00:00:01  * ircretaryquit (Remote host closed the connection)
00:00:09  * ircretaryjoined
00:03:17  <othiym23>tjfontaine: I'm backporting the 0.11 / 0.12 version of Agent to get the keepalive behavior I need, and it's barfing when trying to unref the CleartextStream
00:03:50  * paulfryzelquit (Remote host closed the connection)
00:03:58  <othiym23>it doesn't look like either CleartextStream or its _handle have unref defined in 0.10, but since this is a straight copypasta, I assume everything just works in 0.12
00:04:19  <othiym23>unfortunate, because the unref behavior is what I'm looking for
00:08:59  * euoiaquit (Ping timeout: 240 seconds)
00:09:07  <tjfontaine>I would need to look and see where you need exactly to monkey punch 0.10 ssl streams to make that happen
00:09:28  <tjfontaine>also
00:09:29  <tjfontaine>ZOMG WHY DID IT TAKE US UNTIL 0.11 TO DO UNREF FOR KEEPALIVE
00:09:29  <LOUDBOT>IT'S ESPECIALLY NICE WHEN YOU SET TANK ZOMBIES ON FIRE
00:10:45  <othiym23>tjfontaine: didn't you only just add unref to 0.10 in the first place?
00:11:00  <tjfontaine>well I added some in 0.8, and then finished the job in 0.10
00:11:21  <tjfontaine>(clearly incompletely)
00:12:13  <othiym23>I think I'm probably going to smash agentkeepalive's idle socket reaper behavior onto the side of my backport, which should be good enough for my porpoises
00:13:18  * thlorenzjoined
00:14:16  * wolfeida_joined
00:15:17  <tjfontaine>oh I'm supposed tob e looking at client .. not server
00:15:41  * daviddia_quit (Remote host closed the connection)
00:17:03  * thlorenzquit (Remote host closed the connection)
00:17:16  * wolfeidauquit (Ping timeout: 265 seconds)
00:17:34  * thlorenzjoined
00:18:49  * defunctzombie_zzchanged nick to defunctzombie
00:19:39  <tjfontaine>client.socket.pair.cleartext.socket._handle.unref()
00:19:45  <tjfontaine>not sure if there's a faster path to that or not
00:20:44  * mcavagequit
00:21:04  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
00:21:47  * AvianFluquit (Ping timeout: 264 seconds)
00:21:58  <tjfontaine>client.socket.socket._handle.unref()
00:22:01  * thlorenzquit (Ping timeout: 240 seconds)
00:22:01  <tjfontaine>ya that's saner
00:23:55  * mikealquit (Quit: Leaving.)
00:25:48  * mikealjoined
00:33:31  * zz_karupanerurachanged nick to karupanerura
00:34:45  * paulfryzeljoined
00:38:50  * thlorenzjoined
00:38:59  * paulfryzelquit (Ping timeout: 240 seconds)
00:39:48  * thlorenz_joined
00:41:40  * thlorenz_quit (Remote host closed the connection)
00:42:13  * thlorenz_joined
00:43:23  * thlorenzquit (Ping timeout: 264 seconds)
00:46:21  * euoiajoined
00:46:45  * thlorenz_quit (Ping timeout: 265 seconds)
01:03:18  * eugenewarequit (Remote host closed the connection)
01:12:05  * AvianFlujoined
01:24:32  * jmar777joined
01:35:41  * paulfryzeljoined
01:40:29  * paulfryzelquit (Ping timeout: 269 seconds)
01:57:40  * dap_1quit (Quit: Leaving.)
02:01:25  * dap_joined
02:02:15  * dap_quit (Client Quit)
02:03:11  * brsonquit (Ping timeout: 264 seconds)
02:08:46  * jmar777quit (Remote host closed the connection)
02:09:13  * euoiaquit (Ping timeout: 240 seconds)
02:10:58  * brsonjoined
02:12:28  * jmar777joined
02:12:38  * AWintermanquit (Ping timeout: 240 seconds)
02:14:37  * euoiajoined
02:15:37  * brsonquit (Ping timeout: 240 seconds)
02:21:42  * thlorenzjoined
02:22:38  * brsonjoined
02:28:16  * mikealquit (Quit: Leaving.)
02:30:33  * jmar777quit (Remote host closed the connection)
02:32:01  * brsonquit (Ping timeout: 240 seconds)
02:32:04  * calvinfoquit (Quit: Leaving.)
02:35:59  * benviequit (Ping timeout: 240 seconds)
02:36:25  * paulfryzeljoined
02:36:52  * dominictarrjoined
02:37:29  * defunctzombiechanged nick to defunctzombie_zz
02:40:29  * paulfryzelquit (Ping timeout: 240 seconds)
02:41:20  * bradleymeckjoined
02:41:35  * rosskquit (Ping timeout: 264 seconds)
02:46:58  * mikolalysenkoquit (Ping timeout: 240 seconds)
02:52:18  * dominictarrpart
03:02:36  * calvinfojoined
03:04:51  * calvinfo1joined
03:04:51  * calvinfoquit (Read error: Connection reset by peer)
03:09:04  * bradleymeckquit (Quit: bradleymeck)
03:19:45  * AWintermanjoined
03:21:30  * abraxasjoined
03:24:17  * rosskjoined
03:24:59  * AWintermanquit (Ping timeout: 240 seconds)
03:28:18  * rosskquit (Remote host closed the connection)
03:32:57  * rosskjoined
03:33:05  <othiym23>tjfontaine: I'll poke around some more, thanks for the pointer
03:34:51  * abraxasquit (Remote host closed the connection)
03:35:25  * abraxasjoined
03:39:37  * abraxasquit (Ping timeout: 240 seconds)
03:39:39  * abraxas_joined
03:42:14  * benviejoined
03:45:02  * thlorenzquit (Read error: Connection reset by peer)
04:03:37  * euoiaquit (Ping timeout: 240 seconds)
04:13:08  * abraxas_quit (Remote host closed the connection)
04:13:41  * abraxasjoined
04:17:58  * abraxasquit (Ping timeout: 240 seconds)
04:20:59  * abraxasjoined
04:27:05  * rosskquit (Remote host closed the connection)
04:37:47  * paulfryzeljoined
04:41:20  * mscdexjoined
04:42:29  * paulfryzelquit (Ping timeout: 240 seconds)
04:42:52  <mscdex>is it possible to use a v8 locker with node's isolate in a separate thread or are we required to use uv_async?
04:44:05  <mscdex>i've tried using a locker, but it hangs at locker creation as if node still has the isolate locked
04:44:13  * bradleymeckjoined
04:59:10  <tjfontaine>mscdex: you created the locker in a separate thread? I've not looked into this specifically, they've clearly been working towards an environment with first class threading and locking support but I don't yet know the semantics around it
04:59:58  <mscdex>i just noticed that there's a locker created for the whole duration that the event loop is alive, so that's why it's stuck waiting to get a lock
05:00:00  <mscdex>:\
05:00:30  <tjfontaine>generally because we're not ready for people to do something else (yet)
05:01:04  * defunctzombie_zzchanged nick to defunctzombie
05:01:05  <tjfontaine>to support that concept we also need to support their (in future versions) micro task api
05:01:36  * benviequit (Ping timeout: 252 seconds)
05:02:14  <tjfontaine>mscdex: strictly speaking I'm not sure that locker needs to exist outside the do{}while();
05:02:22  <mscdex>having a uv_async_t, callback, and separate userdata struct for every callback i have in C is going to be a pain in the ass
05:02:36  <mscdex>i'm writing a binding and there's tons of callbacks i need to have happen
05:02:51  <mscdex>that's why i was hoping i could use a locker somehow
05:03:12  * abraxasquit (Remote host closed the connection)
05:03:13  <mscdex>temporarily get access when node wasn't doing anything
05:03:24  <tjfontaine>moving the locker inside the do while could potentially be releasing it often enough to actually get useful locking out of it
05:03:44  <tjfontaine>the problem is we're generally sleeping inside uv_run() when waiting on IO
05:03:48  * abraxasjoined
05:03:51  <tjfontaine>"sleep"
05:04:14  * brsonjoined
05:05:11  <tjfontaine>what you actually want is to move the locker inside the do while, drop the before exit semantics, and always uv_run with nowait
05:05:55  <tjfontaine>that moves node out of sleeping by default, so you need to be sure you're prepared for that
05:07:01  <tjfontaine>mscdex: also you really only need one uv_async_t and or a single pipe used with uv_poll to do your signalling on, user data struct per event would still be necessary yes
05:07:12  <tjfontaine>bbl
05:08:35  * abraxasquit (Ping timeout: 264 seconds)
05:10:05  <mscdex>wouldn't uv_async force me to keep a queue also in case multiple events happen? since according to the docs it says that the callback isn't necessarily called for every uv_async_send?
05:11:29  * rosskjoined
05:17:21  <mscdex>i guess the alternative would be to create a new uv_async_t, call uv_async_init (guarded by a mutex), call uv_async_send, and then inside the callback call uv_close
05:17:33  <mscdex>that sounds inefficient but ...
05:20:37  * abraxasjoined
05:32:22  * defunctzombiechanged nick to defunctzombie_zz
05:33:59  * LeftWingquit (Remote host closed the connection)
05:34:25  * LeftWingjoined
05:38:34  * paulfryzeljoined
05:40:21  <WalrusPony>mscdex: I use async with a queue system, and it works just fine
05:42:24  * LeftWingquit (Remote host closed the connection)
05:42:32  * LeftWingjoined
05:42:59  * paulfryzelquit (Ping timeout: 240 seconds)
05:44:47  * rosskquit (Remote host closed the connection)
05:44:51  <mscdex>what kind of queue is it though? lock free or ?
05:45:36  * mikolalysenkojoined
05:46:56  * nrajlichquit (Quit: ["Textual IRC Client: www.textualapp.com"])
05:50:34  * Kakerajoined
05:51:42  * janjongboomjoined
06:04:05  <WalrusPony>mscdex: I'm just using a normal queue with a mutex
06:04:32  <WalrusPony>I can't be bothered writing a lock free queue
06:05:46  * LeftWingquit (Remote host closed the connection)
06:05:58  * LeftWingjoined
06:09:13  <mscdex>ok
06:13:47  * WalrusPony1joined
06:14:52  * WalrusPonyquit (Ping timeout: 244 seconds)
06:15:07  * LeftWingquit (Remote host closed the connection)
06:15:18  * LeftWingjoined
06:28:59  * Kakeraquit (Ping timeout: 264 seconds)
06:41:57  * AvianFluquit (Remote host closed the connection)
06:43:29  * abraxasquit (Remote host closed the connection)
06:44:02  * abraxasjoined
06:48:18  * abraxasquit (Ping timeout: 240 seconds)
06:53:24  * mikealjoined
06:59:43  * m76joined
07:07:49  * petka_joined
07:11:14  * wolfeida_quit (Remote host closed the connection)
07:16:29  * eugenewarejoined
07:22:33  * eugenewarequit (Ping timeout: 244 seconds)
07:22:33  * rosskjoined
07:24:14  * calvinfo1quit (Quit: Leaving.)
07:24:40  * abraxasjoined
07:30:49  * rosskquit (Ping timeout: 244 seconds)
07:33:40  * sinclair|workjoined
07:40:22  * paulfryzeljoined
07:42:33  * abraxasquit (Remote host closed the connection)
07:43:08  * abraxasjoined
07:43:10  * abraxasquit (Remote host closed the connection)
07:43:24  * abraxasjoined
07:44:25  * paulfryzelquit (Ping timeout: 240 seconds)
07:57:42  * rosskjoined
08:02:15  * rosskquit (Ping timeout: 265 seconds)
08:19:38  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
08:20:25  * mikolalysenkoquit (Ping timeout: 244 seconds)
08:22:55  * rendarjoined
08:40:53  * paulfryzeljoined
08:42:52  * etruscottjoined
08:42:56  <etruscott>hey! :)
08:43:11  * etruscottis coding some stuff with libuv right now
08:43:37  * abraxasquit (Remote host closed the connection)
08:44:12  * abraxasjoined
08:44:35  * janjongboomjoined
08:45:29  * paulfryzelquit (Ping timeout: 240 seconds)
08:46:59  * abraxasquit (Read error: Connection reset by peer)
08:47:16  * abraxasjoined
08:48:18  <etruscott>if i create a uv_thread and have it wait on a uv_sem_t whats the best way to signal it to end?
08:49:55  <etruscott>i guess have like int active; then post to the semaphore to wake the thread up and it'll see active is 0 and terminate
08:55:15  * hzjoined
08:55:34  * calvinfojoined
09:00:17  * calvinfoquit (Ping timeout: 252 seconds)
09:01:57  * raffikiquit (Remote host closed the connection)
09:02:12  * daviddiasjoined
09:06:16  * wolfeidaujoined
09:06:34  * brsonquit (Quit: leaving)
09:10:59  * sinclair|workquit (Ping timeout: 264 seconds)
09:12:01  <indutny>isaacs: hey man
09:12:04  <indutny>you broke npm again
09:12:19  <indutny>https://fullfatdb.npmjs.com/registry/netmask
09:12:26  <indutny>http://isaacs.iriscouch.com/registry/netmask
09:12:44  * sinclair|workjoined
09:13:06  <indutny>and https://skimdb.npmjs.com/registry/netmask
09:13:17  <indutny>somehow the document is just broken in fullfat
09:19:32  * etruscottquit (Ping timeout: 252 seconds)
09:27:45  * guilleiguaranquit (Quit: Connection closed for inactivity)
09:41:52  * paulfryzeljoined
09:45:59  * paulfryzelquit (Ping timeout: 240 seconds)
09:49:55  * etruscottjoined
09:50:03  * dshaw_joined
09:56:15  * calvinfojoined
10:00:29  * calvinfoquit (Ping timeout: 240 seconds)
10:06:18  * abraxasquit (Remote host closed the connection)
10:06:54  * abraxasjoined
10:07:48  * abraxasquit (Read error: Connection reset by peer)
10:08:15  * abraxasjoined
10:16:06  <saghul>etruscott: yeah, that should do
10:19:07  <MI6>joyent/libuv: David Capello master * 97fc348 : windows: Remove unused variable from uv__pipe_insert_pending_socket - http://git.io/6wlxRQ
10:21:07  * abraxasquit (Remote host closed the connection)
10:21:41  * abraxasjoined
10:22:05  * dsantiagoquit (Ping timeout: 252 seconds)
10:25:59  * abraxasquit (Ping timeout: 240 seconds)
10:26:25  * dshaw_quit (Quit: Leaving.)
10:32:54  * abraxasjoined
10:39:39  * etruscottquit (Quit: etruscott)
10:40:09  * WalrusPony1changed nick to WalrusPony
10:42:22  * dsantiagojoined
10:42:27  * hzquit (Read error: Connection reset by peer)
10:42:28  * paulfryzeljoined
10:46:21  * hzjoined
10:46:42  * sinclair|workquit (Quit: ChatZilla 0.9.90.1 [Firefox 27.0.1/20140212131424])
10:47:29  * paulfryzelquit (Ping timeout: 269 seconds)
10:50:14  * daviddiasquit (Remote host closed the connection)
10:50:30  * daviddiasjoined
10:50:46  * janjongboomquit (Ping timeout: 244 seconds)
10:51:54  * daviddiasquit (Remote host closed the connection)
10:52:09  * janjongboomjoined
10:52:50  * daviddiasjoined
10:55:14  <MI6>joyent/libuv: Paul Tan master * 3470009 : unix: workaround broken pthread_sigmask on Android - http://git.io/d88r0w
10:56:00  <indutny>hey people
10:56:08  <saghul>hey!
10:57:00  * calvinfojoined
10:57:02  * dshaw_joined
11:01:12  * calvinfoquit (Ping timeout: 240 seconds)
11:01:18  * dshaw_quit (Ping timeout: 240 seconds)
11:04:57  * wolfeidauquit (Ping timeout: 265 seconds)
11:07:43  * abraxasquit (Remote host closed the connection)
11:08:17  * abraxasjoined
11:08:58  * daviddiasquit (Remote host closed the connection)
11:12:38  * abraxasquit (Ping timeout: 240 seconds)
11:19:41  * dshaw_joined
11:26:42  * dshaw_quit (Ping timeout: 265 seconds)
11:28:02  * abraxasjoined
11:32:57  * rosskjoined
11:36:57  * karupanerurachanged nick to zz_karupanerura
11:37:16  * rosskquit (Ping timeout: 244 seconds)
11:53:31  * abraxasquit (Remote host closed the connection)
11:57:46  * calvinfojoined
12:01:59  * calvinfoquit (Ping timeout: 240 seconds)
12:24:07  * dshaw_joined
12:28:34  * dshaw_quit (Ping timeout: 265 seconds)
12:44:06  * paulfryzeljoined
12:48:29  * paulfryzelquit (Ping timeout: 240 seconds)
12:56:31  * thlorenzjoined
12:58:28  * calvinfojoined
13:00:44  * thlorenzquit (Client Quit)
13:03:01  * dshaw_joined
13:03:11  * calvinfoquit (Ping timeout: 264 seconds)
13:07:13  * dshaw_quit (Ping timeout: 240 seconds)
13:13:13  * daviddiasjoined
13:13:15  * daviddiasquit (Remote host closed the connection)
13:13:45  * daviddiasjoined
13:17:39  * eugenewarejoined
13:19:57  * euoiajoined
13:21:59  * eugenewarequit (Ping timeout: 240 seconds)
13:26:38  * m76quit (Ping timeout: 240 seconds)
13:36:33  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
13:39:22  * janjongboomjoined
13:50:01  * AvianFlujoined
13:50:24  * guilleiguaranjoined
13:54:44  * abraxasjoined
13:59:16  * calvinfojoined
13:59:21  * abraxasquit (Ping timeout: 244 seconds)
13:59:21  * euoiaquit (Ping timeout: 244 seconds)
14:01:13  * euoiajoined
14:03:18  * calvinfoquit (Ping timeout: 240 seconds)
14:03:46  * dshaw_joined
14:08:37  * dshaw_quit (Ping timeout: 265 seconds)
14:17:40  * eugenewarejoined
14:18:48  * guybrushquit (Excess Flood)
14:18:54  * guybrushjoined
14:22:07  * eugenewarequit (Ping timeout: 252 seconds)
14:22:12  * m76joined
14:27:37  * hzquit
14:46:04  * mscdexpart ("Leaving")
14:50:41  * daviddiasquit (Remote host closed the connection)
14:51:11  * janjongboomquit (Ping timeout: 264 seconds)
14:51:11  * daviddiasjoined
14:52:19  * jmar777joined
14:55:41  * daviddiasquit (Ping timeout: 252 seconds)
14:59:37  * sblomquit (Ping timeout: 240 seconds)
15:00:05  * calvinfojoined
15:01:24  * daviddiasjoined
15:04:29  * calvinfoquit (Ping timeout: 240 seconds)
15:04:35  * dshaw_joined
15:09:02  * dshaw_quit (Ping timeout: 265 seconds)
15:25:33  * dshaw_joined
15:28:49  * dshaw_quit (Client Quit)
15:31:36  * paulfryzeljoined
15:36:31  * mikealquit (Quit: Leaving.)
15:39:02  <bradleymeck>tjfontaine: when you do discuss packaging node apps lmk http://bmeck.github.io/
15:42:47  * bradleymeckquit (Quit: bradleymeck)
15:43:13  * bradleymeckjoined
15:43:22  * AvianFluquit (Ping timeout: 265 seconds)
15:48:54  * mikealjoined
15:49:51  * mikealquit (Client Quit)
15:55:44  * abraxasjoined
15:58:46  * paulfryzelquit (Remote host closed the connection)
16:00:11  * abraxasquit (Ping timeout: 264 seconds)
16:00:43  * calvinfojoined
16:04:02  * paulfryzeljoined
16:05:35  * calvinfoquit (Ping timeout: 264 seconds)
16:07:04  <tjfontaine>morn
16:08:05  <indutny>mornin
16:08:32  <indutny>bradleymeck: you could try using my libs
16:08:34  <indutny>elfy and macho
16:08:38  <indutny>to do the same
16:08:48  <indutny>without patching node itself
16:08:53  <bradleymeck>mmm to do self extracting?
16:09:03  <tjfontaine>I kinda want to do an npm+manta query and find out who in the public space is using .on('pipe'
16:09:08  <bradleymeck>I've got it mostly working on windows but will have to fix it later
16:09:23  <indutny>bradleymeck: yes
16:09:39  <indutny>bradleymeck: yeah, you'll need library to parse .exe as well
16:09:51  <tjfontaine>wee pe/coff
16:09:58  <indutny>and also may need to amend elfy and macho to teach them to encode as well
16:10:02  <indutny>right now they are just parsing
16:10:10  <bradleymeck>imma leave it for now, once I add in memory based files it will end with patching node
16:10:34  <bradleymeck>don't like all the things that break when you use in-memory files though
16:10:51  <tjfontaine>ya but extracting out isn't really ideal
16:11:23  <tjfontaine>the cruft that's left over from dead processes is pretty nasty
16:11:33  <bradleymeck>tjfontaine: if you do streaming reads/writes all sorts of bugs occured when we did that in earlier versions
16:11:55  <tjfontaine>that's because you're probably patching fs to try and be smarter than it should be
16:11:56  <bradleymeck>tjfontaine: it only leaves it on abnormal termination like SIGKILL right now
16:12:42  <bradleymeck>tjfontaine: I tried patching fs on JS side and looked at patching libuv to have in memory resources
16:12:59  <tjfontaine>that's what I'm saying, you shouldn't do it that way, it's just wrong :)
16:13:23  <bradleymeck>mmm I'm wondering what other approach there would be, since many programs modify things on disk
16:13:45  <tjfontaine>modules shoudln't really be mucking around on disk like that, that's quite unclean, and breaks down in sandboxed environments
16:14:04  <tjfontaine>the question is "program" vs "rando module"
16:14:27  <tjfontaine>few well behaved programs make determinatons on __dirname, and instead on cwd
16:14:55  <tjfontaine>a random dependency doing fs operations without provocation is actually a security risk anyway
16:15:10  <bradleymeck>I would much rather combine the 2 ideas, started with extracting going to take another attempt at in-memory in a bit
16:15:25  <bradleymeck>you shouldn't rely on a bundler for security
16:15:51  <tjfontaine>you shouldn't be relying on node for security -- but needless to say the easier we make it to do things the more surface area we expose ourselves to
16:16:11  <tjfontaine>a more interesting approach is teaching the module loader how to load a bundle/zip imo
16:16:12  <bradleymeck>that and determining how to do .fork to sub-resource without loading all the resources into resident memory are my big things
16:17:34  <bradleymeck>tjfontaine: already on that heh, but it does break stuff still and requires checks for things like require('/myzip.zip/mysubresource'); generally treating archives as directories
16:18:53  <tjfontaine>ya, I'm not saying its easy, and it would need to come with a lot of tests, but letting a module / set of modules be zip'd and then the module loader switch out into a different mode when in that tree would be incredibly helfpul
16:19:04  <tjfontaine>especially for those people who complain about a small pathmax on winders
16:20:51  <bradleymeck>yea paths having different lengths based on unicode or not is odd in windows
16:25:35  * dshaw_joined
16:26:56  <indutny>saghul: yt?
16:28:02  <saghul>indutny: sup?
16:28:06  <bradleymeck>tjfontaine: if you have a set of requirements for a bundle module loader link it to me and ill try and conform
16:28:16  <indutny>let's handle ENXIO https://github.com/joyent/libuv/pull/1180 :)
16:28:19  <indutny>saghul: ^
16:31:07  <saghul>indutny: replied
16:31:41  <indutny>gotcha
16:31:53  <indutny>that's what BSD says
16:31:54  <indutny> [ENXIO] O_NONBLOCK and O_WRONLY are set, the file is a FIFO, and no process has it open for reading.
16:32:00  <indutny>also
16:32:00  <indutny> [ENXIO] The named file is a character-special or block-special file and the device associated with this special file does not exist.
16:32:25  * Kakerajoined
16:33:32  <saghul>fun
16:33:34  <indutny>yes
16:33:43  <indutny>could you please check that it returns it?
16:33:44  <indutny>:)
16:34:23  <indutny>oh
16:34:24  <indutny> ENXIO O_NONBLOCK | O_WRONLY is set, the named file is a FIFO and no
16:34:24  <indutny> process has the file open for reading. Or, the file is a
16:34:24  <indutny> device special file and no corresponding device exists.
16:34:25  <indutny>http://man7.org/linux/man-pages/man2/open.2.html
16:34:36  <saghul>http://pubs.opengroup.org/onlinepubs/9699919799/
16:35:05  * dshaw_quit (Ping timeout: 265 seconds)
16:35:14  <tjfontaine>bradleymeck: ya, coworker was going to be writing up what we had talked about, I'll check on the status of that
16:35:17  <saghul>actually, this one: http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/errno.h.html
16:35:42  <indutny>ok
16:35:49  <indutny>not really informative :)
16:36:13  <indutny>saghul: force pushed
16:36:42  <saghul>indutny: lgtm
16:37:12  <indutny>thanks
16:37:28  <MI6>joyent/libuv: Fedor Indutny v0.10 * 4f72f21 : error: add ENXIO for O_NONBLOCK FIFO open() - http://git.io/0Awgvw
16:37:51  <saghul>I guess we also need in in master?
16:38:14  <saghul>indutny: can you pl check https://github.com/joyent/libuv/pull/1179
16:38:47  * AlexisMochajoined
16:38:49  * AvianFlujoined
16:39:01  <tjfontaine>what's confusing here is why this didn't at least throw an unknown exception
16:39:19  <indutny>saghul: doing a merge
16:39:36  <saghul>kewl
16:39:37  <indutny>I have to trust you here
16:39:37  <indutny>:)
16:39:44  <indutny>LGTM, but please wait until a merge
16:40:46  <saghul>sure
16:44:42  * rosskjoined
16:46:19  <indutny>ok, land it :)
16:46:22  <MI6>joyent/libuv: Fedor Indutny master * 9aa4831 : Merge remote-tracking branch 'origin/v0.10' (+6 more commits) - http://git.io/VRze6A
16:48:22  <tjfontaine>call in 12 mins?
16:48:31  <tjfontaine>indutny, trevnorris, AlexisMocha, daviddias
16:48:37  <indutny>oh, a call
16:48:37  <indutny>ok
16:49:05  <daviddias>heey
16:49:07  <daviddias>yep
16:49:07  <daviddias>:)
16:49:10  <tjfontaine>indutny: if I do a libuv 0.11 release want to merge it?
16:49:40  <AlexisMocha>yep
16:49:48  <AlexisMocha>for the call :)
16:50:02  * calvinfojoined
16:50:28  * mikealjoined
16:50:54  * AWinterm_joined
16:51:43  <indutny>tjfontaine: hm... yeah
16:52:16  <saghul>indutny: it should just work (tm), I tried it yesterday
16:52:26  <saghul>just remember that uv_cwd patch
16:52:41  <indutny>ok
16:53:42  <tjfontaine>fuck you
16:53:44  <tjfontaine>jenkins
16:54:06  * benviejoined
16:54:18  <tjfontaine>maybe it's azure or something, whatever the fuck is econnreset'ing these slaves is annoying as piss though
16:55:04  * mikealquit (Ping timeout: 252 seconds)
16:55:30  * rosskquit (Remote host closed the connection)
16:56:02  * rosskjoined
16:57:56  * bradleymeckquit (Quit: bradleymeck)
16:59:05  * AWinterm_quit (Remote host closed the connection)
17:00:18  * rosskquit (Ping timeout: 240 seconds)
17:00:31  <indutny>so
17:00:32  <indutny>a call?
17:00:52  <tjfontaine>wtf google, keeps saying I'm not allowed to join this call
17:00:56  <isaacs>indutny: fixed registry/netmask
17:00:58  <tjfontaine>YES I AM GOOGLE, YES I AM
17:00:58  <LOUDBOT>EXCEPT FOR WHERE THEY BELIEVE IN WEIRD SHIT
17:01:09  <indutny>isaacs: thanks man
17:01:11  <indutny>what was that?
17:01:12  <isaacs>indutny: looks like there was a problem fetching the tarballs from manta when it was replicating
17:01:19  <indutny>oh gosh
17:01:25  <indutny>isaacs: you want believe me how I have fixed it :)
17:01:26  <isaacs>indutny: the fullfatdb replicator doesn't put versions into the db that have unfetchable tarballs
17:01:37  * AWintermanjoined
17:01:41  <indutny>I have purged couchdb doc
17:01:44  <AlexisMocha>tjfontaine: we probably need a new link
17:01:44  <isaacs>indutny: i fixed it by clicking "save" on skimdb futon so that it would re-fetch
17:01:46  <indutny>and did _bulk_upload
17:02:01  <indutny>hopefully, it has replicated back from fullfat
17:02:01  <isaacs>indutny: gosh
17:02:06  <indutny>yeah
17:02:12  <indutny>otherwise replication would have been broken
17:02:18  <tjfontaine>ok link is https://plus.google.com/hangouts/_/calendar/dGpmb250YWluZUBub2RlanMub3Jn.7o3iugm5mbjqj81eb56fsoj118
17:02:20  <indutny>thanks anyway
17:02:25  <isaacs>indutny: here's a trick: when that happens, try "npm star $pkg"
17:02:25  <tjfontaine>had to log out and in 3 times
17:02:32  <indutny>ok
17:02:34  <isaacs>indutny: (with the "real" registry)
17:02:36  <indutny>tjfontaine: not allowed
17:02:45  <indutny>to join call
17:02:52  <isaacs>indutny: that'll cause a _rev bump which'll make the replicator see it again
17:02:57  * isaacsis available for call, same invite?
17:03:00  <indutny>thanks for a tip
17:03:20  <tjfontaine>indutny: email invite sent to team@nodejs.org does that work?
17:03:22  <daviddias>tjfontaine: it says that I'm not allowed
17:03:30  * isaacsalso not allowed..
17:03:31  <isaacs>:(
17:03:49  <AlexisMocha>not working for me either
17:03:50  <daviddias>tjfontaine: wants the hangouts just for him
17:04:14  <tjfontaine>if you guys go to google calendar and click the link, does it let you in?
17:04:27  <tjfontaine>there is something crazy going on here in google land
17:04:29  <AlexisMocha>nope
17:04:46  <AlexisMocha>what's your google+ email address? could try to invite on the fly
17:04:54  <tjfontaine>I'd rather bail to skype at this point
17:05:09  <daviddias>want to try some webrtc goodness?
17:05:12  <daviddias>talky.io/nodeteam
17:05:26  <indutny>let's give it a try :)
17:05:49  <tjfontaine>ya, trying
17:05:52  * dap_joined
17:06:06  <othiym23>oh man, good luck guys
17:06:12  * petka_quit (Quit: Connection closed for inactivity)
17:06:21  <othiym23>the last time I tried to get more than 3 people in Talky simultaneously it was... entertaining
17:06:36  <tjfontaine>AlexisMocha: are you able to hit that url?
17:06:47  <AlexisMocha>talky.io/nodeteam ?
17:06:56  <AlexisMocha>yes, but nothing happens other that I can play with a ship
17:07:40  <tjfontaine>what's your skype name AlexisMocha?
17:07:42  <AlexisMocha>supposedly i'm in a call
17:07:45  <AlexisMocha>alexis.campailla
17:08:55  <indutny>hahaha
17:09:05  <indutny>tjfontaine: I forgot that I'm on 3G connection
17:09:12  <indutny>and hosting conversation is really a bad idea
17:09:13  <indutny>:)
17:09:21  <trevnorris>morning.
17:10:26  <isaacs>AlexisMocha: Alexis Campailla can’t be added to this conversation due to his/her privacy settings
17:10:31  <isaacs>AlexisMocha: let us skype you!
17:11:03  <tjfontaine>trevnorris: skype day
17:11:12  <MI6>joyent/libuv: Saúl Ibarra Corretgé master * 4c47fcd : freebsd: use accept4, introduced in version 10 - http://git.io/9SH9NA
17:11:18  <AlexisMocha>try again
17:11:24  <saghul>indutny: bam! ^^
17:11:41  <trevnorris>tjfontaine: you going to use skype?
17:12:18  <dap_>indutny: you said it exactly :) dynamic translators are the idea of putting the translator logic into DOF in the binary, like we do the ustack helper.
17:12:26  <indutny>yeah
17:12:32  <dap_>indutny: apparently I talked with Adam Leventhal about it at dtrace.conf in 2012: http://www.youtube.com/watch?v=CqLcj0lVnp4
17:12:33  <indutny>but I'm too lame to do it
17:12:36  <indutny>:)
17:12:37  <dap_>ha!
17:12:40  <indutny>perhaps you too :)
17:12:40  <dap_>So's everyone else :)
17:13:10  <isaacs>AlexisMocha: nope
17:13:19  <dap_>I almost tried in 2011, but Bryan talked me into doing the ustack helper instead.
17:13:35  <othiym23>dap_: "talked", or "tricked"?
17:13:42  * mikolalysenkojoined
17:13:46  <isaacs>AlexisMocha: Alexis Campailla can’t be added to this conversation due to his/her privacy settings
17:13:49  <isaacs>AlexisMocha: nope
17:13:56  <dap_>othiym23: I think I was tricked into talking :)
17:14:02  <othiym23>heh
17:14:06  <tjfontaine>AlexisMocha: we'll catch up in a second
17:14:36  <AlexisMocha>ok, go ahead
17:14:57  <indutny>dap_: good that you did it though
17:14:58  <indutny>:)
17:15:15  <indutny>I just written a translator for bud
17:15:16  <indutny>https://github.com/indutny/bud/blob/master/src/bud.d
17:15:22  <indutny>and it is using int64 pointers
17:15:29  <indutny>so I thought node could benefit from it
17:15:34  <indutny>but legacy...
17:15:35  <indutny>:)
17:15:42  * TooTallNatejoined
17:16:24  <dap_>Yeah, it would have been nice to clean that up
17:16:32  <AlexisMocha>isaacs: I changed my privacy settings
17:16:56  <tjfontaine>added
17:17:47  * eugenewarejoined
17:19:44  <trevnorris>freaking pulseaudio. ok, got that working.
17:19:54  <trevnorris>tjfontaine: how we meeting up today?
17:20:54  <tjfontaine>skype at the moment
17:21:27  <trevnorris>bummer. don't think i'll be able to join. let me see if skype will magically play nicely w/ linux
17:21:59  * eugenewarequit (Ping timeout: 240 seconds)
17:23:52  <trevnorris>wow, there is magic in the universe.
17:24:21  * janjongboomjoined
17:24:35  <trevnorris>tjfontaine: do you need to send me an invite or something? don't use skype so not quite sure how it works to join a meeting
17:25:00  * janjongboomquit (Client Quit)
17:25:20  * dshaw_joined
17:33:18  * bradleymeckjoined
17:34:08  * dshaw_quit (Ping timeout: 252 seconds)
17:39:07  * dshaw_joined
17:44:56  * brsonjoined
17:56:22  * abraxasjoined
17:57:19  <indutny>dap_: what do you think about deterministic loops in DTrace D scripts?
17:57:33  <indutny>like for (i in [1..10])
17:57:39  <indutny>or whatever syntax it has in D
17:57:52  <indutny>this should bring down size of ustack helper in node
18:01:05  * abraxasquit (Ping timeout: 252 seconds)
18:01:57  * mmaleckiquit (Read error: Connection reset by peer)
18:02:12  * mmaleckijoined
18:02:25  <indutny>gosh
18:02:30  <indutny>my laptop has burned all air in a car
18:02:33  <indutny>need to open window
18:02:52  <indutny>ok, ttyl guys
18:03:01  <dap_>indutny: That's a good question for Bryan. But I don't think it helps. You'd need an OS update to get that feature, so if we made that change to the helper, the helper would stop working on all systems that weren't updated. And if you're willing to upgrade in order to make the big ustack helper work, then we could just change only the dof maxsize tunable. :)
18:14:23  * andrewrkquit (Quit: ZNC - http://znc.in)
18:17:18  * xaqjoined
18:17:42  * eugenewarejoined
18:18:07  * xaq_joined
18:19:09  * dshaw_quit (Quit: Leaving.)
18:21:38  * xaqquit (Ping timeout: 240 seconds)
18:22:01  * eugenewarequit (Ping timeout: 240 seconds)
18:22:26  * sblomjoined
18:22:49  * mikolalysenkoquit (Ping timeout: 240 seconds)
18:28:17  * mikolalysenkojoined
18:31:12  * etruscottjoined
18:36:09  * AlexisMochaquit (Read error: Connection reset by peer)
18:58:28  * mcavagejoined
18:59:52  * hzjoined
19:10:18  * kpdecker1quit (Ping timeout: 240 seconds)
19:14:39  * eugenewarejoined
19:14:58  * mikolalysenkoquit (Ping timeout: 240 seconds)
19:16:21  * defunctzombie_zzchanged nick to defunctzombie
19:18:39  * eugenewa_joined
19:18:58  * eugenewarequit (Ping timeout: 240 seconds)
19:21:20  * eugenewa_quit (Read error: Connection reset by peer)
19:23:06  * felixgequit (Quit: felixge)
19:23:47  * bradleymeckquit (Quit: bradleymeck)
19:28:13  * bradleymeckjoined
19:31:01  * kpdeckerjoined
19:34:15  * eugenewarejoined
19:37:12  <othiym23>tjfontaine: I made pretty good progress backporting 0.11's Agent to earlier versions, but in the process found what I'm reasonably sure is a bug
19:37:45  <othiym23>tjfontaine: pretty sure that https://github.com/joyent/node/blob/master/lib/_http_agent.js#L87 should have "count > self.maxSockets" instead of "count >= self.maxSockets"
19:38:39  <othiym23>newly-created sockets are shift()ed onto the sockets list, so count will always be at least maxSockets when maxSockets is set to 1
19:39:08  <othiym23>and you can do the lil inductive proof necessary to show yourself that this will always lead to cases where the socket doesn't get put on the free list appropriately
19:39:32  <othiym23>I'll take a look at the tests later and file an issue / PR if this just isn't some weird difference between 0.11 and 0.10
19:40:24  <othiym23>looks like this bug came in a commit of isaacs's, so maybe he knows if the current behavior is intentional or an oversight
19:42:17  <othiym23>yeah, looks like he introduced the bug when he added maxFreeSockets support
19:42:22  * benvie_joined
19:42:28  <othiym23>OK, when I get a few minutes I'll put together a PR
19:43:48  * benviequit (Ping timeout: 252 seconds)
19:43:59  * eugenewa_joined
19:47:56  * eugenewarequit (Ping timeout: 265 seconds)
19:51:23  * daviddiasquit (Remote host closed the connection)
19:51:45  * defunctzombiechanged nick to defunctzombie_zz
19:56:24  * kpdeckerquit (Quit: Leaving.)
19:56:41  * kpdeckerjoined
19:57:32  * abraxasjoined
19:58:47  * [m76]joined
20:01:51  * daviddiasjoined
20:01:59  * abraxasquit (Ping timeout: 264 seconds)
20:02:12  * daviddiasquit (Remote host closed the connection)
20:02:28  * eugenewarejoined
20:02:41  * daviddiasjoined
20:03:09  * eugenewarequit (Remote host closed the connection)
20:03:27  * Domenic_quit (Ping timeout: 265 seconds)
20:03:28  * rjequit (Excess Flood)
20:03:28  * eugenewa_quit (Read error: No route to host)
20:03:28  * rjejoined
20:03:28  * m76quit (Ping timeout: 265 seconds)
20:03:35  * Domenic_joined
20:04:23  * gounkojoined
20:07:59  <indutny>dap_: well, I just don't enjoy 7mb DIF code
20:08:07  <indutny>and distributing it over the wire
20:08:35  <indutny>we could introduce loops and in a couple of years it'll be default
20:11:52  * kpdecker1joined
20:13:38  * kpdeckerquit (Ping timeout: 240 seconds)
20:15:42  * kpdeckerjoined
20:16:23  * kpdecker1quit (Ping timeout: 264 seconds)
20:17:25  * rphillipsquit (Ping timeout: 265 seconds)
20:17:54  * nifocquit (Ping timeout: 265 seconds)
20:18:50  * wolfeidaujoined
20:19:06  <othiym23>indutny: added a less cryptic explanation of that agent bug ;)
20:19:22  <indutny>othiym23: haha, and test? ;)
20:19:48  <othiym23>test will come after I get things wrapped up for the thing I'm working on for New Relic
20:19:56  <indutny>ok
20:19:57  <othiym23>but it'll come!
20:20:06  <indutny>I'll take a look at it by then, if you don't mind :)
20:20:17  <othiym23>that's fine
20:20:36  <othiym23>not really sure how to test it, but I'll figure something out
20:20:38  * rphillipsjoined
20:20:56  * nifocjoined
20:21:24  * dshaw_joined
20:21:29  * etruscottquit (Quit: etruscott)
20:23:41  * mikolalysenkojoined
20:25:29  * xaq_quit (Remote host closed the connection)
20:28:25  * gounkoquit (Ping timeout: 240 seconds)
20:28:57  * eugenewarejoined
20:36:38  * dschatzbergjoined
20:38:54  <dschatzberg>hi all, I was looking at unix/getaddrinfo.c and was confused by the use of char* buf. It seems like it is malloced and populated but never used or freed. Am I wrong?
20:40:52  * calvinfoquit (Quit: Leaving.)
20:42:45  * etruscottjoined
20:48:43  <indutny>dschatzberg: I think you are right
20:48:48  <indutny>we are leaking it
20:49:02  <indutny>ah, one sec
20:49:11  <indutny>no, we don't
20:49:33  <indutny>we are assigning it to req->hints or req->service or req->hostname
20:49:37  <indutny>and freeing them above
20:49:41  <indutny>in that order
20:49:56  <dschatzberg>oh I see, didn't catch that
20:52:54  * calvinfojoined
20:54:39  * dschatzbergpart ("Leaving")
20:54:45  <trevnorris>tjfontaine: have to say, that beforeExit event is actually coming in handy running multiple tests that independently depend on asynchronous events firing.
21:01:39  * paulfryzelquit (Write error: Broken pipe)
21:09:22  * dshaw_quit (Quit: Leaving.)
21:22:37  * etruscottquit (Quit: etruscott)
21:26:29  * paulfryzeljoined
21:27:04  * parshapjoined
21:33:34  * jmar777quit (Remote host closed the connection)
21:39:45  * defunctzombie_zzchanged nick to defunctzombie
21:40:32  * hzquit
21:42:36  * AWintermanquit (Remote host closed the connection)
21:42:48  * dshaw_joined
21:43:09  * defunctzombiechanged nick to defunctzombie_zz
21:43:38  * AWintermanjoined
21:43:42  * etruscottjoined
21:43:49  * AWintermanquit (Remote host closed the connection)
21:47:13  * wolfeidauquit (Ping timeout: 240 seconds)
21:47:15  * defunctzombie_zzchanged nick to defunctzombie
21:53:24  <indutny>haha!
21:53:52  <tjfontaine>ok I'm back
21:53:56  <tjfontaine>beforeExit is very scary
21:54:12  <tjfontaine>even though some of it was my idea
21:54:39  <othiym23>having a way to grab crash dumps and / or make a hail mary pass to a network service on processes going down is pretty handy
21:55:47  <tjfontaine>maybe, I expect to see an influx of all sorts of "my node app is hung" type reports from people not realizing they're constantly reviving the loop
21:56:52  <othiym23>alias kill="kill -9"
21:57:00  <othiym23>(o_o)-b
21:57:20  <tjfontaine>:)
21:58:25  * abraxasjoined
22:00:33  <indutny>tjfontaine: they can always do it :)
22:00:39  <indutny>I'm more scared of thread-pool changes
22:00:50  <indutny>especially that it is joining threads
22:00:51  <tjfontaine>what tpool change? :)
22:00:55  <tjfontaine>ah ok
22:00:57  <indutny>at the end of loop
22:00:57  <indutny>yeah
22:01:05  <indutny>users could easily lock it up
22:02:07  <trevnorris>tjfontaine: you know how to dump the gdbjit symbols so they can be used while viewing a core file after the fact?
22:02:23  <tjfontaine>not that this makes it better, but I'd really like to start considering getting separate thread pools (either together with, or in lieu of scaling thread pools)
22:02:36  <trevnorris>i tried gdbjit-dump, but that seems more for debugging gdbjit itself.
22:02:57  * abraxasquit (Ping timeout: 244 seconds)
22:02:59  * Kakeraquit (Ping timeout: 269 seconds)
22:03:08  <tjfontaine>trevnorris: no, I don't really use the gdb stuff because I don't want to get too used to having something that I won't have access to when debugging in production
22:03:31  <trevnorris>fair enough.
22:03:44  <tjfontaine>trevnorris: fwiw you can upload that core and node binary into manta and inspect them after the fact
22:03:45  * beetlesjoined
22:03:51  <tjfontaine>not if it's a debug build though
22:06:47  * etruscottquit (Quit: etruscott)
22:07:34  * AWintermanjoined
22:08:05  <othiym23>I'm going to be publishing Yet Another Keep-Alive Agent™ that's mostly just a backport of what's on node master
22:08:49  <othiym23>it's got some shimming for e.g. unreffing / re-reffing of sockets in 0.10 / not blowing up under 0.8
22:09:07  <tjfontaine>mcavage: ^^
22:09:20  <tjfontaine>exciting times at node.js high
22:09:23  <othiym23>and one feature I'm stealing from agentkeepalive (socket timeout, so I can have pooled keepalive sockets that will naturally time out under 0.8)
22:09:38  <othiym23>anybody have any good ideas for a name? ;)
22:09:40  <tjfontaine>real keep alive -- what a concept
22:10:05  <tjfontaine>how are you unref'ing in 0.8? you're not using the builtin res.setTimeout
22:10:32  <othiym23>tjfontaine: did you see #7260?
22:10:42  <tjfontaine>I did, that seems like a legit bug indeed
22:10:57  <othiym23>I think I know how I'll test it
22:11:00  <othiym23>I'll try to get that in tonight
22:11:11  * AWinterm_joined
22:11:19  <tjfontaine>ya, with a test that will land in a heart beat
22:11:30  <othiym23>tjfontaine: no unref, just setTimeout
22:11:38  <tjfontaine>ah ok
22:11:44  * AWintermanquit (Ping timeout: 244 seconds)
22:12:00  <tjfontaine>I was going tos ay, because there is some concepts around unref'ing sockets in 0.8, but not entirely
22:12:11  <tjfontaine>but there's 0 concept of unrefing a timer
22:12:32  <othiym23>well, I need to get soemthing ready to ship pretty fast, but we can come back to this
22:12:53  <tjfontaine>nod
22:13:00  <othiym23>I'd just use agentkeepalive if it didn't have the same bug as master and it didn't DISABLE NAGLE'S ALGORITHM BY DEFAULT @_@
22:13:20  <tjfontaine>it may be that you just have shorter timeouts in 0.8
22:15:13  <trevnorris>tjfontaine: does compiling w/ --gdb jut make the exe bigger?
22:15:21  <trevnorris>or are there any perf implications?
22:15:47  * beetlesquit (Ping timeout: 264 seconds)
22:15:49  <tjfontaine>I'd have to look more closely, but I'm sure it comes with a certain amount of perf hit -- though I can't quantify it having never done it
22:16:07  <tjfontaine>I can imagine it pinning functions to specific places and maybe even dynamically generating dwarf
22:16:35  * [m76]quit (Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/)
22:16:38  <trevnorris>ooh. that sounds awesome.
22:16:48  * m76joined
22:17:12  <tjfontaine>the definitions of the flags themselves are scary as hell
22:17:36  <trevnorris>you mean of --gdbjit and the like?
22:17:43  <tjfontaine>yes
22:17:49  <tjfontaine>from a performance aspect
22:17:54  <mcavage>tjfontaine: that's cool -- so I can bring YAKAA into restify?
22:17:59  <mcavage>and kill the other one off?
22:18:03  <trevnorris>hehe yeah.
22:18:11  <tjfontaine>mcavage: yup. and have faith that it works like master does
22:18:24  <mcavage>i'm going to party like it's 2007
22:18:36  <tjfontaine>god forbid node have a working http client ;)
22:18:39  <tjfontaine>"working"
22:19:29  <othiym23>I'm probably just going to gank the test/simple tests for the agent off master as well, just so there's *some* test coverage
22:19:38  <othiym23>I'm having to do all this more hastily than I'd like
22:19:59  <tjfontaine>trevnorris: a lot of this seems like extra annotation around these code paths, I don't se (yet) any thing that disable specific optimization passes
22:20:15  * paulfryzelquit (Read error: Connection reset by peer)
22:20:21  <trevnorris>cool
22:20:26  * paulfryzeljoined
22:20:40  <tjfontaine>that's not to say it doesn't have an effect, because if it was a noop they would just do it all the time
22:22:22  * rendarquit
22:22:35  <tjfontaine>trevnorris: looking at gdb-jit.cc it seems like they are indeed dynamically generating dwarf
22:24:11  <trevnorris>interesting.
22:25:14  <tjfontaine>oh hmmmmm
22:25:17  <tjfontaine>I wonder actually
22:26:46  <othiym23>tjfontaine: BTW, using the complete toString of the 'ca' property for the TLS socket name means that the names (just the names) of the sockets are somewhere around 80KiB for New Relic
22:26:57  <tjfontaine>sigh.
22:27:13  <othiym23>NR basically packages its own CA store for portability reasons
22:27:18  <tjfontaine>of course
22:27:27  <tjfontaine>please file that as an issue as well
22:27:40  <othiym23>will do, enthusiastically
22:27:47  <othiym23>it makes debugging very exciting
22:28:14  <othiym23>nothing makes me feel like I'm hacking the gibson like enormous screensful of text scrolling past every time I do anything
22:32:38  <mmalecki>see othiym23, I just have openssl compilation running in a while loop in one of my windows at all times
22:37:03  <othiym23>you must be at one with the matrix now, mmalecki
22:37:06  * m76quit (Read error: Connection reset by peer)
22:37:51  <tjfontaine>indutny: no test for 7263? :)
22:42:07  * paulfryzelquit (Read error: Connection reset by peer)
22:42:37  <othiym23>it's gonna be so sweet / such a trainwreck when WeakMaps land in V8 without a flag
22:42:41  * paulfryzeljoined
22:42:53  <othiym23>we'll be able to do all kinds of terrible and inappropriate things so much more easily
22:46:16  <othiym23>mcavage, tjfontaine: I'm stealing the name yakaa
22:46:23  <othiym23>speak now or forever hold your peace
22:47:03  <mcavage>othiym23: I grant it to you royalty free!
22:47:28  <othiym23>sweet! we can discuss your honorarium at a later date
22:48:06  <mcavage>that's fine. I need to time to commission the right artist anyway.
22:54:28  <indutny>tjfontaine: 7263?
22:54:45  <indutny>I think I added test
22:54:50  <indutny>see `-servername` argument
22:54:53  <indutny>it crashes right now
22:56:17  * wolfeidaujoined
23:04:40  <indutny>tjfontaine: what do you think about moving
23:04:52  <indutny>DEFAULT_CIPHERS and DEFAULT_ECDH_CURVE
23:04:55  <indutny>to crypto.js
23:05:56  * eugenewarequit (Remote host closed the connection)
23:06:09  * daviddiasquit (Remote host closed the connection)
23:06:29  * eugenewarejoined
23:09:16  <mmalecki>tjfontaine: yo. care to make me an op in #node.js? I believe there's no op usually awake on European mornings. if there is, disregard :)
23:10:50  * eugenewarequit (Ping timeout: 252 seconds)
23:15:22  <trevnorris>i still have that panda stress ball from nodeconf.eu. think i've squeezed it's little panda eyeballs out.
23:15:38  <AvianFlu>those things were pretty awesome
23:15:50  <AvianFlu>I think I threw mine at my cat a few too many times and it wound up torn apart
23:15:55  <othiym23>I think after last week's presentation that's part of your Personal Brand™, trevnorris
23:15:57  <indutny>mmalecki: done
23:15:59  <trevnorris>yeah. always have it in my laptop bag. use it a dozen times a day
23:16:28  <trevnorris>othiym23: heh.
23:16:31  <trevnorris>groundwater: ping
23:16:40  <groundwater>trevnorris PONG
23:17:47  * benviejoined
23:18:01  <mmalecki>indutny: thanks
23:18:03  <trevnorris>groundwater: on 7145, check out the add/removable ability in this test: https://github.com/trevnorris/node/blob/al-pass-provider/test/simple/test-asynclistener-multi-context.js
23:18:07  * rosskjoined
23:18:35  <indutny>I wonder how I could persiste it
23:18:36  <indutny>mmalecki: ^
23:18:37  <mmalecki>indutny: tjfontaine still need recurring (ChanServ) OP I think
23:18:39  <trevnorris>mainly the last one. there's one kink i'm about to finish but then it'll be working
23:18:44  <indutny>mmalecki: I did it too
23:18:54  <indutny>well
23:18:58  <indutny>at least it wasn't erroring
23:21:18  * benvie_quit (Ping timeout: 240 seconds)
23:21:20  * eugenewarejoined
23:23:05  <trevnorris>indutny: how does the 3.24 upgrade look? I haven't been keeping track of it.
23:23:12  <indutny>good
23:23:16  <indutny>waiting for one CL to get landed
23:23:42  <trevnorris>wow, feel like a noob asking this. CL?
23:25:53  <indutny>https://codereview.chromium.org/177243012/
23:26:14  <trevnorris>no, sorry. what does CL stand for?
23:26:20  <indutny>actually, I don't know what it stands for :D
23:26:22  <indutny>tjfontaine: do you?
23:26:35  <indutny>ah
23:26:36  <indutny>Change List
23:26:39  <indutny>so simple
23:26:43  <trevnorris>ah ok. cool.
23:26:48  * mikolalysenkoquit (Ping timeout: 240 seconds)
23:29:14  <MI6>joyent/node: Ben Noordhuis v0.10 * 6bd78fd : deps: fix v8 valgrind warning - http://git.io/zMpoCw
23:30:34  <bradleymeck>it lives!
23:30:45  <indutny>?
23:31:30  <bradleymeck>tjfontaine: wanted a module loader extension to be able to read archives
23:31:38  <trevnorris>groundwater: pung. that code is a little obtuse. what it's trying to demonstrate is that each branch of an async call stack is independent of all others.
23:32:01  <bradleymeck>need to make it work w/ nested archive though...
23:32:12  * dranisimovjoined
23:32:13  <trevnorris>groundwater: so you can conditionally remove a listener without affecting any of the other calls.
23:32:51  * seldoquit (Read error: Connection reset by peer)
23:33:13  * calvinfoquit (Quit: Leaving.)
23:35:07  * mikolalysenkojoined
23:36:33  * bradleymeckquit (Quit: bradleymeck)
23:39:27  * seldojoined
23:40:01  * WalrusPonyquit (Ping timeout: 240 seconds)
23:42:27  * benvie_joined
23:43:58  * inolen1joined
23:44:24  * defunctzombiechanged nick to defunctzombie_zz
23:44:27  * WalrusPonyjoined
23:44:29  * benviequit (Ping timeout: 240 seconds)
23:46:23  * inolenquit (Ping timeout: 264 seconds)
23:48:29  * rosskquit (Remote host closed the connection)
23:49:03  * rosskjoined
23:52:16  * wolfeidauquit (Remote host closed the connection)
23:52:24  * AvianFluquit (Remote host closed the connection)
23:53:29  * rosskquit (Ping timeout: 240 seconds)
23:54:53  * rosskjoined
23:56:16  * paulfryzelquit (Remote host closed the connection)
23:57:24  * wolfeidaujoined
23:57:47  * wolfeidauquit (Remote host closed the connection)
23:59:16  * abraxasjoined
23:59:48  * mikolalysenkoquit (Ping timeout: 252 seconds)