[00:00:59] <sktrdie> can anyone help me?
[00:01:08] <sktrdie> i've been trying all night and tried millions different links
[00:01:15] <sktrdie> for 2 router routing setup
[00:01:28] <DreadWingKnight> erf
[00:01:35] <DreadWingKnight> double-nat == annoying as hell
[00:01:42] <sktrdie> i seem to set all the ips and ports fine, still can't get it to work
[00:02:18] <sktrdie> i have a local web server running on 7337
[00:02:30] <sktrdie> which works locally, of course, to test, i've set all ports to 7337
[00:02:37] <sktrdie> but i can't access it with external ip
[00:02:58] <sktrdie> DreadWingKnight: what's annoying about it?
[00:04:02] <DreadWingKnight> when each layer has different requirements for port forwarding
[00:04:22] <sktrdie> oh hrm
[00:04:25] <DreadWingKnight> and loopback tests rarely if ever work
[00:04:28] *** Andrius has quit IRC
[00:04:39] <sktrdie> i think the router closest to the WAN is the one giving me problems
[00:04:39] <DreadWingKnight> so you need a computer or proxy outside your internet connection to test with
[00:04:47] <sktrdie> i have it
[00:05:56] <sktrdie> is it ok to have ports open in the range 1-40000 on both routers?
[00:06:48] <DreadWingKnight> I'd only pinhole one port or a small number of ports I'd actively be using forwarded
[00:07:10] <sktrdie> ok
[00:07:18] <sktrdie> i've tested that as well
[00:07:19] <sktrdie> same result
[00:08:02] <sktrdie> the routher closest to me, has an ip of 192.168.0.1
[00:08:04] <DreadWingKnight> can you safely bridge either layer so it's no longer doing nat?
[00:08:13] <sktrdie> so i use that for the NAT on the other router
[00:08:17] <DreadWingKnight> no
[00:08:29] <DreadWingKnight> you can't forward to another nat's LAN ip
[00:08:47] <sktrdie> hrm
[00:08:56] <sktrdie> what do you mean?
[00:09:13] <DreadWingKnight> 10.43.44.1 <-- outer nat LAN IP (for example)
[00:09:22] <sktrdie> sure
[00:09:35] <sktrdie> you mean
[00:09:36] <DreadWingKnight> 10.43.44.42 <-- inner nat WAN IP
[00:09:41] <sktrdie> both routers will have that sort of ip?
[00:09:47] <DreadWingKnight> 192.168.0.1 <-- inner nat LAN IP
[00:09:59] <DreadWingKnight> 192.168.0.33 <-- computer LAN IP
[00:10:13] <DreadWingKnight> on outer nat, forward the port to 10.43.44.42
[00:10:21] <DreadWingKnight> on inner nat, forward the same port to 192.168.0.33
[00:10:26] <sktrdie> ohhhh
[00:10:27] <sktrdie> hrm
[00:10:42] <DreadWingKnight> because trying to forward the outer nat to 192.168.0.1 just plain won't work
[00:11:05] <sktrdie> so where do i get the inner NAT wan ip?
[00:11:11] <DreadWingKnight> (and yes, all addresses I gave for my example are valid for use in a multi-nat scenario, see RFC1918)
[00:11:20] <DreadWingKnight> from the status page of the inner nat's router
[00:11:39] <sktrdie> IP address: 192.168.0.1
[00:11:50] <DreadWingKnight> lan or wan?
[00:11:55] <sktrdie> OHH
[00:11:58] <sktrdie> wan is IP address: 192.168.1.100
[00:12:01] <sktrdie> i use that?
[00:12:04] <DreadWingKnight> yes
[00:12:06] <sktrdie> but that's the same as my computer :|
[00:12:38] <sktrdie> wait
[00:12:40] <sktrdie> its not
[00:12:41] <sktrdie> FUCK
[00:12:52] <sktrdie> i may love you
[00:12:55] <DreadWingKnight> hence why double-nat is a bitch
[00:13:04] <The_8472> also
[00:13:05] <The_8472> http://portforward.com/help/doublerouterportforwarding.htm
[00:14:09] <sktrdie> well i did it
[00:14:12] <sktrdie> not working yet, it seems
[00:14:24] <sktrdie> on the router closest to me
[00:14:30] <sktrdie> i've set my ip address, right?
[00:14:56] <The_8472> yes
[00:15:05] <sktrdie> well then it's all fine and dandy
[00:15:09] <sktrdie> but not working
[00:15:32] <sktrdie> when i try reaching my public IP on port :80 from a browser, I get the outer routers default page
[00:15:33] <The_8472> you forward to the outwards-facing IP of each successive step
[00:15:45] <The_8472> uh
[00:15:51] <The_8472> that probably won't work
[00:15:56] <sktrdie> what?
[00:16:01] <DreadWingKnight> from inside it won't
[00:16:07] <The_8472> NAT routers have issues with hairneedle routing, so you can't accurately test it yourself
[00:16:17] <The_8472> you need some external IP to test it
[00:16:27] <sktrdie> i do have it :(
[00:16:37] <sktrdie> OMG
[00:16:38] <sktrdie> IT WORKS
[00:16:39] <sktrdie> OMG
[00:16:40] <sktrdie> OMG
[00:16:41] <sktrdie> thanks
[00:17:04] <sktrdie> thanks
[00:17:06] <sktrdie> i guess that was it
[00:17:58] <The_8472> try http://portcheck.transmissionbt.com/<your port number here>
[00:18:02] <The_8472> it should return 1 on success
[00:18:13] <sktrdie> yep it's working
[00:18:19] <sktrdie> thanks a lot
[00:18:32] <sktrdie> i just forgot to test after the last change (on my other computer)
[00:33:54] *** burris has quit IRC
[00:35:41] *** burris has joined #bittorrent
[00:45:31] *** burris has quit IRC
[00:46:34] *** burris has joined #bittorrent
[00:53:46] *** skampler has quit IRC
[01:09:02] *** ajaya has quit IRC
[01:34:00] <DreadWingKnight> ok, digging through my hard drive is frightening me
[01:34:10] <DreadWingKnight> I have BNBT code sitting on my drive from 2004
[01:35:54] <DreadWingKnight> some of this code dates back to 2003
[01:35:55] <DreadWingKnight> damn
[01:47:18] *** goussx has quit IRC
[01:52:14] *** goussx has joined #bittorrent
[02:09:26] *** Waldorf has quit IRC
[02:11:25] *** Waldorf has joined #bittorrent
[02:12:16] *** Waldorf has quit IRC
[02:31:22] *** _rafi_ has quit IRC
[02:36:09] *** bittwist has quit IRC
[03:13:54] *** burris has quit IRC
[03:14:10] *** burris has joined #bittorrent
[03:32:58] *** ProperNoun has joined #bittorrent
[03:33:32] *** bittwist has joined #BitTorrent
[03:41:17] *** The_8472 has quit IRC
[03:41:38] *** The_8473 has joined #bittorrent
[03:41:39] *** The_8473 is now known as The_8472
[04:01:27] *** The_8472 has quit IRC
[04:02:15] *** Switeck has joined #bittorrent
[04:57:28] *** goussx has quit IRC
[05:01:20] *** Mekkis_ has joined #bittorrent
[05:08:59] *** Mekkis_ has quit IRC
[05:09:09] *** Mekkis has quit IRC
[05:09:12] *** Mekkis_ has joined #bittorrent
[05:19:14] *** Mekkis_ has quit IRC
[05:19:27] *** Mekkis_ has joined #bittorrent
[05:31:27] *** goussx has joined #bittorrent
[05:31:39] *** PN has joined #bittorrent
[05:42:16] *** init0_ has joined #bittorrent
[05:43:41] *** ProperNoun has quit IRC
[05:54:45] *** init0 has quit IRC
[06:05:24] *** DWKnight has joined #bittorrent
[06:05:38] *** rrr_ has quit IRC
[06:05:59] *** DWKnight has joined #bittorrent
[06:23:41] *** DreadWingKnight has quit IRC
[06:25:06] *** Switeck has quit IRC
[06:25:15] *** GTHK has quit IRC
[06:29:41] *** Mekkis_ has quit IRC
[06:30:04] *** Mekkis has joined #bittorrent
[06:34:03] *** edigaryev has joined #bittorrent
[06:34:17] *** edigaryev has quit IRC
[06:49:06] *** Mekkis_ has joined #bittorrent
[06:50:06] *** Mekkis_ has quit IRC
[06:50:19] *** Mekkis_ has joined #bittorrent
[06:52:35] *** Mekkis has quit IRC
[07:00:21] *** Mekkis_ has quit IRC
[07:00:34] *** Mekkis_ has joined #bittorrent
[07:11:49] *** uau has quit IRC
[07:16:11] *** rrr_ has joined #bittorrent
[07:18:50] *** rrr_ has quit IRC
[07:19:12] *** rrr_ has joined #bittorrent
[07:19:40] *** rrr_ has quit IRC
[07:25:02] *** rrr_ has joined #bittorrent
[07:39:49] *** _rafi_ has joined #bittorrent
[08:02:43] *** burris has quit IRC
[08:02:57] *** burris has joined #bittorrent
[08:33:26] *** stalled has quit IRC
[08:34:14] *** stalled has joined #bittorrent
[09:07:22] *** kwinz2 has quit IRC
[09:15:08] *** Waldorf has joined #bittorrent
[09:46:47] *** kwinz2 has joined #bittorrent
[10:07:24] *** Waldorf has quit IRC
[10:08:28] *** Waldorf has joined #bittorrent
[10:18:04] *** uau has joined #bittorrent
[10:19:54] *** Andrius has joined #bittorrent
[10:20:47] <_rafi_> in uT - what is net.uTP_receive_target_delay ? it's not in the help file
[10:23:44] <alus> it's just like the send delay target, but for receiving
[10:26:57] <_rafi_> thanks. So does it actually means that uT will stop downloading from uTP pears with less then, say, 100msec  latency ?
[10:27:13] <_rafi_> I mean more then...
[10:27:16] *** Firon has quit IRC
[10:27:34] *** hydri has quit IRC
[10:27:49] <alus> _rafi_: no
[10:28:09] <alus> _rafi_: it means uTP will target that amount of latency
[10:28:26] *** Firon has joined #bittorrent
[10:28:32] <_rafi_> alus: "target" ? how ?
[10:28:43] <alus> i.e. download more from a peer if the recv latency is < X and download less from the peer if the recv latency is > X
[10:28:52] *** hydri has joined #bittorrent
[10:28:54] <alus> the same way uTP targets send latency
[10:29:00] <alus> it adjusts the congestion window
[10:29:17] *** Firon has quit IRC
[10:29:18] *** hydri has quit IRC
[10:29:29] *** Firon has joined #bittorrent
[10:30:36] *** hydri has joined #bittorrent
[10:31:29] <_rafi_> alus: so, if I know that my connection is phisically with ~100msec to Europe, and ~200  avarage latency to us pears is about 200 to US peers, and almost ALL the peers are above this limit - shuold any of those #s be changed ? or uT will "know" to not "adjust" anything in this case ?
[10:32:21] <alus> no, it's latency on top of the minimum latency
[10:33:14] <_rafi_> "minimum" being ?
[10:38:02] <_rafi_> alus: what is actualy  measured from some local uT peer (probably ~30m in my case)
[10:39:21] <alus> minimum being the minimum latency seen for that connection
[10:39:35] <_rafi_> thx
[10:41:10] <_rafi_> alus: btw, I've found out the reason that my uTP seeding/yupload did not work at all... net.uTP_target_delay was set to 1 by mistake ... :P
[10:42:12] <alus> :P
[10:43:17] <_rafi_> if it's by design ... you might want to consider protecting against it...
[10:44:44] <alus> yeah, by removing the Advanced options page? :P
[10:46:41] <_rafi_> or paswword protection on critical parameters... :P ... no, by setting a reasonable minimum.. (like you did in RSS timout)
[10:49:09] <_rafi_> I would say that min of 10-30 would be appropreate here... alus
[10:50:42] <_rafi_> or a global "reset-all-to-defaults" button... :P
[10:51:20] <_rafi_> (+ backup/export...)
[10:57:00] <alus> not on a GB network
[11:01:08] <_rafi_> I'm not considering internal LAN  as a typical uT download environment (if that's what you mean)
[11:07:50] *** kwinz2 has quit IRC
[11:39:35] *** Waldorf has quit IRC
[12:04:59] <sktrdie> hi
[12:05:17] <sktrdie> if i have port forwarding setup, i should be able to connect to anybody right?
[12:05:21] <sktrdie> even those who dont
[12:05:42] <DWKnight> yes, but you can't force the attempts
[12:05:53] <alus> _rafi_: well, some people do use uT there. so why should I enforce a minimum on the advanced setting? just don't mess with it :P
[12:18:08] <_rafi_> alus: right, I'll try... still, both me and those "people" (as well as 'your' uT support team..) could use the set-all-to-defaults  and export add-ons... :)
[12:19:24] <_rafi_> for pin-pointing  issues caused by user errors... you know, we are not all perfect ...
[12:20:01] <alus> there's a little * hint that the setting has been changed, and you can reset the defaults on each setting
[12:38:43] *** Waldorf has joined #bittorrent
[12:52:15] <_rafi_> alus: yes, I know. not very convenient, though, with  a long list, if you want a go-nogo test. export->reset-all->import would have been much simpler ... (and also useful/easier for remote trouble-shooting). Thanks anyways
[13:04:54] <_rafi_> btw, ... and the latest RC5 release does look good (overhead-wise). I will wait that they will add/enhance with the proper GUI for it in the planed follow-up release
[13:19:12] *** The_8472 has joined #bittorrent
[13:32:27] *** kwinz2 has joined #bittorrent
[15:02:41] *** _rafi_ has quit IRC
[15:03:58] *** cyb2063 has joined #bittorrent
[15:32:19] <sktrdie> where do i download the original python bittorren tracker?
[15:37:51] <mpl> bittorrent.com
[15:38:08] <mpl> I guess the tracker's in the tarball.
[15:38:21] <mpl> at least it used to be like that.
[15:45:53] *** kwinz2 has quit IRC
[15:47:29] *** _rafi_ has joined #bittorrent
[16:05:06] *** kwinz2 has joined #bittorrent
[16:05:31] *** Waldorf has quit IRC
[16:12:43] *** Waldorf has joined #bittorrent
[16:17:33] <charles> awfully quiet in here today for some reason... ;)
[16:20:55] <sktrdie> :)
[16:21:11] * cyb2063 makes some noise and quickly goes into hiding again...
[16:31:28] *** razvand has joined #bittorrent
[17:14:59] *** Andrius has quit IRC
[17:17:15] *** Andrius has joined #bittorrent
[17:24:04] *** PN has quit IRC
[17:25:57] *** ProperNoun has joined #bittorrent
[17:56:49] *** GTHK has joined #bittorrent
[18:17:39] <TheSHAD0W> *BELCH*
[18:17:42] <TheSHAD0W> 'Scuse me.
[18:22:12] *** goussx has quit IRC
[18:28:58] *** sumkid has joined #bittorrent
[18:29:07] <sumkid> hey can any one help me?
[18:29:42] *** sumkid has quit IRC
[18:58:05] *** andar2 has joined #bittorrent
[19:03:24] *** goussx has joined #bittorrent
[19:11:27] *** rrr_ has quit IRC
[19:25:24] <TheSHAD0W> sumkid: Sure, wadaya want?  :-P
[19:31:38] *** qemodcdbs has joined #bittorrent
[19:31:38] <qemodcdbs> 
VERSION Learn more at http://free.sweettits.net/

[19:31:38] <qemodcdbs> 
VERSION Learn more at http://free.sweettits.net/

[19:31:38] <qemodcdbs> 
VERSION Learn more at http://free.sweettits.net/

[19:31:38] *** qemodcdbs has quit IRC
[19:43:20] <sktrdie> hi
[19:43:48] <sktrdie> does the official bittorrent tracker use a known library to do HTTP?
[19:44:00] <sktrdie> or does it just implement the protocol from scratch?
[19:44:32] <sktrdie> like many only GET
[19:46:20] *** Miller` has quit IRC
[19:55:42] *** The_8472 has quit IRC
[19:56:03] *** The_8473 has joined #bittorrent
[19:56:05] *** The_8473 is now known as The_8472
[19:59:18] <DWKnight> the python tracker from the mainline code is a very minimalistic implementation
[20:07:12] *** GTHKn has joined #bittorrent
[20:07:12] *** GTHK has quit IRC
[20:11:59] <burris> youd be surprised at how many torrents you can track with it
[20:13:22] <DWKnight> I personally probably wouldn't
[20:13:34] <DWKnight> but yeah, it's pretty hefty
[20:16:54] <sktrdie> hrm ok
[20:17:19] <sktrdie> but is it normal to use existing HTTP servers to implement a tracker?
[20:17:29] <sktrdie> like Apache, say
[20:18:04] <DWKnight> some people do using php-based trackers
[20:18:26] <DWKnight> but there's a scalability limit on those because of php/mysql or the httpd they choose to use
[20:18:38] <sktrdie> why is this?
[20:18:44] <sktrdie> because in Apache, every request is a process?
[20:18:49] <DWKnight> socket handling
[20:19:02] <sktrdie> hrm, what's that have to do with anything?
[20:19:05] <DWKnight> threading sometimes (no apache doesn't do thread per request last I saw)
[20:19:23] <sktrdie> so normal implementations dont use a db to store peers?
[20:19:29] <sktrdie> they just store it in memory?
[20:20:13] <DWKnight> some trackers cache their peers on disk and do primary operation entirely in memory
[20:20:24] <DWKnight> php trackers usually use a mysql database
[20:20:37] <DWKnight> the php/mysql interaction is one of the limiting factors on those trackers
[20:22:17] <sktrdie> but why is it limiting?
[20:22:20] <sktrdie> its just a bit slow
[20:22:22] <sktrdie> but not much
[20:22:30] <sktrdie> i mean, it's just about getting the peers data
[20:22:31] <sktrdie> right?
[20:22:32] <The_8472> <DWKnight> but there's a scalability limit on those because of php/mysql or the httpd they choose to use <- you could use memcached or apd
[20:23:24] <TheSHAD0W> Heh.
[20:23:31] <TheSHAD0W> Offtopic, but...
[20:23:32] <TheSHAD0W> http://mthruf.com/2010/01/22/work-fails-tug-one/
[20:24:04] <DWKnight> it does cause some problems on systems where the memory-based trackers don't see them
[20:24:27] <sktrdie> TheSHAD0W: computer engineers are good with math?
[20:26:08] <The_8472> we're good with math as long as it only requires intuitive/superficial understanding. theoretical mathematics requires a whole lot more
[20:28:25] <swolchok> we can also cope with probability if we really have to, which I usually find counterintuitive
[20:29:11] <The_8472> and i think experienced programmers will have intuitive understanding of algorithmic complexity.
[20:29:26] 
[20:30:33] <The_8472> or "this looks like n! ... but it's only called with n=10 at most, who gives a shit about optimizing it"
[20:30:54] <sktrdie> so what's the difference with computer scientists?
[20:31:43] <The_8472> computer scientists proove things. or come up with "neat" solutions where an engineer will just work around it.
[20:32:06] <The_8472> they create mathematical models, simulations and whatnot for things
[20:33:00] *** PN has joined #bittorrent
[20:34:13] <The_8472> it's sometimes a bit hard to separate.
[20:34:35] <The_8472> bascially it's theoretical computer science vs. applied computer science
[20:40:34] *** ProperNoun has quit IRC
[20:53:29] *** razvand has quit IRC
[20:58:10] <TheSHAD0W> Heh.
[20:58:55] <TheSHAD0W> When I coded selective download in BitTornado, I wrote some code that was O(n^2) and didn't realize it.
[20:59:14] <TheSHAD0W> It didn't show up until someone complained when downloading a 5000 file torrent.
[20:59:34] <The_8472> for each file it did something to all files?
[20:59:42] <TheSHAD0W> Right.
[20:59:48] <The_8472> yeah, sounds familiar ^^
[21:00:14] <TheSHAD0W> Wasn't too hard to code around, but it was so fast you didn't notice it on normal torrents.
[21:06:00] *** cyb2063 has quit IRC
[21:24:12] <swolchok> fwiw, a computer science degree here is actually mostly practical, whereas a computer engineering degree is a cross between EE and CS; you do some circuits (especially digital), some programming, and lots of computer architecture and logic design
[21:24:43] <swolchok> http://www.eecs.umich.edu/eecs/undergraduate/index.html
[21:25:04] <The_8472> then you haven't seen the computer science degree at the university here. it's more math than anything else
[21:25:24] <The_8472> but yeah... we have all that pesky electronics stuff in our computer engineering
[21:25:35] <The_8472> i'm trying to forget about that
[21:26:49] *** kwinz2 has quit IRC
[22:17:53] *** kwinz2 has joined #bittorrent
[22:41:18] *** ajaya has joined #bittorrent
[22:42:08] <wereHamster> are there any clients which support BEP 26 (mDNS peer discovery)?
[22:42:57] *** _rafi_ has quit IRC
[22:47:05] <The_8472> i think someone wrote a plugin for azureus as proof of concept
[22:58:32] *** ryanprior has joined #bittorrent
[23:03:42] *** Chaing1284 has joined #bittorrent





top