00:00:00  * ircretaryquit (Remote host closed the connection)
00:00:08  * ircretaryjoined
00:06:49  * stagasquit (Ping timeout: 240 seconds)
00:06:51  * paulfryzeljoined
00:10:18  * Kakeraquit (Ping timeout: 240 seconds)
00:11:17  * paulfryzelquit (Ping timeout: 246 seconds)
00:24:27  * nickleeflyjoined
00:33:32  * quijotejoined
00:38:19  * quijotequit (Ping timeout: 264 seconds)
00:44:30  * AvianFluquit (Remote host closed the connection)
00:44:51  * brsonjoined
00:59:51  * thlorenzjoined
01:07:34  * paulfryzeljoined
01:10:52  * perldorkjoined
01:11:50  * paulfryzelquit (Ping timeout: 246 seconds)
01:13:48  * perldorkquit (Remote host closed the connection)
01:14:37  * c4milojoined
01:16:04  * AvianFlujoined
01:16:38  * brsonquit (Ping timeout: 252 seconds)
01:19:43  * c4miloquit (Ping timeout: 264 seconds)
01:26:10  * petka_quit (Quit: Connection closed for inactivity)
01:32:40  * brsonjoined
01:34:17  * quijotejoined
01:36:59  * perldorkjoined
01:38:55  * quijotequit (Ping timeout: 264 seconds)
01:38:56  * perldorkquit (Remote host closed the connection)
02:08:19  * paulfryzeljoined
02:12:38  * paulfryzelquit (Ping timeout: 240 seconds)
02:35:12  * quijotejoined
02:39:18  * quijotequit (Ping timeout: 240 seconds)
02:40:01  * brsonquit (Ping timeout: 240 seconds)
02:51:58  * mikolalysenkoquit (Ping timeout: 240 seconds)
02:57:36  * mikolalysenkojoined
03:03:43  * c4milojoined
03:08:01  * c4miloquit (Ping timeout: 240 seconds)
03:09:09  * paulfryzeljoined
03:14:07  * paulfryzelquit (Ping timeout: 268 seconds)
03:22:54  * [1]porchratjoined
03:24:48  * m76joined
03:25:02  * porchratquit (Ping timeout: 252 seconds)
03:25:03  * [1]porchratchanged nick to porchrat
03:31:12  * WalrusPonyjoined
03:34:01  * WalrusPony1quit (Ping timeout: 240 seconds)
03:35:29  * inolenquit (Quit: Leaving.)
03:35:50  * quijotejoined
03:35:56  * inolenjoined
03:40:47  * quijotequit (Ping timeout: 252 seconds)
04:09:31  * mikolalysenkoquit (Ping timeout: 264 seconds)
04:09:53  * paulfryzeljoined
04:11:55  * thlorenzquit (Remote host closed the connection)
04:13:58  * paulfryzelquit (Ping timeout: 240 seconds)
04:19:36  * kevinsimperquit (Remote host closed the connection)
04:34:53  * nickleeflyquit (Quit: Connection closed for inactivity)
04:36:36  * quijotejoined
04:41:08  * quijotequit (Ping timeout: 246 seconds)
04:50:28  * kenperkins_joined
04:53:37  * kenperkinsquit (Ping timeout: 240 seconds)
05:05:24  * mikolalysenkojoined
05:08:47  * bradleymeckjoined
05:08:58  * porchratquit (Ping timeout: 240 seconds)
05:09:45  * mikealjoined
05:09:52  * porchratjoined
05:09:58  * mikolalysenkoquit (Ping timeout: 240 seconds)
05:10:39  * paulfryzeljoined
05:14:33  * bradleymeckquit (Quit: bradleymeck)
05:14:49  * paulfryzelquit (Ping timeout: 240 seconds)
05:27:18  * m76quit (Ping timeout: 240 seconds)
05:37:22  * quijotejoined
05:42:22  * quijotequit (Ping timeout: 252 seconds)
05:58:56  * AvianFluquit (Remote host closed the connection)
05:59:26  * AvianFlujoined
06:04:19  * AvianFluquit (Ping timeout: 268 seconds)
06:20:27  * kevinsimperjoined
06:24:38  * kevinsimperquit (Ping timeout: 240 seconds)
06:29:00  * bajtosjoined
06:38:13  * quijotejoined
06:41:52  * michael-joined
06:43:10  * quijotequit (Ping timeout: 268 seconds)
07:12:18  * paulfryzeljoined
07:16:32  * paulfryzelquit (Ping timeout: 246 seconds)
07:16:52  * [1]porchratjoined
07:17:55  * kenperkinsjoined
07:19:24  * inolen1joined
07:19:26  * inolenquit (Ping timeout: 240 seconds)
07:19:33  * porchratquit (Ping timeout: 268 seconds)
07:19:33  * kenperkins_quit (Ping timeout: 268 seconds)
07:19:35  * [1]porchratchanged nick to porchrat
07:42:37  * c4milojoined
07:47:07  * michael-quit (Quit: michael-)
07:47:55  * c4miloquit (Ping timeout: 268 seconds)
07:51:42  * m76joined
07:59:17  * kevinsimperjoined
08:10:32  * bajtosquit (Quit: bajtos)
08:12:59  * paulfryzeljoined
08:16:40  * janjongboomjoined
08:17:31  * paulfryzelquit (Ping timeout: 268 seconds)
08:20:06  * bajtosjoined
08:27:14  * petka_joined
08:27:44  * quijotejoined
08:43:21  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
08:47:59  * stagasjoined
09:00:40  * sinclair|workjoined
09:13:43  * paulfryzeljoined
09:17:59  * paulfryzelquit (Ping timeout: 246 seconds)
09:33:01  * c4milojoined
09:39:58  * c4miloquit (Ping timeout: 240 seconds)
09:41:26  * c4milojoined
09:48:12  * rmgjoined
09:53:43  * rmgquit (Ping timeout: 268 seconds)
09:55:53  * c4milo_joined
09:56:38  * c4miloquit (Ping timeout: 240 seconds)
10:08:53  * c4milo_quit (Remote host closed the connection)
10:14:42  * paulfryzeljoined
10:18:53  * paulfryzelquit (Ping timeout: 246 seconds)
10:28:37  * janjongboomjoined
10:28:53  * hzjoined
10:39:38  * bajtosquit (Quit: bajtos)
10:46:28  * Kakerajoined
10:47:37  * quijotequit (Quit: quijote)
11:10:42  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
11:15:19  * paulfryzeljoined
11:19:12  * sinclair|workquit (Quit: ChatZilla 0.9.90.1 [Firefox 27.0.1/20140212131424])
11:20:03  * paulfryzelquit (Ping timeout: 268 seconds)
11:27:46  * quijotejoined
11:32:56  * quijotequit (Ping timeout: 265 seconds)
12:00:05  * janjongboomjoined
12:16:03  * paulfryzeljoined
12:20:55  * paulfryzelquit (Ping timeout: 264 seconds)
12:28:38  * quijotejoined
12:33:26  * quijotequit (Ping timeout: 268 seconds)
12:44:25  * AvianFlujoined
13:14:08  * thlorenzjoined
13:16:51  * paulfryzeljoined
13:19:19  * thlorenzquit (Remote host closed the connection)
13:19:52  * thlorenzjoined
13:21:14  * paulfryzelquit (Ping timeout: 246 seconds)
13:23:09  * stagasquit (Ping timeout: 265 seconds)
13:23:24  * stagasjoined
13:24:31  * thlorenzquit (Ping timeout: 264 seconds)
13:26:07  * bajtosjoined
13:29:16  * quijotejoined
13:30:38  * janjongboomquit (Ping timeout: 240 seconds)
13:32:54  * mikolalysenkojoined
13:33:53  * quijotequit (Ping timeout: 252 seconds)
13:36:36  * janjongboomjoined
13:48:07  * bradleymeckjoined
13:54:33  * cphooverjoined
13:59:56  * jmar777joined
14:04:33  * thlorenzjoined
14:04:57  * jmar777quit (Read error: Connection reset by peer)
14:05:26  * jmar777joined
14:09:05  * EM3RYchanged nick to emery
14:17:33  * paulfryzeljoined
14:22:08  * paulfryzelquit (Ping timeout: 246 seconds)
14:24:54  * quijotejoined
14:48:29  * paulfryzeljoined
14:51:26  * mikolalysenkoquit (Ping timeout: 252 seconds)
14:54:15  * prettyrobotsjoined
15:00:02  * mikolalysenkojoined
15:16:09  * brsonjoined
15:20:31  * quijotequit (Quit: quijote)
15:21:13  * prettyrobotsquit (Read error: Connection reset by peer)
15:41:39  * paulfryzelquit (Read error: Connection reset by peer)
15:42:01  * rmgjoined
15:42:04  * paulfryzeljoined
15:42:46  * seldojoined
15:43:10  * WalrusPony1joined
15:45:18  * WalrusPonyquit (Ping timeout: 240 seconds)
15:49:23  * seldoquit (Remote host closed the connection)
15:51:05  * quijotejoined
15:51:29  * seldojoined
15:55:18  * quijotequit (Ping timeout: 240 seconds)
15:55:28  * seldoquit (Remote host closed the connection)
15:55:47  * seldojoined
15:55:49  * quijotejoined
16:04:58  * roxluquit (Quit: leaving)
16:08:12  <bradleymeck>tjfontaine: the bundle loader I am just ensuring reaches minimal spec for my end goal, the encryption does not need to be in core, but we need to figure out the minimal way to allow it
16:09:45  * kevinsimperquit (Remote host closed the connection)
16:11:08  * jmar777quit (Remote host closed the connection)
16:15:09  <indutny>hey hey hey
16:15:12  <indutny>how are you, people?
16:15:31  <bradleymeck>indutny: doing well, yourself?
16:15:39  <indutny>good good
16:17:11  <bradleymeck>indutny: still playing with loaders, but getting more annoyed as time goes on
16:17:18  <indutny>oh
16:17:29  <indutny>good that you have a progress on it
16:17:41  <indutny>have you managed to get through that mach-o problems?
16:18:20  <bradleymeck>indutny: no, using temp files since tj didn't want it to have a custom linker
16:18:23  * dap_joined
16:18:28  <indutny>oh
16:18:30  <indutny>I see
16:18:31  <bradleymeck>will probably revisit once I finish up the loader
16:18:36  <indutny>ok
16:18:57  * sblomjoined
16:19:45  <bradleymeck>I never quite figured out wth was going on with symbol names
16:20:42  * seldoquit (Remote host closed the connection)
16:21:08  * seldojoined
16:21:09  <tjfontaine>the customer linker just isn't scalable, and if jni is doing tempfiles that's reasonable for me
16:21:43  <bradleymeck>tjfontaine: yea just is a lot of edge casey things going on and need to keep ironing them until there aren't problems
16:24:07  <bradleymeck>tjfontaine: as per the issue comments: single executables want those, should have put it more clearly. they do not need to be in core, but able to be done
16:24:21  <indutny>tjfontaine: nah, scales well :)
16:24:26  <indutny>just needs some good npm module
16:24:30  <indutny>that will do it
16:24:35  * seldoquit (Remote host closed the connection)
16:24:42  <indutny>though, I suppose there will be dragons and ponies on windows
16:24:44  <indutny>with all that code signature
16:24:47  <indutny>and friends
16:25:02  <bradleymeck>indutny: the biggest and serious problem is implementing the shared memory spaces some use
16:25:15  <indutny>I'm afraid I don't follow it
16:25:24  <indutny>what memory should be shared?
16:25:31  <bradleymeck>indutny: some mmap shared memory when they load, and share across multiple processes
16:25:40  <indutny>erm
16:25:50  <indutny>and? :)
16:25:50  <bradleymeck>see voltdb... T_T
16:26:01  <indutny>let's say they have mmap'ed something
16:26:05  <bradleymeck>indutny: it differs because it is not actually using mmap
16:26:08  <indutny>what's wrong with it?
16:26:14  <indutny>wait
16:26:20  * bradleymeckdigs
16:26:20  <indutny>aaaah
16:26:31  <indutny>you mean that the db itself is bundled within binary
16:26:41  <indutny>yeah, this complicates the fuck out of it
16:27:02  <indutny>it'll be read-only anyway
16:28:11  <indutny>tjfontaine: ping :)
16:28:12  <indutny>hello
16:28:22  <indutny>where are we now in terms of 0.12?
16:28:30  <indutny>how far are we from branching?
16:28:36  <indutny>what are we currently blocked by?
16:28:46  <tjfontaine>pong, I need you to review a thing to make trevors stuff private, it's pretty straight forward
16:29:00  <tjfontaine>well it's straight forward if you don't review all of his previous changes :)
16:29:10  <indutny>hahah
16:29:11  <indutny>ok
16:29:15  <tjfontaine>https://github.com/tjfontaine/node/compare/al-external
16:29:17  <tjfontaine>just the last 4 commits
16:29:39  <indutny>looking
16:29:42  <indutny>is that all?
16:30:04  <tjfontaine>that plus my inclusion of dtrace-provider which will need rebased after this stuff lands
16:30:15  <tjfontaine>https://github.com/joyent/node/pull/5940/files
16:30:57  <tjfontaine>if you could specifically take care to look at src/util.h change, I may want to backport something like that to v0.10 and v0.8 for Utf8Value
16:31:10  <indutny>hm...
16:31:14  <indutny>ok
16:34:27  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
16:37:12  * rosskjoined
16:39:57  * mikealquit (Quit: Leaving.)
16:41:38  * hzquit
16:42:01  <hueniverse>tjfontaine: can you tell me what this is about: https://github.com/spumko/hapi/issues/1509
16:42:22  * mikealjoined
16:42:25  <tjfontaine>lemme see
16:43:06  * mikealquit (Client Quit)
16:46:07  <tjfontaine>hueniverse: it's particularly odd that it changed in behavior for oyu from hapi1 -> 2
16:46:16  <tjfontaine>hueniverse: I presume this is all on v0.10
16:47:14  <hueniverse>tjfontaine: no clue. I didn't confirm that on the same version of node this happens between versions of hapi. I don't want to setup node on windows...
16:47:30  <hueniverse>tjfontaine: was mainly wondering if there was a node change that explains this
16:47:45  <hueniverse>tjfontaine: so I can ask them to try something out
16:47:47  <tjfontaine>nothing off the top of my head
16:48:43  <tjfontaine>I've commented so I can get notifications going forward
16:49:07  <hueniverse>tjfontaine: thanks
16:49:31  <hueniverse>tjfontaine: I confirmed everything we got works on 0.11 master except good
16:49:56  <tjfontaine>ok I'm still trying to narrow that SOB down
16:50:09  <hueniverse>tjfontaine: and poop but that's because it needs to compile C++ add-on (heapdump)
16:50:48  <hueniverse>1/31 isn't too bad :-)
16:51:01  <tjfontaine>hueniverse: nothing in uv has changed that indicated tcp_nodelay isn't being called
16:51:56  <tjfontaine>hueniverse: btw in your good tests, could you update them to use process.execPath instead of 'node'?
16:52:22  * cphooverquit (Quit: Leaving.)
16:53:11  * thlorenzquit (Remote host closed the connection)
16:53:52  <hueniverse>tjfontaine: sure. can you open an issue?
16:54:41  <tjfontaine>sure
16:55:21  <hueniverse>tjfontaine: thanks. I'm in a self-made coverage hell this morning... now that we track every statement in logical expressions...
16:56:19  <tjfontaine>heh oh man
16:58:50  * mikealjoined
17:14:56  <indutny>tjfontaine: al LGTM
17:15:48  <tjfontaine>ok -- then it's just a matter of squashing trevors stuff down into a few sensible commits
17:17:47  <hueniverse>tjfontaine: I did find one bug thanks to it. hapi was missing 31 tests using this new standard
17:20:14  * quijotequit (Quit: quijote)
17:22:02  * benviequit (Ping timeout: 246 seconds)
17:22:14  <trevnorris>moring
17:23:22  <hueniverse>trevnorris: are we meeting this week?
17:23:47  * thlorenzjoined
17:24:07  <trevnorris>hueniverse: sure. tomorrow or thurs work for me.
17:24:57  <hueniverse>trevnorris: I assume you prefer at Mozilla? you are welcome to drive up to the farm too
17:25:22  <trevnorris>hueniverse: where's the farm? south of cupertino right? if that's the case then that commute would actually be easier.
17:25:30  <trevnorris>going against traffic and all
17:26:58  * cphooverjoined
17:28:06  * stagasquit (Read error: Connection reset by peer)
17:28:41  * thlorenzquit (Ping timeout: 265 seconds)
17:30:06  * thlorenzjoined
17:31:34  * quijotejoined
17:32:59  <indutny>tjfontaine: cool
17:33:04  <indutny>tjfontaine: going to review the last one
17:33:07  <indutny>https://github.com/joyent/node/pull/5940/files
17:33:10  * seldojoined
17:34:58  * bradleymeck_joined
17:35:53  * mikolalysenkoquit (Ping timeout: 252 seconds)
17:36:42  * quijotequit (Quit: quijote)
17:36:48  * benviejoined
17:43:35  * mikealquit (Quit: Leaving.)
17:56:24  * mikolalysenkojoined
18:08:02  <tjfontaine>fuck you v8, fuck you very much -- I don't want to debug both gc collection problems *and* thread locking issues
18:08:13  <tjfontaine>gc and optimization paths are in threads
18:08:19  <tjfontaine>THANKS GUYS
18:08:19  <LOUDBOT>I CALL THIS AN ESTIMATE
18:08:23  <trevnorris>hehe, yup.
18:08:31  <trevnorris>tjfontaine: though I believe there's a flag that disables that
18:08:42  <tjfontaine>doesn't help in production though
18:08:48  <trevnorris>ah, true true
18:09:17  <trevnorris>too bad you can't enable/disable flags on the fly
18:09:26  <trevnorris>that's would be an awesome debugging option.
18:09:26  <tjfontaine>you can, it's just not helpful
18:09:31  <trevnorris>ah, ok
18:09:39  <tjfontaine>by the time you need it, you're probalby already screwed
18:09:45  <trevnorris>hehehe
18:12:25  <tjfontaine>hmm this may be some fallout from the other work you were doing
18:12:34  <tjfontaine>frame #0: 0x0000000100031429 node`node::StreamWrapCallbacks::TryWrite(uv_buf_t**, unsigned long*) [inlined] node::StreamWrapCallbacks::wrap(this=0x0000000101900020, this=0x0000000000000000) const + 4 at stream_wrap.h:137
18:12:40  <tjfontaine>my this pointer really really shouldn't be null
18:12:51  <trevnorris>is this on master?
18:13:00  <tjfontaine>it's master + 7145
18:14:22  <trevnorris>tjfontaine: how are you producing that?
18:14:33  <tjfontaine>just hitting a demo http server with a bunch of requests
18:15:03  <trevnorris>hm. ok. well, I fixed the one issue w/ https://github.com/trevnorris/node/commit/4f531a7f7c1771e5bc4ac74c59085a5d1f2743d1
18:15:29  <tjfontaine>ya tha tcommit is there
18:15:34  <tjfontaine>it's not all of it though
18:15:58  <trevnorris>so I should be able to reproduce w/ a simple hello world and smacking the crap out of it?
18:16:05  <trevnorris>also, can you gist the entire bt?
18:20:40  <tjfontaine>https://gist.github.com/tjfontaine/9746027
18:20:47  <trevnorris>thanks
18:21:55  <trevnorris>tjfontaine: StreamWrapCallbacks doesn't inherit from AsyncWrap
18:21:59  <tjfontaine>there are other evil things going on in this environment
18:22:10  <trevnorris>how do you mean?
18:22:41  <tjfontaine>updated gist with another backtrace
18:22:48  <trevnorris>thanks
18:22:59  <tjfontaine>somehting is very wrong.
18:23:19  <trevnorris>whoa. yeah.
18:23:27  <trevnorris>ok, i'm trying to reproduce now
18:23:41  <tjfontaine>I'm totally ont he wrong platform to be debugging tihs right now
18:26:04  <trevnorris>tjfontaine: what are you using to hit the server?
18:26:32  <trevnorris>e.g. wrk, ab, ?
18:29:42  <trevnorris>i wish js debugger had a way to trace access to a given variable
18:33:08  <tjfontaine>curl
18:33:11  <tjfontaine>and up arrow
18:33:16  <tjfontaine>:)
18:33:25  <tjfontaine>on osx
18:33:35  <trevnorris>hehe
18:34:02  <trevnorris>can you gist your script? I seriously can't reproduce on linux
18:35:05  <tjfontaine>I can later, for now I've gotta prep for a demo and am going to have to do it with 0.10 for now
18:35:12  <trevnorris>coolio
18:37:15  * quijotejoined
18:39:12  <trevnorris>groundwater: you around today?
18:39:47  <groundwater>trevnorris: AHOY
18:40:28  <trevnorris>cool. i'll have questions for you in a minute. right now i'm stepping through the cls tests.
18:41:10  <groundwater>trevnorris: sure thing, i'm about to step out for lunch but i'll followup at some point
18:41:26  <trevnorris>cool
18:41:43  <trevnorris>groundwater: so, it looks like it might be trying to enter the same AL twice,
18:42:00  <trevnorris>but a single AL instance is only allowed to be added once.
18:42:09  <trevnorris>so it's entered twice, then removed twice
18:42:12  * quijotequit (Ping timeout: 268 seconds)
18:42:24  <trevnorris>but the second time it's removed since it was already removed there's no context available
18:44:01  <trevnorris>well, wait. that should be impossible because then the same.... ah wait. ok
18:44:07  <trevnorris>think I might know what's going on
18:48:08  <trevnorris>wtf. running backtrace from debugger causes the program to terminate
18:50:07  * AvianFluquit (Remote host closed the connection)
18:51:01  * quijotejoined
18:55:27  * quijotequit (Ping timeout: 252 seconds)
19:00:00  <Kakera>indutny: any thoughts about that gzip thing?
19:09:31  * cphooverquit (Quit: Leaving.)
19:14:48  * thlorenzquit
19:15:45  <trevnorris>groundwater: so.. i'm not sure why but for some reason for the second test it's running create then after. there's no call to before.
19:17:43  * indexzerojoined
19:19:39  * thlorenzjoined
19:22:38  * quijotejoined
19:28:40  * cphooverjoined
19:34:08  * Kakeraquit (Ping timeout: 252 seconds)
19:42:23  * mikolalysenkoquit (Ping timeout: 252 seconds)
19:51:34  * rmgquit (Remote host closed the connection)
19:53:12  * stagasjoined
19:53:28  * mikolalysenkojoined
19:54:28  * bajtosquit (Quit: bajtos)
20:02:24  * quijotequit (Quit: quijote)
20:03:20  * quijotejoined
20:04:35  * quijotequit (Client Quit)
20:06:36  * quijotejoined
20:09:58  * quijotequit (Quit: Textual IRC Client: www.textualapp.com)
20:10:10  * quijotejoined
20:11:22  * Kakerajoined
20:12:43  * brsonquit (Quit: leaving)
20:12:51  * brsonjoined
20:15:10  * quijotequit (Quit: My MacBook has gone to sleep. ZZZzzz…)
20:17:40  <groundwater>trevnorris: can you gist the result?
20:17:48  * janjongboomjoined
20:21:23  <trevnorris>groundwater: sure, give me a min, just tracing something.
20:22:31  * rmgjoined
20:25:53  <indutny>Kakera: hey
20:25:54  <indutny>what should I say? :)
20:26:42  <trevnorris>groundwater: https://gist.github.com/trevnorris/9748398
20:27:03  <trevnorris>groundwater: see how the before for hello-0 runs, but there's no before for hello-1
20:27:29  <othiym23>hueniverse: did you change any of your dependencies between hapi 2 -> 3?
20:27:38  <trevnorris>groundwater: i guess that'd have to be a problem w/ my code, but still figuring out where.
20:27:56  <othiym23>hueniverse: when I was looking at HTTP agents, I ran across a few that set noDelay = true for "performance reasons"
20:28:07  <othiym23>those who do not understand Nagle's algorithm are doomed to reinvent it or something
20:28:14  <groundwater>trevnorris: the CLS tests often have multiple async-listeners in play, could it be realted to that?
20:28:19  <Kakera>indutny: whatever you think
20:28:31  <trevnorris>groundwater: well, when I only run one test then it finishes fine.
20:28:38  <indutny>I think it is good to be on the par with unzip and others :)
20:28:48  <trevnorris>groundwater: but, it shouldn't be possible to run an after w/o running the before
20:28:56  <indutny>but without breaking existing code
20:29:14  <Kakera>you mean gunzip?
20:29:18  <trevnorris>and, if you check the value of "after" it's 1, meaning the { value: 'hello-1' } that was returned from create somehow didn't get to the after callback
20:29:19  <groundwater>trevnorris: can you post the code you ran?
20:30:06  <trevnorris>sure, one sec.
20:30:13  <hueniverse>othiym23: the odd thing is that we don't touch it or include any module that does
20:30:23  * rmgquit (Remote host closed the connection)
20:30:36  * rmgjoined
20:30:42  <indutny>Kakera: and it too
20:30:44  <indutny>I mean
20:30:53  <indutny>if everyone has a feature
20:30:58  <indutny>we should have it probably too
20:31:05  <othiym23>hueniverse: OK, so much for that hypothesis
20:31:09  <trevnorris>groundwater: here: https://gist.github.com/trevnorris/9748398
20:31:31  <trevnorris>groundwater: i updated the test to show that the create does return hello-1, but it doesn't show up in the after. that's because the before doesn't run.
20:31:39  <Kakera>indutny: and what should it do with trailing garbage?
20:31:51  <indutny>what gunzip does?
20:32:27  <Kakera>ignores it and returns a non-zero result
20:34:11  <Kakera>python's gzip module just throws, which is supposedly a bug
20:36:29  <indutny>I guess it is better to throw
20:36:37  <indutny>sorry, I'm not in the right mood right now :)
20:36:41  <indutny>to decide anything
20:38:08  * mikolalysenkoquit (Ping timeout: 268 seconds)
20:38:44  <Kakera>"Some tar.gz files are padded with zeroes to have a size which is a multiple of a certain block size. This occurs in particular when the compressed tar file is on a device such as a magnetic tape."
20:38:51  <Kakera>from gzip.org
20:40:15  <trevnorris>groundwater: ok. so what it looks like, before for hello-0 runs, which runs the create for hello-1, but then for some reason the after for hello-1 is run.
20:41:00  <trevnorris>groundwater: i'm going to look at how bindEmitter() does its thing
20:42:12  <othiym23>trevnorris: with a lot of grody monkeypatching
20:42:42  <othiym23>trevnorris: http://npmjs.org/package/emitter-listener
20:47:50  <trevnorris>othiym23: ok. somehow the hello-1 create is running, but that attaches to the hello-0 context, so it causes the hello-1 after to run
20:47:59  <trevnorris>othiym23: this is for test/bind-emitter.tap.js
20:49:08  <trevnorris>othiym23: basically, each test should have its own context stack, but they're throwing all the tests on the same context
20:49:16  <trevnorris>so they're not running in the correct order
20:52:16  <othiym23>trevnorris: emitter-listener is different from AL in that it captures the context at *both* the time the listener is added to the EE, *and* at the time emit fires
20:52:33  <othiym23>trevnorris: this all worked fine with the old AL interface
20:56:15  <trevnorris>othiym23: you mean the old AL implementation. which unfortunately had a fundamental flaw. there where changes that had to be made despite the PROVIDER filtering stuff
20:58:03  <othiym23>how is emitter-listener breaking the new AL? it only does its thing synchronously
20:59:58  <trevnorris>othiym23: i'm not even completely sure. I think it's actually how tap is running the tests. basically the create and teardown
21:00:03  <trevnorris>still tracing through that.
21:00:20  <trevnorris>there are a lot of layers
21:02:17  <trevnorris>othiym23: because if I run only one test then everything runs fine.
21:02:57  <othiym23>trevnorris: it probably is the overlapping async listeners, then
21:03:13  <othiym23>because I create a fresh namespace for each test, which in turn creates a fresh AL instance
21:04:06  <trevnorris>yeah. but the output shows that the same namespace is loaded multiple times, so the .active is all messed up
21:04:22  * AvianFlujoined
21:04:25  * mikolalysenkojoined
21:04:27  <trevnorris>anyways. think i'm actually getting somewhere
21:08:38  * mikolalysenkoquit (Ping timeout: 240 seconds)
21:10:06  * hzjoined
21:11:25  * bradleymeck_quit (Quit: bradleymeck_)
21:13:27  <trevnorris>othiym23: part of this might be because prior I tried to support running addAsyncListener from a create/before/after callback. but that created unresolvable edge cases so I removed all the logic to support that case.
21:13:39  <trevnorris>othiym23: because the implementation was fundamentally flawed.
21:14:07  <othiym23>trevnorris: CLS only adds the asyncListener at namespace creation, which should never be happening inside an AL callback
21:16:06  <trevnorris>othiym23: I logged from the create/before/after in test/bind-emitter.js and commented out all but one function. check how many times it ran: https://gist.github.com/trevnorris/9748398#file-al-cls-out-log
21:19:06  <trevnorris>ok. I see that it's destroyed (e.g. removeAL() runs) but the AL continues to trigger.
21:20:05  <trevnorris>othiym23: you know if tap uses timers to run its tests?
21:20:34  <othiym23>trevnorris: yes, it does, to force timeout
21:21:12  <trevnorris>ok
21:30:31  * indexzeroquit (Quit: indexzero)
21:37:36  * stagasquit (Ping timeout: 265 seconds)
21:38:29  * cphooverquit (Ping timeout: 240 seconds)
21:40:42  * hueniversequit (Read error: Connection reset by peer)
21:41:27  * stagasjoined
21:41:30  * m76quit (Read error: Connection reset by peer)
21:53:39  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
22:00:18  * mikolalysenkojoined
22:02:25  * thlorenzquit (Remote host closed the connection)
22:20:40  * hueniversejoined
22:29:46  * jmar777joined
22:32:55  * jmar777quit (Remote host closed the connection)
22:33:47  <trevnorris>othiym23: I think I know what's happening. you won't have to make any changes to your existing code to have it work w/ the PR.
22:34:42  <trevnorris>othiym23: at least, that's a goal. unless I run into something earth shattering, I plan on having all your current cls tests passing before this PR is merged.
22:44:01  <tjfontaine>do you have a timeframe? I thought the grand compromise was that it wasn't strictly necessary for that to happen and thenw e could feature freeze and bug fix
22:48:36  <trevnorris>tjfontaine: i'm taking the rest of the day to do code/commit cleanup. i'm pushing 30 commits and it's become hard to track the order of when/what/why things have changed.
22:48:57  <trevnorris>tjfontaine: let me finish this and i'll get back to you on the other.
22:49:56  <trevnorris>i have a partial idea why this is happening, but my brain needs to go though and do a sweep.
22:50:17  <trevnorris>and sorry, what was our grand compromise?
22:53:34  <trevnorris>tjfontaine: btw, have you been able to reproduce that issue? I haven't yet.
23:00:54  <tjfontaine>trevnorris: the grand compromise is that the majority of the tracing api would be stuck behind a private module, and then exposed through an npm installable 'tracing' module such that we can iterate through the mechanism more easily
23:02:43  <trevnorris>tjfontaine: right now I"m not going to operate under those constraints. When I have the PR at the point where it's functioning as it should then I'll begin defining the points of contact that need to be exposed to the tracing module.
23:03:26  <tjfontaine>the point of the constraints were to aid in getting 0.12 actually out the door though
23:03:32  <tjfontaine>that's part of what the error summit was about
23:05:08  <trevnorris>tjfontaine: i understand. and i'm working furiously to finish a couple fundamentals. there are a few things that will have to be exposed from the JS side from core in order for the external tracing module to work.
23:05:22  <trevnorris>these are different than the bug fixes.
23:05:37  <trevnorris>things like, getting the error handling propagation to work like I'd like.
23:05:39  <trevnorris>those are bugs.
23:06:06  <tjfontaine>can you be specific about what those semantics are and why they're broken so we can iterate as a team on them
23:06:07  <trevnorris>but things like the fact that the NEXTTICK provider is called wholesale is causing serious issues in the case of the cls tests
23:06:54  <tjfontaine>like plop them in 7145's comments
23:07:49  <trevnorris>let me finish this cleanup so we have properly segmented commits between changes that should happen in core before v0.12 (e.g. the PROVIDER stuff in AsyncWrap) and the external stuff like the AsyncListener API
23:08:00  <trevnorris>right now it's all a serious mess.
23:16:23  * brsonquit (Quit: leaving)
23:17:20  <hueniverse>can os.hostname() ever return an empty string?
23:25:06  <trevnorris>tjfontaine: I just want to double check. the issue you're encountering. does that have to do with my PR or can you reproduce w/o my PR?
23:27:58  * bradleymeck_joined
23:32:10  <othiym23>hueniverse: anything is possible
23:32:40  <othiym23>hueniverse: but at least on Linux & SmartOS, os.hostname should always return *something*
23:33:21  * perldorkjoined
23:34:24  <hueniverse>othiym23: I'm in coverage hell, so I just messed with hostname() :-)
23:36:00  * perldorkquit (Remote host closed the connection)
23:39:23  * indexzerojoined
23:46:05  * AvianFluquit (Remote host closed the connection)
23:46:35  * AvianFlujoined
23:48:58  <tjfontaine>I have never seen os.hostname return empty string on a properly configured host
23:49:06  <tjfontaine>trevnorris: I haven't gone down that rabbit hole again
23:49:27  <trevnorris>ok
23:50:30  <trevnorris>from that second bt it didn't look like something from my PR. at least, I didn't see anything. but if you manage to reproduce it properly let me know. I want to make sure it's not.
23:50:55  * AvianFluquit (Ping timeout: 264 seconds)
23:58:50  * bradleymeck_quit (Quit: bradleymeck_)