] <Jay7> or should we have some middleware between slave and master jenkins'es?
[23:45:59] <ka6sox> Jay7, we need a "not connected all the time" plugin.
[23:46:32] <Tartarus> ka6sox, for which part?  Slave can just fail to come up when asked
[23:47:43] <ka6sox> okay so establish JNLP connections to pick up work then go away, connect back up and report?
[23:48:26] <ka6sox> has to be fault tolerant and buffer up stuff till slave is ready
[23:49:09] <Tartarus> Yeah.  I don't use JNLP slaves right now (mutter at VMware), but yes, jenkins will just keep waiting for the slave
[23:49:14] <Tartarus> Or see if another slave can do it
[23:50:00] <ka6sox> probably would be best if periodically the slaves would connect back up and report" I"m still working on it"
[23:50:15] <Jay7> via http post ;>
[23:50:16] <ka6sox> so it doesn't get given away as not being handled.
[23:50:34] <Tartarus> A slave that's alive and registered is a tunnel of sorts
[23:50:37] <ka6sox> Jay7, thats a plugin...I"m hoping not to have to do that.
[23:50:58] <ka6sox> so we have to depend on maintaining the tunnels?
[23:51:06] <ka6sox> or it fails?
[23:51:12] *** webbananaboy has left #jenkins
[23:51:23] <Tartarus> ka6sox, well, the connection, or yes, thart portion of the job will fail
[23:51:24] * Jay7 scared by scheme
[23:51:32] <Tartarus> But I don't think that's a bad thing, per se
[23:51:55] <Tartarus> It just means that the can be controlled via the public jenkins slaves will be up, or not
[23:51:59] <ka6sox> there is distributed and then distributed....
[23:52:08] <ka6sox> like globally
[23:52:10] <Tartarus> And then the firewalled bits will be depending, either r/o or coming up when they can be up
[23:52:13] <ka6sox> and JNLP isn't robust.
[23:52:36] *** elliot has joined #jenkins
[23:53:43] <ka6sox> I was hoping for something that didn't require FT tunnels...otherwise it FAILS..on a 2-3day build that means failure due to lost link loses days of work.
[23:53:43] *** elliot has quit IRC
[23:54:10] <Tartarus> Well, not exactly
[23:54:21] * Jay7 have uplink via pptp/l2tp
[23:54:21] <Tartarus> It depends on how you split the job up
[23:54:40] <Tartarus> for example, if the slave goes away in the middle of 1 part of the matrix job, that 1 part will fail
[23:54:52] <Tartarus> and the remaining parts will either wait for the slave to come back online, or go to another slave
[23:54:55] <Jay7> so it may reconnects sometimes and get other ip
[23:55:11] <sigmonsay> I've got a interesting question here. Is there a way to monitor changes for specific directories from git in hudson?
[23:55:24] <sigmonsay> well jenkins? I guess.
[23:55:36] <Tartarus> Depends on how your git repo is setup, I think
[23:55:44] <Tartarus> You can at the module level
[23:55:59] <ka6sox> Tartarus, we need something that doesn't depend on FT links across nets we don't control.
[23:56:13] <Tartarus> Well, hmm
[23:56:25] <Tartarus> Would a not full time controlled slave be possible, folks?
[23:56:35] <Tartarus> ie just giving periodic "still working" messages + a timeout or something
[23:56:45] * ka6sox hopes.
[23:57:14] <ka6sox> we don't need that tightly coupled...
[23:57:25] *** Teo1 has left #jenkins
[23:57:33] *** larrys has left #jenkins
[23:59:33] <ka6sox> best would be slave connects up, gets work and goes away, sending periodic "I'm still working" messages
[23:59:52] <ka6sox> otherwise...if it fails to report the job is considered compromised and reassigned.

top