00:00:01  * ircretaryquit (Remote host closed the connection)
00:00:02  <mjr____>Meanwhile, I think I'll just add a few more underscores to my nick.
00:00:08  * ircretaryjoined
00:00:18  <isaacs>mjr____: you gotta register, man
00:00:18  <isaacs>srsly
00:00:37  <mjr____>Yeah, I guess so.
00:00:57  <mjr____>I figure once you register then they'll find out about your medical pot card, and pretty soon we'll find out that 9/11 was an inside job.
00:01:05  <isaacs>hahhahh
00:01:13  <isaacs>this is freenode, man. that shit's legal here.
00:01:41  <mjr____>I hear freenode was complicit with faking the moon landings, so I dunno.
00:06:50  * hzquit
00:19:38  * kazuponjoined
00:24:22  * kazuponquit (Ping timeout: 256 seconds)
00:25:03  * amartensquit (Quit: Leaving.)
00:37:32  * st_lukejoined
00:46:40  * hueniversejoined
00:48:50  * mjr____quit (Quit: mjr____)
00:53:17  * dapquit (Quit: Leaving.)
01:02:26  * st_lukequit (Read error: No route to host)
01:04:25  * st_lukejoined
01:20:29  * inolenquit (Quit: Leaving.)
01:20:47  * kazuponjoined
01:22:23  * st_lukequit (Remote host closed the connection)
01:25:27  * kazuponquit (Ping timeout: 260 seconds)
01:32:49  * TooTallNatequit (Quit: ["Textual IRC Client: www.textualapp.com"])
01:43:09  * abraxasjoined
01:51:24  * mmaleckichanged nick to mmalecki[zzz]
01:53:08  * qardquit (Quit: Leaving.)
02:05:13  * inolenjoined
02:10:27  * st_lukejoined
02:16:49  * qardjoined
02:35:14  * st_lukequit (Remote host closed the connection)
02:49:18  * dapjoined
02:49:19  * dapquit (Client Quit)
02:54:03  * kazuponjoined
02:58:59  * kazuponquit (Ping timeout: 255 seconds)
03:30:59  * brsonquit (Quit: leaving)
03:54:41  * kazuponjoined
04:12:24  * perezd_joined
04:12:25  * perezdquit (Read error: Connection reset by peer)
04:12:25  * perezd_changed nick to perezd
04:16:49  * indexzerojoined
04:26:51  <DrPizza>do any of you know if there's a viable spidermonkey equivalent/port of node.js?
04:26:57  <DrPizza>I know there was chatter a while back about creating one
04:29:01  <DrPizza>but I don't know if it materialized, or if the relevant devs just said "eh, jsshell is good enough" or something
05:08:55  * icarotjoined
05:17:29  * qardquit (Quit: Leaving.)
05:19:49  * timoxleyjoined
05:21:54  * inolenquit (Read error: Connection reset by peer)
05:22:12  * inolenjoined
05:28:02  * rendarjoined
05:54:18  * mjr__joined
05:56:33  * icarotquit (Quit: leave())
05:56:53  * icarotjoined
05:56:59  * stolsmajoined
06:00:36  * mjr__quit (Quit: mjr__)
06:00:55  * mjr__joined
06:03:20  <mjr__>isaacs: at your advice, I'm now locked in with just two underscores.
06:03:43  <mjr__>Although, I now believe Black Helicopters and/or chemtrails are now going to target me.
06:07:55  * hueniverse1joined
06:11:17  * hueniversequit (Ping timeout: 248 seconds)
07:22:40  * porcojoined
07:23:20  * porcoquit (Client Quit)
07:25:02  * Siphonblastjoined
07:27:35  * icarotquit (Ping timeout: 260 seconds)
07:34:24  * Siphonblastquit (Ping timeout: 264 seconds)
07:41:31  * `3rdEdenjoined
07:55:03  * kuebkjoined
07:55:18  <kuebk>hi
07:55:37  <kuebk>is there any list with all built-in variables in node-gyp?
08:12:22  * V1joined
08:16:33  * mjr__quit (Quit: mjr__)
08:20:33  * V1quit (Ping timeout: 256 seconds)
08:37:53  * kazupon_joined
08:38:25  * dominictarr_joined
08:46:18  * kazuponquit (*.net *.split)
08:46:18  * dominictarrquit (*.net *.split)
08:46:19  * dominictarr_changed nick to dominictarr
08:54:29  * dsantiag_joined
08:59:04  * indexzeroquit (*.net *.split)
08:59:05  * dsantiagoquit (*.net *.split)
09:01:03  * indexzerojoined
09:01:03  * dsantiagojoined
09:01:55  * dsantiagoquit (Ping timeout: 246 seconds)
09:15:21  * isaacs_joined
09:20:37  * niska`joined
09:21:46  * Raynosquit (Ping timeout: 264 seconds)
09:21:47  * isaacsquit (Ping timeout: 264 seconds)
09:21:48  * niskaquit (Ping timeout: 264 seconds)
09:22:03  * JakeVjoined
09:23:43  * dominictarrquit (Quit: dominictarr)
09:23:45  * rendarquit
10:02:37  * rendarjoined
10:02:46  * abraxasquit (Remote host closed the connection)
10:32:56  * kazupon_quit (Remote host closed the connection)
10:37:38  * hzjoined
10:47:39  * kazuponjoined
10:57:54  * mmalecki[zzz]changed nick to mmalecki
11:12:48  * indexzeroquit (Quit: indexzero)
11:15:21  * mralephquit (Read error: Connection reset by peer)
11:32:06  * inolenquit (Quit: Leaving.)
11:39:44  * c4milojoined
11:50:30  * kazuponquit (Remote host closed the connection)
11:52:12  * mralephjoined
12:39:16  * c4miloquit (Remote host closed the connection)
12:45:15  * c4milojoined
12:49:12  * c4miloquit (Remote host closed the connection)
12:55:44  * kazuponjoined
12:57:19  * bnoordhuisjoined
13:07:48  * kevinswiber-awayjoined
13:19:09  * kevinswiber-awayquit (Read error: Connection reset by peer)
13:19:19  * kevinswiber-awayjoined
13:36:07  * kevinswi_joined
13:38:34  * kevinswiber-awayquit (Ping timeout: 252 seconds)
13:38:38  * c4milojoined
13:40:46  * kevinswi_quit (Ping timeout: 252 seconds)
13:52:37  * defunctzombie_zzchanged nick to defunctzombie
13:59:38  * kevinswiber-awayjoined
14:00:20  * kevinswi_joined
14:03:18  * stolsmaquit (Ping timeout: 272 seconds)
14:04:31  * kevinswiber-awayquit (Ping timeout: 256 seconds)
14:17:29  * mjr__joined
14:22:59  * bnoordhuisquit (Ping timeout: 255 seconds)
14:25:56  * kevinswi_quit (Read error: Connection reset by peer)
14:33:11  * kazuponquit (Remote host closed the connection)
14:35:55  * kevinswiber-awayjoined
15:02:09  * kevinswiber-awayquit (Read error: Connection reset by peer)
15:03:18  * kevinswiber-awayjoined
15:08:29  * `3rdEdenquit (Remote host closed the connection)
15:11:53  * stagasquit (Ping timeout: 252 seconds)
15:21:15  * normanmjoined
15:22:16  * kuebkquit (Quit: Leaving.)
15:26:44  * kevinswiber-awayquit (Ping timeout: 252 seconds)
15:30:52  * defunctzombiechanged nick to defunctzombie_zz
15:31:14  * kevinswiber-awayjoined
15:33:53  * normanmquit (Quit: Computer has gone to sleep.)
15:40:44  * bnoordhuisjoined
15:42:23  * kevinswiber-awayquit (Remote host closed the connection)
15:42:57  * kevinswiber-awayjoined
15:43:39  * kazuponjoined
15:49:17  * kazuponquit (Ping timeout: 252 seconds)
15:49:18  * kevinswiber-awayquit (Ping timeout: 264 seconds)
15:50:13  * kevinswiber-awayjoined
15:55:54  * kevinswiber-awayquit (Ping timeout: 264 seconds)
15:59:29  * luigyjoined
16:06:11  * perezdquit (Quit: perezd)
16:15:50  * TooTallNatejoined
16:20:38  * kevinswiber-awayjoined
16:21:40  * isaacs_changed nick to isaacs
16:22:10  * isaacschanged nick to Guest54043
16:32:58  * stagasjoined
16:36:03  * amartensjoined
16:41:03  * bnoordhuisquit (Ping timeout: 256 seconds)
16:42:42  * sofaerjoined
16:43:21  * TooTallNatequit (Quit: Computer has gone to sleep.)
16:44:58  <sofaer>Hi, I have a question about shutting down streams. If there are pending writes to a stream, how do I make it so the stream closes as soon as the writes go through? I'm calling uv_shutdown right now, thinking that would flush the writes, but it never reaches my shutdown callback and hangs.
16:55:31  * inolenjoined
16:56:08  * normanmjoined
17:00:18  * inolenquit (Ping timeout: 256 seconds)
17:00:43  * TooTallNatejoined
17:12:38  * Guest54043changed nick to isaacs_
17:12:52  <tjfontaine>isaacs_: clearly you need at least one more _
17:13:06  <isaacs_>srsly, wth
17:13:15  <isaacs_>keeps telling me i'm not available :)
17:13:31  <mjr__>All the cool kids have at least one underscore
17:13:53  <mjr__>But seriously isaacs_, you should consider registering your nick.
17:13:55  * qardjoined
17:17:13  <isaacs_>mjr__: >_<
17:17:26  <isaacs_>mjr__: actually, isaacs_ is a registered handle to the isaacs account
17:17:38  <mjr__>ouch, burned.
17:17:54  <isaacs_>17:17 -!- isaacs_ [~isaacs@173.203.97.118]
17:17:54  <isaacs_>17:17 -!- ircname : Isaac Z. Schlueter
17:17:54  <isaacs_>17:17 -!- channels : @#libuv @#Node.js #stackvm #nerdtracker
17:17:54  <isaacs_>17:17 -!- server : wolfe.freenode.net [Manchester, England]
17:17:54  <isaacs_>17:17 -!- away : /
17:17:56  <isaacs_>17:17 -!- hostname : 173.203.97.118 173.203.97.118
17:17:59  <isaacs_>17:17 -!- idle : 0 days 0 hours 0 mins 11 secs [signon: Fri May 10 09:15:15 2013]
17:18:02  <isaacs_>17:17 -!- account : isaacs
17:18:04  <isaacs_>17:17 -!- End of WHOIS
17:18:13  * isaacs_changed nick to isaacs
17:18:37  <isaacs>there we go, that's better
17:18:55  <tjfontaine>and the world was set right
17:33:30  <sofaer>Hi isaacs (or anyone else that can help), I'm trying to fix a bug in Julia, and I need to close a stream after the pending writes have gone out. I tried using uv_shutdown, but my shutdown callback never gets called and the script hangs. Any idea what might cause that or how to debug it?
17:33:44  * AvianFluquit (Remote host closed the connection)
17:40:00  * benoitcquit (Excess Flood)
17:45:44  * perezdjoined
17:46:09  * benoitcjoined
17:46:10  * kevinswi_joined
17:46:12  * kevinswiber-awayquit (Ping timeout: 256 seconds)
17:46:45  <txdv>what do I nee to link to if I am using the libuv.a?
17:46:48  <txdv>-lrt and -m?
17:49:18  <txdv>gcc -o bin/sredirect_mm_i686.so dllapi.o engine_api.o h_export.o meta_api.o sdk_util.o -shared -ldl -lm -lrt -s libuv/libuv.a
17:49:20  <txdv>I hate this
17:49:30  <txdv>I bet there is some bad ordering or something
17:50:55  * kevinswi_quit (Ping timeout: 264 seconds)
17:51:09  * bnoordhuisjoined
17:51:19  * JakeVchanged nick to Raynos
17:54:49  * kevinswiber-awayjoined
17:59:01  * inolenjoined
17:59:49  * perezdquit (Quit: perezd)
18:01:09  * hueniverse1quit (Ping timeout: 248 seconds)
18:01:48  * hueniversejoined
18:02:13  * piscisaureus_joined
18:02:42  * bajtosjoined
18:05:33  * normanmquit (Quit: Computer has gone to sleep.)
18:06:08  <bnoordhuis>sofaer: do you have some code i can look at?
18:07:37  * hueniverse1joined
18:07:38  <bnoordhuis>txdv: -ldl and friends should come after libuv.a (if -Wl,--as-needed is in effect, which is probably the case)
18:08:06  <txdv>so there is a flag
18:08:13  <txdv>so it will resolve it on its own?
18:08:35  * hueniversequit (Ping timeout: 256 seconds)
18:12:41  <bnoordhuis>txdv: yeah, if i understand you correctly. -Wl,--as-needed means 'only link in what previous object/archive files need'
18:13:09  <txdv>rising dependency order
18:16:38  <isaacs>sofaer: bnoordhuis or piscisaureus_ are your best targets
18:19:46  * kevinswi_joined
18:20:12  * `3rdEdenjoined
18:20:31  * dapjoined
18:21:05  * kevinswiber-awayquit (Ping timeout: 252 seconds)
18:25:23  * normanmjoined
18:44:20  * brsonjoined
18:48:50  * bnoordhuisquit (Quit: leaving)
18:51:07  * kevinswi_quit (Ping timeout: 246 seconds)
18:52:00  * bajtosquit (Quit: bajtos)
18:56:22  * piscisaureus_quit (Ping timeout: 276 seconds)
18:57:09  * kevinswiber-awayjoined
19:03:42  * paddybyersquit (Ping timeout: 264 seconds)
19:19:19  * kevinswi_joined
19:20:10  * normanmquit (Quit: Computer has gone to sleep.)
19:21:04  * kevinswiber-awayquit (Ping timeout: 276 seconds)
19:23:30  * benoitcquit (Excess Flood)
19:26:52  * `3rdEdenquit (Remote host closed the connection)
19:27:42  * AvianFlujoined
19:28:01  * dapquit (Quit: Leaving.)
19:28:10  * benoitcjoined
19:29:18  * mralephquit (Ping timeout: 268 seconds)
19:31:29  * kevinswi_quit (Ping timeout: 252 seconds)
19:39:59  * mralephjoined
19:51:08  * paddybyersjoined
19:59:49  * dapjoined
20:04:16  * abraxasjoined
20:08:53  * abraxasquit (Ping timeout: 252 seconds)
20:14:03  * dominictarrjoined
20:14:55  * dapquit (Ping timeout: 264 seconds)
20:24:15  * dominictarrquit (Quit: dominictarr)
20:27:46  * c4miloquit (Remote host closed the connection)
20:27:48  * dominictarrjoined
20:39:13  * dapjoined
20:46:57  * c4milojoined
20:49:29  * dominictarrquit (Quit: dominictarr)
20:51:07  * bajtosjoined
20:51:14  * brsonquit (Ping timeout: 252 seconds)
20:52:34  * brsonjoined
20:53:34  * piscisaureus_joined
21:05:53  <sofaer>Hi piscisaureus_ I was told to ask you about a problem I'm having with uv_shutdown. I'm working on a bug in the julia language where uv_close is getting called on streams with pending writes before the writes go through. I'm trying to call uv_shutdown, then call uv_close in the shutdown callback, but it causes the script to hang, and never gets to the shutdown callback.
21:06:37  <piscisaureus_>sofaer: is this on windows or unix?
21:06:54  <sofaer>piscisaureus_, unix.
21:07:55  <piscisaureus_>sofaer: so, I don't follow. you say you call uv_close from the shutdown callback that never happens?
21:08:36  <piscisaureus_>sofaer: it sounds like you are doing the right thing to me :)
21:09:15  <piscisaureus_>although in this case you could also opt to not call uv_shutdown and call uv_close from the (last) uv_write callback directly
21:09:17  <sofaer>piscisaureus_, yeah, so with my failing patch, uv_close is never called on the stream (I'm pretty sure).
21:10:00  <piscisaureus_>sofaer: so - another thing - the stream you are closing is not expecting any more incoming data?
21:10:25  <sofaer>piscisaureus_, Is there a way to give the stream a callback to be hit on every write?
21:10:31  <piscisaureus_>sofaer: no
21:10:40  <piscisaureus_>sofaer: you specify callbacks for individual writes
21:11:11  <piscisaureus_>sofaer: so just to be clear - a "normal" graceful close is by calling uv_shutdown, then waiting for the uv_read callback to be called with UV_EOF
21:11:17  <piscisaureus_>so you know the other end is finished writing data
21:11:24  <piscisaureus_>and then you uv_close it
21:12:02  <piscisaureus_>It is also possible that you *know* that the other end will not write any more data (like, in the case of http) - in that case you'd just call uv_close immediately after the last uv_write comes back
21:12:28  <piscisaureus_>(calling uv_shutdown in that case isn't bad, it's just unnecessary)
21:13:20  <piscisaureus_>sofaer: maybe just show me your failing patch
21:13:40  <sofaer>OK, there's a little cruft in here: https://github.com/rsofaer/julia/commit/e399fab29d0e74b1c08a0ddaa26f19529921e8d4
21:14:18  <sofaer>Thanks for your help, I can try to make a gist of a reduced example if that's messy.
21:14:34  <sofaer>jl_uv.c is the relevant part.
21:14:51  <sofaer>Do I need to make sure a UV_EOF is being written after the shutdown?
21:15:42  <piscisaureus_>sofaer: uv_shutdown *does* write the EOF (well, FIN packet technically)
21:17:22  <piscisaureus_>sofaer: btw - I think "&shutdownCallback" is probably wrong btw
21:17:37  <piscisaureus_>I can't imagine that would be right. You probably want to drop the &
21:18:28  <sofaer>As opposed to shutdownCallback? I thought so too when I started this, but the C docs I've found all say it doesn't matter, so I'm following the convention already in the file.
21:19:27  <piscisaureus_>maybe you are right in that it doesn't matter. I feel pain looking at it though :)
21:19:33  <piscisaureus_>ok let me try to compile julia
21:22:45  <sofaer>Unfortunately that takes a while the first time
21:23:42  <piscisaureus_>yeah it doesn't really look wrong what you are doing
21:23:46  <piscisaureus_>so I can't tell otherwise
21:27:10  <sofaer>Would it be a problem if uv_shutdown is getting called on a stream more than once? I just realized that might be happening.
21:29:36  <sofaer>yeah, that seems to have fixed the hanging problem. I guess it was trying to drain an already-drained stream.
21:30:31  <piscisaureus_>aah
21:30:33  <piscisaureus_>sofaer: yes that is very likely
21:31:07  <sofaer>piscisaureus_, well, now I'm getting a SIGABRT, so I might be back in a second
21:42:48  <piscisaureus_>sofaer: libuv *should* not allow calling uv_shutdown on a shut stream
21:43:04  <piscisaureus_>sofaer: however the version of libuv you are using is a fork and it's quite old so we can't really "support" that ...
21:43:15  <piscisaureus_>sofaer: do you know how to get around this:
21:44:15  <piscisaureus_>. /bin/sh: 1: ./flisp/flisp: not found
21:44:15  <piscisaureus_>make[2]: *** [julia_flisp.boot] Error 127
21:44:15  <piscisaureus_>make[1]: *** [julia-release] Error 2
21:44:15  <piscisaureus_>make: *** [release] Error 2
21:48:37  * inolenpart
21:49:27  * rendarquit
21:52:58  * stagasquit (Read error: Connection reset by peer)
22:07:33  * icarotjoined
22:08:43  <isaacs>ok, i'm gonna start trying to build a 0.11
22:14:49  <isaacs>piscisaureus_: can you take a look at this? https://github.com/joyent/node/pull/5403
22:15:46  <sofaer>piscisaureus_, I'm really not sure. Something to do with the parser, obviously. I'm making progress, though. I'm hitting an assert in uv_close that tells me not to close a closing stream right now. I think I'm going to figure it out. Thanks.
22:17:55  * dapquit (Quit: Leaving.)
22:21:11  <MI6>joyent/libuv: isaacs master * ab157df : Now working on v0.11.3 (+1 more commits) - http://git.io/9aUpWQ
22:26:18  <MI6>libuv-master: #88 UNSTABLE windows (3/189) osx (1/189) smartos (3/188) linux (1/188) http://jenkins.nodejs.org/job/libuv-master/88/
22:29:20  <MI6>libuv-v0.10-gyp: #18 UNSTABLE smartos-x64 (3/186) windows-ia32 (3/187) windows-x64 (4/187) smartos-ia32 (3/186) http://jenkins.nodejs.org/job/libuv-v0.10-gyp/18/
22:32:04  <MI6>libuv-master-gyp: #28 UNSTABLE smartos-x64 (3/188) windows-ia32 (4/189) smartos-ia32 (6/188) windows-x64 (4/189) linux-x64 (1/188) http://jenkins.nodejs.org/job/libuv-master-gyp/28/
22:38:37  <isaacs>ircretary: tell bnoordhuis file watching on mac stopped providing the filename as of latest libuv. https://gist.github.com/isaacs/5557953
22:38:37  <ircretary>isaacs: I'll be sure to tell bnoordhuis
22:40:59  * paddybyersquit (Ping timeout: 256 seconds)
22:42:49  <piscisaureus_>isaacs: doing it
22:43:00  <isaacs>piscisaureus_: thanks
22:46:26  <piscisaureus_>isaacs: this is for master right?
22:46:52  <isaacs>piscisaureus_: the StringBytes thing?
22:46:54  <isaacs>piscisaureus_: no, v0.10
22:47:42  <isaacs>piscisaureus_: should be no (non-internal) API changes
22:47:50  <piscisaureus_>isaacs: https://github.com/isaacs/node/commit/bf965307e9c85ff8e09c8d394e1bff3f8536ac62 <-- not an api change?
22:48:12  <isaacs>piscisaureus_: push(chunk,encoding) was already there, just not documented
22:48:21  * kazuponjoined
22:48:35  <isaacs>piscisaureus_: i'm just avoiding creating a buffer, only to turn around a turn it back into the exact same kind of string
22:49:41  <isaacs>ie, let's say you do: stream.setEncoding('utf8'); stream.push('some string', 'utf8')
22:51:36  <piscisaureus_>isaacs: oook - right, you changed the function signature but before you were just using arguments[1]
22:51:56  <isaacs>right
22:56:10  * sofaerquit (Quit: Leaving)
22:57:44  <piscisaureus_>isaacs: I guess this [ https://github.com/isaacs/node/commit/bf965307e9c85ff8e09c8d394e1bff3f8536ac62#L2L138 ] strictly qualifies for an API change
22:57:53  <piscisaureus_>but I can't imagine it to cause any fallout
22:58:08  <piscisaureus_>it'd mean that people rely on the second argument to push being ignored
22:58:14  <isaacs>yeah, i mean, it's a bug fix
22:58:15  <isaacs>clearly
22:58:21  <piscisaureus_>I guess we should not be overly zealous
22:58:32  <piscisaureus_>given the importance of this issue
22:58:37  <isaacs>no, we should be just zealous enough :)
22:58:54  <isaacs>where it suits us, and serves the interests of our users
22:59:38  <isaacs>oh, i guess ben is on a plane, huh..
22:59:40  <isaacs>forgot about that
22:59:53  <isaacs>piscisaureus_: any idea why libuv 0.11.2 breaks fs.watch on mac?
23:00:04  <piscisaureus_>isaacs: no clue. one thing at a time?
23:00:27  <isaacs>surely
23:00:39  <piscisaureus_>there hasn't been too many changes on the libuv front. It should be pretty obvious by looking at the commit log?
23:01:38  <tjfontaine>11.2 or 11.3?
23:01:40  <isaacs>there are a few likely culprits, i think
23:01:48  <isaacs>0.11.2
23:01:54  <tjfontaine>right k
23:01:57  <MI6>libuv-node-integration: #48 UNSTABLE windows-ia32 (10/592) windows-x64 (10/592) osx-x64 (2/592) osx-ia32 (2/592) smartos-x64 (2/592) linux-ia32 (1/592) http://jenkins.nodejs.org/job/libuv-node-integration/48/
23:02:00  <tjfontaine>now working on fooled me
23:02:01  <isaacs>but you'er right, one thing at a time
23:02:20  <piscisaureus_>isaacs: btw why are you dropping the "include path in header" rule from cpplint?
23:03:02  <isaacs>piscisaureus_: because it warns on every single file in node.
23:03:09  <piscisaureus_>oh oj
23:03:13  <isaacs>piscisaureus_: so we can either get rid of the rule, or get rid of cpplint
23:09:57  <Raynos>I switched from 0.8 to 0.10 and get "Missing PFX or certificate + private key." on a HTTPS server that previously worked
23:11:04  <Raynos>did some of the HTTPS / openssl stuff change or should I file a regression bug report?
23:11:16  <isaacs>Raynos: your server, or are you the client?
23:11:25  <Raynos>I'm the server
23:11:40  <isaacs>Raynos: and does setting and it won't start up?
23:11:56  <isaacs>er, i mean, the server wont' start up?
23:12:01  <Raynos>the error crashes my process
23:12:07  <tjfontaine>were you not previously setting cert and key?
23:12:12  <isaacs>are you using a PFX cert? or a cert+key?
23:12:16  <isaacs>Raynos: ie, what's your code?
23:12:29  <Raynos>i need to debug it a bit more to see exactly whats wrong
23:12:33  <Raynos>it might be me being naughty
23:12:41  <piscisaureus_>isaacs: did you steal the base64 decoder from somewhere or write it yourself?
23:12:53  <isaacs>piscisaureus_: i stole it from elsewhere in node.
23:12:56  <tjfontaine>I mean that error comes from a pretty trivial check
23:13:01  <Raynos>I did set "SNICallback" so it might be my fault for touching internal APIs
23:13:25  <isaacs>Raynos: are you passing a valid cert and key to the server?
23:13:34  <isaacs>even setting SNICallback should be fine
23:13:50  <Raynos> yes
23:13:59  <Raynos>this is stuff that works in production on 0.8 with no problems atm
23:14:01  <tjfontaine>https://github.com/joyent/node/blob/master/lib/tls.js#L1041
23:14:35  <tjfontaine>something be missing
23:16:01  <Raynos>ill check
23:16:16  <tjfontaine>granted it is a new check to v0.10 https://github.com/joyent/node/commit/8bec26122d6de0b230f74731c1d09da267c95add
23:16:45  <isaacs>Raynos: is it possible that you're seeing them later, rather than in the ctor?
23:16:57  <Raynos>I need to check, Im using http-proxy
23:17:11  <Raynos>Is there an easy way to add console.log to /lib without recompiling node?
23:17:36  <tjfontaine>you could use the debugger
23:18:07  <isaacs>Raynos: no. dtrace or debugger
23:19:08  <Raynos>the options I pass to https
23:19:14  <Raynos>is only the SNICallback
23:19:18  <isaacs>there you go
23:19:24  <isaacs>you have to pass the cert/key as well
23:19:31  <isaacs>presumably you HAVE a cert and a key, yes?
23:19:34  <Raynos>so maybe `if (!self.pfx && (!self.cert || !self.key) && !self.SNICallback) {
23:20:00  <Raynos>https://gist.github.com/Raynos/8175f223c1081a16bd76
23:20:08  <tjfontaine>you should really have a fallback cert for the non sni case though, I thought that was the point of setting it here
23:20:09  <Raynos>I use SNICallback to dynamically serve a cert based on hostname
23:20:33  <Raynos>i dont understand enough about https / tls to know what im supposed to do instead
23:20:37  <isaacs>oh, wait... right, because you use the SNI callback to provide the creds...
23:20:50  * c4miloquit (Remote host closed the connection)
23:20:50  * kazuponquit (Remote host closed the connection)
23:20:59  <isaacs>Raynos: i guess provide the creds for the default server, and also send us a pull request :)
23:21:27  <Raynos>when do I need default certs?
23:22:37  <tjfontaine>SNICallback should only be triggered if the client supports it, if they dont' support it you need credentials for them
23:23:33  <Raynos>I see
23:23:43  <tjfontaine>either our documentation is wrong, or you have been lucky enough to only have clients that support SNI
23:24:22  <Raynos>we only support IE9+
23:24:29  <tjfontaine>right.
23:24:35  <Raynos>and our IE9 windows XP https users we dont care about
23:24:59  <tjfontaine>yes, well -- there are plenty of other platforms and situations where this matters, not just your https server :P
23:25:13  <Raynos>https://github.com/joyent/node/pull/5451
23:25:36  <Raynos>I dont know how you should handle this wrt back compat
23:26:05  <isaacs>Raynos: yes, so, i'm kinda torn
23:26:11  <isaacs>Raynos: because not all clients will send SNI
23:26:18  <isaacs>Raynos: and so you really ought to provide a default
23:26:26  <Raynos>so if I'm correct, if I use a default cert, it will always try SNICallback unless the client does not SNI and then it will use my default?
23:26:32  <isaacs>yes
23:26:46  <Raynos>isaacs: maybe print a fat warning instead of throwing if there is an SNICallback and no cert?
23:26:47  <isaacs>or if your SNICallback function doesn't return a credentials object, i believe
23:26:56  <isaacs>Raynos: or throw a better error message :)
23:27:23  <Raynos>+100% "set SNICallback but did not provide default credentials"
23:27:49  <Raynos>my gut reaction was "shitty provider gave me malformed credential files"
23:27:59  <tjfontaine>ya I know that feeling as well
23:28:06  <isaacs>right
23:28:26  <tjfontaine>pulling useful errors out of openssl can be frustrating
23:29:33  <isaacs>ha! first guess, it was * | 145f7b3 Ben Noordhuis darwin: look up file path with F_GETPATH (4 weeks ag
23:30:14  <tjfontaine>nice
23:35:17  <isaacs>oh, lol
23:35:20  <isaacs>i'm dumb
23:35:33  <isaacs>it ADDED support for the filename, and the test was verifying that it was nul
23:35:46  <tjfontaine>ah
23:35:51  <tjfontaine>bug fixed
23:35:53  <isaacs>ircretary: tell bnoordhuis Nevermind, the test in node doesn't realize that you fixed a bug. My mistake *^_^*
23:35:54  <ircretary>isaacs: I'll be sure to tell bnoordhuis
23:36:20  <Raynos>isaacs, tjfontaine: thanks for helping out with my certs thing!
23:36:30  <tjfontaine>no problem
23:36:33  <isaacs>Raynos: np
23:37:53  <Raynos>I'll open an issue for node-http-proxy ( https://github.com/nodejitsu/node-http-proxy#using-two-certificates )
23:37:59  <Raynos>they are the root cause of this issue!
23:38:04  <tjfontaine>there is already one I think
23:39:18  <isaacs>Raynos: i think the issue here is that if you just tried a raw tls connection to that server, you'd get invalid garbage.
23:39:25  <isaacs>Raynos: ie, by ip address
23:39:33  <tjfontaine>indeed
23:39:45  <tjfontaine>actually, the connection would drop/reset
23:40:16  <MI6>joyent/node: isaacs master * 5e9c7a9 : test: Darwin file watcher has paths now (+1 more commits) - http://git.io/mCZ5Ew
23:46:49  * philipsquit (Changing host)
23:46:49  * philipsjoined
23:49:04  * groundwaterquit (Quit: groundwater)
23:50:29  <MI6>nodejs-master: #204 UNSTABLE smartos-x64 (5/592) osx-x64 (1/592) linux-ia32 (1/592) linux-x64 (2/592) http://jenkins.nodejs.org/job/nodejs-master/204/
23:53:04  <Raynos>tjfontaine: your right
23:53:33  <tjfontaine>on the drop/reset?
23:54:39  <Raynos>on there being an open issue
23:54:48  <tjfontaine>ah right
23:54:53  <Raynos>just to confirm, if I set both a key/cert and SNICallback it will try SNICallback first then the fallback cert?
23:55:49  <tjfontaine>I haven't traced the code, but yes that's what's supposed to happen, if that doesn't happen there's a legit bug there
23:56:11  <tjfontaine>Raynos: presuming of course the client advertises it supports SNI
23:56:24  <Raynos>cool.
23:56:55  <tjfontaine>later
23:59:36  * icarotquit (Ping timeout: 264 seconds)