00:00:01  * ircretaryquit (Remote host closed the connection)
00:00:09  * ircretaryjoined
00:00:33  <trevnorris>groundwater: so, all you're saying there is that you want to know when a new event is created?
00:00:46  <trevnorris>and know just before a callback is fired?
00:01:57  <trevnorris>man. this is so freakishly simple compared to getting AL to automatically continue its context chain.
00:02:07  <groundwater>trevnorris basically, also there are after events
00:02:22  * jmar777quit (Remote host closed the connection)
00:02:37  <trevnorris>yeah, and error, right?
00:03:13  <groundwater>trevnorris i forget what we were thinking for errors, i'd double check with the notorious tjfontaine
00:03:25  <trevnorris>but anyways. it's possible that, say an event happens before the EVENT_CREATE is turned on. then context._continuation wouldn't exist when EVENT_BEFORE fires
00:04:59  <trevnorris>groundwater / tjfontaine: is there a rough document about what the current API is?
00:05:22  <tjfontaine>not a document, so much as lib/tracing.js
00:05:28  <trevnorris>ok
00:05:44  <tjfontaine>so I'm finishing up here, and will be able to catch up and do the fucking releases :/
00:06:02  <tjfontaine>after I get the releases out I'll get a PR with lib/tracing.js that includes ben's v8 work
00:06:05  <tjfontaine>jsut that work
00:06:11  <tjfontaine>and you can rubberstamp that
00:06:16  <trevnorris>tjfontaine: i'm sorry dude. seriously feel like 30% of your time is for coding. the rest is for everything else that I'd hate to touch :P
00:06:19  <tjfontaine>and we can get you to move AL to that interface
00:06:44  <trevnorris>now that my head is wrapping around this more we need to have more of a discussion
00:07:03  <tjfontaine>ok, discussions can be had, do you want to do in virtual or meat space
00:07:40  <trevnorris>hm... this might be hard. thursday i'm in meetings all afternoon. friday i get lasik in the morning.
00:07:56  <trevnorris>then the following week i'm at netflix.
00:08:12  <tjfontaine>will you be able to discuss this weekend?
00:08:28  <trevnorris>still won't be able to drive from the lasik
00:08:49  <tjfontaine>but we can do a hangout/skype
00:09:10  <trevnorris>the gist of it is, i had in my head that these tracing probes would work like AL, where they continue on the async execution chain
00:09:21  <trevnorris>but groundwater told me they're mot like a switch that you turn on and off
00:09:25  <trevnorris>is that right?
00:10:56  <trevnorris>groundwater: in your script I see .onAsync(), but that method doesn't exist in lib/tracing.js
00:11:02  <trevnorris>i'm confused, :(
00:11:44  <trevnorris>tjfontaine: so, yeah, are probes things that you just switch on and get alerted by, then switch off and stop getting alerts?
00:11:48  <groundwater>trevnorris that script worked against the commit hash i pasted to you
00:11:55  <trevnorris>erm probes meaning tracing as a whole
00:11:59  <trevnorris>ah ok
00:12:10  <trevnorris>groundwater: so not against https://github.com/joyent/node/pull/5940
00:12:34  <tjfontaine>https://github.com/joyent/node/pull/5940/files#diff-d057cd3202b378894d5ae0ba345d537dR234
00:12:49  <trevnorris>coolio
00:12:57  <trevnorris>oh. wait. ok.
00:13:03  <trevnorris>i'm looking at the first of your two commits
00:13:04  <trevnorris>heh
00:13:11  <tjfontaine>right
00:13:17  <trevnorris>tjfontaine: anyways. is that how those tracers are going to work?
00:13:26  * AvianFluquit (Ping timeout: 240 seconds)
00:13:27  <tjfontaine>so we're in bikeshed area, these two apis are not incompatible fwiw
00:13:52  <tjfontaine>they just need to twiddle the same flags for when providers/probes/events fire
00:14:04  <trevnorris>yeah. i'm just getting a flurry of pure awesomeness. haven't felt like this in a while.
00:14:37  <tjfontaine>I'm glad I guess :)
00:14:44  <groundwater>the probes have a very tight scope
00:14:53  <trevnorris>so. giving users the ability to say, hey I want to know when X happens starting now, and ending now
00:15:02  <trevnorris>it'll be so freakin easy
00:17:21  <trevnorris>tjfontaine: so, i'm going to ask that you create lib/tracing.js with only ben's work included.
00:17:36  <tjfontaine>yes I started on that last night
00:17:39  <trevnorris>but seriously. this will be so awesome i'm feeling giddy.
00:17:42  <trevnorris>ok cool.
00:18:13  <trevnorris>there are like dozens of things I want to spew out about what could be done.
00:19:55  <trevnorris>tjfontaine: ok. so... help me understand what's going on inside of node_dtrace, why it matters, how it's used, etc.
00:20:33  * brsonjoined
00:20:38  <tjfontaine>it's basically going ot be compiled for all platforms, but they'll be noops, and it will be trivial to make they never jump to it from js
00:21:34  <tjfontaine>we basically need that on platforms where dtrace/etw/systemtap is available so those static probes will fire
00:22:23  * tjkrusinskijoined
00:23:37  <trevnorris>tjfontaine: i mean. I see things like DTRACE_NET_STREAM_END, right? so all it's doing is telling c++ when something in js happens. correct?
00:24:28  * daviddiasjoined
00:25:20  <tjfontaine>ok, so these are static probes that we fire in js
00:25:32  <trevnorris>tjfontaine: heck. want to chat right now?
00:25:47  <tjfontaine>I can't right this second, hang on other channel
00:25:54  <trevnorris>cool
00:26:53  <trevnorris>tjfontaine: this is going to sound absurd, but a noop call to c++ is pretty much free. i mean, sub 10 ns call.
00:27:06  <trevnorris>so, in macro terms, it's pretty much free
00:27:09  <tjfontaine>good then I won't do anything to remove that? :)
00:27:25  <trevnorris>it all depends on how data is communicated.
00:27:57  * mikolalysenkojoined
00:28:10  <tjfontaine>so the flow is (as refactored) js -> c++ -> if enabled -> js to marshal arguments (where we have more state and easier access) ... return value used for systemtap/etw/dtrace
00:28:53  * daviddiasquit (Ping timeout: 245 seconds)
00:29:42  <trevnorris>but that's if the call is from js. in the case of MakeCallback then the path could be different.
00:30:10  <tjfontaine>I'm not able to give you my full cycles righ tnow, but I'm confused why makecallback has anything to do with it
00:30:15  <tjfontaine>it's using Function::Call directly
00:30:23  <trevnorris>tjfontaine: also, say we want to know when the server connection is closed. do we use the current, and imo incorrect, way of triggering that before it's actually closed, or do we wait for libuv close callback return?
00:30:51  <tjfontaine>we should discuss it, I am open to libuv close cb
00:31:15  * zz_karupanerurachanged nick to karupanerura
00:31:17  <tjfontaine>we should not conflate AL and our static probes though in our conversation at the moment
00:31:27  <trevnorris>yeah. forget AL
00:31:46  <tjfontaine>our static probes are surgical pieces we can sprinkle throughout our code
00:31:48  <trevnorris>i'm only thinking about the implementation details that pertain to AL that relate to the same parts
00:32:30  <trevnorris>pseudo code here:
00:36:00  <trevnorris>require('tracing').on('tcp', 'write', ...); I assume we want to trigger that on the #write() JS function. not when the callback is about to go out.
00:36:12  <trevnorris>groundwater: feel free to chime in. :)
00:36:40  <tjfontaine>trevnorris: but see that's my point you're conflating static probes and al
00:36:48  <tjfontaine>and why I called it provider_tcpwrap
00:36:49  <groundwater>trevnorris: i think while eventually we may merge the AL probes with teh static probes, we should keep them separate for now
00:36:51  <tjfontaine>and on_write
00:37:08  <groundwater>static probes should just fire wherever they're currently firing, on stack
00:37:56  <trevnorris>tjfontaine: no. i'm not confusing them. the static probe fires when the js function itself occurs. AL only fires when an async object is created, or when the return callback is about to be fires
00:37:58  <trevnorris>*fired
00:38:27  <trevnorris>AL doesn't give the ability to tell me that a specific method in TCP#* is being called
00:39:44  <trevnorris>so I get the two main differences between the two. which are:
00:39:47  * mikolalysenkoquit (Ping timeout: 272 seconds)
00:40:05  <trevnorris>1) probes are toggled while AL encapsulates an execution chain
00:41:00  <trevnorris>2) probes let me know when some mechanism of what i'm watching is about to fire, while AL only lets me know when it's been created, or when it's done (e.g. when the return callback is about to fire, etc.)
00:41:15  <trevnorris>and you can replace probes with tracing
00:42:57  <trevnorris>groundwater: take your examples. if you were to, tracing.onAsync('TCP', '*', 'EVENT_CREATE', ...) there are still a bunch of internal mechanisms that have to fire to keep the execution chain alive
00:43:35  <groundwater>trevnorris the probes don't have to keep track, i handle that myself
00:43:42  <trevnorris>groundwater: while just tracing.on('TCP', '*', ...) doesn't have to do squat internally other than check a few flags, because once it's off, it's off.
00:43:45  <trevnorris>yeah
00:44:19  <trevnorris>groundwater: wait, how do you mean, you'll handle that yourself?
00:45:12  <groundwater>the async probe has no concept of what execution chain it's in, it just fires always or never
00:45:45  <trevnorris>do you mean w/ the syntax you currently have in your script?
00:45:56  * mikolalysenkojoined
00:46:03  <groundwater>well, i mean a specific probe will either always fire, or never fire
00:46:13  <groundwater>it does not depend on what execution chain it's in
00:46:19  <groundwater>probes are context free
00:46:29  <trevnorris>yeah, I got that. :)
00:46:46  <groundwater>cool, just treading carefully
00:46:51  <groundwater>what's next?
00:47:16  <trevnorris>i'm writing up an example script to show what i'm talking about for an AL specific, and how it differs from probes
00:48:25  <trevnorris>groundwater: do you have an example using tracing.on() ?
00:48:39  <groundwater>trevnorris tracing.on() is just for static probes
00:48:44  <groundwater>dtrace stuff
00:48:48  <trevnorris>yes, I know. :)
00:48:56  <groundwater>okay, no i don't have an example
00:49:00  <trevnorris>ok
00:54:15  <trevnorris>groundwater: here: https://gist.github.com/trevnorris/8570711
00:54:27  <trevnorris>let me know if that doesn't make sense.
00:56:17  <groundwater>trevnorris so, the create callback should fire once in that example
00:56:41  <trevnorris>groundwater: crap. I forgot to have the server actually listen on a port
00:56:59  <groundwater>oh, even if server is listening and picking up connections
00:57:03  <trevnorris>groundwater: then say, I receive N connections, the setImmediate inside the connection callback will also fire
00:57:08  <groundwater>once the timer is removed, it never fires
00:57:30  <groundwater>trevnorris no, it should never fire again after tracing.removeAsync(al)
00:57:48  <trevnorris>yes. it should. hence why I used .onAsync() and not just .on()
00:58:33  <groundwater>addAsyncListener(...) behaves that way, but the probes don't
00:58:43  <groundwater>even the onAsync(...) probes
00:59:25  <trevnorris>ok... there's a disconnect. so let's go back and review to make sure we're both on the same page.
00:59:32  <trevnorris>man this stuff is confusing and all meta
00:59:46  <groundwater>sure
01:00:32  <trevnorris>there are "PROVIDERS" (e.g. tcp, upd, timers)
01:00:32  <trevnorris>there are "EVENTS" (e.g. create, before, after)
01:00:32  <trevnorris>there are "PROBES" (e.g. write, read, shutdown)
01:01:09  * mcavage_joined
01:01:30  <trevnorris>what i'm talking about is folding the process.*AsyncListener API into this. it solves a problem that i've been aware of since day 1
01:02:08  * mcavagequit (Read error: Connection reset by peer)
01:02:17  <trevnorris>AsyncListeners are only aware of PROVIDERS and EVENTS. It couldn't care about PROBES.
01:02:28  * kazuponjoined
01:02:51  * mikolalysenkoquit (Ping timeout: 272 seconds)
01:03:01  <trevnorris>one API that AsyncListeners are missing, and a part that I've always wanted in, is the ability to select which callback is fired for which PROVIDER in an execution chain.
01:03:13  <tjfontaine>sure
01:03:17  <tjfontaine>certainly
01:04:01  <tjfontaine>I'd like to keep AL focused on the C++ / JS transition, and fill in gaps with static probes where those probes don't fit for people writing monitoring agents
01:04:06  <tjfontaine>as a pure hypothetical situation
01:04:11  <trevnorris>so, in my mind, we can remove the process.*AsyncListener() API completely, and replace it w/ tracing.onAsync(<PROVIDER>, <EVENT>, callback)
01:04:52  <tjfontaine>I liek that idea, except you want _CREATE and _ERROR return values to mean something, so it's an interesting part of the bikeshed
01:05:17  <tjfontaine>what about the implementaton of .removeAL such taht paths aren't watched, if we go this route you don't have that context?
01:05:28  <trevnorris>real quick. does "bikeshed" just mean to shoot around different ideas?
01:05:34  <trevnorris>or does it mean something more than that?
01:06:03  <tjfontaine>it means we all get to argue over the color we want to paint the bikeshed
01:06:11  <trevnorris>haha. ok. got it. :)
01:06:16  <tjfontaine>some of these questions are more like: "should we build the bikeshed"
01:06:32  <trevnorris>tjfontaine: ok. check this gist: https://gist.github.com/trevnorris/8570711
01:07:18  <trevnorris>my color of the bikeshed should be that onAsync() still returns an AsyncListener, and even if you remove it then it'll continue to fire withing the wrapped execution chain.
01:07:36  <tjfontaine>hrm
01:08:08  <trevnorris>then leave the probe stuff to tracing.on()
01:08:26  * paulfryzelquit (Remote host closed the connection)
01:08:33  <trevnorris>i mean. technically, "create" could still be considered a "probe" right?
01:08:37  <groundwater>my suggestion is to just add two versions, and figure out which we like better over time
01:08:44  <groundwater>but i'm not implementing, nor am i the maintainer
01:10:20  <tjfontaine>I'm not opposed to some of this
01:11:16  <groundwater>yah i'll totally defer to ya'll here
01:12:08  * kazuponquit (Remote host closed the connection)
01:12:15  * kazuponjoined
01:12:33  <trevnorris>i know othiym23 would hate me, but i'm totally fine removing process.*AsyncListener(). it does sort of feel strange there.
01:12:47  * kellabytejoined
01:13:03  * kellabytequit (Changing host)
01:13:03  * kellabytejoined
01:13:03  * kellabytequit (Changing host)
01:13:03  * kellabytejoined
01:13:12  <othiym23>my assumption was that it was going to move to tracing.*AsyncListener() for 0.12
01:13:28  <tjfontaine>there may be some tweaks to what it looks liek is all
01:13:39  <trevnorris>tjfontaine: anyways. i'll do some sort of writeup about it all. there are several points that keep getting confused.
01:13:47  <tjfontaine>ok
01:14:00  <tjfontaine>you can email team@ I'd prefer that to gist, because I never get notified of that nonsense
01:14:10  <tjfontaine>absolutely use gist for code examples though :)
01:14:14  <trevnorris>heh, ok :)
01:14:49  <trevnorris>tjfontaine: seems the biggest thing is where the module "tracing" draws the line with its "probes"
01:15:02  * c4milo_quit (Remote host closed the connection)
01:15:43  <trevnorris>imo it gets complicated because triggering on the js side to some c++ thing can lead to incorrect results.
01:16:17  <trevnorris>e.g. closing a fd, where in js we trigger that it's been closed even though libuv hasn't had a chance to finish that operation.
01:16:35  <groundwater>tjfontaine trevnorris can you cc jacob/forrest@newrelic.com
01:17:03  <trevnorris>tjfontaine: so, if we just want to alert that some js function is running, then cool. but if we want some sort of accuracy in these then we'll have to bring down the level of abstraction.
01:17:04  <Ralith>vigith: pre-existing heap corruption
01:17:10  <trevnorris>groundwater: sure
01:17:12  <Ralith>oh wow I was way scrolled up
01:19:07  * daviddiasjoined
01:19:16  * c4milo_joined
01:21:02  <trevnorris>tjfontaine: final, and almost unrelated, thing. I don't like a lot of the de-macro-ification that's happened. e.g. utils.Is*() is useless to me in hot code paths. I hardly ever use it.
01:22:02  <trevnorris>imo macro-like things in js are fine and every time I write util.IsUndefiend() I throw-up a little.
01:22:29  <groundwater>i've gotta head out! ttyl
01:22:37  <trevnorris>see ya, and thanks :)
01:23:12  <trevnorris>tjfontaine: it also prevent me from doing the ToUint32() that I've had in my repo for months because that cost is ridiculous vs just doing x = x >>> 0; inline
01:23:26  * daviddiasquit (Ping timeout: 240 seconds)
01:24:08  <trevnorris>anyways. all I have for now. i'll get the rest written up w/ code examples. and i'll be looking for your tracing patch that includes ben's work. will be fun to mess around w/ that stuff.
01:24:22  * kazupon_joined
01:26:38  * kazupon__joined
01:27:40  * kazuponquit (Ping timeout: 252 seconds)
01:28:45  * kazupon_quit (Ping timeout: 245 seconds)
01:29:45  * mikolalysenkojoined
01:32:51  * dshaw_joined
01:36:09  * abraxasjoined
01:39:14  * Benviequit (Ping timeout: 264 seconds)
01:41:45  * TooTallNatequit (Quit: Computer has gone to sleep.)
01:50:22  * c4milo_quit (Remote host closed the connection)
01:51:34  * dshaw_quit (Quit: Leaving.)
01:52:21  * c4milojoined
01:59:01  * mikealquit (Quit: Leaving.)
02:01:27  * mcavage_quit (Remote host closed the connection)
02:05:12  * mikealjoined
02:07:26  * Ralithquit (Read error: Operation timed out)
02:07:33  * rmgquit (Read error: Connection reset by peer)
02:08:12  * rmgjoined
02:09:29  * paulfryzeljoined
02:12:08  * mikealquit (Quit: Leaving.)
02:13:45  * octetcloudquit (Ping timeout: 252 seconds)
02:13:53  * daviddiasjoined
02:15:33  * mikolalysenkoquit (Ping timeout: 245 seconds)
02:17:29  * paulfryzelquit (Ping timeout: 240 seconds)
02:18:28  * daviddiasquit (Ping timeout: 245 seconds)
02:30:21  * dshaw_joined
02:33:57  * c4miloquit (Remote host closed the connection)
02:35:25  * dshaw_quit (Ping timeout: 272 seconds)
02:37:05  * Ralithjoined
02:37:11  * mikolalysenkojoined
02:38:29  * tjkrusinskiquit (Ping timeout: 272 seconds)
02:55:11  * dshaw_joined
03:00:18  * dshaw_quit (Quit: Leaving.)
03:07:50  * mikolalysenkoquit (Ping timeout: 240 seconds)
03:08:12  * kpdeckerquit (Ping timeout: 252 seconds)
03:08:57  * daviddiasjoined
03:09:57  * kpdeckerjoined
03:11:41  * kazupon__quit (Remote host closed the connection)
03:13:09  * daviddiasquit (Ping timeout: 252 seconds)
03:14:30  * paulfryzeljoined
03:18:29  * paulfryzelquit (Ping timeout: 240 seconds)
03:45:42  * paulfryzeljoined
03:46:08  * dshaw_joined
03:46:59  * Benviejoined
03:49:59  * paulfryzelquit (Ping timeout: 240 seconds)
03:58:23  * dshaw_quit (Ping timeout: 272 seconds)
03:58:32  * tjkrusinskijoined
03:59:29  * jmar777joined
04:00:32  <tjfontaine>trevnorris: funny, macros in js make me rip my eyeballs out :)
04:00:50  <tjfontaine>anyway, finally at my desk to do the v0.10.25 release
04:03:37  * daviddiasjoined
04:04:22  * kazuponjoined
04:04:28  * AvianFlujoined
04:08:25  * daviddiasquit (Ping timeout: 272 seconds)
04:09:43  * tjkrusinskiquit (Ping timeout: 245 seconds)
04:11:48  * tjkrusinskijoined
04:13:08  <tjfontaine>jesus libuv build for v0.10 is broken on osx 10.9
04:18:05  * tjkrusinskiquit (Quit: Lost terminal)
04:23:53  * kazuponquit (Remote host closed the connection)
04:24:00  * kazuponjoined
04:46:34  * paulfryzeljoined
04:50:57  * Benviequit (Ping timeout: 252 seconds)
04:50:59  * paulfryzelquit (Ping timeout: 240 seconds)
04:52:29  * dshaw_joined
04:57:17  * dshaw_quit (Ping timeout: 272 seconds)
04:58:08  * daviddiasjoined
05:02:38  * daviddiasquit (Ping timeout: 245 seconds)
05:03:53  <MI6>joyent/node: tjfontaine created branch v0.10.25-release - http://git.io/6Gv5fw
05:06:33  <MI6>joyent/node: Timothy J Fontaine v0.10.25-release * b0e5f19 : src: lint lib/net.js - http://git.io/Q5v4cQ
05:10:00  * mcavagejoined
05:13:46  <MI6>node-review-windows: #8 UNSTABLE windows-x64 (7/610) windows-ia32 (7/610) http://jenkins.nodejs.org/job/node-review-windows/8/
05:14:38  * mcavagequit (Ping timeout: 264 seconds)
05:15:55  * brsonquit (Ping timeout: 272 seconds)
05:17:51  * bradleymeckjoined
05:20:17  <MI6>node-review-unix: #155 UNSTABLE smartos-ia32 (9/610) linux-ia32 (2/610) osx-ia32 (3/610) smartos-x64 (9/610) osx-x64 (6/610) linux-x64 (6/610) http://jenkins.nodejs.org/job/node-review-unix/155/
05:20:20  <tjfontaine>sigh
05:26:55  <othiym23>I didn't do it!
05:27:06  <othiym23>it was on fire when I lay down on it!
05:27:39  <tjfontaine>heh
05:30:29  * dshaw_joined
05:35:17  * dshaw_quit (Ping timeout: 272 seconds)
05:37:21  * brsonjoined
05:48:13  * brsonquit (Ping timeout: 272 seconds)
05:48:16  * thlorenzjoined
05:52:53  * daviddiasjoined
05:54:44  * brsonjoined
05:57:36  * daviddiasquit (Ping timeout: 252 seconds)
06:00:07  * AvianFluquit (Remote host closed the connection)
06:04:43  * thlorenzquit (Remote host closed the connection)
06:17:27  <othiym23>tjfontaine: somebody on #node.js acting like the Walmart bug might be liked to a DoS they've fond
06:18:13  <othiym23>*linked
06:24:09  * mikealjoined
06:24:34  * abraxas_joined
06:25:01  * bradleymeckquit (Quit: bradleymeck)
06:26:03  * abraxasquit (Ping timeout: 260 seconds)
06:34:51  * jmar777quit (Remote host closed the connection)
06:35:01  <MI6>nodejs-v0.10-windows: #454 UNSTABLE windows-ia32 (8/610) windows-x64 (13/610) http://jenkins.nodejs.org/job/nodejs-v0.10-windows/454/
06:36:31  * mikolalysenkojoined
06:37:22  * dap_quit (Quit: Leaving.)
06:41:38  <hueniverse>trevnorris: dgram send() clearly shows that the callback gets (err, bytes) so 0.11 breaks that.
06:45:24  * defunctzombiechanged nick to defunctzombie_zz
06:46:16  * Benviejoined
06:47:38  * daviddiasjoined
06:47:44  * indexzerojoined
06:48:06  * paulfryzeljoined
06:48:22  * Benvie_joined
06:50:20  * m76joined
06:50:58  * Benviequit (Ping timeout: 256 seconds)
06:52:13  * daviddiasquit (Ping timeout: 245 seconds)
06:52:29  * paulfryzelquit (Ping timeout: 240 seconds)
07:05:47  * bajtosjoined
07:07:08  * abraxas_quit (Remote host closed the connection)
07:16:52  * bradleymeckjoined
07:20:23  * indexzeroquit (Quit: indexzero)
07:27:37  * TooTallNatejoined
07:28:27  * abraxasjoined
07:31:21  * TooTallNatequit (Read error: Connection reset by peer)
07:31:28  * nrajlichjoined
07:42:23  * daviddiasjoined
07:46:48  * daviddiasquit (Ping timeout: 245 seconds)
07:48:44  * paulfryzeljoined
07:52:59  * paulfryzelquit (Ping timeout: 240 seconds)
07:56:28  * kazuponquit (Read error: Connection timed out)
07:57:00  * kazuponjoined
07:59:02  * rendarjoined
08:04:23  * nrajlichquit (Quit: ["Textual IRC Client: www.textualapp.com"])
08:18:44  * hzjoined
08:21:26  * AWintermanquit (Read error: Connection reset by peer)
08:22:33  * AWintermanjoined
08:23:05  * groundwaterquit (Ping timeout: 252 seconds)
08:23:28  * MI6quit (Ping timeout: 245 seconds)
08:23:40  * vigithquit (Ping timeout: 246 seconds)
08:23:57  * vigithjoined
08:25:30  * groundwaterjoined
08:28:00  * mikolalysenkoquit (Ping timeout: 252 seconds)
08:28:29  * dshaw_joined
08:29:35  * ircretaryquit (Ping timeout: 245 seconds)
08:37:08  * daviddiasjoined
08:38:14  * dshaw_quit (Quit: Leaving.)
08:41:23  * daviddiasquit (Ping timeout: 245 seconds)
08:51:19  * calvinfojoined
08:53:51  * mikolalysenkojoined
08:55:49  * AWintermanquit (Read error: Connection reset by peer)
08:56:41  * AWintermanjoined
08:57:08  * brsonquit (Quit: leaving)
08:57:20  * Benvie_quit (Ping timeout: 256 seconds)
08:59:07  * mikolalysenkoquit (Ping timeout: 272 seconds)
08:59:56  * mikealquit (Quit: Leaving.)
09:04:21  <indutny>hey people!
09:08:21  * bradleymeckpasses out
09:13:21  * calvinfoquit (Quit: Leaving.)
09:19:00  * janjongboomjoined
09:19:51  * sinclair|workquit (Quit: ChatZilla 0.9.90.1 [Firefox 26.0/20131205075310])
09:20:36  * rmgquit (Ping timeout: 265 seconds)
09:27:37  * janjongboomquit (Ping timeout: 252 seconds)
09:28:30  * janjongboomjoined
09:31:54  * daviddiasjoined
09:36:23  * daviddiasquit (Ping timeout: 245 seconds)
09:43:01  * karupanerurachanged nick to zz_karupanerura
09:45:51  * mordy__quit (Ping timeout: 252 seconds)
09:47:23  * mordy__joined
09:50:19  * paulfryzeljoined
09:54:29  * paulfryzelquit (Ping timeout: 240 seconds)
09:54:42  * mikolalysenkojoined
09:59:36  * mikolalysenkoquit (Ping timeout: 252 seconds)
10:26:40  * daviddiasjoined
10:30:58  * daviddiasquit (Ping timeout: 245 seconds)
10:37:22  * daviddiasjoined
10:42:01  * bajtosquit (Quit: bajtos)
10:46:59  * MI6joined
10:51:02  * paulfryzeljoined
10:55:17  * paulfryzelquit (Read error: Operation timed out)
10:55:25  * mikolalysenkojoined
11:01:45  * mikolalysenkoquit (Ping timeout: 252 seconds)
11:04:11  * inolenquit (Quit: Leaving.)
11:19:23  * kazuponquit (Remote host closed the connection)
11:47:17  * bajtosjoined
11:51:48  * paulfryzeljoined
11:55:59  * paulfryzelquit (Ping timeout: 240 seconds)
11:57:59  * mikolalysenkojoined
12:03:21  * mikolalysenkoquit (Ping timeout: 252 seconds)
12:05:16  * m76quit (Read error: Connection reset by peer)
12:30:13  * kazuponjoined
12:34:43  * kazuponquit (Ping timeout: 245 seconds)
12:51:57  * thlorenzjoined
12:52:29  * paulfryzeljoined
12:56:42  * thlorenzquit (Ping timeout: 252 seconds)
12:56:59  * paulfryzelquit (Ping timeout: 240 seconds)
12:58:46  * mikolalysenkojoined
13:00:22  * m76joined
13:00:29  * abraxasquit (Remote host closed the connection)
13:02:23  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
13:02:36  * thlorenzjoined
13:03:50  * mikolalysenkoquit (Ping timeout: 264 seconds)
13:07:32  * thlorenzquit (Remote host closed the connection)
13:27:31  * janjongboomjoined
13:29:19  <indutny>heya
13:31:54  <mmalecki>heya indutny
13:31:55  <mmalecki>janjongboom: heya
13:32:02  <janjongboom>mmalecki: yooo
13:32:10  <indutny>how are you?
13:32:14  <mmalecki>janjongboom: it's been a while since last beer!
13:32:25  <mmalecki>indutny: good, what about yourself?
13:32:30  <indutny>good good
13:32:33  <indutny>triaging github issues
13:32:39  <janjongboom>mmalecki: yeah, I know!
13:32:51  <janjongboom>mmalecki: I'll be in Oslo & paris next two weeks, but after that I'm all up for it
13:33:00  <janjongboom>mmalecki: isnt michal coming to Ams this weekend?
13:34:15  <mmalecki>janjongboom: yes, he is, this weekend. he's crashing on my couch. I already schedule a polish vs british drinking contest (which he doesn't know about yet)
13:34:28  <janjongboom>mmalecki: \o/
13:34:31  <janjongboom>haha
13:36:35  * thlorenzjoined
13:53:18  * paulfryzeljoined
13:57:29  * paulfryzelquit (Ping timeout: 240 seconds)
13:59:29  * mikolalysenkojoined
14:04:45  * mikolalysenkoquit (Ping timeout: 272 seconds)
14:16:23  * AvianFlujoined
14:24:50  * janjongboomquit (Ping timeout: 264 seconds)
14:25:34  * janjongboomjoined
14:27:33  * hzquit (Ping timeout: 272 seconds)
14:29:34  * c4milojoined
14:30:04  * thlorenzquit (Remote host closed the connection)
14:30:38  * thlorenzjoined
14:30:49  * hzjoined
14:33:21  * thlorenz_joined
14:33:47  * thlorenz_quit (Remote host closed the connection)
14:35:07  * thlorenzquit (Ping timeout: 260 seconds)
14:38:27  * hzquit (Ping timeout: 252 seconds)
14:40:29  * kazuponjoined
14:46:45  * defunctzombie_zzchanged nick to defunctzombie
14:51:32  * bajtosquit (Quit: bajtos)
14:51:37  * m76quit (Read error: Connection reset by peer)
14:52:46  * jmar777joined
14:54:07  * paulfryzeljoined
14:57:38  * kazuponquit (Read error: Connection reset by peer)
14:57:46  * kazuponjoined
14:58:29  * paulfryzelquit (Ping timeout: 240 seconds)
14:59:12  * hzjoined
14:59:44  * AvianFluquit (Remote host closed the connection)
15:00:18  * mikolalysenkojoined
15:01:22  * abraxasjoined
15:05:19  * mikolalysenkoquit (Ping timeout: 252 seconds)
15:05:47  * abraxasquit (Read error: Operation timed out)
15:07:26  * thlorenzjoined
15:12:21  * othiym23quit (Quit: WeeChat 0.3.8)
15:17:36  <MI6>nodejs-master: #881 FAILURE linux-x64 (2/695) osx-x64 (3/695) linux-ia32 (3/695) osx-ia32 (1/695) smartos-x64 (2/695) http://jenkins.nodejs.org/job/nodejs-master/881/
15:20:28  * pachetjoined
15:20:28  * pachetquit (Client Quit)
15:21:09  * pachetjoined
15:21:09  * pachetquit (Changing host)
15:21:09  * pachetjoined
15:24:59  * janjongboomquit (Ping timeout: 240 seconds)
15:26:37  * janjongboomjoined
15:48:28  * AvianFlujoined
15:50:02  * mikolalysenkojoined
15:53:56  * paulfryzeljoined
16:16:03  * kpdeckerquit (Quit: Leaving.)
16:16:15  * kpdeckerjoined
16:20:45  * jmar777quit (Remote host closed the connection)
16:24:45  <indutny>so
16:24:47  <indutny>apparently
16:24:51  <indutny>ipc on windows is totally fucked
16:24:54  * defunctzombiechanged nick to defunctzombie_zz
16:24:57  <indutny>reading the code in libuv now
16:25:29  <mmalecki>indutny: s/ipc on /g/
16:25:37  <indutny>no
16:25:39  <indutny>I didn't mean that
16:25:48  <indutny>and actually
16:25:51  <indutny>it might be not
16:25:59  <mmalecki>yes, you did, just gotta admit that to yourself ;)
16:25:59  <indutny>I'm just not getting some things from first look
16:26:07  <indutny>okay
16:26:35  * kazuponquit (Remote host closed the connection)
16:27:02  <tjfontaine>in what way fedor?
16:32:28  * jmar777joined
16:33:35  * calvinfojoined
16:34:42  * kpdeckerquit (Quit: Leaving.)
16:35:49  * kpdeckerjoined
16:38:15  <indutny>tjfontaine: nvm :)
16:38:24  <indutny>but the way it's done
16:38:25  <indutny>gosh
16:38:33  <indutny>it's sad that there are no actual OS support
16:39:44  <MI6>joyent/node: Fedor Indutny master * c79c304 : tls: process accumulated input - http://git.io/UEVHqA
16:40:20  <indutny>tjfontaine: you know that win is framing all IPCs
16:40:26  <indutny>adding headers to every write
16:44:18  * daviddiasquit (Remote host closed the connection)
16:44:22  * calvinfoquit (Quit: Leaving.)
16:44:49  * daviddiasjoined
16:46:15  * Benviejoined
16:49:08  <rendar>indutny, really? why a frame for every write?
16:49:17  <MI6>nodejs-master: #882 UNSTABLE linux-x64 (2/696) osx-x64 (2/696) linux-ia32 (3/696) osx-ia32 (1/696) smartos-ia32 (3/696) smartos-x64 (2/696) http://jenkins.nodejs.org/job/nodejs-master/882/
16:49:18  * daviddiasquit (Ping timeout: 245 seconds)
16:53:02  * bajtosjoined
17:02:30  * abraxasjoined
17:04:34  <AlexisMocha>is anyone gonna be on the status call today?
17:05:10  <tjfontaine>AlexisMocha: sorry I am sending an email, and will be reviewing and landing some of your tests
17:05:56  <tjfontaine>sigh why are releases so hard right now :/
17:06:01  <tjfontaine>LINK : fatal error LNK1181: cannot open input file 'winmm.lib' [g:\jenkins\workspace\nodejs-msi\d73b0901\deps\v8\tools\gyp\mksnapshot.vcxproj]
17:06:44  <AlexisMocha>tjfontaine: so no call? :)
17:06:48  * abraxasquit (Ping timeout: 252 seconds)
17:06:50  <tjfontaine>no, no call
17:07:06  <tjfontaine>AlexisMocha: thanks for your work you're doing, did you see the ocmment I left about v0.10 on the one test?
17:10:46  * daviddiasjoined
17:11:25  <daviddias>Yo tjfontaine where can I find the hangout link for this week? :)
17:11:47  <tjfontaine>you'll be getting an email shortly
17:11:51  <tjfontaine>:)
17:12:06  <daviddias>thank you :)
17:15:19  * rmgjoined
17:17:52  <tjfontaine>bbiab
17:18:29  <daviddias>but do we have meeting today tjfontaine?
17:22:10  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
17:23:35  * dap_joined
17:26:52  <indutny>tjfontaine: hey man
17:27:00  <indutny>tjfontaine: can't send an email to the team@ group
17:27:01  <indutny>I think
17:27:09  <indutny>perhaps you forgot to add fedor@indutny.com
17:29:36  <AlexisMocha>tjfontaine: yes I saw your comment. will that block the PR?
17:29:49  * daviddiasquit (Remote host closed the connection)
17:30:21  * octetcloudjoined
17:30:25  * daviddiasjoined
17:30:27  * TooTallNatejoined
17:32:55  * janjongboomjoined
17:34:43  * daviddiasquit (Ping timeout: 245 seconds)
17:38:18  * vigithquit (Ping timeout: 252 seconds)
17:41:43  * dshaw_joined
17:42:07  * mikealjoined
17:44:17  * mikealquit (Client Quit)
17:45:14  * vigithjoined
17:46:16  <vigith>Ralith: thank you.. i had got it fixed :)
17:48:49  <hueniverse>running the nipple module tests with 0.11 master: [Error: 47662526030976:error:0607907F:digital envelope routines:EVP_PKEY_get1_RSA:expecting an rsa key:../deps/openssl/openssl/crypto/evp/p_lib.c:288:
17:48:50  <hueniverse>]
17:51:44  <indutny>hueniverse: already fixed
17:51:50  <indutny>see https://github.com/joyent/node/pull/6949
17:51:57  <indutny>ttyl
17:51:59  <indutny>rebooting to windows
17:53:14  <MI6>libuv-master: #450 FAILURE http://jenkins.nodejs.org/job/libuv-master/450/
17:54:42  * Benviequit (Ping timeout: 252 seconds)
17:57:51  <hueniverse>indutny: great. will test again when it lands in master
18:03:47  <tjfontaine>AlexisMocha: no it won't block it, I just want to understand the difference
18:05:37  * mcavagejoined
18:07:48  * Benviejoined
18:11:15  * mikolalysenkoquit (Ping timeout: 245 seconds)
18:11:38  * brsonjoined
18:16:34  * inolenjoined
18:34:24  <indutny>ok, time to reboot to windows
18:34:38  <tjfontaine>you still have access to the scratch vm
18:44:49  * hueniversequit (Read error: Connection reset by peer)
18:44:54  <MI6>nodejs-master-windows: #669 UNSTABLE windows-x64 (23/696) windows-ia32 (17/696) http://jenkins.nodejs.org/job/nodejs-master-windows/669/
18:47:03  * c4miloquit (Remote host closed the connection)
18:50:11  * mikolalysenkojoined
18:53:57  * piscisaureusjoined
18:55:11  * mikolalysenkoquit (Ping timeout: 252 seconds)
18:55:35  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
19:00:14  * piscisaureusquit (Ping timeout: 264 seconds)
19:04:23  * c4milojoined
19:04:24  <tjfontaine>yay, I think I have the windows buildbot fixed so I can actually finish the v0.10.25 release
19:04:26  * dshaw_quit (Quit: Leaving.)
19:04:41  * txdvquit (Read error: Connection reset by peer)
19:05:30  * kpdeckerquit (Quit: Leaving.)
19:05:41  * txdvjoined
19:10:50  * AvianFlu_joined
19:11:12  * AvianFluquit (Disconnected by services)
19:11:15  * AvianFlu_changed nick to AvianFlu
19:11:25  * dshaw_joined
19:11:34  * defunctzombie_zzchanged nick to defunctzombie
19:13:55  * superjoe30joined
19:15:20  * mikealjoined
19:16:18  * piscisaureusjoined
19:17:57  <Ralith>vigith: what'd it turn out to be?
19:24:01  * othiym23joined
19:28:03  * kpdeckerjoined
19:32:44  * superjoe30part ("Leaving")
19:36:10  * m76joined
19:38:31  * kenperkins_quit (Quit: Computer has gone to sleep.)
19:41:00  * bajtosquit (Quit: bajtos)
19:43:32  <MI6>joyent/node: tjfontaine created tag v0.10.25 - http://git.io/Ubjg7g
19:44:40  <MI6>joyent/node: Timothy J Fontaine v0.10 * 5aebc73 : blog: Post for v0.10.25 (+5 more commits) - http://git.io/GMoooQ
19:50:51  * mikolalysenkojoined
19:53:10  <MI6>nodejs-v0.10-windows: #455 UNSTABLE windows-ia32 (7/610) windows-x64 (12/610) http://jenkins.nodejs.org/job/nodejs-v0.10-windows/455/
19:53:32  <MI6>nodejs-v0.10: #1734 UNSTABLE osx-x64 (6/610) linux-ia32 (3/610) smartos-x64 (8/610) smartos-ia32 (8/610) linux-x64 (2/610) osx-ia32 (5/610) http://jenkins.nodejs.org/job/nodejs-v0.10/1734/
19:55:58  <indutny>piscisaureus: hey man
19:56:01  <indutny>you there?
19:56:11  * mikolalysenkoquit (Ping timeout: 260 seconds)
19:56:13  <piscisaureus>I'm working. sup?
19:57:51  * inolenquit (Quit: Leaving.)
19:57:57  <TooTallNate>indutny: hey man, i'm having some weird behavior popping up after https://github.com/joyent/node/commit/1e066e4a4a88f97af865d965f104b5fe8136797f
19:58:04  <TooTallNate>see gist: https://gist.github.com/TooTallNate/8584477
19:58:10  <indutny>piscisaureus: oh, I need some help with windows :)
19:58:17  <indutny>piscisaureus: let me know if you'll have a free minutes
19:58:21  <TooTallNate>(sorry for external deps, but hopefully the NODE_DEBUG output will be helpful enough)
19:58:36  <piscisaureus>indutny: not a lot of time. but what's your question?
19:58:56  <piscisaureus>indutny: iow, I can do talking but no serious code writing or debugging
19:59:12  <indutny>unfortunately I'm in need of latter one
19:59:13  <indutny>https://github.com/joyent/libuv/pull/1056
19:59:20  <indutny>so
19:59:21  <indutny>ok
19:59:24  <indutny>I'll figure it out by myself
19:59:34  * inolenjoined
19:59:35  <indutny>seems to be building
19:59:40  <indutny>but somehow tests are failing
19:59:45  <indutny>anyway
19:59:52  <indutny>TooTallNate: you must be kidding :)
20:00:00  <indutny>tls over tls again?
20:00:07  <TooTallNate>hahah
20:00:11  <TooTallNate>ya
20:00:15  <indutny>TooTallNate: what happens?
20:00:17  <indutny>in a few words
20:00:20  <indutny>if posssible
20:00:32  <TooTallNate>indutny: something like a double "end" event or something
20:00:34  <TooTallNate>i get Error: stream.push() after EOF
20:00:46  <TooTallNate>on v0.11.10
20:00:49  <indutny>hm...
20:00:54  <indutny>interesting
20:00:56  <indutny>what about v0.10 ?
20:01:06  <tjfontaine>on all platforms or only windows?
20:01:29  <indutny>tjfontaine: is it a question to me or nate?
20:01:38  <tjfontaine>nate
20:01:46  <piscisaureus>indutny: src/win/pipe.c scares me
20:01:54  <indutny>me too
20:01:58  <TooTallNate>tjfontaine: all, indutny is working on a separate win issue
20:02:02  <TooTallNate>my thing is new :)
20:02:04  <indutny>haha
20:02:13  <indutny>yep
20:02:14  <TooTallNate>indutny: v0.10 works fine
20:02:14  <tjfontaine>TooTallNate: I know, we see the streams push after eof a lot in windows tests :)
20:02:23  * defunctzombiechanged nick to defunctzombie_zz
20:02:26  <TooTallNate>ahhh, didn't realize that :)
20:02:36  <indutny>and
20:02:40  <indutny>are you on windows?
20:02:44  <indutny>or does it happen on unixes as well?
20:02:44  <TooTallNate>indutny: OS X
20:02:47  <indutny>k
20:02:47  <TooTallNate>all
20:03:06  <piscisaureus>indutny: but as far as I can tell you're on the right path
20:03:14  <indutny>piscisaureus: thank you
20:03:17  <indutny>for looking
20:03:28  <indutny>I'm still not sure if it was my build that got borked
20:03:32  <indutny>or if it is really borked
20:03:39  <indutny>because after checking out master branch
20:03:43  <indutny>build has started to fail :)
20:03:44  <piscisaureus>where "right" is the most reasonable solution achievable without rewriting the whole bloody thing
20:03:50  <piscisaureus>indutny: what's the error you get?
20:03:55  <indutny>oh
20:03:59  <indutny>a lot of different unrelated
20:04:04  <indutny>starting from python not found
20:04:06  <indutny>out of sudden
20:04:12  <indutny>and everything down that path
20:04:14  <tjfontaine>magic
20:04:16  <indutny>yeah
20:04:32  <indutny>anyway
20:04:35  <indutny>I was going to reboot
20:04:43  <indutny>but MBP rebooted back into OSX
20:04:44  <tjfontaine>we have the scratch vm
20:04:49  <tjfontaine>you can RDP
20:04:54  <indutny>hm...
20:04:58  <indutny>thank you
20:05:01  <indutny>I'll keep it in mind
20:05:25  <indutny>TooTallNate: could you just add some checks like
20:05:30  <indutny>if (!this._readableState.ended)
20:05:33  <indutny>before .push(null)
20:05:40  <indutny>I think it is pretty stupid race condition...
20:05:46  * mikolalysenkojoined
20:06:10  <indutny>oh
20:06:12  <indutny>btw
20:06:35  <indutny>double eof is not possible with _stream_writable.js
20:07:42  <indutny>what should I install to make it happen?
20:08:17  <indutny>TooTallNate: ^
20:08:55  <TooTallNate>indutny: npm install proxy https-proxy-agent
20:08:58  <TooTallNate>i think are the two
20:09:02  <indutny>oh, ok
20:09:03  <indutny>thank you
20:09:24  <TooTallNate>indutny: you need to run the proxy.js script first, then run test.js <serverPort> <proxyPort>
20:09:36  <TooTallNate>(the error happens in the test.js script)
20:10:08  * defunctzombie_zzchanged nick to defunctzombie
20:10:20  <indutny>oh
20:10:24  <indutny>it comes through an http layer
20:10:27  <indutny>what a surprise
20:10:55  <TooTallNate>maybe this one should be called test-https-inception.js :p
20:11:07  <TooTallNate>i need to write a standalone test case though...
20:11:53  <indutny>:)
20:13:06  * piscisaureusquit (Read error: Operation timed out)
20:14:14  * perintquit (Ping timeout: 252 seconds)
20:17:56  * bradleymeckquit (Quit: bradleymeck)
20:21:02  <indutny>TooTallNate: if I'll fix it
20:21:06  <indutny>will you write the test? :)
20:21:22  <indutny>I could even explain you what's wrong with it
20:22:25  <TooTallNate>indutny: i'm trying to think of how to write up a good test... if you have a patch i can probably write a test to trigger it
20:22:31  <indutny>TooTallNateyep
20:22:32  <trevnorris>indutny: cpplint much? ;P
20:22:35  <indutny>that would be great
20:22:41  <indutny>trevnorris: uuuh?
20:22:50  <trevnorris>indutny: c79c304e, tls_wrap.cc line 211
20:22:52  <indutny>TooTallNate: basically the problem
20:23:06  <indutny>trevnorris: could you give me a proper link? :)
20:23:15  <trevnorris>indutny: run make cpplint
20:23:19  <indutny>oooh
20:23:29  <indutny>tadam
20:23:35  <indutny>trevnorris: could you please fix it?
20:23:38  <trevnorris>sure
20:23:44  <indutny>I think it is not worth reviewing it again
20:24:09  <indutny>TooTallNate: so what happens here
20:24:10  <indutny>is that
20:24:28  <indutny>one of your tls sockets is receiving close_notify
20:24:38  <indutny>i.e. "end" notification from other peer
20:24:45  <indutny>and then cleartext read happens
20:25:32  <TooTallNate>ok makes sense
20:26:26  <indutny>yeah
20:26:31  <indutny>and it think that it was cleartext that "end"ed
20:26:39  <indutny>and since it disallows half-open connections
20:26:42  <indutny>it closes both sides
20:26:47  <indutny>and then
20:26:48  <TooTallNate>indutny: so to trigger it, do tls-inception, end the first tls connection, then write some data on the second one?
20:26:56  <indutny>I have no idea :)
20:27:02  <TooTallNate>haha, ok i'll try it out
20:27:08  <MI6>joyent/node: trevnorris created branch v8-3.22.24.17 - http://git.io/ZfSKJQ
20:27:10  <TooTallNate>i think that'll do the trick ;)
20:27:13  <trevnorris>tjfontaine: quick check ^
20:27:17  <tjfontaine>looking
20:27:38  <indutny>TooTallNate: oh, actually
20:27:40  <indutny>hm...
20:27:53  <indutny>perhaps, I am wrong
20:27:56  <indutny>and it should close it
20:28:03  <indutny>because it receives 0 read
20:28:07  <indutny>on cleartext
20:28:25  <indutny>shit
20:30:56  <indutny>TooTallNate: I think it could be something more sophisticated
20:31:05  <indutny>something on http layer
20:35:59  <tjfontaine>indutny: https://github.com/joyent/node/issues/6784
20:36:07  <tjfontaine>as another potential example
20:36:09  <indutny>similar
20:36:16  <indutny>that's quite similar to what I see
20:36:19  <indutny>looks like a streams bug
20:36:21  <indutny>honestly saying
20:36:23  <indutny>or
20:36:24  <indutny>net.js
20:36:27  <indutny>not sure yet
20:36:29  <tjfontaine>we have an isaacs :)
20:36:38  <indutny>so
20:36:41  <indutny>what happens
20:36:42  <indutny>is that
20:36:44  <indutny>tls_legacy.js
20:36:46  <indutny>push(data)
20:36:48  <indutny>push(null)
20:37:07  <indutny>and somehow that gets messed up
20:37:09  <indutny> and reversed
20:37:15  <indutny>hm...
20:37:20  <indutny>that issue is about http_client.js too
20:37:21  <tjfontaine>that's not good
20:38:06  <indutny>oh
20:38:08  <indutny>I have an idea
20:38:12  <indutny>no
20:38:17  <indutny>I didn't
20:38:35  <trevnorris>tjfontaine: v8 upgrade lgty?
20:39:48  <tjfontaine>waiting on the review builds atm
20:39:49  <indutny>hahahaha
20:39:52  <indutny>gosh
20:39:57  <tjfontaine>trevnorris: http://jenkins.nodejs.org/job/node-review-unix/157/
20:40:09  <trevnorris>tjfontaine: thanks
20:40:33  <indutny>that's it https://github.com/joyent/node/blob/master/lib/_http_incoming.js#L101
20:40:41  <indutny>oh god
20:41:01  <tjfontaine>we need to be careful here
20:41:05  <indutny>well
20:41:08  <indutny>it may be fine
20:41:13  <indutny>but _tls_legacy.js sets it too eraly
20:41:15  <indutny>early*
20:41:17  <tjfontaine>right
20:41:22  <indutny>it better not set it
20:41:27  <indutny>and leave it to streams2
20:41:42  <tjfontaine>that sounds quite reasonable, presuming you are signalling in the normal way
20:41:43  <indutny>yeah
20:41:46  <indutny>yep
20:41:49  <indutny>it worked out
20:42:16  <indutny>now
20:42:21  <indutny>need to write test :)
20:42:50  <indutny>TooTallNate: could you please open an issue?
20:43:04  <TooTallNate>indutny: ya
20:43:07  <tjfontaine>tracking++
20:43:09  <tjfontaine>indutny++
20:43:11  <tjfontaine>TooTallNate++
20:43:44  <TooTallNate>got distracted by a video of an elephant painting a picture of an elephant
20:43:50  <tjfontaine>so meta
20:43:58  <TooTallNate>i guess that's a thing
20:44:14  <indutny>what was elephant in picture doing?
20:44:21  <indutny>writing C++?
20:44:26  <tjfontaine>java
20:44:28  <indutny>ooooh
20:44:47  <MI6>joyent/node: Alexis Campailla v0.10 * 42cd468 : test: fix http-incoming-pipelined-socket-destroy (+1 more commits) - http://git.io/jx-rQg
20:45:23  <indutny>they are torturing elephants
20:45:23  <TooTallNate>indutny: picking an apple from a tree
20:45:24  <indutny>what a shame
20:45:27  <indutny>TooTallNate: oh god
20:47:32  * defunctzombiechanged nick to defunctzombie_zz
20:47:59  <MI6>node-review-unix: #157 UNSTABLE smartos-ia32 (2/696) linux-ia32 (1/696) osx-ia32 (2/696) smartos-x64 (2/696) osx-x64 (4/696) linux-x64 (2/696) http://jenkins.nodejs.org/job/node-review-unix/157/
20:49:31  <MI6>node-review-windows: #10 UNSTABLE windows-x64 (12/696) windows-ia32 (17/696) http://jenkins.nodejs.org/job/node-review-windows/10/
20:49:35  <trevnorris>tjfontaine: ok. build looks good. want me to wait for windows as well...
20:49:39  <trevnorris>or... not
20:49:55  <tjfontaine>:)
20:50:02  <trevnorris>ok. pushing
20:50:44  * janjongboomjoined
20:50:51  <MI6>joyent/node: Trevor Norris master * f78e5df : v8: upgrade to 3.22.24.17 - http://git.io/YEnYtA
20:51:20  <tjfontaine>thanks
20:51:49  <trevnorris>:)
20:53:48  <trevnorris>wow, building v8 w/ icu support, clang filled my scrollback buffer with warnings. :P
20:53:54  <MI6>nodejs-v0.10-windows: #456 UNSTABLE windows-ia32 (6/610) windows-x64 (11/610) http://jenkins.nodejs.org/job/nodejs-v0.10-windows/456/
20:54:24  <tjfontaine>heh
20:58:06  <trevnorris>indutny: fyi stuff like this CHECK(Buffer::HasInstance(args[0])); isn't necessary just before using Buffer::Data() because the same check is done in Buffer::Data()
20:58:10  * indexzerojoined
20:58:15  <indutny>oh, ok
20:59:35  <trevnorris>well, and the HasInstance check is a bit of a hack. :P since it's impossible to do an instanceof check in C++. (that I know of)
21:00:01  <tjfontaine>not without caching constructors
21:00:09  * octetcloudquit (Ping timeout: 252 seconds)
21:01:14  <indutny>so
21:01:20  <indutny>fixing that bug was easy
21:01:22  <indutny>writing test is not
21:01:23  <indutny>:(
21:01:31  <trevnorris>heh, got to love those.
21:03:02  <MI6>joyent/node: Trevor Norris master * 74656ca : lint: use reinterpret_cast, not C-style casts - http://git.io/g8m1mQ
21:03:15  <indutny>yay
21:03:15  <tjfontaine>erm
21:03:28  <indutny>is 80 column limit preserved?
21:03:33  <trevnorris>yeah
21:03:36  <MI6>nodejs-v0.10: #1735 UNSTABLE osx-x64 (2/610) linux-ia32 (3/610) smartos-x64 (10/610) smartos-ia32 (8/610) linux-x64 (2/610) osx-ia32 (3/610) http://jenkins.nodejs.org/job/nodejs-v0.10/1735/
21:03:37  <indutny>ok
21:03:40  <indutny>tjfontaine: ?
21:03:59  <tjfontaine>just surprised by the change
21:04:16  <trevnorris>just shutting up cpplint
21:04:17  * bajtosjoined
21:04:43  * mikolalysenkoquit (Ping timeout: 245 seconds)
21:04:49  <tjfontaine>are you familiar with the debian openssl issue? :)
21:04:58  <trevnorris>me? no
21:05:00  <MI6>joyent/node: isaacs v0.10 * 485fcf9 : npm: Upgrade to v1.3.25 - http://git.io/wtjIVw
21:06:01  <tjfontaine>https://www.schneier.com/blog/archives/2008/05/random_number_b.html
21:06:33  <trevnorris>tjfontaine: btw, wife visited the doc today. the baby's popping out in the next week.
21:06:51  <tjfontaine>i.e. ANY TIME, GET THE CATCHERS MITT
21:06:58  <trevnorris>yup. pretty much.
21:08:01  <tjfontaine>blah that commit is going to make my merge a pain
21:08:29  <tjfontaine>isaacs: btw, it'd be great if you did a bit of heads up before a code drop :)
21:08:35  <MI6>node-review-unix: #158 UNSTABLE smartos-ia32 (3/696) linux-ia32 (2/696) osx-ia32 (3/696) smartos-x64 (2/696) osx-x64 (3/696) linux-x64 (1/696) http://jenkins.nodejs.org/job/node-review-unix/158/
21:08:37  <indutny>tjfontaine: oh
21:08:39  <indutny>trevnorris: oh
21:08:43  <indutny>congrats !:)
21:08:48  <indutny>didn't know that you have a wife
21:08:48  <trevnorris>thanks :)
21:08:51  <indutny>:)
21:08:53  <isaacs>tjfontaine: sorry, it's ok if that doesn't get in
21:09:04  <trevnorris>heh, yeah.
21:09:16  <isaacs>tjfontaine: it's minor
21:09:19  * bajtosquit (Quit: bajtos)
21:09:33  <MI6>node-review-windows: #11 UNSTABLE windows-x64 (11/696) windows-ia32 (19/696) http://jenkins.nodejs.org/job/node-review-windows/11/
21:09:39  <MI6>nodejs-master-windows: #670 UNSTABLE windows-x64 (13/696) windows-ia32 (16/696) http://jenkins.nodejs.org/job/nodejs-master-windows/670/
21:09:41  <isaacs>tjfontaine: mostly just a good-faith compliance with some IP licensing crap.
21:09:42  <tjfontaine>isaacs: ok, just want to coordinate around this part a little better so we're aware of what is necessary
21:09:46  <isaacs>tjfontaine: sure sure
21:10:09  <isaacs>tjfontaine: this just removes the gubblebum blocky font from the distribution of the npm client
21:10:20  <MI6>nodejs-master: #883 UNSTABLE linux-x64 (2/696) osx-x64 (2/696) linux-ia32 (3/696) osx-ia32 (3/696) smartos-ia32 (2/696) smartos-x64 (3/696) http://jenkins.nodejs.org/job/nodejs-master/883/
21:10:30  <isaacs>since one could make the argument that npm is not (or won't soon be) strictly "not for profit"
21:10:37  <tjfontaine>isaacs: right
21:10:48  <isaacs>but no substantive change as far as node's concerned
21:11:20  <tjfontaine>ok, thanks
21:11:48  * bradleymeckjoined
21:12:05  <indutny>tjfontaine: gosh
21:12:12  <indutny>I know why it isn't reproducible on v0.10
21:12:21  <indutny>the http client is using .ondata
21:12:29  <tjfontaine>ah right
21:12:31  <indutny>yeah
21:12:54  <indutny>that means that fix will go only in master
21:12:57  <indutny>for good or bad
21:13:45  <MI6>nodejs-v0.10-windows: #457 UNSTABLE windows-ia32 (6/610) windows-x64 (11/610) http://jenkins.nodejs.org/job/nodejs-v0.10-windows/457/
21:14:30  <tjfontaine>6 test failures on windows
21:14:31  <tjfontaine>FUCK YA
21:14:32  <LOUDBOT>"AMERICAN MOVIE GUY JOHN C. RAMBO"
21:14:37  <tjfontaine>AlexisMocha: <3 thanks so much
21:15:09  <indutny>so soon
21:15:15  <indutny>we will have zero failure CI results :)
21:15:22  <tjfontaine>that's what we're aiming for
21:15:32  <tjfontaine>so we can actually trust the damn thing
21:15:50  <isaacs>indutny tjfontaine: omg.
21:16:00  * mikolalysenkojoined
21:16:02  <isaacs>it was good to hand off node to this tj guy.
21:16:05  <indutny>never seen it in ages
21:16:07  <indutny>:)
21:16:11  <isaacs>zero test failures wouldmake me soo happy
21:16:18  <indutny>haha
21:16:20  <indutny>rm -rf test/
21:16:28  <indutny>it was so close
21:16:48  <MI6>nodejs-v0.10: #1736 UNSTABLE osx-x64 (4/610) linux-ia32 (3/610) smartos-x64 (8/610) smartos-ia32 (8/610) linux-x64 (2/610) osx-ia32 (4/610) http://jenkins.nodejs.org/job/nodejs-v0.10/1736/
21:18:00  <indutny>ok, will continue tomorrow
21:19:00  <tjfontaine>isaacs: haha, well it's good to have someone looking into the failures on windows
21:19:13  <tjfontaine>I'm excited to get alexis on to "real" things asap
21:19:18  * indexzeroquit (Quit: indexzero)
21:19:59  <MI6>nodejs-master: #884 FAILURE linux-x64 (1/696) osx-x64 (1/696) linux-ia32 (1/696) osx-ia32 (1/696) smartos-x64 (3/696) http://jenkins.nodejs.org/job/nodejs-master/884/
21:20:26  <tjfontaine>now I just need to fix gyp
21:20:27  <tjfontaine>...
21:22:36  * indexzerojoined
21:26:25  <MI6>nodejs-master-windows: #671 UNSTABLE windows-x64 (13/696) windows-ia32 (18/696) http://jenkins.nodejs.org/job/nodejs-master-windows/671/
21:30:59  * indexzeropart
21:33:12  * jmar777quit (Remote host closed the connection)
21:33:58  * superjoe30joined
21:34:19  <MI6>libuv-master-windows: #30 FAILURE http://jenkins.nodejs.org/job/libuv-master-windows/30/
21:35:25  * octetcloudjoined
21:41:18  <MI6>libuv-v0.10-windows: #19 FAILURE http://jenkins.nodejs.org/job/libuv-v0.10-windows/19/
21:48:38  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
21:49:53  * janjongboomjoined
21:51:33  * mcavagequit (Remote host closed the connection)
21:51:36  * defunctzombie_zzchanged nick to defunctzombie
21:54:08  * dshaw_quit (Quit: Leaving.)
21:55:28  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
22:01:51  * brsonquit (Ping timeout: 252 seconds)
22:03:12  * kenperkinsjoined
22:03:40  * brsonjoined
22:05:18  * pachetquit (Quit: leaving)
22:06:02  * kenperkinsquit (Client Quit)
22:06:21  * dshaw_joined
22:09:12  * pachetjoined
22:10:50  * piscisaureusjoined
22:15:05  * kpdeckerquit (Quit: Leaving.)
22:17:34  * janjongboomjoined
22:21:19  * AvianFluquit (Remote host closed the connection)
22:24:28  * rendarquit (Quit: Leaving)
22:27:31  * bradleymeckquit (Ping timeout: 252 seconds)
22:28:53  * superjoe30quit (Ping timeout: 248 seconds)
22:29:58  * bradleymeckjoined
22:32:31  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
22:33:11  * superjoe30joined
22:33:40  * janjongboomjoined
22:35:29  * Ralithquit (Ping timeout: 240 seconds)
22:40:02  * m76quit (Read error: Connection reset by peer)
22:42:11  * piscisaureusquit (Ping timeout: 252 seconds)
22:45:40  <trevnorris>tjfontaine: I'll want to get https://github.com/joyent/node/pull/6954 in sooner than later
22:45:47  * mcavagejoined
22:45:50  <trevnorris>it's a subset of what I've been working on
22:47:03  * AvianFlujoined
22:50:17  * piscisaureusjoined
22:54:59  * janjongboomquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
22:58:46  <TooTallNate>indutny: creating a standalone test case has been difficult to come up with so far :p
23:06:41  * Ralithjoined
23:14:01  * TooTallNatequit (Quit: Computer has gone to sleep.)
23:15:42  * mcavagequit (Remote host closed the connection)
23:19:18  * robertkowalskiquit (Ping timeout: 245 seconds)
23:20:46  * robertkowalskijoined
23:22:14  * thlorenzquit (Remote host closed the connection)
23:22:36  * mikolalysenkoquit (Ping timeout: 252 seconds)
23:27:29  * kpdeckerjoined
23:27:37  * kpdeckerquit (Client Quit)
23:27:54  * kpdeckerjoined
23:28:03  * AvianFluquit (Ping timeout: 272 seconds)
23:29:42  <trevnorris>othiym23: ping
23:30:14  * piscisaureusquit (Ping timeout: 264 seconds)
23:31:12  <AlexisMocha>tjfontaine: good stuff :) I am looking forward to the "real" stuff too
23:36:13  * mcavagejoined
23:36:51  * mikolalysenkojoined
23:37:54  * bradleymeckquit (Quit: bradleymeck)
23:43:13  * superjoe30quit (Quit: Leaving)
23:44:26  * indexzerojoined
23:47:50  * pachetquit (Quit: leaving)
23:48:05  <vigith>Ralith: using already freed memory in uv_write i suppose
23:56:11  * indexzeroquit (Quit: indexzero)
23:59:12  <MI6>libuv-master-gyp: #408 UNSTABLE smartos-ia32 (3/204) smartos-x64 (4/204) osx-ia32 (1/205) http://jenkins.nodejs.org/job/libuv-master-gyp/408/