00:00:01  * ircretaryquit (Remote host closed the connection)
00:00:09  * ircretaryjoined
00:10:00  * mikealquit (Quit: Leaving.)
00:10:27  <isaacs>asdf12: it's usually better form to do server.close(), and then the test also prevents regressions from leaving things open
00:11:00  <isaacs>asdf12: setting a flag on the first request and then closing after the second request, is pretty ok.
00:12:45  <wolfeidau>TooTallNate: It is still running 0.8.5 node-gyp, is this something to do with me running 0.8.x node?
00:13:29  * indexzerojoined
00:17:41  <TooTallNate>wolfeidau: what's the problem?
00:17:49  <sblom>What causes uv_tcp_endgame() to get called?
00:17:59  <asdf12>isaacs: ok i just wanted to make the test less about recursively calling a http request, because it's not needed to simulate the error
00:18:01  <sblom>(Like, in general.)
00:18:17  <asdf12>isaacs: it was just so the test didn't call an external site
00:18:55  <wolfeidau>TooTallNate: For some reason even after running npm remove node-gyp -g && rm -r node_modules then npm install node-gyp -g I am still getting gyp ERR! node-gyp -v v0.8.5
00:19:26  <TooTallNate>wolfeidau: https://github.com/TooTallNate/node-gyp/wiki/Updating-npm's-bundled-node-gyp
00:20:18  <tjfontaine>sblom: looks like when a cluster is shutting down it's going through and cleaning up all the sockets?
00:20:35  <wolfeidau>TooTallNate: OK new error,
00:20:57  <tjfontaine>sblom: that was just my inital impression looking over it
00:21:12  <asdf12>isaacs: isn't this a bug? after calling server.close() the last http.request hangs, should the socket know the server closed the connection?
00:21:25  <asdf12>shouldn't*
00:21:32  <wolfeidau>TooTallNate: https://gist.github.com/wolfeidau/5121655
00:22:03  <wolfeidau>TooTallNate: BTW that tip worked a treat, now that I am using 0.9.0 it is having an issue forking something by the looks
00:22:10  <TooTallNate>wolfeidau: ok, you know… i'm getting that too on solaris… we gotta figure that out…
00:22:55  <tjfontaine>sblom: http://stackoverflow.com/a/14570411
00:22:57  <wolfeidau>TooTallNate: haha yeah I will dig into it now
00:23:11  <TooTallNate>wolfeidau: it's really weird though… i think it's just trying to spawn python
00:23:17  <TooTallNate>not sure why that wouldn't work…
00:23:30  <wolfeidau>TooTallNate: I will dtrace what commands it is running!
00:23:41  <TooTallNate>wolfeidau: <3
00:23:42  <tjfontaine>sblom: so if ther were pending requests when the loop was wanting to shutdown?
00:25:38  <sblom>tjfontaine: Okay. That helps a little.
00:25:54  <sblom>At least I think I understand what _should_ be happening here.
00:25:57  <wolfeidau>TooTallNate: Gist is updated https://gist.github.com/wolfeidau/5121655 note the full command line is clipped, going to look at that last one and see what is up with the ../ stuff
00:26:11  <tjfontaine>sblom: unfortunately that code path is only on winders it seems :)
00:26:23  * kazuponjoined
00:26:55  <TooTallNate>wolfeidau: i think that's ok, that's just isaacs doing path stuff
00:27:18  <TooTallNate>wolfeidau: but so it doesn't look like you're capturing the command that's failing
00:27:37  <wolfeidau>TooTallNate: OK I need to tweak my dtrace :) as it is only getting success :P
00:31:30  * kazuponquit (Ping timeout: 260 seconds)
00:32:47  <TooTallNate>wolfeidau: it works fine on osx and linux, so this is weird…
00:33:55  <tjfontaine>whats up?
00:37:29  <wolfeidau>TooTallNate: Yeah gotta change to tracing the execvp cause that is what is failing..
00:37:49  <TooTallNate>tjfontaine: talking to me?
00:38:00  <tjfontaine>both of you, what's going on? :)
00:38:11  <TooTallNate>tjfontaine: deep in build-system hell :)
00:38:18  <TooTallNate>tjfontaine: but you should be familiar with that by now :)
00:38:25  <tjfontaine>too true :/
00:38:54  <TooTallNate>tjfontaine: actually, we're in the process of making node-gyp *not* build "thin" static archives by default…
00:39:01  <TooTallNate>since it seems to cause problems on older platforms
00:39:13  <tjfontaine>ah
00:39:30  <tjfontaine>and something gyp is generating is not working on smartos?
00:40:15  <TooTallNate>tjfontaine: well… after a recent change (i have a commit in mind), simply cp.exec() ing "python" is giving us https://gist.github.com/wolfeidau/5121655
00:40:31  <TooTallNate>which is, well, strange
00:40:38  <TooTallNate>since python *is* there :)
00:42:05  <tjfontaine>heh, ya needless to say I spawn(python) without issue on on smartos :)
00:42:20  <tjfontaine>well have fun figuring that out :)
00:43:18  <wolfeidau>TooTallNate: I found some exec calls returning -1 just need to remember how to convert the char* of the syscall to human
00:47:53  <isaacs>asdf12: server.close() just stops accepting new connections
00:48:00  <isaacs>asdf12: it doesn't close all existing connections
00:49:07  * bradleymeckjoined
00:49:57  * EhevuTovquit (Quit: This computer has gone to sleep)
00:50:39  <sblom>Alright--I updated https://github.com/joyent/node/issues/4959 with my debugging notes. I feel like we know what's broken at this point, but I need a libuv-win32 architect to help me figure out what fix to apply where.
00:51:55  <isaacs>sblom: awesome
00:52:23  * bradleymeckquit (Client Quit)
00:54:19  <sblom>isaacs: can you skim over the notes and see if they're coherent enough for piscisaureus_ to have some hope of being able to take them and turn them into a fix when he wakes up in the morning?
00:54:51  <sblom>(I made some edits up until about 2 minutes ago, but it's stable now)
00:55:08  * trevnorrisquit (Quit: Leaving)
00:56:50  <asdf12>https://gist.github.com/anonymous/5121856 isn't this a bug?
00:57:23  <asdf12>the socket hangs, but shouldn't closing the server automatically end the open socket?
01:07:52  <isaacs>asdf12: no.
01:07:55  <isaacs>asdf12: see above
01:08:02  <isaacs>asdf12: server.close() != socket.close()
01:08:10  <isaacs>asdf12: server.close() just stops listening
01:08:23  <isaacs>asdf12: but existing connections still need to be finished up
01:08:33  <asdf12>ah sorry, was detached from screen
01:08:38  <isaacs>asdf12: np :)
01:09:04  <isaacs>asdf12: that's why i am always /away, so that i can walk away and then when i come back, view the awaylog
01:10:14  <asdf12>oh yea, i wrote this client, and never got around to making it log while away, am just waiting for some time to rewrite it in erlang
01:12:32  <isaacs>sblom: this is a great writeup
01:12:44  <isaacs>asdf12: you should just use irssi
01:12:48  <isaacs>asdf12: it's pretty good
01:12:58  <asdf12>oh it's a web client
01:13:32  <asdf12>it's better this way, so i can log in anywhere and not rely on ssh (since i'd want to keep the client running)
01:19:02  <isaacs>sblom: how do you feel about the other stuff in windows-test-fixing?
01:20:16  <isaacs>sblom: https://github.com/isaacs/node/compare/joyent:master...windows-test-fixing
01:22:04  * AvianFlujoined
01:24:19  <TooTallNate>wolfeidau: any luck?
01:25:28  * hzjoined
01:28:55  <wolfeidau>TooTallNate: Only hint at this stage is python is not being run as far as I can tell, it is failing before that point
01:30:02  <wolfeidau>For some reason that ../ command is returning -1
01:31:42  <TooTallNate>wolfeidau: that's just node-gyp
01:31:49  <TooTallNate>wolfeidau: i think i figured it out though…
01:32:17  <wolfeidau>TooTallNate: Is it inside node-gyp ?
01:32:24  <TooTallNate>but of course…
01:35:19  <TooTallNate>wolfeidau: try v0.9.1
01:35:27  <wolfeidau>Only issue with dtrace is you can go off on a journey of discovery and disapear for hours
01:36:11  <TooTallNate>hahaha
01:36:18  <TooTallNate>i would like to learn me some dtrace
01:37:40  <wolfeidau>TooTallNate: We have a winner
01:38:13  <wolfeidau>TooTallNate: If you plan to dive deeper into build systems it would be a wise move
01:38:48  <wolfeidau>TooTallNate: In short IT WORKS! without change I can now build rvagg leveldown
01:38:52  <TooTallNate>wolfeidau: cool, well we'll let people upgrade… see if anybody reports problems with non-thin archives
01:42:26  <wolfeidau>TooTallNate: Awesome, hopefully I was of some assistance :P
01:42:35  <TooTallNate>wolfeidau: yes definitely! thanks
01:44:40  <wolfeidau>TooTallNate: Happy to help, I just learnt more about dtrace which is a bonus
01:44:49  <isaacs>TooTallNate: you up for a bit of zlib reviewing by any chance?
01:44:57  <isaacs>TooTallNate: (it's easy, and you're the best Transform user atm)
01:45:15  <TooTallNate>isaacs: ya i was taking a look ealier… let me do a thourough one
01:45:21  <isaacs>kewl
01:45:23  <isaacs>thanks :)
01:47:15  * sblomquit
01:47:36  <wolfeidau>TooTallNate: This is the dtrace script i used to watch all the exec system calls btw, slightly tweaked old script by Brendan Gregg https://gist.github.com/wolfeidau/5122096 chmod +x it and run it in a root shell then in another shell run node-gyp
01:57:40  * dapquit (Quit: Leaving.)
02:15:30  * hzquit
02:20:58  * pooyaquit (Read error: Connection reset by peer)
02:31:34  <MI6>joyent/node: isaacs windows-test-fixing * b98ccd4 : test: Fail faster in simple/test-cluster-bind-twice-v2 Crashing on windo - http://git.io/IgnvCw
02:42:21  * indexzeroquit (Quit: indexzero)
02:44:58  * indexzerojoined
02:57:24  <MI6>joyent/node: isaacs master * 6e34dfd : test: Fail faster in simple/test-cluster-bind-twice-v2 Crashing on windo (+9 more commits) - http://git.io/NBz2aQ
03:00:12  <MI6>joyent/node: isaacs master * 061a7dd : Merge remote-tracking branch 'ry/v0.8' into master Conflicts: AUTHORS (+7 more commits) - http://git.io/yVMZXw
03:09:42  * TooTallNatequit (Quit: Computer has gone to sleep.)
03:09:56  <MI6>nodejs-master: #88 UNSTABLE osx-x64 (1/555) windows-ia32 (5/555) windows-x64 (5/555) osx-ia32 (1/555) http://jenkins.nodejs.org/job/nodejs-master/88/
03:21:07  * c4milojoined
03:22:26  <MI6>nodejs-master: #89 UNSTABLE windows-ia32 (6/555) windows-x64 (5/555) http://jenkins.nodejs.org/job/nodejs-master/89/
03:31:39  * c4miloquit (Remote host closed the connection)
03:35:09  * c4milojoined
04:01:02  * brsonquit (Quit: leaving)
04:05:08  * qmx|awaychanged nick to qmx
04:06:15  * qmxchanged nick to qmx|away
04:19:00  * TooTallNatejoined
04:21:49  * piscisaureus__joined
04:21:58  * piscisaureus__changed nick to piscisaureus_
04:25:38  <isaacs>piscisaureus_: wb
04:25:59  <isaacs>piscisaureus_: must've been a good birthday party, if you're just getting back at 4:30 in the morning
04:27:32  <MI6>joyent/node: isaacs master * 7c9ff8e : lint (+1 more commits) - http://git.io/30MFvQ
04:40:08  <MI6>nodejs-master: #90 UNSTABLE linux-x64 (1/555) windows-ia32 (6/555) windows-x64 (6/555) http://jenkins.nodejs.org/job/nodejs-master/90/
04:48:59  * brsonjoined
05:06:49  * c4milo_joined
05:07:10  * c4miloquit (Read error: Connection reset by peer)
05:52:46  <MI6>joyent/node: isaacs master * 43c1830 : Now working on 0.11.0 - http://git.io/TAex2g
05:53:48  <isaacs>weird, no MI6 for v0.10 branch
06:05:32  <MI6>nodejs-master: #91 UNSTABLE osx-x64 (1/555) windows-ia32 (5/555) windows-x64 (5/555) http://jenkins.nodejs.org/job/nodejs-master/91/
06:05:48  * c4milo_quit (Remote host closed the connection)
06:07:36  * TooTallNatequit (Quit: ["Textual IRC Client: www.textualapp.com"])
06:10:08  <isaacs>tjfontaine: <3 having the test names in the jenkins output.
06:20:48  * AvianFluquit (Remote host closed the connection)
06:38:04  <tjfontaine>isaacs: so much more helpful right?
06:41:16  <tjfontaine>got them to appear in the test overview but it broke actually viewing a configurations results, and the urls they generated were awful
06:45:04  * mikealjoined
06:53:15  * brsonquit (Quit: leaving)
06:55:52  <MI6>nodejs-v0.10: #1 UNSTABLE windows-ia32 (5/555) windows-x64 (5/555) http://jenkins.nodejs.org/job/nodejs-v0.10/1/
06:55:58  <tjfontaine>weee
07:12:03  * stagasjoined
07:48:57  * rendarjoined
07:51:44  * dominictarrquit (Quit: dominictarr)
07:56:50  * dominictarrjoined
08:40:48  * mmalecki[zzz]changed nick to mmalecki
08:50:49  * benoitcquit (Excess Flood)
08:52:58  * benoitcjoined
09:07:39  * `3rdEdenjoined
09:20:42  * Raltjoined
09:21:21  <Ralt>isaacs: hi
09:21:52  <Ralt>isaacs: on npm issues, do you mind having people "taking care" of old issues? even though they can't close or something.
09:34:28  * felixgejoined
09:34:51  * jguerreroquit (Quit: jguerrero)
09:43:54  * `3rdEdenquit (Remote host closed the connection)
10:40:37  * `3rdEdenjoined
10:51:02  * Kakerajoined
10:51:02  <Ralt>is st-luke on irc?
10:59:16  * dominictarrquit (Quit: dominictarr)
11:09:06  * benoitcquit (Excess Flood)
11:16:00  * benoitcjoined
11:38:25  * `3rdEdenquit (Remote host closed the connection)
11:54:03  * indexzeroquit (Quit: indexzero)
12:09:28  * benoitcquit (Excess Flood)
12:12:34  * Raltquit (Remote host closed the connection)
12:13:32  * benoitcjoined
12:32:26  * `3rdEdenjoined
13:09:50  * benoitcquit (Excess Flood)
13:12:31  * benoitcjoined
13:34:53  * Raltjoined
13:56:11  * benoitcquit (Excess Flood)
13:56:53  * Raltquit (Remote host closed the connection)
14:00:31  * benoitcjoined
14:19:33  * hzjoined
14:35:22  * hzquit
14:39:05  * piscisaureus_quit (Ping timeout: 260 seconds)
14:39:50  <isaacs>ircretary: Yes, that's lovely. If someone does a really good job of this and is trustworthy, I sometimes make then a contributor.
14:39:50  <ircretary>isaacs: I'm not sure what to do with that command. Ask for help in PM.
14:39:58  <isaacs>ircretary: tell Ralt Yes, that's lovely. If someone does a really good job of this and is trustworthy, I sometimes make then a contributor.
14:39:58  <ircretary>isaacs: I'll be sure to tell ralt
14:52:15  * jguerrerojoined
15:00:32  <MI6>joyent/node: isaacs v0.10 * 80472bc : domain: Fix double-exit on nested domains Minor oversight in fix for #49 - http://git.io/lfEFVw
15:01:18  <indutny>time to reboot to windows :)
15:01:19  <indutny>brb
15:17:14  <MI6>nodejs-v0.10: #2 UNSTABLE windows-ia32 (5/555) osx-ia32 (1/555) windows-x64 (6/555) http://jenkins.nodejs.org/job/nodejs-v0.10/2/
15:18:19  * `3rdEdenquit (Remote host closed the connection)
15:48:53  * `3rdEdenjoined
15:57:15  * `3rdEdenquit (Ping timeout: 260 seconds)
16:17:19  <isaacs>indutny: feel like tracking down a windows bug in libuv?
16:17:30  <indutny>not yet
16:17:33  <indutny>installing visual studio express
16:17:35  * indutnypart
16:17:37  <isaacs>k
16:17:39  * indutnyjoined
16:17:49  <isaacs>probably best to save for piscisaureus anyway :)
16:17:58  <isaacs>i think https://github.com/joyent/node/issues/4959 is the source of our curent windows test failures
16:18:02  <isaacs>most of them, anyway
16:18:18  <tjfontaine>hrh
16:18:21  <tjfontaine>er heh
16:19:18  <isaacs>tjfontaine: bless you
16:19:46  <tjfontaine>indeed, I need to clean the screen now
16:20:17  <MI6>joyent/node: Andreas Madsen v0.10 * 7becf15 : timers: consistent this keyword in setImmediate When calling setImmediat - http://git.io/0TybVg
16:21:16  <tjfontaine>so many brnaches, so little time
16:21:53  <indutny>:)
16:29:26  * stagasquit (Ping timeout: 256 seconds)
16:33:39  <MI6>nodejs-v0.10: #3 UNSTABLE windows-ia32 (7/556) windows-x64 (6/556) osx-x64 (1/556) http://jenkins.nodejs.org/job/nodejs-v0.10/3/
16:35:26  * stagasjoined
16:36:31  <isaacs>hrm. http://jenkins.nodejs.org//job/nodejs-v0.10/3/DESTCPU=x64,label=osx//tapTestReport/test.tap-29/
16:36:36  <isaacs>indeed, it is failing sometimes
16:36:59  <isaacs>bnoordhuis was right
16:37:03  <indutny>yes
16:37:04  <indutny>I told ya
16:37:09  <indutny>bnoordhuis?
16:37:24  <indutny>I found that issue long ago, when I was fixing cluster issues
16:37:30  <indutny>with sockets send list
16:37:46  <indutny>it seems to be some weird osx bug
16:37:53  <indutny>and isn't really reproducible outside test
16:42:44  <isaacs>indutny: for a while i was getting it pretty reliably by running the script 100 times or so
16:42:51  <isaacs>indutny: but then it stopped being reproducible.
16:42:59  <isaacs>so i closed the ticket, and then bnoordhuis opened it again
16:43:35  <isaacs>trying again with this double-free-parser patch.
16:43:47  <isaacs>looks good so far.
16:46:56  <MI6>joyent/node: hc v0.10 * 5757ce4 : http: check if incoming parser has already been freed Fix #4948 This ad - http://git.io/ZW4grw
16:47:02  <isaacs>asdf12: thanks ^
16:48:01  * stagasquit (Ping timeout: 248 seconds)
16:49:09  * piscisaureus_joined
16:49:19  <piscisaureus_>ircretary: notes
16:49:25  <piscisaureus_>isaacs: why is pipelining a bad idea?
16:50:05  <isaacs>piscisaureus_: becuase it's annoying. the only client that actually does it irl (that is, outside of tests) that i've ever seen, is couchdb's replicator, and that's why couchdb's replicator is shit.
16:50:13  <piscisaureus_>hehe
16:50:15  <isaacs>if an error happens, you have no idea why, or where.
16:50:33  <piscisaureus_>you can turn it on in chrome too, though
16:50:49  <isaacs>yeah, but "can turn it on" === "not in real life, only in tests"
16:50:53  <indutny>AFAIK its turned on in opera
16:51:00  <indutny>at least someone told me so
16:51:05  <isaacs>opera gets less usage than chrome's dev options.
16:51:34  <indutny>piscisaureus_: man, after all years
16:51:49  <isaacs>in order for pipeline-style apis to not be shit, you need transaction ids.
16:52:16  <isaacs>and even then, it's unnecessary optimization that isn't worth the complexity overhead.
16:52:16  <indutny>I want to ask you... how do you use this non-unix OS
16:52:47  <isaacs>piscisaureus_: so, anyway... yeah. https://github.com/joyent/node/issues/4959
16:52:47  <indutny>isaacs: SPDY
16:52:50  <isaacs>piscisaureus_: that's the big one.
16:53:03  <isaacs>indutny: i'm not super familiar with spdy
16:53:13  <indutny>spdy = multiplexing
16:53:23  <indutny>a good answer to pipelining
16:53:58  <piscisaureus_>isaacs: yeah, I saw the issue
16:54:20  <isaacs>piscisaureus_: it smells like sblom might have gotten close to the root cause of some of the other failures.
16:54:33  <isaacs>piscisaureus_: it looks like we're sometimes killing pipes and sockets before we should.
16:54:43  <piscisaureus_>that's not good eh :)
16:54:48  <isaacs>not so much
16:54:59  <isaacs>but, they do all have a pretty similar pattern.
16:55:15  <isaacs>things are going along nicely, then you try to call uv_write2 or uv_write, and you get an EINVAL because it's shut down already.
16:55:24  <isaacs>that's the only one that actually blows an assert.
16:55:49  <piscisaureus_>isaacs: do you know which tests make this happen
16:56:14  <isaacs>piscisaureus_: https://github.com/joyent/node/issues?milestone=14&state=open <-- all the ones that say "windows" in the title.
16:56:23  <isaacs>piscisaureus_: basically, all the windows test failures on the v0.10 branch
16:56:32  <isaacs>piscisaureus_: they're all like this.
16:56:42  <isaacs>well, except for the "stdio is async" problem.
16:56:47  <isaacs>but as you know, that's not new.
16:56:57  <isaacs>and i'm ok with punting that out another stable family.
16:57:05  <isaacs>if it's not easy to fix
16:57:37  <piscisaureus_>isaacs: ok let me drill down one bug
16:57:45  <piscisaureus_>isaacs: after that I have other work to do
16:57:53  <piscisaureus_>I'll do the assertion error
16:57:53  <isaacs>kewl
16:57:57  <isaacs>\o/
16:58:10  <indutny>:)
17:02:37  <MI6>nodejs-v0.10: #4 UNSTABLE windows-ia32 (5/557) windows-x64 (5/557) http://jenkins.nodejs.org/job/nodejs-v0.10/4/
17:10:32  * benoitcquit (Excess Flood)
17:12:32  <isaacs>piscisaureus_: should i branch libuv at this point as well?
17:12:37  <isaacs>piscisaureus_: add a v0.10 branch, i mean
17:14:57  <piscisaureus_>isaacs: not yet, do it after the release
17:15:10  <isaacs>ok
17:15:17  <piscisaureus_>isaacs: ben is on holiday this weekend so nothing gets landed anyway unless I do it
17:15:19  <piscisaureus_>and I won't
17:15:46  <isaacs>piscisaureus_: well, i mean, i can just create a branch, but not land any commits on it
17:16:31  * benoitcjoined
17:16:41  <piscisaureus_>isaacs: I would say - do it immediately before or after the release. It's just inconvenient to do it now.
17:16:51  <isaacs>k
17:16:55  <piscisaureus_>ok, me /& trying to figure out the bug
17:17:35  <piscisaureus_>BTW - who turned on asserts in the release branch in teh first place... ?
17:18:33  <tjfontaine>how did we do that? I blame gyp
17:19:08  <isaacs>piscisaureus_: yeah, ti's odd. *some* asserts are definitely not there in the release build.
17:19:17  <isaacs>piscisaureus_: also: i'm going to lcase NPM in the MSI.
17:19:19  <isaacs>that's a typo
17:19:35  <piscisaureus_>The asserts used to be off on windows
17:19:47  <piscisaureus_>Although I know ben prefers them on so he has them on on unix
17:19:59  <isaacs>generally, i prefer them to be on
17:20:00  <piscisaureus_>It's probably best to have them on anyway
17:20:05  <isaacs>they are cheap, and they catch problems.
17:20:17  <piscisaureus_>well - they might be cheap but who knows :)
17:20:24  <piscisaureus_>I think I have a couple expensive ones there as well
17:21:04  <piscisaureus_>isaacs: lcase npm all that you want.
17:21:11  <isaacs>yeah
17:22:15  * hzjoined
17:22:20  <MI6>joyent/node: isaacs v0.10 * 12713c3 : win/msi: Fix typos (+1 more commits) - http://git.io/CKmrxQ
17:25:29  <MI6>joyent/libuv: Bert Belder master * f59dc22 : win: fix potential HANDLE corruption due to incorrect cast Closes #738. - http://git.io/9prfZg
17:29:49  <MI6>libuv-master: #45 UNSTABLE linux (8/183) osx (7/183) smartos (10/183) http://jenkins.nodejs.org/job/libuv-master/45/
17:33:18  * AvianFlujoined
17:37:10  <MI6>nodejs-v0.10: #5 UNSTABLE windows-ia32 (5/557) osx-ia32 (1/557) windows-x64 (5/557) osx-x64 (1/557) http://jenkins.nodejs.org/job/nodejs-v0.10/5/
17:46:30  * piscisaureus_quit (Ping timeout: 250 seconds)
17:55:51  * hzquit
18:02:47  * hzjoined
18:10:11  * c4milojoined
18:16:19  * philips-joined
18:17:09  * abraxasjoined
18:18:27  * philips-quit (Excess Flood)
18:21:15  * philips_quit (Quit: http://ifup.org)
18:21:16  * abraxasquit (Ping timeout: 245 seconds)
18:24:26  * philips_joined
18:24:27  * philips_quit (Excess Flood)
18:24:42  * AvianFluquit (Remote host closed the connection)
18:28:41  * piscisaureus_joined
18:29:19  * philips_joined
18:31:49  <indutny>piscisaureus_: how can I run tests on windows?
18:31:56  <tjfontaine>Release\run-tests
18:32:09  <indutny>thanks
18:32:09  <tjfontaine>oh that's libuv
18:32:14  <indutny>ah
18:32:15  <indutny>I gues
18:32:21  <indutny>python test/run-tests
18:32:22  <tjfontaine>for node you python tools\test.py
18:32:48  <indutny>yep
18:32:53  <tjfontaine>that will run them all, `make test` is equivalent to python tools\test.py simple message
18:32:54  <indutny>thanks
18:33:40  <tjfontaine>yw
18:33:55  <indutny>so many failing tests
18:34:04  <tjfontaine>on what branch?
18:34:07  <indutny>master
18:34:24  <tjfontaine>you're seeing more than 5?
18:34:49  <indutny>yes
18:34:50  <indutny>9
18:34:57  <tjfontaine>in the ideal setup you should only have http://jenkins.nodejs.org/job/nodejs-master/DESTCPU=ia32,label=windows/lastCompletedBuild/tapTestReport/
18:34:58  <indutny>and that's only 48%
18:35:03  <indutny>ah
18:35:04  <indutny>no curl
18:35:08  <indutny>nvm
18:35:11  <tjfontaine>ya put git in your path :)
18:35:16  <tjfontaine>*msysgit
18:37:05  <tjfontaine>bbl errands
18:38:58  * dapjoined
18:40:23  * philips_quit (Remote host closed the connection)
18:44:40  * `3rdEdenjoined
18:54:10  * philipsjoined
18:58:33  * mikealquit (Quit: Leaving.)
19:15:50  * c4miloquit (Remote host closed the connection)
19:26:02  <MI6>joyent/libuv: Bert Belder master * 5462dab : win/tcp: don't enable iocp sync bypass when iocp emulation is used When - http://git.io/aWTPPg
19:28:21  <MI6>libuv-master: #46 UNSTABLE linux (11/183) osx (10/183) smartos (12/183) http://jenkins.nodejs.org/job/libuv-master/46/
19:29:57  <piscisaureus_>indutny: vcbuild release x64 test-all
19:30:22  <indutny>thanks
19:30:26  <indutny>I'm on osx now :)
19:32:35  * jguerreroquit (Quit: jguerrero)
19:33:13  * TooTallNatejoined
19:36:57  * mikealjoined
19:47:36  * mikealquit (Quit: Leaving.)
19:51:23  * c4milojoined
19:56:22  <isaacs>piscisaureus_: !!! testing now
19:56:47  <piscisaureus_>isaacs: it's probaby a fix for this failure only though...
19:56:52  <isaacs>oh, o
19:56:53  <isaacs>k
19:57:02  <piscisaureus_>i think 4966 is separate
19:59:20  <MI6>joyent/node: isaacs windows-test-fixing * 2098c4b : uv: Upgrade to 5462dab8 - http://git.io/7bYzcg
20:00:40  <isaacs>still, this one is more frightening anyway
20:00:57  * tjfontainetriggers oneoff
20:01:09  <isaacs>thanks!
20:01:13  * dapquit (Quit: Leaving.)
20:07:53  * AvianFlujoined
20:21:48  <isaacs>piscisaureus_: lgtm
20:21:58  <isaacs>bind-twice still getting EINVAL
20:21:58  <piscisaureus_>cool
20:22:06  <piscisaureus_>yeah - next
20:22:06  <isaacs>piscisaureus_: and debugger-client still getting ECONNRESET
20:25:22  * c4miloquit (Remote host closed the connection)
20:28:07  <piscisaureus_>isaacs: I know I know but I can only fix one bug at a time
20:28:17  <piscisaureus_>or, maybe more, but 1 at a time tends to work better
20:28:23  <isaacs>:)
20:28:32  <isaacs>piscisaureus_: you should do what i do and just have other people fix them for you.
20:28:41  <isaacs>piscisaureus_: not as efficient, but it scales better.
20:28:45  <piscisaureus_>isaacs: hah
20:29:21  <piscisaureus_>isaacs: hah
20:29:48  <piscisaureus_>isaacs: true - although with only isaacs there would be no node
20:29:56  <isaacs>piscisaureus_: yes.
20:29:59  <piscisaureus_>or - nothing at all even
20:30:02  <isaacs>piscisaureus_: i'm saying you should get someone ELSE to do it
20:30:15  <isaacs>piscisaureus_: not me, i'm already busy
20:30:18  <isaacs>:)
20:30:36  <piscisaureus_>isaacs: pleeeease ::puppyeyes::
20:31:24  * mikealjoined
20:33:31  * c4milojoined
20:42:05  * mikealquit (Quit: Leaving.)
20:42:41  * stagasjoined
20:44:56  <isaacs>piscisaureus_: we can trade. you write the 0.10 blog post, i'll fix windows ;P
20:45:11  <piscisaureus_>haha
20:45:12  <piscisaureus_>hmm
20:45:22  <piscisaureus_>isaacs: no I prefer to be a minion
20:45:27  <tjfontaine>heh
20:45:30  <piscisaureus_>isaacs: minion > concierge
20:45:44  <isaacs>hahaha
20:45:49  <isaacs>minions are nice.
20:45:53  <isaacs>except the plotting kinds.
20:45:56  <isaacs>fuck those guys
20:51:33  * c4miloquit (Remote host closed the connection)
21:03:01  * TooTallNatequit (Quit: Computer has gone to sleep.)
21:18:15  <indutny>:)
21:18:49  * indutnytopic: bert - the minion "liberal utopian vacation ~ http://logs.libuv.org/libuv ~ http://groups.google.com/group/libuv"
21:28:46  * `3rdEdenquit (Remote host closed the connection)
21:30:54  * c4milojoined
21:37:53  * c4miloquit (Remote host closed the connection)
21:52:45  * moneypennyjoined
21:53:05  * moneypennyquit (Client Quit)
22:00:04  * isaacsheading out ot get some food and enjoy the lovely day a little bit.
22:00:17  <isaacs>it's springtime in california! it's so warm and lovely here! everyone should be here!
22:00:28  <tjfontaine>agreed.
22:08:33  <indutny>shit
22:08:35  <indutny>spdy works on 0.10
22:08:39  <indutny>and broke on 0.8
22:08:41  <indutny>:)
22:09:42  * isaacs_mobilejoined
22:09:55  <piscisaureus_>man this triple-stacked child process issue is hell to debug :(
22:10:23  <indutny>haha
22:10:26  <indutny>I believe you
22:13:41  <isaacs>indutny: it's a feature!
22:13:46  <isaacs>indutny: get people to update!
22:13:46  <indutny>nooo
22:13:50  <indutny>0.8 crashes on it
22:13:51  <isaacs>just bump the major version.
22:13:53  <indutny>with assertion error
22:13:59  <isaacs>or use require('readable-stream')
22:14:07  <indutny>meh :)
22:14:09  <indutny>oh
22:14:14  <indutny>that might be a good idea
22:14:15  <isaacs>ok, gotta run
22:14:15  <indutny>but
22:14:16  <indutny>meh
22:16:04  <indutny>oh gosh
22:16:13  <indutny>calling .onend() might crash node
22:16:19  <indutny>if done in incorrect place
22:16:47  <indutny>ok
22:16:49  <indutny>time to sleep
22:16:52  <indutny>c ya
22:26:36  * isaacs_mobilequit (Ping timeout: 240 seconds)
22:27:34  * mmaleckichanged nick to mmalecki[zzz]
22:34:03  * rendarquit
22:45:49  * stagasquit (Ping timeout: 245 seconds)
22:47:15  * stagasjoined
22:51:42  <Kakera>I'm trying to compile a simple project that links to libuv using GYP with mingw
22:51:57  <Kakera>mkdir: cannot create directory `out/Debug/obj.target/libuv/libuv\\src" out': No such file or directory
22:52:05  <Kakera>libuv itself compiles fine
22:52:16  <Kakera>any suggestions?
22:52:47  <tjfontaine>how the f did you convince it to mix paths like that
22:53:11  <Kakera>I'd like to know too
22:53:45  <Kakera>one min I'll create a smaller test case
22:54:01  <Kakera>also what's with the `"'
22:57:39  <Kakera>receiving objects at 13KiB/s -_-
23:05:01  <Kakera>so here's my steps
23:05:10  <Kakera>1) build libuv (successfully)
23:05:41  <Kakera>2) generate a makefile for my project with: gyp --depth=. -Dlibrary=static_library -fmake
23:05:50  <Kakera>3) run make
23:06:28  <Kakera>and I get this http://pastebin.com/fV1vthkS
23:06:35  <tjfontaine>I'm sure gyp is all kinds of confused by the fact that you're generating a makefile for mingw
23:07:24  <tjfontaine>what does your projects gyp look like?
23:07:43  <tjfontaine>and why do you need to pass in --depth?
23:07:56  <Kakera>http://pastebin.com/eH3068NE
23:08:10  <tjfontaine>oh I guess we do that as well
23:08:25  <Kakera>because otherwise I get: gyp: Could not automatically locate src directory. This isa temporary Chromium feature that will be removed. Use--depth as a workaround.
23:08:32  <tjfontaine>right
23:09:12  * hzquit (Ping timeout: 264 seconds)
23:13:31  * hzjoined
23:17:01  * TooTallNatejoined
23:20:06  * piscisaureus_quit (Excess Flood)
23:21:32  * AvianFluquit (Remote host closed the connection)
23:22:30  <Kakera>so is linking to libuv.a directly my best bet?
23:22:47  * piscisaureus_joined
23:28:09  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
23:55:00  * dapjoined