00:00:00  * ircretaryquit (Remote host closed the connection)
00:00:02  * euoia_quit (Ping timeout: 260 seconds)
00:00:09  * ircretaryjoined
00:02:02  * avalanche123joined
00:06:09  * mikealjoined
00:14:57  * zz_karupachanged nick to karupa
00:17:22  * dshaw_quit (Quit: Leaving.)
00:17:29  * brsonquit (Quit: leaving)
00:24:42  * euoia_joined
00:26:00  * dshaw_joined
00:29:31  * Soarezchanged nick to Soarez|afk
00:32:36  * dap_1quit (Quit: Leaving.)
00:33:07  * abraxasjoined
00:34:43  * dsantiagojoined
00:37:44  * abraxasquit (Ping timeout: 250 seconds)
00:40:17  * dshaw_quit (Quit: Leaving.)
00:46:50  * quijotejoined
00:51:04  * quijotequit (Ping timeout: 250 seconds)
00:57:46  * kellabytejoined
01:01:38  * euoia_quit (Ping timeout: 260 seconds)
01:02:32  * inolenquit (Quit: Leaving.)
01:04:25  * avalanche123quit (Remote host closed the connection)
01:04:51  * avalanche123joined
01:09:25  * avalanche123quit (Ping timeout: 260 seconds)
01:09:56  * jgiquit (Quit: jgi)
01:12:51  * kazuponjoined
01:23:47  * avalanche123joined
01:27:44  * bradleymeck_joined
01:29:06  * kazuponquit (Remote host closed the connection)
01:31:07  * bradleymeckquit (Ping timeout: 272 seconds)
01:31:08  * bradleymeck_changed nick to bradleymeck
01:35:15  * avalanche123quit (Remote host closed the connection)
01:35:30  * rmgquit (Remote host closed the connection)
01:37:53  * kazuponjoined
01:40:07  * rosskquit
01:47:27  * quijotejoined
01:52:16  * quijotequit (Ping timeout: 260 seconds)
01:53:33  * a3fquit (Ping timeout: 240 seconds)
02:07:38  * euoia_joined
02:07:52  * cjihrigpart
02:08:14  * cjihrigjoined
02:09:39  * abraxasjoined
02:27:48  * andrehjrquit (Quit: Computer has gone to sleep.)
02:34:59  * c4miloquit (Remote host closed the connection)
02:43:25  * sh1mmerquit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
02:48:21  * quijotejoined
02:49:16  * swajjoined
02:49:21  * swajquit (Client Quit)
02:49:29  * swajjoined
02:50:27  * Left_Turnquit (Remote host closed the connection)
02:52:51  * inolenjoined
02:52:56  * quijotequit (Ping timeout: 244 seconds)
02:54:54  * TooTallNatequit (Quit: Computer has gone to sleep.)
02:55:10  * cjihrigquit (Quit: Leaving.)
02:57:00  * rmgjoined
03:05:46  * janjongboomjoined
03:11:52  * bradleymeckquit (Remote host closed the connection)
03:19:01  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
03:24:42  * rmgquit (Remote host closed the connection)
03:25:17  * rmgjoined
03:27:14  * euoia_quit (Ping timeout: 260 seconds)
03:29:58  * rmgquit (Ping timeout: 264 seconds)
03:38:46  * brsonjoined
03:43:45  * rmgjoined
03:44:08  * sh1mmerjoined
03:45:11  * brsonquit (Quit: leaving)
03:48:12  * sh1mmerquit (Ping timeout: 245 seconds)
03:49:02  * quijotejoined
03:54:04  * quijotequit (Ping timeout: 272 seconds)
04:16:53  * kazuponquit (Remote host closed the connection)
04:29:16  * avalanche123joined
04:36:22  * kellabytequit (Quit: Connection closed for inactivity)
04:48:03  * avalanche123quit (Remote host closed the connection)
04:49:47  * quijotejoined
04:51:15  * c4milojoined
04:54:17  * quijotequit (Ping timeout: 255 seconds)
04:55:37  * c4miloquit (Ping timeout: 245 seconds)
04:56:44  * kazuponjoined
04:57:34  * rvaggquit (Quit: ta ta)
04:59:11  * avalanche123joined
05:03:36  * Ralithjoined
05:12:08  * kenperkinsquit (Quit: Computer has gone to sleep.)
05:12:30  * SquirrelCZECHjoined
05:23:57  * rmg_joined
05:24:36  * rmgquit (Read error: Connection reset by peer)
05:25:03  * tumdedumquit (Ping timeout: 240 seconds)
05:25:11  * avalanche123quit (Remote host closed the connection)
05:25:55  * tumdedumjoined
05:35:48  * avalanche123joined
05:37:19  * avalanche123quit (Remote host closed the connection)
05:37:46  * avalanche123joined
05:42:34  * avalanche123quit (Ping timeout: 260 seconds)
05:44:31  * jeayejoined
05:46:04  * Damn3d_quit (Ping timeout: 260 seconds)
05:47:34  * Damn3djoined
05:50:42  * quijotejoined
05:55:24  * quijotequit (Ping timeout: 260 seconds)
05:56:11  * andrehjrjoined
05:57:53  * rendarjoined
06:13:38  * andrehjrquit (Quit: Computer has gone to sleep.)
06:19:49  <dainis>tjfontaine, about setTimer firing sooner than expected, v0.10 branch works as expected, timer is fired after 500-502ms
06:21:13  * avalanche123joined
06:27:02  * bajtosjoined
06:28:14  * TooTallNatejoined
06:37:13  * rmgjoined
06:37:33  * rmg_quit (Read error: Connection reset by peer)
06:39:32  * c4milojoined
06:43:54  * c4miloquit (Ping timeout: 250 seconds)
06:49:31  * TooTallNatequit (Quit: Computer has gone to sleep.)
06:51:18  * quijotejoined
06:52:39  * jgijoined
06:53:05  * quijotequit (Read error: Connection reset by peer)
06:53:26  * quijote_joined
06:58:17  * inolenquit (Quit: Leaving.)
07:09:07  * a_lequit (Remote host closed the connection)
07:10:13  * rmgquit (Read error: Connection reset by peer)
07:10:25  * rmgjoined
07:13:43  * inolenjoined
07:15:24  * avalanche123quit (Remote host closed the connection)
07:15:51  * avalanche123joined
07:17:33  * a_lejoined
07:20:32  * avalanche123quit (Ping timeout: 255 seconds)
07:24:30  * guybrushquit (Excess Flood)
07:25:06  * guybrushjoined
07:25:24  * janjongboomjoined
07:29:56  * bajtosquit (Quit: bajtos)
07:30:06  * jgiquit (Quit: jgi)
07:30:38  * bajtosjoined
07:30:52  * dignifiedquirejoined
07:32:14  * Left_Turnjoined
07:47:07  * a_lequit (Remote host closed the connection)
07:49:21  * Left_Turnquit (Ping timeout: 260 seconds)
07:58:10  <jeaye>I'm building a product with libuv that uses a custom version instead of the system-installed/provided version of libuv. Compilation goes well, but, when I run, the binary looks for the libuv shared object in only the system directories. I can resolve this by describing LD_LIBRARY_PATH, but is there a way I can instead do it at compile-time?
08:00:07  <indutny>jeaye: I don't think that you could do it this way
08:00:39  <txdv>just compile it a static library into the app
08:00:43  <jeaye>Why does libuv need a shared object?
08:00:44  <txdv>it as a
08:01:00  <txdv>jeaye: for FFI bindings
08:01:08  <txdv>there is also a static library version of it
08:01:09  * Soarez|afkchanged nick to Soarez
08:01:13  * inolenquit (Quit: Leaving.)
08:01:13  <jeaye>I _am_ statically linking to libuv. Apaprently not all of it is coming in.
08:01:33  <txdv>then your linking instructions are badly written or something
08:01:45  <txdv>indutny: i see you have done that thing with returning ENOMEM after all?
08:03:30  <indutny>txdv: yes
08:03:45  * petka_joined
08:04:51  <txdv>you could have mentioned that in the comment, because once I saw your commit I pulled it and wanted to make the change and then I was surprised that it already was there
08:05:54  <jeaye>txdv: Hmm, I was fucking up somewhere. All fixed.
08:06:09  <txdv>thats what I said
08:06:27  <jeaye>uv.h has a function declaration uv_fs_mkdtemp, which has a param called 'template', leading to a syntax error in C++.
08:06:39  <txdv>already fixed in master
08:07:27  <jeaye>#define template fuckmylife\n#include <uv.h>\n#undef template
08:07:43  <txdv>2014-08-06 11:47 Iñaki Baz Castillo o include: avoid using C++ 'template' reserved word
08:07:55  <jeaye>Cheers.
08:08:48  <txdv>to what?
08:09:20  <jeaye>Successful compilations.
08:09:44  <txdv>what beverage are we toasting with?
08:10:11  <jeaye>Your choice. I have water and a cold chai latte.
08:12:56  <rendar>txdv: FFI?
08:13:03  * quijote_quit (Ping timeout: 240 seconds)
08:13:14  <txdv>foreign function interface
08:13:25  <rendar>i see
08:14:00  <txdv>for example in C# you can import functions from shared libraries
08:14:04  <rendar>txdv: well, FFI are achievable with bot .so shared objects and .a static libs, right?
08:14:12  <rendar>both*
08:19:04  <txdv>dynamically loading static libraries is not possible
08:19:08  <txdv>they are for static linking
08:19:51  <txdv>and if you have a language in some bytecode, like .net/mono, then you'll have to load it dynamically, since you can't link it in
08:21:56  <rendar>well yeah, i was considering the "interface" part of the FFI acronym, you don't link static libraries at runtime, but you still have some kind of interface exported by them that you include at compile time
08:23:50  <txdv>basically everything has to be exposed as a function
08:24:07  * jeayepart ("WeeChat 0.4.3")
08:27:39  * c4milojoined
08:28:32  * Soarezchanged nick to Soarez|afk
08:32:26  * c4miloquit (Ping timeout: 260 seconds)
08:55:03  * rmgquit (Remote host closed the connection)
08:55:36  * rmgjoined
08:59:57  * rmgquit (Ping timeout: 240 seconds)
09:03:29  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
09:09:23  * quijotejoined
09:13:56  * quijotequit (Ping timeout: 255 seconds)
09:17:32  * quijotejoined
09:20:37  * txdvquit (Read error: Connection reset by peer)
09:21:38  * txdvjoined
09:22:54  * Left_Turnjoined
09:23:41  * sinclair|workjoined
09:27:54  * Soarez|afkchanged nick to Soarez
09:34:18  * kazuponquit (Remote host closed the connection)
09:37:54  * kazuponjoined
09:41:29  * Soarezchanged nick to Soarez|afk
10:05:15  * Soarez|afkchanged nick to Soarez
10:09:11  * quijote_joined
10:09:15  * quijotequit (Read error: Connection reset by peer)
10:16:04  * c4milojoined
10:17:35  * avalanche123joined
10:20:32  * c4miloquit (Ping timeout: 255 seconds)
10:21:53  * avalanche123quit (Ping timeout: 255 seconds)
10:23:43  * avalanche123joined
10:26:48  * quijote_quit (Ping timeout: 246 seconds)
10:30:43  * avalanche123quit (Ping timeout: 272 seconds)
10:41:14  * seishunjoined
10:47:00  * SquirrelCZECHquit (Ping timeout: 250 seconds)
10:51:40  * SquirrelCZECHjoined
10:52:40  * Alex7Komjoined
10:55:24  * quijotejoined
10:55:46  * abraxasquit (Remote host closed the connection)
10:57:15  * karupachanged nick to zz_karupa
10:59:49  * quijotequit (Ping timeout: 255 seconds)
11:03:40  * quijotejoined
11:04:23  * c4milojoined
11:19:15  * andrehjrjoined
11:19:29  * quijotequit (Ping timeout: 255 seconds)
11:21:13  * Soarezchanged nick to Soarez|afk
11:27:15  * sblom_joined
11:27:34  * sblomquit (Ping timeout: 264 seconds)
11:34:48  * andrehjrquit (Quit: ["Textual IRC Client: www.textualapp.com"])
11:59:02  * sblom_quit (Read error: Connection reset by peer)
12:03:38  * bajtosquit (Quit: bajtos)
12:16:07  * quijotejoined
12:20:24  * quijotequit (Ping timeout: 260 seconds)
12:24:22  * c4miloquit (Remote host closed the connection)
12:31:43  * skebciojoined
12:39:46  * c4milojoined
12:41:55  * cjihrigjoined
12:44:43  * euoia_joined
12:49:52  * cjihrigquit (Ping timeout: 240 seconds)
12:50:42  * janjongboomjoined
12:51:28  * euoia_quit (Read error: No route to host)
12:54:17  * Alex7Komquit (Read error: Connection reset by peer)
12:54:22  * Alex7Kom_joined
12:56:32  * abraxasjoined
13:01:42  * abraxasquit (Ping timeout: 260 seconds)
13:04:11  * Alex7Kom_quit (Read error: Connection reset by peer)
13:04:39  * Alex7Komjoined
13:16:25  * cjihrigjoined
13:16:53  * quijotejoined
13:21:02  * quijotequit (Ping timeout: 245 seconds)
13:21:05  * cjihrigquit (Ping timeout: 272 seconds)
13:25:07  * kazuponquit (Remote host closed the connection)
13:27:19  * avalanche123joined
13:31:36  * kenperkinsjoined
13:31:48  * avalanche123quit (Ping timeout: 260 seconds)
13:32:44  * seishunquit (Ping timeout: 240 seconds)
13:38:59  * quijotejoined
13:42:35  * seishunjoined
13:47:11  * cjihrigjoined
13:48:53  * Soarez|afkchanged nick to Soarez
13:51:56  * cjihrigquit (Ping timeout: 272 seconds)
13:53:14  * cjihrigjoined
14:02:50  * wavdedjoined
14:13:27  * inolenjoined
14:18:02  * inolenquit (Ping timeout: 250 seconds)
14:29:58  * mikealquit (Quit: Leaving.)
14:36:02  * quijotequit (Ping timeout: 245 seconds)
14:37:08  * wavdedquit (Ping timeout: 240 seconds)
14:37:15  * kenperkinsquit (Quit: Computer has gone to sleep.)
14:57:48  * abraxasjoined
15:02:08  * abraxasquit (Ping timeout: 250 seconds)
15:03:20  * kazuponjoined
15:04:04  * quijotejoined
15:06:04  * bradleymeckjoined
15:12:29  * a_lejoined
15:17:55  * dignifiedquirequit (Quit: dignifiedquire)
15:19:56  * mikealjoined
15:25:02  * wavdedjoined
15:29:06  * kazuponquit (Remote host closed the connection)
15:31:15  * mikealquit (Quit: Leaving.)
15:34:49  * daviddiasjoined
15:38:15  * a_lequit (Remote host closed the connection)
15:41:07  * kenperkinsjoined
15:43:47  * dignifiedquirejoined
15:44:55  * jgijoined
15:49:23  * bajtosjoined
15:56:41  * Soarezchanged nick to Soarez|afk
15:58:35  * rmgjoined
16:06:15  * mikealjoined
16:08:56  * mcavagejoined
16:10:05  * Alex7Komquit (Read error: Connection reset by peer)
16:10:11  * Alex7Kom_joined
16:12:24  * piscisaureusjoined
16:15:27  * mcavagequit (Remote host closed the connection)
16:16:13  * dap_joined
16:17:50  * kazuponjoined
16:24:21  * piscisaureusquit (Quit: ~ Trillian Astra - www.trillian.im ~)
16:26:23  * Alex7Kom_quit (Read error: Connection reset by peer)
16:26:29  * Alex7Komjoined
16:29:13  * quijote_joined
16:29:14  * quijotequit (Ping timeout: 250 seconds)
16:30:23  <AlexisMocha>@tjfontaine: does node still support Windows Server 2003?
16:34:38  * wavdedquit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
16:36:43  * mikealquit (Quit: Leaving.)
16:37:05  * mikealjoined
16:41:54  * Alex7Komquit (Read error: Connection reset by peer)
16:41:59  * Alex7Kom_joined
16:45:42  * wavdedjoined
16:45:44  * a_lejoined
16:56:11  * wavdedquit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
16:57:39  * wavdedjoined
16:58:35  * abraxasjoined
17:01:13  * dignifiedquirequit (Quit: dignifiedquire)
17:03:05  * wavdedquit (Quit: Textual IRC Client: www.textualapp.com)
17:03:26  * abraxasquit (Ping timeout: 260 seconds)
17:03:32  * brsonjoined
17:04:41  * kenperkinsquit (Quit: Computer has gone to sleep.)
17:05:33  * quijote_quit (Ping timeout: 255 seconds)
17:09:21  * avalanche123joined
17:10:44  * rmgquit (Remote host closed the connection)
17:11:18  * rmgjoined
17:11:31  * quijotejoined
17:12:58  <tjfontaine>AlexisMocha: that's more a question for you? though I think we proposed making a change that drops support for xp
17:15:23  <AlexisMocha>tjfontaine: I thought you'd say something like that :) I guess I was wondering if that decision had been made before... it sounds like it hasn't.
17:16:02  * rmgquit (Ping timeout: 260 seconds)
17:17:02  <tjfontaine>AlexisMocha: I think the answer is actually the changes for getaddrinfo means we can't support < vista
17:17:08  <rendar>tjfontaine: why a change that drop support for xp is needed?
17:17:15  <tjfontaine>rendar: see above
17:17:31  <tjfontaine>the passing of v4mapped etc didn't exist in an xp world
17:17:43  <rendar>hmm, v4mapped?!
17:18:09  <tjfontaine>rendar: http://msdn.microsoft.com/en-us/library/windows/desktop/ms737530(v=vs.85).aspx
17:18:15  * inolenjoined
17:18:36  * kazuponquit (Remote host closed the connection)
17:18:49  <rendar>oh, the AI_V4MAPPED flag
17:19:15  * c4miloquit (Remote host closed the connection)
17:19:30  * Soarez|afkchanged nick to Soarez
17:19:52  <rendar>but isn't node using external libraries (c-ares?) for get address info?
17:20:16  * rosskjoined
17:20:17  <tjfontaine>no, we defer mostly to getaddrinfo so we can get what the system thinks is the right answer
17:20:25  <tjfontaine>dns.lookup === getaddrinfo
17:20:36  <tjfontaine>dns.resolve === cares
17:20:57  <rendar>ok
17:20:59  * Soarezchanged nick to Soarez|afk
17:21:11  <AlexisMocha>I wonder how much is actually broken on xp
17:22:13  <rendar>tjfontaine: with dns.lookup we're speaking about getting addrinfo structures on a given "hostname.com" string, otherwise with dns.resolve we mean actually getting a valid ipv4/6 address?
17:22:56  <saghul>AlexisMocha, tjfontaine: on libuv at least, the plan is to drop if after 0.12
17:23:37  <AlexisMocha>I'm assuming we are considering win2k3==xp
17:25:00  <saghul>yes
17:26:02  * saghulchanged nick to saghul|afk
17:27:36  <tjfontaine>AlexisMocha: I am, does azure have 2k3?
17:28:20  <tjfontaine>rendar: well, dns.resolve we mean talk to nameservers, dns.lookup ask libc
17:28:32  <AlexisMocha>tjfontaine: not supported by azure, though you can in theory hack your way through
17:28:44  <tjfontaine>frustrating :)
17:28:55  <tjfontaine>AlexisMocha: what would that look like?
17:29:14  <AlexisMocha>prepping and uploading your own .vhd
17:29:16  <rendar>tjfontaine: got it
17:29:17  <tjfontaine>is it easier to just do a local vm for now to test against and find out how bad the test suite is?
17:29:31  <rendar>tjfontaine: dns.resolve is the "our" implementation, dns.lookup is the system's
17:29:48  <AlexisMocha>tjfontaine: yeah, I am looking into this. Unless you already have a VM lying around
17:30:13  <tjfontaine>AlexisMocha: I don't currently, but I could start building one in the background
17:30:22  <tjfontaine>rendar: more or less, yes
17:30:58  <rendar>tjfontaine: hmm, i see, but why this difference? i mean, if there is dns.resolve which works, why i'd need dns.lookup?
17:31:03  * jreyno40joined
17:31:15  * dignifiedquirejoined
17:31:30  <tjfontaine>rendar: there are more record types that we can get by plumbing dns itself, that libc's don't normally give you access to
17:31:54  <rendar>tjfontaine: oh, got it
17:32:06  * wavdedjoined
17:33:58  <AlexisMocha>tjfontaine: if your VM were to be part of Jenkins, then it would be more useful than one in MSOT's lab
17:34:35  <AlexisMocha>but to do an initial test run, I think I can find one
17:35:09  * rmgjoined
17:35:17  <tjfontaine>AlexisMocha: ok, I think I can get a 2k3 into the jenkins infra, but in the mean time if you could do an initial test that'd be fantastic
17:35:46  <AlexisMocha>tjfontaine: sounds like a plan
17:39:40  * quijotequit (Read error: Connection reset by peer)
17:39:46  * quijote_joined
17:45:14  * quijote_quit (Read error: Connection reset by peer)
17:45:47  * quijotejoined
17:46:41  * wavded_joined
17:48:19  * wavdedquit (Ping timeout: 244 seconds)
17:51:23  * rendar_joined
17:52:56  * jreyno40quit (Quit: jreyno40)
17:53:23  * mikealquit (Quit: Leaving.)
17:54:01  * SquirrelCZECHquit (Ping timeout: 244 seconds)
17:55:03  * rendarquit (Ping timeout: 240 seconds)
18:00:29  * mikealjoined
18:01:02  * c4milojoined
18:05:37  * c4miloquit (Ping timeout: 245 seconds)
18:12:52  * quijotequit (Ping timeout: 240 seconds)
18:13:46  * kellabytejoined
18:21:09  * c4milojoined
18:29:23  * kazuponjoined
18:31:35  * Alex7Kom_quit
18:33:39  * kazuponquit (Ping timeout: 246 seconds)
18:36:48  * dignifiedquirequit (Quit: dignifiedquire)
18:37:25  * mcavagejoined
18:39:21  * quijotejoined
18:41:36  * c4miloquit (Remote host closed the connection)
18:43:33  * quijotequit (Ping timeout: 240 seconds)
18:45:40  * qard-appnetapart
18:45:55  * bajtosquit (Quit: bajtos)
18:48:34  * qard-appnetajoined
18:49:52  * c4milojoined
18:59:27  * abraxasjoined
19:04:01  * abraxasquit (Ping timeout: 255 seconds)
19:05:17  * Soarez|afkchanged nick to Soarez
19:11:12  * quijotejoined
19:13:55  * mcavagequit (Remote host closed the connection)
19:15:13  * brsonquit (Quit: leaving)
19:20:34  <jgi>saghul|afk: feel free to ping me if you have any issue with https://github.com/joyent/libuv/pull/1422 when running the tests on Jenkins
19:21:35  * Left_Turnquit (Quit: Leaving)
19:23:54  * Soarezchanged nick to Soarez|afk
19:24:52  * rmgquit (Ping timeout: 240 seconds)
19:26:13  * Soarez|afkchanged nick to Soarez
19:27:17  <jgi>I have a question regarding the cluster module
19:27:22  <jgi>this test fails on Jenkins: http://jenkins.nodejs.org/job/smartos-repro/28/DESTCPU=ia32,label=ubuntu-12.04/tapTestReport/simple.tap-81/
19:27:35  <jgi>it fails *most of the time*, that is not always
19:27:49  <jgi>it also fails generally on Ubuntu, not that much on other platforms
19:28:37  <jgi>it seems the problem comes from the handler of the event ‘exit’ on workers’ processes
19:28:50  <jgi>this handler deletes the worker from the list of workers
19:29:07  * Soarezchanged nick to Soarez|afk
19:29:10  * mcavagejoined
19:29:14  <jgi>and then, when the ‘disconnect’ handler of the cluster tries to do some resource cleanup
19:29:19  <jgi>it can’t find the worker anymore
19:31:39  <jgi>Removing the deletion of the worker in the worker’s exit event handler fixes this specific issue. However, in other cases I guess only the exit event on the worker process could fire, and thus we wouldn’t delete the worker from the workers list.
19:32:12  * Left_Turnjoined
19:32:41  * rmgjoined
19:33:52  * quijotequit (Ping timeout: 240 seconds)
19:34:04  <jgi>So my question is: is it safe to assume that the disconnect event will always fire on the cluster?
19:34:50  * Soarez|afkchanged nick to Soarez
19:39:37  * avalanche123quit (Remote host closed the connection)
19:40:10  * avalanche123joined
19:43:44  * indutnyquit (Ping timeout: 260 seconds)
19:44:33  * avalanche123quit (Ping timeout: 240 seconds)
19:44:54  * indutnyjoined
19:52:01  * Soarezchanged nick to Soarez|afk
19:52:28  * indutnyquit (Ping timeout: 250 seconds)
19:54:22  * mikealquit (Quit: Leaving.)
19:54:47  * avalanche123joined
19:56:52  * rmgquit (Ping timeout: 245 seconds)
20:01:41  * indutnyjoined
20:10:58  * jreyno40joined
20:27:06  * Soarez|afkchanged nick to Soarez
20:31:59  * quijotejoined
20:32:24  <jreyno40>So I have a program that keeps failing on uv__req_unregister inside of src/unix/udp.c on line 97
20:32:41  <jreyno40>(inside of the libuv-master source code)
20:33:31  <jreyno40>and I have no idea why it’s failing. This is the message I’m getting: “Assertion failed: (uv__has_active_reqs(handle->loop)), function uv__udp_run_completed, file src/unix/udp.c”
20:35:11  <tjfontaine>jgi: disconnect will always fire on the cluster worker or master?
20:35:13  * EhevuTovjoined
20:35:36  <tjfontaine>jreyno40: same stuff from yesterday? do you have test code that you can show?
20:36:01  <jreyno40>tjfontaine: No. But I’m following the memory even into libuv and I can’t see any memory corruption
20:36:13  * hueniversequit (Quit: Leaving.)
20:36:13  <jreyno40>tjfontaine: I may code one up
20:36:14  * quijotequit (Ping timeout: 244 seconds)
20:36:28  * hueniversejoined
20:42:51  * dshaw_joined
20:43:11  * SquirrelCZECHjoined
20:52:49  * sblomjoined
20:59:46  <jgi>tjfontaine: on the master
21:00:19  * abraxasjoined
21:00:57  * inolen1joined
21:03:03  * inolenquit (Ping timeout: 240 seconds)
21:04:44  * abraxasquit (Ping timeout: 240 seconds)
21:05:24  * mcavagequit (Remote host closed the connection)
21:08:39  <jreyno40>tjfontaine: I found something kindof odd in libuv
21:08:46  <jreyno40>tjfontaine: I’ve been looking at source
21:08:53  <tjfontaine>jgi: so the question is should the death of a worker always fire a disconnect event on the master?
21:09:08  <jgi>tjfontaine: exactly
21:09:12  <tjfontaine>jreyno40: what might that be?
21:09:46  <tjfontaine>jgi: that would seem like something we would be expecting to happen, such that we know when to try and spawn a replacement
21:10:35  <jreyno40>tjfontaine: uv__run is called which calls uv__run_pending, which calls uv__udp_io which checks for outgoing messages, and calls uv__udp_sendmsg(handle) if it finds messages waiting to send.
21:11:27  <jreyno40>tjfontaine: uv_udp_sendmsg() removes requests from the queue as it handles them & calls sendmsg()
21:11:42  <jreyno40>tjfontaine: AFTER which uv_udp_run_completed is called
21:12:01  * inolen1quit (Read error: Connection reset by peer)
21:12:19  * inolenjoined
21:12:27  <jreyno40>tjfontaine: and inside uv_udp_run_completed is this: uv_req_unregister()
21:13:00  <jreyno40>and in uv_req_unregister, the faulting line: “assert(uv__has_active_reqs(loop));"
21:13:47  <jreyno40>tjfontaine: now I *probably* misread something, but it seems to me that requests are pulled off of the queue as handled, then we try to see if any are left later with an assert?
21:18:33  * avalanche123quit (Remote host closed the connection)
21:19:36  * wavded_quit (Quit: Textual IRC Client: www.textualapp.com)
21:19:52  <jreyno40>tjfontaine: assuming I’m wrong, here’s another question: should EACH request sent use it’s own callback function?
21:20:00  <jreyno40>tjfontaine: or is it okay to share callback functions
21:20:29  <tjfontaine>up to the implementor, in practice for node we use a single cb, differentiating by the data for the request, which stores the actual js cb
21:20:56  * avalanche123joined
21:21:14  * mikealjoined
21:25:04  <tjfontaine>jreyno40: seems like you're reusing the same req if you're blowing this assertion
21:25:40  <jreyno40>tjfontaine: That’s what it would seem like
21:26:13  <tjfontaine>things in the udp layer have changed recently though
21:27:41  <tjfontaine>jreyno40: I'm curious to know if 817c09257d5b842a9efedac72e6ebd9aa313f109 and 41891222bca4e985bef45515fe131fbdbec3f969 are to blame here
21:27:49  * mcavagejoined
21:28:25  <jreyno40>tjfontaine: What are those?
21:28:45  <jreyno40>recent commits?
21:28:48  <tjfontaine>indeed
21:29:00  <jreyno40>tjfontaine: I mean I’m a pretty decent programmer
21:29:05  * rmgjoined
21:29:11  <jreyno40>and I’ve traced the memory all the way through my program
21:29:14  <tjfontaine>jreyno40: https://github.com/joyent/libuv/commit/41891222bca4e985bef45515fe131fbdbec3f969
21:29:18  <jreyno40>and it seems we’re NOT reusing requests
21:29:27  <tjfontaine>jreyno40: try and use libuv v0.11.26
21:31:43  <jreyno40>okay, where can I find that?
21:31:43  <jreyno40>sorry
21:32:02  <tjfontaine>you're just using our git repo? you can checkout the tag
21:32:06  <jreyno40>tjfontaine: and truth be told, putting my programming skills to test, I’m really starting to wonder whether this is a libuv issue or not as well
21:32:29  <jreyno40>I’m a total git noob
21:32:43  * quijotejoined
21:32:57  <tjfontaine>jreyno40: git checkout v0.11.26
21:33:58  <jreyno40>tjfontaine: Ahh, okay. This will take a second; I’ll have to recompile the library assembly file for static linking with the open source proj
21:34:04  <jreyno40>tjfontaine: bear with me if you don’t mind
21:34:17  <tjfontaine>I've got plenty to do, I'm not going anywhere for a while :)
21:36:23  <jreyno40>tjfontaine: I really hope this works :D
21:37:37  * quijotequit (Ping timeout: 272 seconds)
21:39:23  <jreyno40>tjfontaine: got an assertion failed in make check
21:39:28  <jreyno40>but for tcp
21:41:28  <tjfontaine>make check is ok
21:46:52  * mcavagequit (Remote host closed the connection)
21:48:03  <jreyno40>tjfontaine: unfortunately I’m still getting it
21:48:20  <jreyno40>tjfontaine: So it probably isn’t a libuv issue
21:51:05  * kenperkinsjoined
21:54:00  * kenperkins_joined
21:54:02  * kenperkinsquit (Read error: Connection reset by peer)
21:54:04  * kenperkins_quit (Client Quit)
21:56:58  * yunongjoined
22:04:44  * mcavagejoined
22:11:17  <rmg>tjfontaine: are there plans for another 0.11.x release before a 0.12 rc?
22:11:43  * mikealquit (Quit: Leaving.)
22:12:10  * mikealjoined
22:12:39  <tjfontaine>rmg: I would want the next 0.11 release to be the rc
22:13:39  <rmg>thanks for confirming. that's what I thought I read
22:14:26  * mikealquit (Client Quit)
22:15:20  <rmg>I ask because it is easier to automate testing against tagged node releases than arbitrary sha1's, and its been a few months since v0.11.13
22:16:11  <tjfontaine>ok
22:17:51  * a_lequit (Remote host closed the connection)
22:18:26  * a_lejoined
22:18:31  * c4miloquit (Remote host closed the connection)
22:19:32  <rmg>v0.12 is the milestone for tracking blockers and planned changes, correct?
22:20:18  <tjfontaine>to some definitions, it's pretty close
22:20:29  * mcavagequit (Remote host closed the connection)
22:20:31  <tjfontaine>the biggest blockers are me reviewing trevors AL removal stuff
22:20:38  <tjfontaine>and intl
22:21:32  <rmg>that's what I figured.. was just going to take a look if there was anything lower hanging to clean up
22:22:42  * a_lequit (Ping timeout: 245 seconds)
22:22:51  <indutny>tjfontaine: yay!
22:25:03  <tjfontaine>rmg: the best thing is to test against an existing code base
22:25:28  <tjfontaine>the bug that jgi found for libuv was one that clears up the other outstanding smartos bugs so I feel safer
22:26:04  * mcavagejoined
22:29:03  <rmg>tjfontaine: like the existing 76 npm module jobs in my CI that I want to test on a tag newer than 0.11.13? :-P
22:31:50  <jreyno40>tjfontaine: We have scavenged our program to no avail; we still think it may be in libuv
22:31:54  * cjihrigquit (Quit: Leaving.)
22:31:58  <jreyno40>tjfontaine: can you suggest any other versions we might try?
22:32:03  * mikealjoined
22:33:38  * quijotejoined
22:34:03  * mikealquit (Client Quit)
22:37:52  * quijotequit (Ping timeout: 240 seconds)
22:42:07  * rendar_quit
22:47:59  * EhevuTovquit (Quit: This computer has gone to sleep)
22:49:52  * dsantiagoquit (Ping timeout: 240 seconds)
22:50:34  <jreyno40>tjfontaine: so we’re changed a few things in our code (the biggest of which was no longer using UV_RUN_NOWAIT
22:50:46  <jreyno40>tjfontaine: and thus far it appears to be working, leading us to believe there is definitely a bug in libuv somewhere
22:52:53  * EhevuTovjoined
22:54:10  <tjfontaine>rmg: anytag? I'll make you a tag ;) but that won't help npm
22:54:34  <tjfontaine>jreyno40: filing a bug with a minimal test case is ideal
22:55:51  <othiym23>it would be super super SUPER handy to have a testing release of node that had the crypto fix for hashes in it so that shasum works properly again
22:55:52  <rmg>tjfontaine: the lack of tag was in reference to http://nodejs.org/dist/nightlies/sprint/v0.11.14/ which node-gyp can't find source code for, so it doesn't even attempt to build addons
22:56:30  <othiym23>that has been a large pain point for npm, because people have been publishing using 0.11.x and then ending up with uninstallable packages because the package digests don't match
22:57:22  * EhevuTovquit (Ping timeout: 240 seconds)
22:58:09  <tjfontaine>rmg: but node gyp needs it to exist in proper /dist
22:58:10  <othiym23>also, fun fact (which I came here to share): process.kill(-1, 0) returns true on OS X
22:58:23  <tjfontaine>rmg: node gyp should be taught to use what we distribute in $PREFIX/include/node
22:58:24  <tjfontaine>:)
22:58:54  <tjfontaine>othiym23: because of overflow?
22:59:04  * dshaw_quit (Quit: Leaving.)
22:59:32  <tjfontaine>TJs-MacBook-Pro:libuv tjfontaine$ kill -0 "-1"
22:59:32  <tjfontaine>TJs-MacBook-Pro:libuv tjfontaine$ echo $?
22:59:32  <tjfontaine>0
22:59:50  <rmg>tjfontaine: yes, but there wasn't even a tag for that build, so I can't provide the source for it manually
23:00:07  <mmalecki>we should `assert(pid > 0)` imo
23:00:07  <tjfontaine>fair
23:00:15  <tjfontaine>pid == 0 is valid
23:00:34  <othiym23>tjfontaine: yeah, it's probably overflow, it just amuses me
23:00:35  <tjfontaine>kill(0xFFFFFFFF, 0x0, 0x1) = 0 0
23:00:43  * dshaw_joined
23:00:48  <mmalecki>ah, scheduler
23:00:57  <tjfontaine>according to dtruss when I watch the command line
23:01:06  <mmalecki>or swapper
23:01:11  <tjfontaine>mmalecki: process group leader
23:01:14  <mmalecki>or w/e that was
23:01:18  * abraxasjoined
23:01:46  <tjfontaine> If pid is zero:
23:01:47  <tjfontaine> Sig is sent to all processes whose group ID is equal to the process group ID of the sender, and for which the process has permission; this is a variant of killpg(2).
23:02:00  <mmalecki>ah, I see
23:02:02  <othiym23>right, but 0 is the signal, not the pid
23:02:11  <tjfontaine> If pid is -1:
23:02:11  <tjfontaine> If the user has super-user privileges, the signal is sent to all processes excluding system processes and the process sending the signal. If the user is not the super user, the signal is sent to all processes with the same uid as the user, excluding the process sending the signal. No error is returned if any process could be signaled.
23:02:16  <tjfontaine>for mr othiym23
23:02:23  <othiym23>i c
23:02:23  * sblomquit (Read error: Connection reset by peer)
23:02:26  <othiym23>magic!
23:02:33  <othiym23>it's not just for badly-written JS libraries
23:02:37  <tjfontaine>that's correct
23:02:41  * cjihrigjoined
23:02:41  <tjfontaine>:)
23:02:42  <tjfontaine>magic
23:03:20  <othiym23>anyway, I would appreciate a 0.11.14, if only so that people who are just using nvm install 0.11 don't end up poisoning the npm registry forever
23:03:30  <tjfontaine>heh
23:03:51  <tjfontaine>I was really hoping to get the libuv fix in, but saghul|afk's last comment in libuv suggest that won't happen before sunday
23:03:55  * a_lejoined
23:04:09  * wolfeidauquit (Remote host closed the connection)
23:04:37  * EhevuTovjoined
23:05:32  <jreyno40>tjfontaine: Okay; if I have the time to create a test case I will
23:05:36  * abraxasquit (Ping timeout: 246 seconds)
23:06:29  <jreyno40>tjfontaine: but really all we did was initialize a uv loop as uv_default_loop() then run it with uv_run(UV_NO_WAIT)
23:06:41  <jreyno40>tjfontaine: then send stuff repeatedly
23:06:51  <jreyno40>tjfontaine: so the issue most probably has to do with UV_NO_WAIT
23:07:08  * seishunquit (Ping timeout: 240 seconds)
23:09:18  * cjihrigquit (Ping timeout: 250 seconds)
23:11:28  * a_lequit (Remote host closed the connection)
23:16:16  * EhevuTovquit (Quit: This computer has gone to sleep)
23:17:44  * EhevuTovjoined
23:20:57  * EhevuTovquit (Client Quit)
23:34:09  * quijotejoined
23:35:15  * Soarezchanged nick to Soarez|afk
23:35:40  * quijotequit (Read error: Connection reset by peer)
23:36:02  * quijotejoined
23:36:38  * jreyno40part
23:37:31  * cjihrigjoined
23:37:34  * quijotequit (Read error: Connection reset by peer)
23:37:51  * quijotejoined
23:40:56  * c4milojoined
23:41:52  * cjihrigquit (Ping timeout: 240 seconds)
23:42:08  * quijotequit (Ping timeout: 250 seconds)
23:45:37  * c4miloquit (Ping timeout: 245 seconds)
23:46:10  * SquirrelCZECHquit (Ping timeout: 255 seconds)
23:46:30  * a_lejoined
23:50:08  * SquirrelCZECHjoined
23:58:49  * dsantiagojoined
23:59:00  * SquirrelCZECHquit (Ping timeout: 260 seconds)