01:34:45  <CIA-53>libuv: Ryan Dahl master * r2f18cf5 / uv.gyp : Fix gyp build - http://git.io/3uNKMA
01:34:51  <ryah>pquerna: --^
02:29:43  * ericktjoined
02:55:05  * isaacsquit (Quit: isaacs)
03:50:32  * dmkbotquit (Remote host closed the connection)
03:50:37  * dmkbotjoined
04:03:55  * piscisaureus__joined
04:23:09  * mikealjoined
04:25:31  * isaacsjoined
04:33:37  <CIA-53>node: Ryan Dahl domains * r517604b / lib/dns_uv.js : Fix test-domains.js - http://git.io/hQpqRg
04:33:38  <CIA-53>node: Ryan Dahl domains * re1eee5b / (lib/domains.js lib/timers_uv.js test/simple/test-domains2.js): domains: handle not being removed correctly - http://git.io/sNFZTA
04:33:38  <CIA-53>node: Ryan Dahl domains * rafa08ad / (lib/domains.js lib/timers_uv.js src/node.js): Fix test-domains2.js - http://git.io/RSb6NQ
04:35:17  <ryah>^-- :D
04:35:20  <ryah>kind of working now
04:35:33  <ryah>i've got it under a switch, so I think I'm just going to merge...
04:37:29  * mikealquit (Ping timeout: 252 seconds)
04:37:44  <piscisaureus__>nice
04:47:10  <pquerna>ryah: oh, damn, i was wondering why the existing -framework CoreServices wasn't in the link line either
04:53:15  * mikealjoined
04:54:25  * ericktquit (Quit: erickt)
04:59:38  <bnoordhuis>piscisaureus__: (mees is nog wat grieperig, vandaar)
04:59:45  <bnoordhuis>err
05:00:04  <piscisaureus__>bnoordhuis: Ik heb het eerste stuk van je verhaal gemist denk ik
05:00:15  <bnoordhuis>ja, dat had ik ge-im'd
05:00:44  <bnoordhuis>bertje, ik ga waarschijnlijk kings of code skippen vandaag / hele nacht opgeweest, beetje geradbraakt
05:00:47  <bnoordhuis>^ dat
05:00:47  <ryah>it's like misspelled german :)
05:01:05  <bnoordhuis>more like properly spelled german >:(
05:01:28  <ryah>ich hab heute erste stuck von jedre (verhaal?) (gemist?) denke ich
05:01:34  <ryah>^-- is this right?
05:01:50  <bnoordhuis>it's... creative
05:02:12  <ryah>bnoordhuis: you know german - or?
05:02:18  <bnoordhuis>yes
05:02:27  <ryah>i think all dutch can understand german more or less?
05:02:41  <bnoordhuis>yes, mostly
05:02:47  <ryah>like californians and spanish :)
05:02:51  <bnoordhuis>heh
05:03:06  <piscisaureus__>Ich habe das erste (stuk?) von dein geschichte gemisst, denke ich
05:03:14  <piscisaureus__>That's probably all wrong
05:03:33  <ryah>sounds okay to me
05:03:37  <bnoordhuis>you'd get away with it
05:09:02  <piscisaureus__>I hate preparing talks
05:09:29  <piscisaureus__>I think today I am going to insult my audience
05:27:12  * piscisaureus__quit (Ping timeout: 260 seconds)
05:54:03  * bnoordhuisquit (Ping timeout: 260 seconds)
06:03:52  * mralephjoined
07:02:13  * mralephquit (Quit: Leaving.)
07:05:49  * isaacsquit (Quit: isaacs)
08:31:09  * pieternjoined
08:32:11  * pieternquit (Client Quit)
08:38:22  * piscisaureus_joined
08:57:41  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
09:00:00  * pieternjoined
10:56:25  * pieternquit (Ping timeout: 260 seconds)
13:49:45  * ericktjoined
14:09:40  * bnoordhuisjoined
14:29:19  <CIA-53>node: Ben Noordhuis master * r243c218 / lib/tls.js : tls: remove superfluous setOptions() call - http://git.io/BcPlzQ
14:34:09  * isaacsjoined
14:44:53  <CIA-53>node: Ben Noordhuis master * r3aa1975 / wscript : build: make wscript work with python 2.4 - http://git.io/cqNDug
14:52:44  * ericktquit (Quit: erickt)
14:57:54  * isaacsquit (Quit: isaacs)
14:59:14  <CIA-53>libuv: Ben Noordhuis master * r8f6f324 / common.gypi :
14:59:15  <CIA-53>libuv: build: use -m32 only when host_arch != target_arch
14:59:15  <CIA-53>libuv: Unbreaks the gyp build on x86_64 linux systems. - http://git.io/TxMHDA
15:12:21  <CIA-53>libuv: Erick Tryzelaar master * r70e1032 / (src/unix/core.c test/test-getaddrinfo.c):
15:12:21  <CIA-53>libuv: unix: Fix uv_getaddrinfo from deleting invalid data
15:12:21  <CIA-53>libuv: If the uv_getaddrinfo_t handle is owned by its
15:12:21  <CIA-53>libuv: data pointer, deleting the data in the callback
15:12:21  <CIA-53>libuv: could cause uv_getaddrinfo_done to call freeaddrinfo
15:12:22  <CIA-53>libuv: on an invalid pointer. - http://git.io/c4kcwQ
15:23:51  <CIA-53>libuv: Ben Noordhuis master * r37e8bf2 / src/unix/core.c : unix: fix dangling pointer free() bug - http://git.io/gdaLRQ
15:24:32  * pieternjoined
15:36:59  <bnoordhuis>pietern: i think bert's been chasing you to sign the cla
15:37:18  <pietern>bnoordhuis: haven't heard from him
15:37:28  <bnoordhuis>well, by proxy then :)
15:37:36  <bnoordhuis>it's okay, we only want your first born
15:37:41  <pietern>bnoordhuis: but I think we can put the couple of things I did under his name without problems
15:37:46  <pietern>hehehe
15:37:57  <bnoordhuis>no, it's this thing: http://nodejs.org/cla.html
15:38:11  <bnoordhuis>essentially that you're allowed to submit the code
15:39:04  <pietern>bnoordhuis: not sure what the implications are if I will contribute something later from my job at vmw
15:39:15  * mikealquit (Quit: Leaving.)
15:39:20  <pietern>so basically there are 2 CLA's, and I can personally sign the first one
15:39:28  <bnoordhuis>okay
15:39:29  <pietern>but I want to check what's up with the second one first
15:39:38  <bnoordhuis>i think we have a vmware guy on the list already
15:40:29  <pietern>bnoordhuis: well, I'd rather check first how things are done
15:40:32  <bnoordhuis>oh no, sorry - different company, same branch of business
15:40:36  <bnoordhuis>sure
15:40:38  <pietern>I see
15:40:51  <pietern>and I'm fine with bert using the few win patches I did under his name
15:40:55  <pietern>they're pretty minor
15:41:25  <bnoordhuis>oh, it's not an attribution thing
15:41:41  <bnoordhuis>it's a copyright / indemnification thing
15:44:05  <pietern>bnoordhuis: hmm, I don't really get it
15:44:29  <pietern>it's MIT, right? so why add a personal "no warranty" clause?
15:46:19  <bnoordhuis>pietern: oh dear, i'm not an american copyright lawyer
15:46:24  <pietern>bnoordhuis: hehe,
15:46:39  <pietern>I can stop whining and just sign the damn thing, but I rather understand it first ;)
15:47:01  <bnoordhuis>i think that the main thing is that joyent wants to ensure that people are allowed to send in the patches they submit
15:47:17  <bnoordhuis>if you write a patch in your company's time, the copyright belongs to the company
15:47:40  <bnoordhuis>if you submit it and we merge, joyent could be liable
15:47:47  <pietern>ah, I see
15:48:23  <pietern>but that does mean that if I attribute these patches to my own time, and write other patches in the company's time
15:48:33  <pietern>I need to sign it once more with the extra clause
15:49:19  <bnoordhuis>yes
15:58:36  * mikealjoined
16:25:05  * isaacsjoined
16:26:26  <CIA-53>node: Ryan Dahl domains * r9acb93f / src/node.cc : Do not try-catch in MakeCallback when using domains - http://git.io/QolMQw
16:26:26  <CIA-53>node: Ryan Dahl domains * rb8023a7 / (5 files in 2 dirs): Stop lazy loading domains - http://git.io/TWGIHg
16:26:26  <CIA-53>node: Ryan Dahl domains * r7d5b38e / (lib/domains.js lib/net_uv.js lib/timers_uv.js): surround a bunch of stuff with domains.enable branches - http://git.io/t1w8fw
16:26:26  <CIA-53>node: Ryan Dahl domains * r983d421 / test/simple/test-domains3.js : add another domain test - broken - http://git.io/MdyfXw
16:30:35  * isaacsquit (Quit: isaacs)
16:43:12  * mikealquit (Quit: Leaving.)
16:44:19  * isaacsjoined
16:45:38  * isaacspart
16:46:00  * isaacsjoined
16:53:11  * ericktjoined
17:13:34  * igorzijoined
17:23:17  <isaacs>beginning to regret fixing the automatic updater in my windows vm
17:23:21  <isaacs>it always wants to restart now
17:27:13  * brsonjoined
17:35:33  * dmkbotquit (Remote host closed the connection)
17:35:40  * dmkbotjoined
17:36:13  * dmkbotquit (Remote host closed the connection)
17:36:17  * dmkbotjoined
17:36:49  * dmkbotquit (Remote host closed the connection)
17:36:53  * dmkbotjoined
17:37:13  <ryah>pietern: hey - i can probably answer CLA questions
17:37:25  * dmkbotquit (Remote host closed the connection)
17:37:29  * dmkbotjoined
17:38:01  * dmkbotquit (Remote host closed the connection)
17:38:06  * dmkbotjoined
17:38:38  * dmkbotquit (Remote host closed the connection)
17:38:42  * dmkbot1joined
17:39:13  <igorzi>ryah: isaacs: bnoordhuis: piscisaureus: could we have the meeting 30 min later today?
17:39:44  * dmkbot1quit (Remote host closed the connection)
17:40:00  * dmkbotjoined
17:40:31  <ryah>igorzi: fine with me
17:40:32  * dmkbotquit (Remote host closed the connection)
17:40:37  * dmkbotjoined
17:40:40  * dmkbotquit (Remote host closed the connection)
17:42:53  * dmkbotjoined
17:43:26  * dmkbotquit (Remote host closed the connection)
17:43:30  * dmkbotjoined
17:43:36  * dmkbotquit (Remote host closed the connection)
17:44:37  <ryah>" build: make wscript work with python 2.4"
17:44:40  <ryah>^-- bnoordhuis thansk!
17:46:47  * dmkbotjoined
17:47:20  * dmkbotquit (Remote host closed the connection)
17:47:23  * dmkbotjoined
17:48:11  * dmkbotquit (Remote host closed the connection)
17:48:16  * dmkbotjoined
17:48:18  * dmkbotquit (Remote host closed the connection)
17:53:03  * dmkbotjoined
17:53:35  * dmkbotquit (Remote host closed the connection)
17:53:37  <bnoordhuis>call at 20:30?
17:53:40  * dmkbotjoined
17:53:42  * dmkbotquit (Remote host closed the connection)
17:53:50  <bnoordhuis>err... half an hour from now, i mean?
17:54:51  <igorzi>bnoordhuis: looks like piscisaureus can't do 30 min from now.. he can do 1 hour from now.
17:55:02  <bnoordhuis>oh, that's fine too
17:55:27  <bnoordhuis>right, just got the email
17:55:33  <igorzi>thanks.. sorry about rescheduling
17:56:20  <bnoordhuis>np, igorzi
18:03:44  <ryah>12-12:20p it is
18:13:29  * ryahlocks ./node test/simple/test-regress-GH-819.js
18:13:47  <ryah>have you guys been using the debugger?
18:13:49  <ryah>it's really nice
18:21:32  * dmkbot1joined
18:22:14  * dmkbot1quit (Remote host closed the connection)
18:22:20  * dmkbotjoined
18:58:13  <bnoordhuis>With today's head, 'make test' yields 100% of the test cases failing with: `--- CRASHED ---`. <- what?
18:58:20  <bnoordhuis>https://github.com/joyent/node/issues/1401#issuecomment-2137076
18:59:52  <bnoordhuis>ryah: https://github.com/joyent/node/issues/1739 <- produces "Error: ENOENT, Success '/tmp/DDDD'" on linux, what does it do on os x?
19:08:21  * graydonjoined
19:08:37  * graydonquit (Client Quit)
19:11:11  <ryah>shit
19:11:14  <ryah>i frogot the meeting
19:11:36  * piscisaureus_joined
19:11:37  <bnoordhuis>so... now?
19:11:55  <piscisaureus_>yes!
19:12:06  <piscisaureus_>I'm waiting for it :-)
19:12:11  <ryah>piscisaureus_: ringing you on skype
19:12:51  <piscisaureus_>skype doesn't collaborate it seems :-/
19:12:57  <ryah>my skype crashed
19:12:59  <ryah>one sec
19:14:24  <piscisaureus_>restaurant
19:15:01  * brsonquit (Ping timeout: 240 seconds)
19:18:54  <piscisaureus_>bnoordhuis: are you better again btw?
19:19:47  <piscisaureus_>igorzi: the raw iocp bench is
19:19:58  <piscisaureus_>still on the map?
19:23:00  <bnoordhuis>piscisaureus_: yeah, it was mostly mees who was not feeling well
19:23:27  <bnoordhuis>on my part, it's the sleepless night that did me in
19:25:48  <bnoordhuis>heh, now my skype did crash
19:26:01  <piscisaureus_>Oh shit
19:26:10  <piscisaureus_>Isaac or me is going to be next
19:26:23  <piscisaureus_>It's like final destination :-/
19:26:28  <bnoordhuis>or that movie, the ring
19:26:30  <bnoordhuis>igorzi: you finished with the file watcher stuff?
19:27:01  <igorzi>bnoordhuis: yep - https://github.com/joyent/libuv/tree/file_watcher
19:27:12  <piscisaureus_>Okay. I am going to leave this shady restaurant
19:27:24  <piscisaureus_>I am so sad that I did not get to have a fancy dinner with ben
19:27:31  <bnoordhuis>aw sorry, piscisaureus_
19:27:36  <bnoordhuis>maybe later this week?
19:28:06  <piscisaureus_>maybe, let me get back to you about that
19:28:14  <bnoordhuis>igorzi: the only thing to decide on is the final api?
19:28:18  <piscisaureus_>today was just right for this :-)
19:28:25  <igorzi>bnoordhuis: yep
19:29:18  <bnoordhuis>ryah: burden of responsibility: file watcher api, how?
19:40:41  <piscisaureus_>ryah: yes - land the file watcher
19:42:31  <piscisaureus_>er
19:42:37  <piscisaureus_>the tty wrap
19:42:41  <piscisaureus_>I am such an idiot
19:45:24  <piscisaureus_>http://code.google.com/p/v8/source/detail?r=9328 <-- mraleph++
19:48:32  * brsonjoined
19:52:00  * piscisaureus_quit (Ping timeout: 260 seconds)
20:03:55  * mralephjoined
20:22:53  <igorzi>piscisaureus: https://gist.github.com/1227502
20:23:44  <igorzi>^------ uv server, which repros the truncated write
20:24:06  <igorzi>(ab -n 100 http://localhost/)
20:28:35  * bnoordhuisis back in a couple of hours
20:33:11  * bnoordhuisquit (Ping timeout: 258 seconds)
20:42:23  <CIA-53>node: Ryan Dahl master * r6326b04 / lib/net_uv.js :
20:42:23  <CIA-53>node: Fix test-regress-GH-819
20:42:23  <CIA-53>node: Problem was exposed in 12798c6. - http://git.io/sCcSbQ
20:44:46  <ryah>mraleph: congradulations
20:45:08  <ryah>*congratulations
20:45:21  <ryah>mraleph: will we see this in trunk soon?
20:45:53  * pieternquit (Quit: pietern)
20:46:54  <mraleph>ryah: thanks. not sure
20:47:06  <mraleph>ryah: buildbots (debug ones) are pretty red
20:47:21  <mraleph>ryah: we will be hammering them down first
20:48:25  <ryah>mraleph: can it GC 4gigs in 15ms?
20:48:40  <mraleph>ryah: should be able to
20:49:00  <ryah>mraleph: got any stats that i can give to people?
20:49:26  <mraleph>ryah: no numbers
20:50:06  <mraleph>ryah: also last time I checked there were degradations on http_simple_bench; that also has to be investigates separately.
20:50:18  <mraleph>*investigated
20:50:30  <mraleph>there are still a lot of work and tuning and fixing ahead
20:50:58  <ryah>mraleph: this is still single threaded, stop the world - right?
20:51:52  <mraleph>yes it is
21:11:54  <piscisaureus>mraleph: does this mean that I now have to file a bug "2mb memory limit"?
21:12:11  <piscisaureus>or are you going to WONTFIX that anyway?
21:15:55  <mraleph>piscisaureus: what do you mean?
21:16:05  <mraleph>piscisaureus: there is no limit
21:16:14  <piscisaureus>mraleph: hmm. Erik Corry says there is
21:16:30  <mraleph>piscisaureus: that's an example
21:16:31  <piscisaureus>mraleph: http://code.google.com/p/v8/issues/detail?id=847
21:16:41  <piscisaureus>oh - ok
21:16:42  <piscisaureus>nice
21:16:51  <piscisaureus>It's not clear from the message
21:16:57  <mraleph>yeah
21:17:12  <mraleph>it also says "The limit is not known on 64 bit."
21:17:15  <mraleph>but yep
21:17:19  <mraleph>that's confusing
21:17:59  <piscisaureus>mraleph: so the limit is just determined by how much continuous virtual address space v8 can reserve?
21:18:43  <piscisaureus>(when I read the bug comments I really get disappointed in humanity btw)
21:20:20  <mraleph>piscisaureus: well… not continiuos
21:20:37  <mraleph>the limit is determined by how much OS is ready to give you
21:20:46  <mraleph>that's basically all
21:20:53  <mraleph>it will not be reserved at once
21:20:59  <mraleph>it will grow as the heap gorws
21:24:34  * mralephback to danish homework
21:25:43  <piscisaureus>Ah, er han at lære dansk. Det er ust om tid.
21:26:55  <piscisaureus>s/ust/bare/
21:28:05  <piscisaureus>He can finally buy a knackebrod
21:28:13  <piscisaureus>Han kan endelig købe en Knækbrød
21:28:31  <mraleph>rofl
21:29:03  <mraleph>You are Dutch. I should be easier for you than for me.
21:29:16  <mraleph>I don't like languages with fixed word order :-)
21:29:31  <piscisaureus>You like tenses?
21:29:43  * piscisaureushits the wall
21:29:53  <mraleph>tenses are fine. perfects are not so fine but ok.
21:30:08  <piscisaureus>hmm. I don't mean tenses
21:30:16  <piscisaureus>erm. what's that called
21:30:37  <mraleph>ah I understand. forms of nouns
21:31:04  <mraleph>grammatical cases
21:31:18  <piscisaureus>"Falle" in german
21:31:28  <piscisaureus>exactly
21:32:47  <mraleph>I've never tried to learn foreign language with them like German or Estonian or what have you. I am fine with them in Russian, but thats because it's in my blood :-)
21:33:52  <piscisaureus>Да я это понимаю.
21:33:59  <piscisaureus><3 google
21:39:53  * pieternjoined
21:45:59  <piscisaureus>igorzi: is this the error you get:
21:45:59  <piscisaureus>apr_socket_recv: Software caused connection abort. (730053)
21:46:03  <piscisaureus>?
21:47:13  <igorzi>piscisaureus: apr_recv: An established connection was aborted by the software in your host machine. (730053)
21:47:44  <piscisaureus>yes - that's WSAECONNABORTED
21:48:37  <piscisaureus>igorzi: that happens because we close the connection while the client is trying to send something
21:48:57  <piscisaureus>igorzi: you must wait for the http request to arrive first
21:51:23  <igorzi>oh crap... so we need to rewrite this server a bit so that it reads the request first before sending the response
21:52:36  <igorzi>piscisaureus: were you getting dropped requests from wcat when you were running you prefork server?
21:53:34  <piscisaureus>igorzi: not really, maybe a few (like < 10 total) errors or so
21:53:45  <piscisaureus>igorzi: that could be timing related too
21:54:30  <piscisaureus>igorzi: but ECONNABORTED or ECONNRESET is expected when the peer closes the socket before you have written everyhing
21:54:39  <piscisaureus>it is not a bug
21:55:39  <igorzi>piscisaureus: yes. sorry for the false alarm. i should have debugged this further before declaring this was the return of the truncated write bug.
21:55:42  <piscisaureus>normally you should do a graceful shutdown - but this close-after-write is really a http-specific optimization
21:55:51  <piscisaureus>igorzi: np
21:56:01  <piscisaureus>igorzi: I do that sometimes too
21:57:37  <igorzi>piscisaureus: but, we should probably rerun prefork with the modified server that reads the request first before writing back the response. also with larger backlog.
21:57:44  <piscisaureus>yes
21:57:48  <piscisaureus>absolutely
21:58:36  <piscisaureus>igorzi: do you mind if I do that tomorrow, had to get up at 4:30am today, I'm pretty worn off now
21:59:09  <igorzi>piscisaureus: yeah np
21:59:24  <piscisaureus>igorzi: sorry - I know I have been a little unresponsive lately
21:59:39  <piscisaureus>I wil report my findings when you wake up ;-)
21:59:58  <piscisaureus>is the cluster booted into windows again btw?
22:00:01  <igorzi>piscisaureus: ok :)
22:00:41  <igorzi>piscisaureus: it's still booted into linux.. i'll either boot it into windows, and let you know how to do that before i take off
22:00:53  <piscisaureus>igorzi++
22:30:55  * pieternquit (Quit: pietern)
23:04:06  <CIA-53>libuv: Igor Zinkovsky master * ree115bf / src/win/process.c : windows: fail uv_spawn() if options.file is NULL - http://git.io/8UpESA