00:00:28  * pietern_quit (Client Quit)
00:00:45  * pieternquit (Read error: Connection reset by peer)
00:01:02  * pieternjoined
00:02:45  * pfox___quit (Quit: leaving)
00:03:58  * brsonquit (Quit: leaving)
00:06:50  <txdv_>If I have a tcp socket listening on one loop, can I create a tcp socket on another loop and accept the incomming connections?
00:09:36  * brsonjoined
00:10:45  * brsonquit (Client Quit)
00:13:26  * coderaritychanged nick to coderarity_away
00:21:56  <isaacs>TooTallNate: You should put a LICENSE file on node-gyp
00:22:16  <isaacs>TooTallNate: i personally could literally not care less. however, i know that i'll get an email about it eventually from the debiants.
00:24:40  * AvianFluquit (Ping timeout: 240 seconds)
00:26:24  <TooTallNate>isaacs: i have one at the bottom of the README, but i'll make a file as well
00:26:31  * pfox___joined
00:30:48  * TooTallNatequit (Quit: Linkinus is updating...)
00:31:01  <isaacs>oh, ok
00:31:03  * TooTallNatejoined
00:41:20  * coderarity_awaychanged nick to coderarity
00:53:11  * dapquit (Quit: Leaving.)
01:02:37  * AvianFlujoined
01:13:01  <txdv_>the uv_spawn takes a char **argv, so it is an array of char arrays
01:13:17  <txdv_>does the last entry in the first dimension have to be 0?
01:23:10  * pieternquit (Ping timeout: 240 seconds)
01:27:11  * perezdquit (Quit: perezd)
01:47:30  * dshaw_quit (Quit: Leaving.)
01:57:55  * philipsquit (Excess Flood)
01:58:10  * philipsjoined
02:21:57  * pfox___quit (Ping timeout: 246 seconds)
02:23:23  * pfox___joined
02:33:09  * dylukesjoined
02:52:08  * pfox___quit (Ping timeout: 244 seconds)
03:36:12  * TooTallNatequit (Quit: Linkinus - http://linkinus.com)
04:48:48  * dshaw_joined
04:59:16  * pfox___joined
05:06:20  * isaacsquit (Remote host closed the connection)
05:11:22  * CoverSli1ejoined
05:13:11  * CoverSlidequit (Read error: Connection reset by peer)
06:05:18  * pfox___quit (Ping timeout: 276 seconds)
06:50:05  * rendarjoined
07:19:43  * dylukesquit (Quit: Pipes are broken. Sending packets via Fedex.)
07:22:14  * txdvjoined
07:26:00  * txdv_quit (Ping timeout: 252 seconds)
07:31:57  * paddybyersquit (Quit: paddybyers)
07:32:32  * paddybyersjoined
07:45:13  * dshaw_quit (Quit: Leaving.)
08:10:48  * mjr_joined
08:37:52  * mjr_quit (Quit: mjr_)
08:38:19  * piscisaureus_joined
08:44:44  <mmalecki>piscisaureus_: hey
08:45:02  <piscisaureus_>mmalecki: het
08:45:02  <piscisaureus_>*hey
08:45:14  <mmalecki>piscisaureus_: I'm arriving to Amsterdam on 12th
08:45:18  <piscisaureus_>mmalecki: I am in a train and in a tunnel
08:45:18  <piscisaureus_>Don't expect to be able to talk to me :-)
08:45:23  <mmalecki>lol ok
08:46:09  <piscisaureus_>mmalecki: beer on the 12th would work for me best I think
08:46:15  <piscisaureus_>mmalecki: also depends on what ben has to say
08:46:41  <mmalecki>piscisaureus_: if your down for a beer after 10 PM or so :)
08:46:46  <mmalecki>*you're
08:46:53  <piscisaureus_>mmalecki: what time are you arriving?
08:47:01  <mmalecki>piscisaureus_: 10 PM
08:47:24  <piscisaureus_>mmalecki: aww that is late.
08:47:24  <piscisaureus_>mmalecki: I don't mind at all
08:47:30  <piscisaureus_>mmalecki: but for ben that might turn out to be a little uncomfy
08:47:49  <piscisaureus_>since he has to take the train home and it doesn't run after 12pm or so
08:47:52  <mmalecki>piscisaureus_: 13th?
08:48:33  <piscisaureus_>mmalecki: maybe that'll work.
08:48:40  <piscisaureus_>mmalecki: I'll ping you back next week about it
08:48:49  <piscisaureus_>it would be nice I guess
08:48:53  <piscisaureus_>to do it on friday
08:50:24  <mmalecki>piscisaureus_: sure
08:50:31  <piscisaureus_>mmalecki: where are you staying btw?
08:50:31  <mmalecki>piscisaureus_: I'm running out of battery anyway
08:50:42  <mmalecki>piscisaureus_: some airbnb indexzero found
08:50:55  <mmalecki>piscisaureus_: I'm staying with him btw :)
08:52:09  <piscisaureus_>mmalecki: ah, right
08:54:27  <mmalecki>but he's staying there longer
08:54:30  <mmalecki>I'm leaving on Monday
08:54:45  <mmalecki>you know why he's staying there, right?
08:57:24  * piscisaureus_quit (Ping timeout: 250 seconds)
09:01:40  * mmaleckichanged nick to mmalecki[plane]
09:03:26  * paddybyersquit (Quit: paddybyers)
11:01:02  * AndreasMadsenjoined
11:29:00  * AndreasMadsenquit (Remote host closed the connection)
11:33:31  * AndreasMadsenjoined
12:23:13  * AndreasMadsenquit (Remote host closed the connection)
12:25:17  * AndreasMadsenjoined
12:29:55  * AndreasMadsenquit (Remote host closed the connection)
12:40:29  * paddybyersjoined
13:15:55  * philipsquit (Excess Flood)
13:16:16  * philipsjoined
13:36:01  * paddybyersquit (Quit: paddybyers)
13:37:15  * bnoordhuisjoined
13:42:54  * paddybyersjoined
13:48:05  * paddybyersquit (Quit: paddybyers)
13:51:54  * paddybyersjoined
14:05:00  * paddybyersquit (Quit: paddybyers)
14:45:53  * paddybyersjoined
14:59:16  * pfox___joined
15:14:35  * Ariajoined
15:30:38  * Ariaquit (Remote host closed the connection)
15:31:14  * Ariajoined
15:33:38  * Ariaquit (Read error: No route to host)
15:34:04  * Ariajoined
15:42:35  * Ariaquit (Remote host closed the connection)
15:51:00  * isaacsjoined
15:56:41  <isaacs>igorzi, bnoordhuis: hey. i started working on a js-only api-adding Domains implementation last night. https://github.com/isaacs/node/compare/domains-wip
16:11:45  * mmalecki[plane]changed nick to mmalecki
16:13:11  * pfox___quit (Ping timeout: 260 seconds)
16:34:23  * mikealquit (Quit: Leaving.)
16:41:37  * dylukesjoined
16:46:46  * mikealjoined
17:01:36  * pfox___joined
17:17:21  * pfox___quit (Ping timeout: 245 seconds)
17:27:55  * dshaw_joined
17:30:25  * dshaw_quit (Client Quit)
17:42:12  * perezdjoined
17:45:17  * TooTallNatejoined
18:33:22  * pieternjoined
18:40:20  * pieternquit (Quit: pietern)
19:14:06  * piscisaureus_joined
19:45:40  * mikealquit (Quit: Leaving.)
20:12:14  * mikealjoined
20:36:35  <TooTallNate>bnoordhuis: yt?
20:41:55  * dylukesquit (Quit: Computer has gone to sleep.)
20:42:06  * pfox___joined
20:44:34  * dylukesjoined
20:48:59  <TooTallNate>bnoordhuis: when you get a chance: https://github.com/joyent/node/issues/3072
20:53:03  <bnoordhuis>TooTallNate: here
20:53:38  <TooTallNate>bnoordhuis: hey
20:54:00  <TooTallNate>bnoordhuis: so what do you make of this? https://gist.github.com/681256e480d4e8eae52d
20:54:20  <bnoordhuis>TooTallNate: that you have a very fancy command prompt
20:54:39  <TooTallNate>bnoordhuis: the read.sh thing was from npm :p
20:54:54  <TooTallNate>but trying to apply the same technique to `node` doesn't seem to work
20:55:13  <TooTallNate>(or do you just mean my PS1?)
20:55:56  <bnoordhuis>yeah :) zsh, right?
20:56:01  <bnoordhuis>your example works for me btw
20:56:20  <TooTallNate>bash actually :p
20:56:27  <TooTallNate>and no shit? maybe it's just OS X
20:56:31  <bnoordhuis>echo "node < /dev/tty" | sh <- starts a repl, 1+1 prints 2, c-d closes it
20:56:38  * TooTallNatefiring my Ubuntu VM
20:56:54  <TooTallNate>then it's a bug
20:57:06  <bnoordhuis>yeah
20:57:21  <bnoordhuis>i don't think there's a separate tty code path for darwin
20:57:42  <TooTallNate>is that an easy fix or no?
20:58:42  <bnoordhuis>depends on what the cause is obviously :)
20:58:48  <bnoordhuis>let me fire up my macbook
20:58:53  <TooTallNate>bnoordhuis: indeed, works perfectly on ubuntu :) :(
21:01:13  <bnoordhuis>okay, need to rebuild node first
21:01:34  <mmalecki>yeah, I mean, she can
21:01:52  <mmalecki>derp, wrong channel
21:01:52  * TooTallNatequit (Read error: Connection reset by peer)
21:02:07  * TooTallNatejoined
21:02:14  <bnoordhuis>i thought it was a cryptic twss joke :)
21:02:43  <mmalecki>bnoordhuis: that'd make a *really* cryptic joke!
21:03:32  <bnoordhuis>TooTallNate: wfm when i ssh into my macbook
21:04:39  <bnoordhuis>TooTallNate: also works in iterm
21:04:41  <benvie>that's what she said
21:04:47  <bnoordhuis>*drum roll*
21:05:00  <benvie>it works with anything =D
21:06:25  <TooTallNate>hmmm...
21:06:27  <TooTallNate>what?
21:06:46  <benvie>nothing...just jokes
21:06:59  <TooTallNate>bnoordhuis: so it works for you? what OS X version?
21:07:05  <bnoordhuis>TooTallNate: lion
21:07:13  <TooTallNate>WWWWTTTTTFFFFFFF
21:07:33  <TooTallNate>i'm gonna reboot
21:07:39  <TooTallNate>i haven't done that in like a month
21:07:52  <mmalecki>but it should Just Work (tm)!
21:08:12  <benvie>I remember back using Windows 95 when it was a major feat to get through the day
21:08:28  * TooTallNatequit (Quit: Linkinus - http://linkinus.com)
21:10:47  <mmalecki>I'm really happy I'm going home tomorrow btw
21:11:25  <mmalecki>web5 was awesome, but France wasn't
21:13:42  * TooTallNatejoined
21:13:52  <TooTallNate>bnoordhuis: did you try only over ssh?
21:14:02  <bnoordhuis>TooTallNate: no. iterm was local
21:14:18  <TooTallNate>still doesn't work for me; tried on Terminal and iTerm, on two machines...
21:14:24  <bnoordhuis>mmalecki: what's so bad about france?
21:14:28  <bnoordhuis>except the people, that is
21:14:48  <bnoordhuis>TooTallNate: do you see anything out of the ordinary when you trace it with dtruss?
21:15:10  <mmalecki>bnoordhuis: shame you didn't go. you'd love talking about programming language design over a bottle of shine
21:15:38  <mmalecki>bnoordhuis: and they close all the shops at 6 5PM or so
21:16:32  <mmalecki>so you can't find any food
21:17:54  <txdv>how do I run specific tests? ./run-tests test-spawn aint working
21:19:12  <bnoordhuis>txdv: `grep spawn test/test-list.h`, then e.g. `./run-tests spawn_and_kill`
21:19:33  <txdv>o the specific test name
21:19:35  <bnoordhuis>yep
21:20:26  <txdv>i tried all the time the file name :(
21:20:44  <txdv>im a genius
21:21:58  <TooTallNate>bnoordhuis: does this mean anything to you? https://gist.github.com/54cd3922e762cf37a5f0
21:22:09  <TooTallNate>bnoordhuis: i see the repl prompt get written, then some error?
21:25:06  <bnoordhuis>TooTallNate: kevent(0x3, 0x7F87D8400200, 0x0) = -1 Err#4 <- that's EINTR, because you pressed ^C
21:25:09  * mjr_joined
21:25:28  <bnoordhuis>that 'out of scratch space' error is probably because dtruss tried to log a long string
21:25:45  <TooTallNate>hmmm http://stackoverflow.com/questions/1673100/application-hanging-in-semwait-signal
21:26:13  * mikealquit (Read error: Connection reset by peer)
21:26:22  <TooTallNate>answer doesn't seem to be relevant...
21:26:22  <bnoordhuis>TooTallNate: it doesn't look like it's hanging, looks like business as usual
21:26:45  <TooTallNate>bnoordhuis: when i mash on the keyboard, nothing was coming out of node or dtruss
21:26:54  <TooTallNate>(from the node tab, of course)
21:27:22  <bnoordhuis>TooTallNate: dtruss interferes with stdio
21:27:33  <bnoordhuis>i think -o will make it log to a file instead of stdout/stderr
21:27:50  <bnoordhuis>anyway, another thing you can try is to ssh to yourself and see if that works
21:27:57  <TooTallNate>bnoordhuis: but i'm running dtruss in another tab
21:28:23  <bnoordhuis>TooTallNate: oh, you're tracing by pid?
21:28:43  <TooTallNate>bnoordhuis: sudo dtruss -n node -a
21:29:02  <mjr_>bnoordhuis: Did you ever get resolution on the TLS hangs we are seeing?
21:29:20  <mmalecki>bnoordhuis: btw, does Friday work for you?
21:29:23  <bnoordhuis>TooTallNate: ah right, in that case ignore what i said about stdio :)
21:31:12  <bnoordhuis>mjr_: the ms guys sent a patch that is probably related to what you're seeing but it landed in master, not in v0.6
21:31:21  <bnoordhuis>mmalecki: sure
21:31:44  <mjr_>bnoordhuis: We get these hangs about once a day, and it's starting to be one of our larger problems.
21:32:03  <mjr_>bnoordhuis: is there any more information I can provide that would help track this down?
21:32:11  <mmalecki>bnoordhuis: great. I'll give you more details when I coordinate with MR INDEX
21:32:13  <bnoordhuis>mjr_: core dumps + binaries would help a great deal
21:32:21  * mikealjoined
21:32:57  <mjr_>OK, I've got one. I'll figure out how to get it to you.
21:33:51  <mjr_>Is there an npm module that does, "serve the files in the current directory via HTTP" ?
21:34:21  <TooTallNate>mjr_: check out `serve`
21:34:25  * mjr_part
21:34:36  * mjr_joined
21:34:41  <bnoordhuis>mjr_: try `python -m SimpleHTTPServer`
21:34:50  <mjr_>Yeah, I knew about the python thing, but I wanted a node solution.
21:34:53  <mjr_>I'll try serve.
21:36:01  <bnoordhuis>TooTallNate: i don't know what's causing your tty trouble - but it's heartening to know that it's a local issue :)
21:36:08  <mmalecki>npm -g i http-server
21:36:38  <mjr_>oh man, serve worked without reading the directions.
21:36:39  <mjr_>http://prod-2181.voxer.com:3000/
21:36:46  <mjr_>That is my kind of software
21:37:16  <mmalecki>then just `http-server` :)
21:38:23  <mjr_>bnoordhuis: there is a core file in there, node binary, output from truss, and output from dtrace jsstack.
21:38:44  <mjr_>Let me know when you have them, so I can stop this server.
21:39:40  <bnoordhuis>mjr_: will do. steadily downloading right now
21:40:22  <TooTallNate>bnoordhuis: interesting. it worked on my mom's Snow Leopard MBP...
21:40:55  <mmalecki>TooTallNate: want some OS X testing?
21:41:09  <TooTallNate>mmalecki: ya definitely
21:41:17  <mmalecki>I got a fairly fresh MBA here
21:41:23  <TooTallNate>mmalecki: does `node < /dev/tty` work for you?
21:41:29  <TooTallNate>mmalecki: i.e. get a usable REPL?
21:41:30  <mmalecki>there's a cool story behind this laptop btw.
21:41:38  <mmalecki>TooTallNate: yeah, I do
21:41:44  * TooTallNatewant's a MBA
21:41:46  <TooTallNate>mmalecki: thanks
21:42:20  <mmalecki>TooTallNate: yeah, MBA is quite cool, I didn't think I'll like it
21:42:45  <mmalecki>I should probably tell you guys the story
21:42:52  <bnoordhuis>mjr_: done
21:42:52  <TooTallNate>did you win it?
21:43:06  <mmalecki>TooTallNate: haha, no
21:43:12  <mjr_>GET /core.13107 200 212045ms
21:43:16  <mjr_>That's a lot of milliseconds.
21:43:35  <mmalecki>so, yeah, I was at this web5 conference. I had a talk on my old MBP, all went well
21:43:44  * mikealquit (Quit: Leaving.)
21:43:54  <mmalecki>then drinks
21:44:00  <bnoordhuis>mjr_: i'm looking at the truss trace. why is it calling getpid() all the time?
21:44:21  <mjr_>bnoordhuis: I dunno. Accidental SmartOS inefficiency?
21:44:26  <mmalecki>then I try to turn the MBP on
21:44:29  <mmalecki>WELL
21:44:35  <mmalecki>flashing directory icon
21:44:47  <bnoordhuis>mjr_: do you guys use any native add-ons?
21:44:54  <mjr_>hiredis
21:45:00  <mmalecki>it's saying "fuck you"
21:45:12  <mjr_>mmalecki: that is the saddest if icons
21:45:15  <bnoordhuis>mjr_: oh right, you've told me that before :)
21:45:36  <mjr_>bnoordhuis: we use hardly any modules from npm, for exactly this reason.
21:45:51  <TooTallNate>bnoordhuis: how freshly compiled is your node?
21:45:54  <TooTallNate>mmalecki: you too?
21:46:01  <mmalecki>and you know, I do this devops stuff, so I kinda need some machine
21:46:02  <TooTallNate> which versions?
21:46:18  <bnoordhuis>TooTallNate: just now, master
21:46:19  <mjr_>If anybody else cares, here's the flame graph for my "node gets stuck in TLS" bug: http://ranney.com/stuck.svg
21:46:30  <mmalecki>the very next day I'm like "ok fuck this shit, I'm totally getting a new one"
21:46:43  <mmalecki>(I plan to get the MBP fixed in Poland)
21:46:51  <TooTallNate>mmalecki: what version of node are you running?
21:46:57  <mmalecki>TooTallNate: master
21:47:03  <TooTallNate>compiled when?
21:47:10  <mmalecki>TooTallNate: today, I think?
21:47:15  <TooTallNate>wow
21:47:16  <mmalecki>sec
21:47:17  <TooTallNate>incredible
21:47:30  <TooTallNate>bnoordhuis: it may have been something fixed very recently
21:47:50  <bnoordhuis>TooTallNate: you mean you're not running the current master?
21:47:53  <mmalecki>so I leave wanting to buy some cheap linux laptop
21:48:02  <mjr_>bnoordhuis: isn't it also interesting how often the time() sys call gets called? Like, WAY too often for something with low resolution like that. It'd be better to just have a 1 second interval timer and stash the value for the whole process to share.
21:48:06  <TooTallNate>bnoordhuis: was on v0.7.7 tag
21:48:30  <bnoordhuis>mjr_: i suspect that's libev's doing
21:48:52  <mmalecki>TooTallNate: yeah, "Already up-to-date."
21:49:12  <mjr_>That's gotta be slowing everybody down, but perhaps we can look at that optimization later.
21:49:28  <bnoordhuis>mjr_: unless you have code in your application somewhere that calls e.g. `+new Date()` all the time
21:50:07  <mjr_>bnoordhuis: We do Date.now() quite a bit, but that shouldn't use time(2), because it's not accurate enough.
21:50:11  <TooTallNate>oh man, make -j8 rules
21:50:32  <mmalecki>anyway, when I enter the shop I notice this 11" MBA
21:51:19  <mmalecki>so, yeah, that's the part when I bump my credit card limit and buy it
21:51:26  <mjr_>TooTallNate: wait till you see make -j24 on a 24 core machine with SSDs. It is winful.
21:51:26  <bnoordhuis>eip 0x0 0 :(
21:51:42  <bnoordhuis>iow, the core doesn't point to anything useful
21:51:46  <TooTallNate>mjr_: holy cow
21:52:10  <mjr_>bnoordhuis: could that be because the node binary doesn't match?
21:52:12  <TooTallNate>mmalecki: is 11" good? i feel like i need to wait for 15"
21:52:51  <bnoordhuis>mjr_: that's with the node binary i downloaded from you
21:53:07  <bnoordhuis>mjr_: could be because /voxer/deploy/server/node_modules/hiredis/build/Release/hiredis.node is missing
21:53:33  <mjr_>bnoordhuis: sure, but I'm wondering if perhaps we upgraded node before I copied the binary.
21:54:07  <mmalecki>TooTallNate: sure it's good!
21:54:21  <bnoordhuis>mjr_: it reports itself as v0.6.14
21:54:29  <mmalecki>TooTallNate: I didn't think it'll be this good, actually
21:55:16  <mjr_>bnoordhuis: on the same machine where the core was taken, I still get this: #0 0x00000000 in ?? ()
21:55:23  <mjr_>like the core file is bad or something
21:55:32  <bnoordhuis>mjr_: what does `info registers` print?
21:56:04  <mjr_>https://gist.github.com/2332381
21:56:43  <mjr_>seems broken
21:56:45  <bnoordhuis>mjr_: right, that's the same i'm seeing
21:56:55  <bnoordhuis>how was the core generated?
21:57:03  <mjr_>On that machine it can load that hiredis.node file, so that's not it.
21:57:10  <mjr_>gcore
21:57:29  <bnoordhuis>hm, should just work
21:57:37  <mjr_>[root@prod-2181 ~]# history | grep gcore
21:57:38  <mjr_> 147 gcore 13107
21:58:20  <mjr_>OK, let's scrap this one and try again the next time it locks up.
21:58:33  <bnoordhuis>mjr_: one more thing that's odd
21:58:48  <bnoordhuis>the core captured two threads, which is more or less expected, but they both have eip == 0
21:59:00  <mjr_>Seems like the core is just mangled.
21:59:16  <mjr_>Or not captured or loaded correctly.
21:59:59  <mjr_>So the next time it happens, I'll gcore it. Does it help to do truss or some dtrace magic on it as well while it's in the stuck state?
22:00:03  <bnoordhuis>mjr_: you know, i think gcore messes up
22:00:12  <mjr_>It's clearly a very busy process.
22:00:23  <bnoordhuis>i just tried it on a smartos machine and while it dumps the core file, it's not usable afterwards
22:01:00  <mjr_>Maybe we need some help from our more Solaris-y friends?
22:01:41  <bnoordhuis>mjr_: can you try this? set `ulimit -c unlimited` before starting the process
22:01:53  <bnoordhuis>and when it happens kill it with `kill -ABRT <pid>`
22:01:59  <mjr_>I'm not sure that ulimits work right on SmartOS.
22:02:05  <mjr_>I think you need to use that other interface.
22:02:21  <bnoordhuis>oh? it seems to wfm
22:02:22  <mjr_>prctl and friends.
22:02:26  <TooTallNate>bnoordhuis: *shrug* i don't feel like bisecting it, but it was fixed in between v0.7.7 and master
22:02:34  <TooTallNate>bnoordhuis: so thanks, got it working :)
22:02:37  <mjr_>Yeah, it seems to, but I find that it doesn't actually change the value where it counts.
22:03:18  <bnoordhuis>TooTallNate: good :)
22:03:54  <bnoordhuis>mjr_: okay. in that case you should maybe bug some joyent people
22:04:04  <TooTallNate>bnoordhuis: lol, was the typo intentional? https://github.com/joyent/node/commit/17524432ff50c7e2ba53aa71b61c2f7409d5e02a
22:04:16  <bnoordhuis>TooTallNate: yeah :)
22:04:21  <TooTallNate>hahaha
22:05:15  <piscisaureus_>bnoordhuis: So basically you replaced English by math-speak
22:05:59  <bnoordhuis>piscisaureus_: yes. i'm doing my part in moving the world to a purer place
22:06:07  <bnoordhuis>next up is replacing all that messy javascript with lisp
22:06:14  <mjr_>bnoordhuis: I think we have the max setting in there already:
22:06:17  <mjr_>[root@prod-2181 ~]# prctl -P 13198 | grep core
22:06:18  <mjr_>process.max-core-size privileged 9223372036854775807 max deny -
22:06:18  <mjr_>process.max-core-size system 9223372036854775807 max deny -
22:06:31  <bnoordhuis>that should be enough :)
22:06:35  <mjr_>8EB
22:06:40  * piscisaureus_is away
22:07:02  <bnoordhuis>if you send the process a SIGABRT, it should dump out a usable core file
22:07:17  <mjr_>OK, let me try a test.
22:09:03  <mjr_>bnoordhuis: it does make a usable core file on ABRT. Good registers and stack.
22:09:08  <mjr_>So I guess I'll do that next time.
22:09:16  <bnoordhuis>cool, thanks
22:09:34  <bnoordhuis>i suppose i should look into why gcore doesn't work, it's quite inconvenient
22:09:49  <bnoordhuis>but i don't particularly want to dive into gdb's internals :)
22:10:13  <bnoordhuis>for one, i'm allergic to the gnu coding style...
22:10:30  * piscisaureus_quit (Read error: Operation timed out)
22:10:43  * mikealjoined
22:12:21  * rendarquit (Ping timeout: 252 seconds)
22:15:42  <TooTallNate>bnoordhuis: the winner is: https://github.com/joyent/node/commit/b6d6a54f8057d0adad13d4bcc11eb3f443079ddc
22:15:50  <TooTallNate>bnoordhuis: so compiler bug?
22:17:06  <TooTallNate>bnoordhuis: that's the commit that adds "-mmacosx-version-min=10.5" to the CFLAGS
22:19:39  <bnoordhuis>TooTallNate: linker issue maybe, i believe -mmacosx-version-min only touches the linker settings
22:19:48  <bnoordhuis>it works for you when you revert that commit?
22:20:24  * brsonjoined
22:20:28  <TooTallNate>bnoordhuis: no, that's the commit that makes it start working
22:20:41  <TooTallNate>i.e. tagged v0.7.7 doesn't work
22:20:42  <TooTallNate>try it
22:20:49  <bnoordhuis>oh right. all is well in that case, isn't it?
22:20:56  <TooTallNate>yup :)
22:30:53  * dylukesquit (Quit: Computer has gone to sleep.)
22:41:57  * dylukesjoined
22:54:37  * mikealquit (Quit: Leaving.)
22:56:05  * mikealjoined
23:21:45  <CIA-99>node: isaacs v0.6 * re81a5e9 / LICENSE : Remove stray comment from LICENSE file (+5 more commits...) - http://git.io/JEw3EQ
23:29:31  * travis-cijoined
23:29:31  <travis-ci>[travis-ci] joyent/node#704 (v0.6 - e81a5e9 : isaacs): The build is still failing.
23:29:31  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/37ac5e5...e81a5e9
23:29:31  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/1041690
23:29:31  * travis-cipart
23:44:31  <isaacs>bnoordhuis: hm... test/simple/test-file-read-noexist.js is failing on v0.6 now. any ideas?
23:44:57  <bnoordhuis>isaacs: didn't igorzi enable that test recently?
23:45:28  <bnoordhuis>oh wait no, that's test-file-largefile
23:45:53  <isaacs>yeah, this is just doing fs.readFile("does not exist"...) and asserting that there's an error.
23:45:58  <isaacs>but instead the cb never gets called.
23:46:01  <isaacs>which seems rather ominoux
23:46:03  <isaacs>*ominous
23:46:16  <isaacs>worked on 0.6.154
23:46:21  <isaacs>*on 0.6.14
23:46:32  <isaacs>bisecting
23:47:21  * perezdquit (Quit: perezd)
23:47:23  <bnoordhuis>isaacs: it's commit 40b7302
23:47:27  * paddybyersquit (Quit: paddybyers)
23:48:36  <isaacs>interesting
23:49:30  <isaacs>yes, bisect agrees
23:51:40  <isaacs>hm. yes, so if it doesn't have an fd, then destroy will close() on the 'open' event, which never occurs, because it was an error opening.
23:54:41  <isaacs>got a fix that makes the test pass. running make test now
23:56:38  <isaacs>bnoordhuis: review? https://github.com/isaacs/node/commit/45b772d8cb3fb2c3a4a07f30056879bc054ee627
23:57:38  <bnoordhuis>isaacs: is !fd correct?
23:58:00  <isaacs>i suppose you could be running fs.readFile on stdin
23:58:15  <isaacs>but normally, yes, fd will be null, or a positive integer
23:58:19  <isaacs>once the file is opened
23:58:33  <isaacs>this regression happens if the error occurs before the file is opened, in which case, there will never be a close event
23:58:55  <bnoordhuis>yes, lgtm
23:58:58  <isaacs>if there is a read error, but the file is open, then we need to close it before calling the cb
23:59:02  <isaacs>great, thanks
23:59:16  * stephankquit (Ping timeout: 260 seconds)
23:59:51  <CIA-99>node: isaacs v0.6 * r57d7220 / doc/community/index.html : website: Add October Sky JS and Node.js Hispano - http://git.io/T-kGvg
23:59:51  <CIA-99>node: isaacs v0.6 * r08e2e57 / (Makefile tools/changelog-head.sh tools/email-footer.md): Make rules for release blog post and email message - http://git.io/1Pw6zw
23:59:51  <CIA-99>node: isaacs v0.6 * r45b772d / lib/fs.js :
23:59:51  <CIA-99>node: fs.readFile: Emit 'close' if we don't expect a close later
23:59:51  <CIA-99>node: This fixes a regression introduced by 40b7302 - http://git.io/5XqA_Q