logs | stats
   July 27, 2009  
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 | 29 | 30 | 31

[00:31:52] *** Lewisham has joined #hudson
[01:13:43] *** davidstrauss has joined #hudson
[03:19:10] *** admc has joined #hudson
[03:29:55] *** rromanchuk has quit IRC
[03:31:25] *** rromanchuk has joined #hudson
[03:57:33] *** admc has quit IRC
[06:17:54] *** davidstrauss has quit IRC
[06:37:46] *** davidstrauss has joined #hudson
[07:07:42] *** ppawel has quit IRC
[07:19:45] *** AgentIcarus has quit IRC
[07:19:45] *** glass-eye has quit IRC
[07:19:45] *** cowmix has quit IRC
[07:19:45] *** huimies has quit IRC
[07:24:55] *** cowmix has joined #hudson
[07:24:55] *** glass-eye has joined #hudson
[07:24:55] *** huimies has joined #hudson
[07:24:55] *** AgentIcarus has joined #hudson
[08:05:48] *** khmarbaise has joined #hudson
[08:10:11] *** khmarbaise has quit IRC
[08:14:55] *** admc has joined #hudson
[08:15:27] *** Lewisham has quit IRC
[08:21:00] *** davidstrauss has quit IRC
[08:31:37] *** davidstrauss has joined #hudson
[08:49:37] *** dsd_ has joined #hudson
[09:39:27] *** akostadinov has joined #hudson
[10:10:32] *** timp_ has joined #hudson
[10:11:26] *** dsd_ has left #hudson
[10:18:59] *** dvrzalik has joined #hudson
[10:27:53] *** admc_ has joined #hudson
[10:39:01] *** admc_ has quit IRC
[10:40:40] *** admc has quit IRC
[10:42:28] <AgentIcarus> hmm, I'm seeing the same behaviour as ppawel
[10:44:17] <AgentIcarus> (ssh slaves plugin throwng a NPE)
[10:44:34] <AgentIcarus> is there an easy way to go back a version of a plugin?
[10:45:14] <AgentIcarus> (I've tried removing the plugin directory but that didn't do anything)
[10:45:47] <AgentIcarus> ah, found 0.5.2
[11:28:34] <rtyler> ahahahah, somebody didn't do a build before checking in
[11:28:36] <rtyler> we has failures
[11:29:01] *** eskatos has joined #hudson
[11:29:11] <eskatos> 'morning
[11:29:15] <rtyler> ahoy
[11:34:13] <eskatos> I've got an issue running hudson 1.317 standalone, with the embedded winstone servlet container
[11:34:15] <eskatos> I've got an issue running hudson 1.317 standalone, with the embedded winstone servlet container
[11:34:16] <eskatos> I'm using maven projects with subversion pooling
[11:34:18] <eskatos> when a build run fine everything go well, failed unit tests are correctly reported too
[11:34:19] <eskatos> but when a build fails, it seems to be just ignored by hudson and at the end of the console output I get the following with no error in the hudson logs : http://pastie.org/560035
[11:34:21] <eskatos> any clue ?
[11:34:49] <rtyler> you see anything in the terminal that you started hudson with?
[11:34:52] <rtyler> or its logs?
[11:35:22] <rtyler> this might have something to do with abayer's work, does everything work fine with 316?
[11:36:08] <eskatos> I have only this in the hudson logs : http://pastie.org/560037
[11:36:52] <eskatos> mmm in fact I'm running 1.316 and this issue is present since several versions
[11:39:45] <eskatos> unfortunately I cannot say for sure which version started to show this defect
[11:45:51] <rtyler> eskatos: do you have the time to jump back a couple versions and isolate when it appeared?
[11:46:26] <eskatos> oh, I forgot to say that's hudson is running behind apache
[11:46:57] <eskatos> rtyler: is there somewhere a versions timeline so I can target the right ones ?
[11:46:58] <rtyler> I don't /think/ that should matter
[11:47:02] <rtyler> yeah, hold on
[11:47:36] <rtyler> https://hudson.dev.java.net/servlets/ProjectDocumentList?folderID=2761
[11:48:30] <eskatos> perfect :)
[11:48:34] <rtyler> <3
[11:48:46] <rtyler> funny, I have another french user in another channel that I'm helping as well
[11:48:55] <rtyler> you don't work for the ministry of education per chance :P
[11:53:29] <eskatos> not at all, I'm working in a java shop :)
[11:54:00] <rtyler> aha, then you two won't know each other :P
[11:54:04] <eskatos> but I'm happy to learn that the ministry of education use hudson :)
[11:54:16] <rtyler> dunno if they do
[11:54:21] <rtyler> they use Cheetah though
[11:55:24] <eskatos> oh ok, I don't know Cheetah (except the famous monkey)
[11:55:35] <rtyler> python-based templating engine
[11:56:32] <eskatos> k
[11:57:20] <rtyler> are you on the users@ list?
[11:57:49] <eskatos> yes, if I need time to find the version causing pb I'll certainly post there
[11:58:03] * rtyler nods
[11:58:04] <rtyler> wonderful
[11:59:33] <eskatos> when I set up the server running hudson I installed it through the hudson's debian package version 1.301 and then upgraded through the webui ... now I realize this is a mess /o\
[11:59:50] <rtyler> hah
[11:59:56] <rtyler> yeah, a bit easier to get hudson direct
[12:00:02] <rtyler> we do releases about every week
[12:00:26] <eskatos> yeah I know, I upgrade about every two weeks :)
[12:02:43] <eskatos> it appears that the available debian packages are up to date (1.317), are they published in sync with the releases ?
[12:02:58] <rtyler> I believe so, they might lag a couple days
[12:03:04] <rtyler> I know kohsuke does something or another with debian
[12:20:59] *** AgentIcarus has quit IRC
[12:20:59] *** huimies has quit IRC
[12:20:59] *** glass-eye has quit IRC
[12:20:59] *** cowmix has quit IRC
[12:21:21] *** akostadinov has quit IRC
[12:22:13] *** cowmix has joined #hudson
[12:22:13] *** glass-eye has joined #hudson
[12:22:13] *** huimies has joined #hudson
[12:22:13] *** AgentIcarus has joined #hudson
[13:51:01] *** akostadinov has joined #hudson
[14:26:20] *** velo has joined #hudson
[14:32:55] <eskatos> rtyler: I've found what caused my issue : I set up the build to keep 0 previous builds ... setting this to 1 fixed the problem
[14:33:03] <rtyler> o_O
[14:33:06] <rtyler> 0 builds
[14:33:07] <rtyler> lolz
[14:33:15] <eskatos> :)
[14:33:49] <eskatos> I tought that hudson will keep the results of just the last build but not archive the artifacts
[14:33:51] <eskatos> my bad
[14:34:22] <eskatos> so, shall putting 0 in this field be allowed ?
[14:34:34] <eskatos> BTW now everything works
[14:35:39] <eskatos> mm, with 0 in this field, everything works well with successfull or unstable builds, only failed build were screwed up
[14:49:23] *** jdolan_ has joined #hudson
[14:54:16] *** abayer__ has joined #hudson
[14:56:32] <rtyler> eskatos: interesting, want to file a P3-5 bug for that?
[15:11:51] *** abayer has quit IRC
[15:20:55] <rtyler> abayer__: you're up awfully early
[15:28:11] *** abayer__ has left #hudson
[15:28:38] *** abayer__ has joined #hudson
[15:28:43] *** abayer__ is now known as abayer
[15:29:16] <abayer> I'm actually always up this early - I work with people on the east coast and in India, etc, so I tend to work ridiculously early hours.
[15:29:48] <rtyler> ah, good to know
[15:29:58] <rtyler> in case I ever need company to join me at denny's
[15:30:01] <rtyler> >:D
[15:30:06] <abayer> Denny's? Good god, man.
[15:30:39] <rtyler> first thing that popped into my head, I usually just go hungry
[15:30:49] <abayer> heh
[15:34:08] <rtyler> bagdad cafe supposedly is open 24 hours, but they've no wifi and sometimes just decide they don't wanna stay open
[15:34:25] <abayer> I had no idea they were even supposed to be.
[15:35:06] <rtyler> http://www.bagdadcafesf.com/
[15:35:10] <rtyler> LIARS!
[15:38:04] <abayer> It just says they're open 24 hours - it doesn't say WHICH 24 hours.
[15:39:04] <rtyler> you win again Mr. I Heart Words
[15:39:14] <abayer> Of course I do.
[15:39:15] <abayer> =)
[15:40:21] <abayer> Why the hell are you up this early, anyway?
[15:40:55] <rtyler> been up all night
[15:41:04] <abayer> Ow.
[15:41:54] *** velo has quit IRC
[15:42:02] <rtyler> sleep issues, I haz them
[15:42:26] <abayer> I used to have nasty insomnia. Finally went away a few years ago.
[15:42:35] <rtyler> I don't think anybody at the hackathon really understood the feat that I had accomplished by waking up before noon on a saturday
[15:42:37] <abayer> Of course, now I go to bed at like 10:30.
[15:42:44] *** keshureddyp1 has quit IRC
[15:42:46] <rtyler> I slept from 8am until 5pm yesterday
[15:42:50] <abayer> Yeesh.
[15:42:53] *** jumpkick has quit IRC
[15:43:01] <abayer> Have you considered a regular sleeping schedule?
[15:43:08] <abayer> I hear they're hip with the kids these days.
[15:43:16] <rtyler> pshaw!
[15:43:19] <abayer> also, my cat says hi.
[15:43:50] <rtyler> I can actually have a regularly sleeping schedule, it unfortunately is quite regular from 7-ish until 4pm-ish
[15:43:58] <abayer> owie.
[15:44:01] <rtyler> those are the hours I stayed steady at back in the day when I did consulting
[15:45:28] <abayer> You should totally move to the Far East - you seem to already be on their time zone.
[15:46:01] <rtyler> $SHE might not appreciate that, and I do like being awake when SF is, there's people to talk to
[15:46:06] <rtyler> I really wish I just didn't require sleep
[15:46:17] <abayer> Same, but hey.
[15:46:31] <abayer> Biological existence has its price.
[15:46:46] *** velo has joined #hudson
[15:47:23] <rtyler> that said, given some of the joys of biological existence *cough*, I'll deal and just nap every now and again :P
[15:48:18] <rtyler> wow, I really do like it when unit testing pays off
[15:48:33] <rtyler> particularly when one refactors code
[15:48:41] * rtyler notes that somebody broke the Hudson build
[15:48:50] <abayer> Me?
[15:48:56] <abayer> I would not be shocked. =)
[15:48:57] <rtyler> don't think so
[15:49:03] <abayer> Ok, actually, I would be.
[15:49:14] <abayer> I haven't committed anything for a couple days.
[15:49:28] *** rtyler changes topic to "http://hudson-ci.org | Paste: http://hudson.pastebin.com || Logs: http://echelog.matzon.dk/?hudson || Twitter: http://twitter.com/hudsonci || Did you break the build? http://hudson.glassfish.org/job/hudson/ || Committers should have voice"
[15:49:43] <rtyler> because of kohsuke insane setup for the Hudson job, it's hard to tell
[15:49:49] <abayer> heh
[15:49:55] * rtyler still doesn't grok why svn polling is so difficult
[15:50:09] <abayer> *shrug*
[15:50:48] <rtyler> somebody seriously just didn't compile before committing
[15:50:50] <rtyler> >_<
[15:50:55] <rtyler> /home/hudson/workspace/hudson/hudson/plugins/backup/src/main/java/org/jvnet/hudson/plugins/backup/utils/BackupTask.java:106: doQuietDown() in hudson.model.Hudson cannot be applied to (org.kohsuke.stapler.StaplerResponse)
[15:50:59] <rtyler>             Hudson.getInstance().doQuietDown(FakeObject.getStaplerResponseFake());
[15:51:01] <abayer> Dude, you see that all the time.
[15:51:02] <rtyler>                                 ^
[15:51:05] <rtyler> /home/hudson/workspace/hudson/hudson/plugins/backup/src/main/java/org/jvnet/hudson/plugins/backup/utils/BackupTask.java:137: doCancelQuietDown() in hudson.model.Hudson cannot be applied to (org.kohsuke.stapler.StaplerResponse)
[15:51:07] <abayer> Well, I do, at least. At work.
[15:51:09] <rtyler>             Hudson.getInstance().doCancelQuietDown(FakeObject.getStaplerResponseFake());
[15:51:12] <rtyler> haha
[15:51:14] <rtyler> I don't :P
[15:51:17] <rtyler> well
[15:51:18] <rtyler> occasionally
[15:51:21] <rtyler> and I usually flip my shit
[15:51:48] <rtyler> we byte-compile the entire site prior to a push, so if somebody has a dumbass SyntaxError checked in, it breaks our ability to push the site
[15:51:54] <rtyler> (that's when I flip my shit)
[15:52:21] <rtyler> especially when there's a bloody Hudson job that's started failing well in advance because of said SyntaxError
[15:52:47] <rtyler> probably best I don't deal with java devs and do releng, I probably would have bashed some skulls by now
[15:52:52] <abayer> heheh
[15:53:15] <abayer> You are definitely not mellow enough to do my job.
[15:53:32] <rtyler> heh
[15:55:03] <eskatos> rtyler: I'll fill the issue, which subcomponent ? core ?
[15:55:17] <rtyler> eskatos: yeah, I think that should work
[15:55:30] <rtyler> will you CC me on it? just in case you're not around I can likely help explain to kohsuke
[15:56:12] <eskatos> rtyler: yep !
[15:56:46] <eskatos> rtyler: could you tell me what's the english name for the incriminated field ? :)
[15:57:26] <rtyler> huh o_O
[15:57:37] <rtyler> perhaps you're referring to the "assigned" field?
[15:58:29] <eskatos> I mean, the field that was set to 0 and causing my troubles .. maybe "Maximum builds to keep" ?
[15:58:49] <rtyler> Number of builds to keep
[15:59:44] <eskatos> thanks
[16:01:52] <eskatos> rtyler: https://hudson.dev.java.net/issues/show_bug.cgi?id=4110
[16:05:04] <btQuark> hello everyone
[16:05:15] <rtyler> hellp
[16:05:40] <btQuark> since the update to 1.316 i am not able to get all config options for jobs
[16:05:59] <btQuark> it stops filling the site after half the stuff, plus issues with sshslaves
[16:07:07] <rtyler> JS error?
[16:12:54] <btQuark> fear so. should not be that way
[16:20:39] *** kstreith has joined #hudson
[16:27:25] *** statlor has joined #hudson
[16:33:38] *** davidstrauss has quit IRC
[16:33:38] *** kohsuke has quit IRC
[16:39:07] *** kohsuke has joined #hudson
[16:39:07] *** davidstrauss has joined #hudson
[16:39:07] *** irc.freenode.net sets mode: +o kohsuke
[16:45:05] <statlor> <- committers should have voice
[16:47:28] *** rtyler sets mode: +v statlor
[16:49:48] *** davidstrauss has quit IRC
[16:50:47] <rtyler> statlor: you do realize the point of it right?
[16:50:56] <rtyler> make sure people can identify and harass developers ;)
[16:51:03] *** rtyler sets mode: +o abayer
[16:51:16] *** rtyler sets mode: +v jieryn-w
[16:51:20] *** rtyler sets mode: +v kstreith
[16:51:24] * rtyler hinks that's it
[16:53:05] *** rromanchuk has quit IRC
[16:53:32] <statlor> oh, I thought it was so we could +m and then discuss our plans for world domination in peace
[16:53:58] <rtyler> well not if you keep blabbing about it while *they're* around ....
[16:53:59] <rtyler> >_>
[16:54:00] <rtyler> <_<
[16:54:25] <statlor> admittedly, my piece of the world domination would be small (it would, in fact, be Otisburg)
[16:56:02] * statlor 's Superman 2 reference falls flat.
[16:56:10] <statlor> my what pointy eyes you have
[16:56:53] *** akostadinov has quit IRC
[16:59:23] *** jumpkick has joined #hudson
[17:02:59] <chetan-> what's the best way to bring a c/c++ project into hudson?
[17:04:57] <rtyler> how do you build normally, and what do you do with tests?
[17:09:16] <chetan-> we have a Makefile.PL in place of configure/autoconf and then we use make as usual
[17:09:39] <chetan-> at the moment we just have integration tests that get run via maven
[17:10:20] *** barmeier has joined #hudson
[17:10:21] <chetan-> so I want to somehow do a build of the c/c++ stuff before maven builds the java stuff
[17:11:58] <barmeier> Hi, anybody out there who has successfully used LDAP with group access rules ?
[17:15:41] *** mart_ has joined #hudson
[17:16:06] *** mart has quit IRC
[17:17:19] *** mart has joined #hudson
[17:18:16] *** davidstrauss has joined #hudson
[17:20:23] *** keshureddyp has joined #hudson
[17:31:25] <statlor> chetan, you could configure your build job to fire the maven integration test job after it builds (use the "Build other projects" option in the post-build actions of the build job)
[17:34:38] *** mart_ has quit IRC
[17:38:46] *** mart has quit IRC
[17:40:50] <barmeier> when using unix autentication I get : SIGSEGV (0xb) at pc=0xb47d4281, pid=17697, tid=3032075152 any ideas ?
[17:41:14] *** rromanchuk has joined #hudson
[17:42:39] <statlor> hm seems like I saw a email about that on the mailing list in the past couple days
[17:44:04] <barmeier> do you know if there is a workaround available ß
[17:44:05] <barmeier> ?
[17:58:48] <statlor> http://www.nabble.com/Hudson-Crashes-while-using-Linux-PAM-based-authentication-method-td24564295.html#a24564295
[17:59:26] <statlor> as far as I can tell, an issue hasn't been created yet on jna.dev.java.net's issue tracker
[18:00:36] <barmeier> I use debian lenny maybe a debian problem ?
[18:01:07] <statlor> what architecture?
[18:01:17] <statlor> x86_64?
[18:02:01] <barmeier> yes
[18:04:28] *** davidstrauss has quit IRC
[18:05:01] <statlor> dunno, it might be worth download the jna source and build jna.jar to run the unit tests (see the JNA Homepage Building section)
[18:05:08] <statlor> downloading even
[18:17:45] <barmeier> I am not sure that I am able to do this, but I will try.
[18:20:32] *** barmeier has left #hudson
[18:23:20] *** akostadinov has joined #hudson
[18:24:15] *** jumpkick_ has joined #hudson
[18:29:22] *** akostadinov has quit IRC
[18:37:53] *** eskatos has quit IRC
[18:39:15] *** lkundrak has joined #hudson
[18:40:24] *** jumpkick has quit IRC
[18:51:30] *** jumpkick has joined #hudson
[18:52:01] *** davidstrauss has joined #hudson
[18:52:07] *** timp_ has quit IRC
[19:04:18] *** akostadinov has joined #hudson
[19:07:55] *** alexlod has joined #hudson
[19:08:04] *** jumpkick_ has quit IRC
[19:16:15] *** mrooney1 is now known as mrooney|w
[19:30:29] *** akostadinov has quit IRC
[19:35:09] <statlor> committers: is there a setting in the issue tracker that makes new issues for a given plugin get auto-assigned to you?
[19:35:20] <abayer> It's called kohsuke. =)
[19:35:27] <statlor> I took over the accurev plugin from stephenconnolly, and someone just opened a new issue and it got assigned to him
[19:35:29] <kohsuke> which component to whom?
[19:35:37] <statlor> accurev plugin to statlor
[19:35:38] <kohsuke> And your java.net ID is statlor?
[19:35:40] <kohsuke> Got it
[19:35:41] <statlor> yep
[19:35:48] <statlor> thanks :)
[19:36:01] <statlor> also I'll go ahead and say thank from stephen as well
[19:36:10] <statlor> er thanks
[19:36:19] *** khmarbaise_ has joined #hudson
[19:37:38] <statlor> I got lucky in that someone updated one of my tickets and mentioned the new ticket, otherwise I would have missed it
[19:37:46] <statlor> I'm used to hitting the "my issues" link now
[19:38:01] <kohsuke> done
[19:38:14] * statlor cheers
[19:42:34] *** Lewisham has joined #hudson
[19:43:51] *** rtyler sets mode: +v Lewisham
[19:53:43] <statlor> hm anyone know about another memory leak with the UserRequest object?
[19:53:49] <statlor> and remoting
[19:54:06] <statlor> had someone email me directly about that
[19:56:31] <statlor> oh.. I think it's 4008
[19:59:37] <statlor> yeah same guy that emailed me
[20:15:14] <jumpkick> is there a way to make a 0 size build que?
[20:15:27] *** Lewisham has quit IRC
[20:16:03] <jumpkick> queue
[20:28:22] <calculus> yay, auto-installer is so much fun: Error 1311.Source file not found(cabinet): C:\Documents and Settings\LocalService\Application Data\Sun\Java\jdk1.6.0_13\st160130.cab.  Verify that the file exists and that you can access it.
[20:28:34] <calculus> and that file does not exist
[20:28:43] <calculus> there is a sd and ss file but no st
[20:33:30] <calculus> kohsuke: ideas? ^^
[20:34:01] <kohsuke> More output lines?
[20:35:26] <calculus> kohsuke: jdk installer log: http://hudson.pastebin.com/m3016f300
[20:37:06] <calculus> console output just says that the auto installer failed and to look at the log file (which I pasted above)
[20:37:10] *** khmarbaise_ has quit IRC
[20:37:11] <kohsuke> If you can file this as a bug that would be great.
[20:37:21] <kohsuke> I'm not sure what went wrong
[20:38:33] <calculus> what subcomponent would this be?
[20:39:16] *** khmarbaise has joined #hudson
[20:40:13] *** Lewisham has joined #hudson
[20:42:00] *** rtyler sets mode: +v Lewisham
[20:43:28] *** Lewisham has quit IRC
[20:44:16] *** jumpkick_ has joined #hudson
[20:46:21] <calculus> kohsuke: issue 4118
[21:03:53] *** jumpkick has quit IRC
[21:08:18] <jumpkick_> kohsuke: did you see my comment about using "Depends: daemon, adduser, java-virtual-machine, default-jre | default-jre-headless" for the deb in instead of depending on java2-runtime?
[21:16:19] <jumpkick_> I have a public hudson instance that I'm using the publisher plugin to send jobs up to @ http://mixxx.org:8080/ ...
[21:16:40] <jumpkick_> that hudson public instance has 0 executors (cause I only want to use it as a dashboard)
[21:16:55] <jumpkick_> because the private nodes that do the building use SCM Polling
[21:17:16] <jumpkick_> jobs are starting to queue up in the public server's build queue
[21:17:31] <jumpkick_> is there a way to disable that queue?
[21:18:37] <statlor> in your build executor status on the left, does your public instance show up with one executor?
[21:21:03] <jumpkick_> statlor: nope, no executors, BZR isn't even installed on the public instance (you can have a look at the builds and queue on the site http://mixxx.org:8080/)
[21:21:51] <statlor> hm that's odd, is the public instance a slave of the master?
[21:22:08] <statlor> I take it jobs that aren't tied to a node are getting picked up here
[21:22:11] <jumpkick_> no, there is no master
[21:22:30] <statlor> you using the forwarding plugin?
[21:23:27] <jumpkick_> http://wiki.hudson-ci.org/display/HUDSON/Build+Publisher+Plugin
[21:23:36] <statlor> that's what I meant, sorry
[21:24:04] <jumpkick_> I'm pushing build results up to the public instance to server as a dashboard for people to see build results
[21:26:57] <statlor> right, so are there actual jobs configured on teh public side (sorry I don't know anything about build publisher)
[21:27:32] <statlor> am wondering why stuff thinks it should queue there, if it's not otherwise connected to the internal instances
[21:37:24] *** davidstrauss has quit IRC
[21:39:07] <jumpkick_> not exactly
[21:40:01] <jumpkick_> the publisher plugin copies everything from the jobs/$JOB folder from the Private server to the Public server...
[21:40:15] <jumpkick_> the jobs are replicated
[21:40:51] <jumpkick_> they aren't actually supposed to execute on the Public server, however because the job definition that is replicated contains a SCM Poll -> @hourly
[21:45:09] *** dvrzalik has quit IRC
[21:54:33] *** davidstrauss has joined #hudson
[21:59:33] <statlor> ahh ok that makes more sense
[21:59:55] <statlor> hm seems like jboss folks would have had this problem
[22:00:58] <statlor> and would have some option when publishing to at least turn off polling
[22:15:14] <bobo_> is there any faster way to test your plugins on hudson then to build-->upload-->restart-->use plugin?
[22:15:39] <bobo_> ie do all the stuf with the webgui
[22:18:41] *** Lewisham has joined #hudson
[22:19:14] <kohsuke> jumpkick_: where did that comment about the debian package go?
[22:19:48] <jumpkick_> I pinged you about it yesterday
[22:19:54] <jumpkick_> I still have it in the buffer
[22:20:13] <kohsuke> pinged me here?
[22:20:20] <jumpkick_> yeah
[22:20:34] <jumpkick_> you were afk
[22:20:58] <jumpkick_> it was just to say that depending on java2-runtime sucks in all the UI stuff required for JDK6
[22:21:05] <jumpkick_> pulseaudio, x libs, etc
[22:21:26] <kohsuke> I'm not really a debian expert, so if you can assure me that you know what you are talking about, I'm happy to apply whatever changes.
[22:21:38] <kohsuke> just send me the diff and it'll be done.
[22:21:40] <jumpkick_> you can replace "java2-runtime" with "default-jre | default-jre-headless"
[22:21:50] <kohsuke> Or better yet, I'm happy to make you a committer.
[22:22:17] <jumpkick_> which will allow either  default-jre or default-jre-headless to satisfy the build-dep
[22:22:34] <jumpkick_> what SCM does it use?
[22:22:43] <kohsuke> Subversion
[22:23:08] <jumpkick_> okay, I can work that one
[22:23:27] <jumpkick_> where do I find the code, I'll do a patch and build
[22:23:49] <kohsuke> I'm bit confused though --- java2-runtime is a virtual package provided by a few things, including sun-java6-jre
[22:23:55] <jumpkick_> yeah
[22:23:56] <kohsuke> and that's the one I really want people to use.
[22:24:15] <kohsuke> but default-jre also provides java2-runtime
[22:24:26] <kohsuke> so isn't the current dependency already correct?
[22:24:37] <kohsuke> I'm looking at http://packages.debian.org/unstable/virtual/java2-runtime
[22:26:14] <jumpkick_> the problem is that one can't use sun-java-6-jre-headless
[22:26:58] <jumpkick_> if you run a server instance with no X-windows, you don't want sun-java-6-jre pulling in all of X windows so it can display swing apps locally
[22:27:40] <kohsuke> I thought Hudson does actually need X libraries
[22:27:53] <kohsuke> for fonts and etc., that it uses for generating graphs
[22:28:18] <kohsuke> (it doesn't need window managers and so on --- does headless mean no X libraries at all?)
[22:28:42] <jumpkick_> openjdk-6-jre-headless
[22:29:08] <jumpkick_> that's the one I'm running it on now
[22:29:41] <jumpkick_> http://packages.debian.org/lenny/openjdk-6-jre-headless
[22:29:52] <jumpkick_> (though I'm actually on ubuntu)
[22:32:07] <jumpkick_> looks like default-jre-headless doesn't properly depend on openjdk-6-jre-headless in Debian
[22:32:13] <jumpkick_> http://packages.ubuntu.com/jaunty/default-jre-headless -- has it right
[22:32:58] <jumpkick_> so I guess the dep should be "java2-runtime | openjdk-6-jre-headless" for max compatibility
[22:34:28] <statlor> bobo_: I think you can do hudson-dev:run ?
[22:34:37] <statlor> uses jetty
[22:35:06] <statlor> kohsuke, if using netbeans can you just hit run and it will launch jetty?
[22:35:10] <statlor> on a plugin
[22:35:41] <kohsuke> Debian doesn't have OpenJDK at all AFAIK, so how about just "java2-runtime-headless" ?
[22:35:46] <statlor> the wiki says you can hit debug but not run
[22:36:16] <kohsuke> When I depend on a virtual package, how does APT choose which one to install?
[22:36:52] <statlor> oh bobo: I think you want to do mvn hpi:run
[22:37:04] <statlor> I think I am looking at the right wiki page now (plugin tutorial)
[22:37:40] <statlor> and it sounds like view changes are picked up automatically without having to redeploy
[22:43:52] <jumpkick_> kohsuke: I don't think it  does pick, I think it tells you to install something to satisfy that dep
[22:44:04] <kohsuke> Ah
[22:44:06] *** khmarbaise has quit IRC
[22:44:23] <kohsuke> Should we depend on java2-runtime-headless and recommend sun-java-6-jre?
[22:50:04] <statlor> hm I see hpi:run on one wiki page and hudson-dev:run on another.. I wonder what the difference is, if any
[22:50:40] *** kstreith has quit IRC
[22:56:05] <statlor> will play around with them tonight
[23:06:26] <jumpkick_> kohsuke: there is no java2-runtime-headless package which is why I think it needs to be "java2-runtime | openjdk-6-jre-headless" perhaps with recommends "sun-java-6-jre"
[23:07:14] <jumpkick_> or maybe that would force the install of openjdk-6-jre-headless ...   arh
[23:07:19] <kohsuke> http://packages.ubuntu.com/jaunty/java2-runtime-headless
[23:07:44] <jumpkick_> http://packages.debian.org/search?searchon=names&keywords=java2-runtime-headless
[23:07:59] <kohsuke> but this search also fails for java2-runtime
[23:08:01] <jumpkick_> breaks on debian (debian is retarted ... arrrgh!)
[23:08:16] <jumpkick_> oh so it does...
[23:08:44] <jumpkick_> does that mean the deb won't install on debian?
[23:09:23] <kohsuke> I think it just means the search is broken: http://packages.debian.org/sid/java2-runtime-headless
[23:09:42] <jumpkick_> lol
[23:10:06] <jumpkick_> yeah so then java2-runtime | java2-runtime-headless it is
[23:10:20] <jumpkick_> that would be good
[23:11:53] <kohsuke> Just "java2-runtime-headless" would do, no?
[23:12:19] <kohsuke> The packages that provide java2-runtime seems to all provide java2-runtime-headless
[23:12:24] <kohsuke> which makes sense.
[23:13:03] <jumpkick_> cool, yeah...
[23:13:09] <jumpkick_> headless should work then
[23:13:16] <kohsuke> ok, I'll make a change. Thanks for your help
[23:13:32] <jumpkick_> thanks for putting up with my blathering.... lol
[23:19:07] <statlor> bbl
[23:19:10] *** statlor has quit IRC
[23:20:38] *** admc has joined #hudson
[23:24:14] *** velo has quit IRC
[23:36:39] *** jumpkick_ has quit IRC
[23:49:34] *** jdolan_ has quit IRC

top