[00:03:01] *** resmo has quit IRC
[01:58:51] *** velo_ has joined #hudson
[02:06:18] *** velo has quit IRC
[02:11:42] *** jtong_ has joined #hudson
[02:28:20] *** jtong has quit IRC
[03:23:55] *** jieryn-w has joined #hudson
[03:50:25] <jieryn-w> any devs around?
[03:50:32] <abayer> What's up?
[03:50:51] <jieryn-w> http://dpaste.com/71456/
[03:50:55] <jieryn-w> was told to report problem
[03:51:01] <abayer> Ah, that one.
[03:51:08] <abayer> I just stumbled across it again.
[03:51:11] <abayer> The error message, I mean.
[03:51:38] <jieryn-w> ah
[03:51:41] <abayer> and that getArtifactsDir one too.
[03:51:46] <jieryn-w> well, i do run m2 extra steps to -U deploy
[03:53:43] <abayer> 'k, yeah, something obviously crapped out in there, but I'm honestly not sure what/why.
[03:54:56] <abayer> Have you ever gotten this to run without a problem, or is this the first time?
[03:56:18] *** jtong has joined #hudson
[03:56:26] <abayer> I'm wondering if this is something specific to the m2 extra steps situation, basically.
[03:58:58] *** jbriguet has quit IRC
[03:59:06] *** jbriguet has joined #hudson
[04:01:22] <jieryn-w> oh, it's been running great for weeks
[04:01:30] <jieryn-w> even on 1.317 for past few days, it's done several successful builds
[04:01:37] <abayer> Good. =)
[04:01:47] <jieryn-w> problem is that it takes about 20 minutes to get me to the point where i can use m2 extra steps :-b
[04:01:49] <abayer> I always prefer a nice situational bug to one of those pesky systemic ones. =)
[04:02:06] <jieryn-w> i'd prefer a systemic one, it would be quicker and easier to reproduce
[04:02:25] <jieryn-w> these timing window type things are really bad at making my hairs gray
[04:02:43] <abayer> True, true.
[04:03:43] <abayer> Yeah, that's a strange bug. It's apparently barfing trying to get the parent item's name, which is why I wondered if it had ever worked - thought there might be some confusion trying to get the parent of one of the extra steps.
[04:04:28] <rtyler> afternoon chaps
[04:04:53] <rtyler> abayer: I came down to On The Corner down Haight, the fog/mist is really blowing in >_<
[04:05:32] <abayer> I was out in the Sunset - headed back around 5 and yeah, it was doing that rushing-onset-of-fog thing.
[04:05:50] <abayer> I get a lot less of it down here at the foot of the hill than you do up there, though.
[04:06:01] <rtyler> I've had to ride into it on my way home every day this week
[04:06:07] <rtyler> summer sucks >_<
[04:06:11] <abayer> heh
[04:06:27] <jieryn-w> i got my feets propped up and am sucking down a few brews
[04:06:36] <rtyler> mmmm brews
[04:06:38] <abayer> I'm just glad we haven't had any of the annoying 90+ degree heat waves yet.
[04:06:45] <jieryn-w> and doing java hacking, watching my builds aggregate on hudson.. it's nice
[04:07:01] <abayer> I'm digesting rather excellent chile verde.
[04:07:13] <jieryn-w> does anyone test hudson on massive SMP systems? i think i have a bug in hudson :-(
[04:07:16] <jieryn-w> mm
[04:07:22] <rtyler> jieryn-w: how massive?
[04:07:27] <abayer> Y'know, that's probably a good point.
[04:07:28] <jieryn-w> 10+ cpus
[04:07:33] <rtyler> o_O
[04:07:40] <jieryn-w> i have 10 here and hudson barfs at start up
[04:07:47] <abayer> I've only got 2 cpu boxes.
[04:07:49] <jieryn-w> i think it has to do with the "starting XX parallel threads"
[04:07:56] <rtyler> jieryn-w: if you ship me some evaluation hardware, I'd be happy to isolate :D
[04:07:58] <jieryn-w> i see in the hudson log
[04:08:01] <jieryn-w> :-b
[04:08:06] <jieryn-w> it's on my corporate network
[04:08:13] <rtyler> we have some monster Sun machines that are 16 cores/128GB of RAM
[04:08:24] <jieryn-w> the machine has 64 CPUs, i only get 10 virtual ones :) mainframe
[04:08:26] <rtyler> unfortunately, those have to serve up MySQL and I can't borrow any for hudson
[04:08:29] <rtyler> MAINFRAME FTL
[04:08:31] <rtyler> ;)
[04:08:34] <jieryn-w> yah, no shit
[04:08:41] <jieryn-w> heh
[04:08:45] <jieryn-w> actually i like mainframes
[04:08:46] * rtyler hands jieryn-w some punch cards and directs him to the teletype
[04:08:49] <abayer> Technically all my machines are on gigantic massive servers, but, well, VMs.
[04:08:52] <jieryn-w> yooo!!
[04:08:59] <jieryn-w> i earn my keep coding z/os :)
[04:09:03] <jieryn-w> i do java work on the side
[04:09:04] <rtyler> O_O
[04:09:05] <jieryn-w> for fun
[04:09:07] <rtyler> crikey
[04:09:09] <rtyler> YOU WIN
[04:09:11] * rtyler bows
[04:09:15] <jieryn-w> i have actual punch cards :)
[04:09:22] <abayer> heheh
[04:09:40] <rtyler> jieryn-w: do you run all local executors?
[04:09:52] <jieryn-w> yes, just one master hudson node
[04:10:07] <jieryn-w> every time i restart, all but 1 executor is Dead
[04:10:23] <rtyler> just one node o_O
[04:10:26] <jieryn-w> and i have to manually Manage Hudson > Reload Configuration From Disk     'cuz all but a couple of my jobs, random, are missing
[04:10:47] <rtyler> nevermind then, I was going to convince you to check out the concurrent-build brnach
[04:11:08] <jieryn-w> well, it's really annoying to have to restart hudson twice every time i want to get the full 5 executors i have defined
[04:12:52] *** velo_ has quit IRC
[04:14:22] <rtyler> abayer: so have you become a developer yet? I see you committing stuff and pooking around..
[04:14:27] <abayer> ?
[04:14:36] *** jtong_ has quit IRC
[04:14:41] <abayer> What's the distinction between "developer" and "not developer"?
[04:14:53] <rtyler> I think it's self-determined
[04:15:00] <abayer> Then yes, I am a developer.
[04:15:03] <rtyler> YAY
[04:17:19] *** ChanServ sets mode: +o rtyler
[04:17:22] *** rtyler sets mode: +o kohsuke
[04:17:25] *** rtyler sets mode: +o abayer
[04:18:40] <abayer> oooo. I'm an op.
[04:18:55] * rtyler hopes he doesn't regret it
[04:19:07] <rtyler> well, in most channels I've lurked in before, they denote developers with a +v
[04:19:09] <rtyler> for example
[04:20:25] <abayer> I'm gonna kickban everyone!
[04:20:33] <abayer> Or not.
[04:20:37] <rtyler> heh
[04:20:41] <abayer> Probably not.
[04:21:04] <abayer> It's just...the power! The pure, unvarnished power1
[04:21:11] <abayer> and typos.
[04:21:19] * rtyler woots
[04:21:19] <jieryn-w> heh :)
[04:21:22] <abayer> (yes, I was alt-tabbing while typing)
[04:21:26] <jieryn-w> i made my first commit to hudson the other day
[04:21:32] *** rtyler sets mode: +v jieryn-w
[04:21:33] <rtyler> :D
[04:21:37] *** rtyler sets mode: +vv mrooney1 Lewisham
[04:21:48] *** rtyler sets mode: +v calculus
[04:21:55] <jieryn-w> booyakasha http://fisheye4.atlassian.com/changelog/hudson/trunk/hudson?cs=20078
[04:21:58] * rtyler is trying to think of other commiters
[04:22:41] *** rtyler changes topic to "http://hudson-ci.org | Paste: http://hudson.pastebin.com || Logs: http://echelog.matzon.dk/?hudson || Twitter: http://twitter.com/hudsonci || Committers should have voice"
[04:23:04] <abayer> heh. I tried to fix a spelling mistake that gets on my nerves a couple days ago - "upstrem" for "upstream". But it turned out that refrences depending on that spelling mistake were *everywhere*.
[04:23:08] <rtyler> I'm quite pleased with this channel, it and the twitter account have both taken off more than I expected
[04:23:17] <rtyler> hahaha
[04:23:24] <rtyler> we've had that in our codebase a couple times
[04:24:59] <jieryn-w> abayer: btw - i just re-ran the job, and it succeeded no probs
[04:25:12] <abayer> Network barf, maybe.
[04:26:01] <jieryn-w> is there any way to disable fingerprinting?
[04:26:21] <jieryn-w> i really don't care about that sort of thing
[04:26:33] <abayer> It's necessary for any upstream/downstream stuff.
[04:26:49] <jieryn-w> doesn't seem it would be necessary in m2 builds
[04:27:24] <abayer> It's apprently how hudson determines relationships between builds.
[04:28:01] <jieryn-w> dang ok
[04:28:07] <rtyler> pwnt
[04:28:14] <abayer> I just learned that yesterday. =)
[04:28:52] <abayer> It's just md5 hashs anyway.
[04:30:38] <jieryn-w> do you guys nice(1) your builds in some way?
[04:30:53] <abayer> Nope.
[04:30:56] <jieryn-w> this is probably bad, but i run hudson on the same machine that we have some other internal prime-time functions on
[04:31:01] <rtyler> booo
[04:31:10] <abayer> heheheh
[04:31:20] <jieryn-w> it would be nice if hudson didn't abuse all the cpu
[04:33:12] <rtyler> that's kind of the point though
[04:36:54] * jieryn-w smells NICE plugin
[04:37:03] <jieryn-w> wonder if that is even possible
[04:37:12] <abayer> heh, good luck. =)
[04:37:36] * rtyler isn't sure why you'd want one
[04:37:48] <rtyler> why not just nice the Hudson instance itself?
[04:39:39] <rtyler> abayer: oh man, the fog is impressive now, I can't see much above the building tops now at all
[04:44:10] <abayer> heh
[04:50:34] <rtyler> best part about the fog: even easier to tell who's a tourist
[04:50:40] <rtyler> take off those flipflops you idiot
[05:15:12] <Lewisham> wow
[05:15:20] <Lewisham> I got my first enhancement request ever today
[05:15:21] <Lewisham> https://hudson.dev.java.net/issues/show_bug.cgi?id=4083
[05:15:27] <Lewisham> shame I don't have the time to work on it ;)
[05:15:40] <rtyler> even with Hudosn, or ever ever
[05:16:15] <Lewisham> ever ever
[05:16:19] <rtyler> HOORAY
[05:16:24] <rtyler> you're a real man now
[05:16:33] <Lewisham> does this mean I can buy a *beer*?
[05:16:36] <rtyler> why won't you have time?
[05:16:55] <Lewisham> the plugin was one in a longer road
[05:17:05] <Lewisham> of making the Continuous Integration Game
[05:17:12] <Lewisham> but super-charged
[05:17:19] <rtyler> *zap*
[05:17:24] <Lewisham> and it was more a learning experience than an end
[05:17:32] <Lewisham> so without me really dogfooding it
[05:17:39] <Lewisham> I can't justify the time to the man that pays me :)
[05:17:58] <Lewisham> I'm supposed to be working on the game
[05:18:58] <rtyler> pshaw
[05:20:14] <Lewisham> sadly, pshaw doesn't cut it in today's California public funding climate :)
[05:20:24] <rtyler> PSHAW
[05:20:29] <rtyler> caps helps though right?
[05:20:30] <Lewisham> the department got another letter this week telling it it had to cut more
[05:20:34] <Lewisham> hmm
[05:20:34] <Lewisham> yes
[05:20:43] <Lewisham> but how does one Caps Lock in public?
[05:20:46] <rtyler> learn python, come to SF :D
[05:20:50] <Lewisham> hah :)
[05:20:53] * Lewisham knows the Python
[05:20:55] <rtyler> Lewisham: IT'S EASY
[05:20:58] <rtyler> JUST RAISE YOUR VOICE
[05:21:01] <rtyler> LIKE THIS
[05:21:09] <Lewisham> HOW DO YOU PSHAW LOUDER WITHOUT SPITTING IN THEIR FACE
[05:21:24] <rtyler> IT DOESNT REALLY MATTER, THEY HAD IT COMING
[05:21:26] <rtyler> JERKS
[05:22:03] <Lewisham> :D
[05:22:08] <Lewisham> your boss must *love you*
[05:24:48] <rtyler> everybody loves me
[05:24:56] <rtyler> I'm like the tall blond version of the Fonz
[05:25:00] <rtyler> heeyyyyyyyy
[05:26:32] <Lewisham> haha
[05:26:35] <Lewisham> well
[05:26:41] <Lewisham> there are some things you just can't argue with
[05:26:46] <Lewisham> and we'll say that is one of them
[05:28:41] *** Lewisham_ has joined #hudson
[05:28:41] *** Lewisham has quit IRC
[06:23:31] *** Lewisham_ has quit IRC
[07:22:28] <rtyler> hmmm, pondering hudosn as a platform some more
[07:22:52] <rtyler> too bad *everything* you can do via the interface isn't exposed via the API
[07:22:56] <rtyler> (JSON that is)
[07:26:32] *** Lewisham has joined #hudson
[07:43:06] *** jieryn has joined #hudson
[07:56:03] *** jieryn-w has quit IRC
[07:58:59] *** rromanchuk has quit IRC
[08:00:39] *** rromanchuk has joined #hudson
[08:02:55] *** Lewisham has quit IRC
[08:02:55] *** cowmix has quit IRC
[08:02:55] *** glass-eye has quit IRC
[08:02:55] *** huimies has quit IRC
[08:02:55] *** AgentIcarus has quit IRC
[08:28:27] *** Lewisham has joined #hudson
[08:28:27] *** AgentIcarus has joined #hudson
[08:28:27] *** huimies has joined #hudson
[08:28:27] *** glass-eye has joined #hudson
[08:28:27] *** cowmix has joined #hudson
[08:35:36] *** Lewisham has quit IRC
[09:15:00] <ppawel> with 1.317 my slaves don't launch :(
[09:15:02] <ppawel> Unexpected error in launching a slave. This is probably a bug in Hudson.
[09:15:02] <ppawel> java.lang.NullPointerException
[09:15:02] <ppawel> 	at hudson.plugins.sshslaves.SSHLauncher.openConnection(SSHLauncher.java:428)
[09:16:39] <rtyler> rhu roh
[09:16:45] <rtyler> have you filed a bug already?
[09:16:48] <rtyler> or pinged the list?
[09:17:36] <ppawel> no I've just noticed that myself
[09:17:46] <ppawel> thought there is a hardware problem with slaves but no
[11:24:21] *** khmarbaise has joined #hudson
[11:26:41] *** akostadinov has joined #hudson
[12:00:35] *** akostadinov has quit IRC
[13:39:15] *** resmo has joined #hudson
[13:51:19] *** resmo has quit IRC
[14:14:09] *** khmarbaise has quit IRC
[14:15:25] *** akostadinov has joined #hudson
[14:25:22] *** khmarbaise has joined #hudson
[15:39:41] *** ppawel has quit IRC
[16:11:45] *** khmarbaise_ has joined #hudson
[16:27:58] *** khmarbaise has quit IRC
[16:35:10] *** ppawel has joined #hudson
[17:07:38] *** akostadinov has quit IRC
[17:54:25] *** jieryn has quit IRC
[20:00:03] *** jumpkick has joined #hudson
[21:25:32] <jumpkick> hmm... anyone use the scp plug-in succesfully?
[21:26:15] <bobo_> i belive so
[21:27:02] <jumpkick> is "/var/cache/pbuilder/result/**" a valid ANT Glob?
[21:27:23] <bobo_> hm, no
[21:27:33] <bobo_> add a /* and i think it is
[21:27:50] <bobo_> or hm
[21:28:13] <bobo_> im not sure, im to tired =(
[21:28:44] <jumpkick> it seems I get Expecting Ant GLOB pattern, but saw '/var/cache/pbuilder/result/*' for that one too
[21:28:46] <jumpkick> :(
[21:29:01] <bobo_> is that for the source? or destination?
[21:29:27] <jumpkick> Files to upload: Source
[21:29:31] <bobo_> ok
[21:29:55] <jumpkick> maybe it has to be based inside the workspace?
[21:29:59] <bobo_> i have **/*.war i belive as source
[21:30:16] <bobo_> it looks everywhere inside the workspace for something matching your pattern
[21:30:48] <jumpkick> okay...  that might be my problem, pbuilder puts it somewhere else outside the workspace by default
[21:31:31] <bobo_> =(
[21:32:18] <jumpkick> pbuilder should be fixable...
[21:33:08] <jumpkick> its just annoying that the example here in the screenshot: http://wiki.hudson-ci.org/display/HUDSON/SCP+plugin
[21:33:16] <jumpkick> is not in the workspace
[21:33:24] <jumpkick> actually nevermind
[21:33:26] <jumpkick> it is
[21:33:28] <jumpkick> duh
[21:33:40] <bobo_> :-)
[21:33:41] <jumpkick> helps if I look at the right screenshot
[21:37:45] <jumpkick> while I wait for my build to try that again, I need to read up on master/slave stuff
[21:38:25] <jumpkick> I'm going to try to set-up builds for all the arches for Mixxx DJ software and push the builds up to mixxx.org
[21:38:51] <jumpkick> I must say hudson is very nice indeed
[21:39:01] <bobo_> yeh, its almost to easy
[21:40:51] <bobo_> http://wiki.hudson-ci.org/display/HUDSON/Swarm+Plugin thats probably what you want to use if your gonna have slaves
[21:42:36] <jumpkick> Actually I was looking a build publisher
[21:42:43] <jumpkick> http://wiki.hudson-ci.org/display/HUDSON/Build+Publisher+Plugin
[21:42:55] <jumpkick> the slaves will be at different peoples houses
[21:43:19] <jumpkick> UDP ain't gonna do it
[21:43:37] <bobo_> ok, then i recomend adding slaves to your master. and then the other people can use the jnlp slave thing
[21:43:53] <bobo_> i use that at work and it works ok
[21:44:22] <bobo_> then you still dont have to do much configuration on the slaves
[21:45:08] <bobo_> but now its time for sleep!
[21:46:04] <jumpkick> might not go that route cause I would want them to be independant
[21:46:24] <jumpkick> maybe not slaves but hudsons publishing up to a master to act as a dashboard
[21:46:38] <jumpkick> thanks for your help bobo_
[21:47:27] <jumpkick> scp worked, yay!
[22:08:27] *** BigAllan has joined #hudson
[22:35:25] <calculus> the swarm slaves has an option to specify the master, so the UDP limit is avoided
[22:37:35] <calculus> it works pretty nicely here, and makes it so I don't have to pre-specify the slaves
[22:46:11] *** khmarbaise_ has quit IRC
[23:05:51] <jumpkick> calculus: that's cool, I think I'm going to go the private hudson -> public hudson route...  that avoids all the problems with firewalls and potential if the public is compromised
[23:06:11] <jumpkick> potential problem
[23:06:41] <jumpkick> I got publishing working
[23:07:03] <jumpkick> now I'm just looking into security...   looks like PAM on Linux isn't going to be a very easy thing to setup
[23:17:16] <jumpkick> er...
[23:17:48] <jumpkick> kohsuke, if you get this, would you consider changing the .deb build deps to
[23:17:52] <jumpkick> Depends: daemon, adduser, java-virtual-machine, default-jre | default-jre-headless
[23:18:26] <jumpkick> java2-runtime sucks in all kinds of stuff that is not needed to run hudson
[23:19:12] <jumpkick> (i.e. pulse-audio, x.org stuff....  which are needed for UI based java)
[23:28:38] *** BigAllan has quit IRC
[23:36:49] <calculus> debian needs USE flags like gentoo!
[23:36:51] * calculus runs
[23:42:05] *** jieryn-w has joined #hudson

top