15:03:12  * mikolalysenkojoined
15:08:21  * mikolalysenkoquit (Ping timeout: 272 seconds)
15:11:12  * shamajoined
15:12:35  * kevino80joined
15:15:49  * shamaquit (Ping timeout: 248 seconds)
15:30:16  * hemanthquit (Quit: This computer has gone to sleep)
15:33:54  <ogd>chrisdickinson: ok h0ke figured out that doing 'git config http.postBuffer 524288000' on the client fixes the hanging pushes
15:34:38  <ogd>chrisdickinson: and i found this post about how its not possible to have the server tell the client to up the limit, the client has to set it http://permalink.gmane.org/gmane.comp.version-control.git/233234
15:34:54  <ogd>chrisdickinson: it seems like the code in git for doing FS IO is not as good at streaming as node
15:35:27  <ogd>chrisdickinson: which leads me to my next question: what parts of `git-receive-pack --stateless-rpc` are already implemented in JS? packfile parsing, right?
15:44:51  * ednapiranhajoined
15:46:54  <rowbit>substack, pkrumins: A developer is waiting in the queue for chrome/23.0
15:48:53  <rowbit>substack, pkrumins: A developer is waiting in the queue for chrome/27.0
15:51:05  * kenperkinsjoined
15:55:20  <rowbit>Hourly usage stats: [developer: 6, free: 32]
16:00:16  * AvianFlujoined
16:01:03  * AvianFlu_joined
16:02:00  * AvianFluquit (Disconnected by services)
16:02:16  * AvianFlu_changed nick to AvianFlu
16:03:53  * mikolalysenkojoined
16:06:44  * tobiequit (Ping timeout: 245 seconds)
16:08:47  * mikolalysenkoquit (Ping timeout: 250 seconds)
16:10:50  * tobiejoined
16:11:21  * daleharveyquit (Ping timeout: 274 seconds)
16:11:21  * mrb_bkquit (Ping timeout: 253 seconds)
16:11:54  * shamajoined
16:12:39  <jjjohnny>is this me or NPM https://gist.github.com/NHQ/7854107a09cfb019b826
16:13:10  * daleharvey_joined
16:13:14  * substack_quit (Ping timeout: 252 seconds)
16:13:26  * substack_joined
16:13:36  * mrb_bkjoined
16:13:53  <rowbit>substack, pkrumins: Encoders down: 50.57.72.69 (free7)
16:15:02  * kevino80quit (Remote host closed the connection)
16:15:23  * h0kejoined
16:16:46  * shamaquit (Ping timeout: 260 seconds)
16:17:45  * kevino80joined
16:19:15  <pfraze>jjjohnny: why is NPM switching to PUT on line 213?
16:19:45  * jcrugzzjoined
16:19:48  <pfraze>are you publishing?
16:20:13  <pfraze>oh, yeah :|
16:25:09  * dguttmanjoined
16:26:16  <jjjohnny>pfraze: see anything?
16:26:31  <jjjohnny>it looks like npm is trying to parse html as json
16:27:22  <pfraze>yeah, looks like your publish command failed!
16:27:26  * shamajoined
16:27:49  <pfraze>you get a 403 on line 211 before the 400 on 214
16:28:07  <pfraze>did you check your creds?
16:29:01  <jjjohnny>what would have changed?
16:29:17  <jjjohnny>i've been publishing from this machine for years
16:30:09  <jjjohnny>oh perhaps its cuz there is a module already called sorta
16:30:54  <pfraze>oh yeah, if sorta isnt yours I'd bet the same
16:31:26  <pfraze>hey, speaking of, congrats to npm http://blog.npmjs.org/post/76320673650/funding
16:32:47  <jjjohnny>zjwow
16:38:21  * contrahaxjoined
16:45:08  * defunctzombie_zzchanged nick to defunctzombie
16:55:20  <rowbit>Hourly usage stats: [developer: 8, free: 46]
16:55:31  * mikolalysenkojoined
17:07:02  * Maciek416joined
17:24:53  * calvinfojoined
17:32:46  * ceejbotjoined
17:41:24  * dguttmanquit (Quit: dguttman)
17:46:44  * phatedjoined
17:53:20  * ralphtheninjajoined
17:54:13  * phatedquit (Ping timeout: 248 seconds)
17:55:20  <rowbit>Hourly usage stats: [developer: 5, free: 44]
17:55:30  * ralphtheninjapart
17:57:30  * ednapiranhaquit (Remote host closed the connection)
17:57:37  * ryan_stevensjoined
18:05:27  <chrisdickinson>ogd: unsure about the `--stateless-rpc` part, but packfile parsing + fetch pack are complete
18:06:58  * ryan_stevenspart
18:11:01  <ogd>chrisdickinson: nice i think the rpc layer should be easy enough to write in JS
18:13:01  <chrisdickinson>stateless-rpc is just the "don't work over a duplex stream" version, yes?
18:14:12  * mikolalysenkoquit (Ping timeout: 260 seconds)
18:16:00  * mikolalysenkojoined
18:17:03  <ogd>chrisdickinson: no its still duplex, it just means to expose the receive pack over a stateless duplex stream e.g. http
18:17:39  <chrisdickinson>it's expected that multiple http requests will happen for a single operation?
18:19:21  <ogd>chrisdickinson: the way it works now in substacks git-http-backend module is the git push client makes a post, the backend spawns a --stateless-rpc --advertise-refs process for the repo, pipes it back to the response, then the client makes another request which has the packfile data, then the backend spawns the same thing without --advertise-refs
18:19:45  <chrisdickinson>yep, that jives with my understanding of smart http remotes
18:20:12  <chrisdickinson>fetch-pack negotiation with most remotes doesn't separate between advertise-refs and "send the pack"
18:20:48  <ogd>chrisdickinson: 'doesnt separate' meaning it does it in one request and not two?
18:20:58  <chrisdickinson>yep, one tcp/ssh connection
18:21:12  <ogd>chrisdickinson: oh right, but for smart http its always two?
18:21:21  <chrisdickinson>for pushes yes
18:21:27  <chrisdickinson>unsure about pull
18:21:32  <ogd>cool
18:21:34  <chrisdickinson>i would *guess* that it's still two for pull
18:22:07  <ogd>so what i wanna do is try and have node receive and unpack the packfile instead of git because git has this issue of forcing the client to increase the postBuffer or else the push hangs
18:22:31  <ogd>ostensibly because when the postBuffer size is too small the client switches to chunked transfer encoding and im guessing that is causing some weird issue somewhere
18:24:51  * contrahaxquit (Ping timeout: 250 seconds)
18:26:24  * contrahaxjoined
18:31:01  * dguttmanjoined
18:31:30  * phatedjoined
18:39:49  * phatedquit (Remote host closed the connection)
18:46:26  * daleharvey_quit (Ping timeout: 245 seconds)
18:50:24  * daleharvey_joined
18:52:21  <ogd>hmm i think level-live-stream doesn't work over multilevel
18:53:24  * peutetrequit (Quit: peutetre)
18:55:21  <rowbit>Hourly usage stats: [developer: 2, free: 26]
19:01:41  * peutetrejoined
19:09:01  * dguttmanquit (Ping timeout: 245 seconds)
19:09:02  <rowbit>substack, pkrumins: A developer is waiting in the queue for chrome/31.0
19:10:31  <rowbit>substack, pkrumins: A developer is waiting in the queue for chrome/31.0
19:12:36  * defunctzombiechanged nick to defunctzombie_zz
19:13:01  <rowbit>substack, pkrumins: A developer is waiting in the queue for chrome/31.0
19:15:31  <rowbit>substack, pkrumins: A developer is waiting in the queue for chrome/18.0
19:16:19  * dguttmanjoined
19:17:31  <rowbit>substack, pkrumins: A developer is waiting in the queue for chrome/31.0
19:18:31  <rowbit>substack, pkrumins: A developer is waiting in the queue for explorer/9.0
19:19:25  * tilgovijoined
19:22:32  <rowbit>substack, pkrumins: A developer is waiting in the queue for chrome/31.0
19:23:06  * tim_smartquit (Ping timeout: 245 seconds)
19:26:45  * dguttmanquit (Ping timeout: 272 seconds)
19:27:31  <rowbit>substack, pkrumins: A developer is waiting in the queue for chrome/31.0
19:29:34  * tim_smartjoined
19:30:31  <rowbit>substack, pkrumins: A developer is waiting in the queue for chrome/31.0
19:30:32  <rowbit>substack, pkrumins: A developer is waiting in the queue for chrome/31.0
19:34:01  * phatedjoined
19:34:27  * dguttmanjoined
19:36:39  <groundwater>substack any ssh-key modules i could use for doing non-browser client/server auth?
19:36:55  <chrisdickinson>ogd: for an example of how to use the unpack stream see https://github.com/chrisdickinson/js-git-demo
19:37:16  <chrisdickinson>you can throw the various objects into leveldb if you want (!!!)
19:42:14  * tim_smartquit (Ping timeout: 264 seconds)
19:46:36  * tim_smartjoined
19:49:28  * daleharvey_changed nick to daleharvey
19:55:20  <rowbit>Hourly usage stats: [developer: 25, free: 38]
19:57:43  * dguttman_joined
19:58:27  * dguttmanquit (Ping timeout: 250 seconds)
19:58:28  * dguttman_changed nick to dguttman
19:58:55  * AvianFluquit (Read error: Connection reset by peer)
19:59:13  * AvianFlujoined
20:02:15  * phatedquit (Remote host closed the connection)
20:03:28  * tilgoviquit (Remote host closed the connection)
20:22:19  * dguttmanquit (Ping timeout: 265 seconds)
20:22:43  * dguttmanjoined
20:23:05  * tilgovijoined
20:23:47  * funkytekjoined
20:38:04  * rsolejoined
20:45:36  * contrahaxquit
20:47:29  * phatedjoined
20:47:44  * contrahaxjoined
20:48:34  * dominictarrjoined
20:54:51  * mikolalysenkoquit (Ping timeout: 245 seconds)
20:55:20  <rowbit>Hourly usage stats: [developer: 3, free: 27]
20:57:59  * defunctzombie_zzchanged nick to defunctzombie
20:59:59  * jxsonjoined
21:06:29  * phatedquit (Remote host closed the connection)
21:13:05  * phatedjoined
21:13:58  * mikolalysenkojoined
21:17:20  * tilgoviquit (Read error: Connection reset by peer)
21:18:40  <defunctzombie>https://github.com/defunctzombie/polyfill-middleware
21:21:40  * jxsonquit (Remote host closed the connection)
21:22:14  * jxsonjoined
21:22:18  * jxsonquit (Remote host closed the connection)
21:23:00  * jxsonjoined
21:25:02  <rowbit>substack, pkrumins: A developer is waiting in the queue for firefox/10.0
21:25:28  * peutetrequit (Quit: peutetre)
21:30:57  * mikolalysenkoquit (Ping timeout: 265 seconds)
21:31:42  * peutetrejoined
21:31:50  * tilgovijoined
21:34:00  * mikolalysenkojoined
21:38:59  * mikolalysenkoquit (Ping timeout: 250 seconds)
21:40:19  * phatedquit (Remote host closed the connection)
21:44:26  * tilgoviquit (Remote host closed the connection)
21:51:45  * dguttman_joined
21:52:13  * dguttmanquit (Ping timeout: 265 seconds)
21:52:13  * dguttman_changed nick to dguttman
21:55:21  <rowbit>Hourly usage stats: [developer: 4, free: 25]
22:06:17  * mikolalysenkojoined
22:10:06  * rsolequit (Ping timeout: 265 seconds)
22:13:23  <owen1>defunctzombie: sweet!
22:18:46  * kevino80quit (Remote host closed the connection)
22:18:52  * rsolejoined
22:22:57  * crankquit (Read error: Connection reset by peer)
22:24:45  * djcoinquit (Quit: WeeChat 0.4.2)
22:25:41  * ednapiranhajoined
22:32:32  * ednapira_joined
22:32:33  * ednapiranhaquit (Read error: Connection reset by peer)
22:34:17  * funkytekquit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
22:35:20  * ednapira_quit (Client Quit)
22:36:23  * matthavenerjoined
22:36:28  * matthavenerpart
22:37:59  * ednapiranhajoined
22:46:02  <rowbit>substack, pkrumins: Encoders down: 184.106.106.46 (dev-ie8-1)
22:46:02  * peutetrequit (Quit: peutetre)
22:52:21  * jcrugzzquit (Ping timeout: 248 seconds)
22:54:08  * dominictarrquit (Remote host closed the connection)
22:54:32  <rowbit>substack, pkrumins: Encoders down: 184.106.106.46 (dev-ie8-1)
22:55:20  <rowbit>Daily usage stats: [developer: 49, free: 39]
22:55:20  <rowbit>Hourly usage stats: [developer: 10, free: 26]
23:04:27  * peutetrejoined
23:05:04  * maksimlinjoined
23:15:50  * mikolalysenkoquit (Ping timeout: 260 seconds)
23:16:35  * mikolalysenkojoined
23:16:54  <mikolalysenko>I have a kind of n00b question about web site stuff:
23:17:12  <mikolalysenko>how do big websites like twitter/facebook/google/etc. handle routing for requests?
23:17:30  <mikolalysenko>do they just have one server which is the big router for everything and passes all the information back and forth?
23:17:38  <mikolalysenko>or do they do something more fancy?
23:19:22  <jjjohnny>node: ../deps/uv/src/unix/core.c:149: uv_close: Assertion `0' failed.
23:20:14  <jjjohnny>thanks for sharing jjjohnny
23:23:22  * defunctzombiechanged nick to defunctzombie_zz
23:27:29  <chrisdickinson>mikolalysenko: this may be wholly inaccurate, but my understanding is that they usually have a beefy hardware load balancer
23:28:23  <mikolalysenko>chrisdickinson: so it is just one super big ass machine?
23:28:55  * dominictarrjoined
23:28:56  <pfraze>a big ass-machine
23:29:01  * thealphanerdjoined
23:29:07  <chrisdickinson>come to think of it, they may also have DNS setup so that different parts of the world go to different load balancers
23:29:14  <chrisdickinson>(sort of the CDN setup)
23:29:23  <ogd>yea round-robin dns is a thing, though i dont know how truly round-robin it is
23:29:39  <mikolalysenko>that would kind of make some sense, but I don't quite understand how you do it
23:30:19  <mikolalysenko>it seems a bit silly to me that you put in a ton of work building a distributed system on the backend, only to stick one gigantic single point of failure router on the front
23:30:43  <chrisdickinson>different DNS servers would be configured to either give different answers based on originating IP (as a guess) or cycle through a set of IPs
23:31:17  <mikolalysenko>ah, so the way it works is you connect to one of n-randomly selected big ass routers
23:31:25  <ogd>just buy the model 840 for $50,000 https://www.barracuda.com/products/loadbalancer/models
23:31:28  <mikolalysenko>and hope that the load amongst them is uniformly distributed?
23:31:40  <pfraze>mikolalysenko: isn't the answer to that, "the DNS is simple enough to be reliable and high throughput"
23:31:49  <pfraze>rather, the router is*
23:32:23  <pfraze>I'm guessing, if anybody knows better
23:32:50  * jxsonquit (Remote host closed the connection)
23:33:03  <mikolalysenko>pfraze: maybe, but it bothers me a little bit since it seems like you are really limited ultimately by how much you can push through your router
23:34:40  <pfraze>mikolalysenko: true. I'm sure google has got some wild setup that's like the matrix spinal tap into the internet
23:35:05  <ogd>lol
23:35:39  * rsolequit (Ping timeout: 265 seconds)
23:35:56  <mikolalysenko>I guess I can understand that a big enough router solves the problem, but doesn't that seem weird to you guys?
23:36:38  <mikolalysenko>I mean your bandwidth for your site would be limited by how much stuff you can push through that one box
23:38:33  * ceejbotquit (Remote host closed the connection)
23:43:03  * ceejbotjoined
23:47:46  <rowbit>/!\ ATTENTION: (default-local) [email protected] successfully signed up for developer browserling plan ($20). Cash money! /!\
23:47:47  <rowbit>/!\ ATTENTION: (default-local) paid account successfully upgraded /!\
23:53:27  * dominictarrquit (Ping timeout: 260 seconds)
23:54:19  * peutetrequit (Quit: peutetre)
23:55:20  <rowbit>Hourly usage stats: [developer: 0, free: 26]
23:55:59  * dominictarrjoined
23:57:23  * ins0mniaquit (Read error: Operation timed out)
23:59:00  * kriskowaljoined