1 2012-05-11 00:02:34 <jgarzik> it looks like testnet spits out 2-3 blocks per hour. not too bad...
2 2012-05-11 00:02:41 <jgarzik> _someone_ is mining on there at least
3 2012-05-11 00:04:13 <jgarzik> hrm
4 2012-05-11 00:04:15 <jgarzik> 05/11/12 02:03:47 version message: version 40000, blocks=47476
5 2012-05-11 00:04:25 <jgarzik> 05/11/12 02:03:48 version message: version 50100, blocks=47476
6 2012-05-11 00:04:34 <jgarzik> these nodes seem stuck... curious
7 2012-05-11 00:04:43 <jgarzik> different node, different version, same testnet block
8 2012-05-11 00:05:05 <sipa> jgarzik: finally figured out why my version.cpp macro magic didn't work anymore for exported archives
9 2012-05-11 00:05:14 <sipa> (the problem reported by helo above)
10 2012-05-11 00:05:23 <sipa> you changed the version macros to consts :)
11 2012-05-11 00:05:30 <jgarzik> 05/11/12 02:04:48 version message: version 32400, blocks=47476
12 2012-05-11 00:05:35 <jgarzik> another one
13 2012-05-11 00:05:39 <sipa> (and i didn't pay attention...)
14 2012-05-11 00:06:17 <jgarzik> sipa: doh
15 2012-05-11 00:10:41 <luke-jr> jgarzik: I don't have any trouble doing TNIAB + Eloipool + BFGMiner, FWIW
16 2012-05-11 00:25:44 <gribble> New news from bitcoinrss: sipa opened pull request 1255 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/pull/1255>
17 2012-05-11 00:30:45 <gribble> New news from bitcoinrss: sgaltsev reopened issue 1231 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/issues/1231>
18 2012-05-11 00:57:16 <splatster> Any easy way to get a recent version of bitcoind on debian?
19 2012-05-11 00:58:23 <luke-jr> splatster: no, I brought it to their attention and the guy complained that it wasn't a bug.
20 2012-05-11 00:58:35 <luke-jr> nevermind that the version they have in there has known exploits etc
21 2012-05-11 00:58:39 <luke-jr> -.-
22 2012-05-11 00:59:00 <splatster> Umm, that sucks.
23 2012-05-11 00:59:22 <splatster> Anyone want to put together a aptitude source for debian?
24 2012-05-11 01:00:01 <splatster> luke-jr: ^ ;)
25 2012-05-11 01:00:11 <luke-jr> splatster: not really, no
26 2012-05-11 01:00:46 <splatster> Well, what are you doing right now?
27 2012-05-11 01:01:10 <splatster> IRC vs. helping out debian bitcoin users
28 2012-05-11 01:01:11 <luke-jr> splatster: it seems I have some major changes to make to BIP 22
29 2012-05-11 01:01:24 <splatster> :(
30 2012-05-11 01:32:25 <nanotube> splatster: there's a ppa on launchpad already
31 2012-05-11 01:33:18 <splatster> nanotube: I thought the PPA was for bitcoin-qt, not bitcoind.
32 2012-05-11 01:33:46 <luke-jr> more importantly, the PPA is Ubuntu, not Debian
33 2012-05-11 01:34:52 <nanotube> splatster: dunno, i thought it had both.
34 2012-05-11 01:34:57 <nanotube> luke-jr: i'm sure it'll work on debian just fine.
35 2012-05-11 01:35:13 <nanotube> well, s/sure/think that it will with a high probability/ :)
36 2012-05-11 01:35:34 <luke-jr> did BlueMatt update it since 0.6.0 yet?
37 2012-05-11 01:48:10 <splatster> Well, nvm, I just built it myself so w/e.
38 2012-05-11 01:50:46 <nanotube> dunno, i don't use the ppa myself, luke-jr :)
39 2012-05-11 01:59:28 <freewil> luke-jr, yes
40 2012-05-11 01:59:32 <freewil> the ppa is at 0.6.2
41 2012-05-11 02:08:23 <gmaxwell> splatster: we don't do builds of "bitcoin-qt, not bitcoind"
42 2012-05-11 02:08:48 <gmaxwell> (unless mac is still that way, but I thought we fixed that)
43 2012-05-11 02:22:24 <luke-jr> gmaxwell: all OS packaging is of course that way
44 2012-05-11 02:22:32 <luke-jr> bitcoind and bitcoin-qt are separate packages
45 2012-05-11 02:22:43 <Diablo-D3> luke-jr: but they're both built from the same parent package
46 2012-05-11 02:23:04 <Diablo-D3> (in debian, this would be a deb-src)
47 2012-05-11 02:23:23 <luke-jr> Diablo-D3: that's somewhat silly, since they don't share object files
48 2012-05-11 02:24:05 <Diablo-D3> no, thats what lib packages are for.
49 2012-05-11 02:24:13 <Diablo-D3> or common for non-object shared files
50 2012-05-11 02:37:37 <BlueMatt> luke-jr: ppa is 0.6.2
51 2012-05-11 02:38:01 <BlueMatt> luke-jr: as a sidenote, Im looking at making a debian package archive (now that Im on debian...)
52 2012-05-11 03:06:50 <luke-jr> anyone know what endian txids are in, off the top of their head?
53 2012-05-11 03:07:13 <Diablo-D3> uh
54 2012-05-11 03:07:15 <TuxBlackEdo> little
55 2012-05-11 03:07:21 <Diablo-D3> oh, WHICH endian
56 2012-05-11 03:07:24 <Diablo-D3> its little
57 2012-05-11 03:07:30 <Diablo-D3> bitcoin is fucktarded and does everything in little
58 2012-05-11 03:08:11 <gmaxwell> Diablo-D3: http://frank.harvard.edu/~coldwell/toolchain/bigendian.html
59 2012-05-11 03:08:41 <Diablo-D3> gmaxwell: yeah, but network endian is big
60 2012-05-11 03:09:33 <luke-jr> Diablo-D3: no, Bitcoin does some things in little, some things in big, and some things in "big for 32-bit chunks, but little for the whole thing"
61 2012-05-11 03:09:53 <gmaxwell> Yes, because SUN used to be very important when many things were created (and even before sun BE was popular for historical reasons, but it's not anymore). OTOH, most IETF protocols are text based and avoid endianness entirely.
62 2012-05-11 03:10:13 <gmaxwell> The inconsistency is worse than not using BE itself.
63 2012-05-11 03:10:16 <Diablo-D3> gmaxwell: yeah, but text is big endian,
64 2012-05-11 03:10:22 <Diablo-D3> oh, and that url of yours?
65 2012-05-11 03:10:30 <Diablo-D3> it brings up POWER bizzaro ordering
66 2012-05-11 03:10:38 <forrestv> the only big thing is port in addresses, right?
67 2012-05-11 03:11:05 <luke-jr> gmaxwell: you disappoint.
68 2012-05-11 03:11:17 <luke-jr> gmaxwell: are you seriously arguing LE is better because it works with bugs?
69 2012-05-11 03:11:47 <luke-jr> forrestv: no
70 2012-05-11 03:11:51 <forrestv> then what?
71 2012-05-11 03:11:52 <gmaxwell> luke-jr: it's easier to write width independant code for LE. It takes less code, and it results in less bugs.
72 2012-05-11 03:12:10 <luke-jr> there is no "width independent code"
73 2012-05-11 03:12:33 <gmaxwell> If you really want to complain, argue about the redundancy in the varints... esp inside stuff that gets signed, as that presents a risk that if you deseralize and reseralize that you'll get something different.
74 2012-05-11 03:12:39 <Diablo-D3> LE is better because this is what the majority of CPUs use
75 2012-05-11 03:12:49 <Diablo-D3> virtually all arms and mips are sold as LE
76 2012-05-11 03:12:58 <Diablo-D3> and thats like 95% of cpus manufed right there
77 2012-05-11 03:13:28 <luke-jr> BE is better because the majority of CPUs use LE
78 2012-05-11 03:13:30 <luke-jr> <.<
79 2012-05-11 03:13:44 <luke-jr> (and therefore BE forces people to write non-buggy code)
80 2012-05-11 03:13:57 <gmaxwell> luke-jr: ... I just spent the last 4 years writing a 50kloc codec which works on devices where int is 16,32,64 and where char is 8,16 and it does this with almost no IFDEFs, and without excessive width mandating everything. There really is width independant code.
81 2012-05-11 03:14:27 <Diablo-D3> gmaxwell: /me salutes you
82 2012-05-11 03:14:40 <Diablo-D3> gmaxwell: btw, why the fuck isnt ssize_t defined on linux?
83 2012-05-11 03:15:55 <gmaxwell> (it also works with BE/LE too, but thats pretty easy when you only read and write byte chunks)
84 2012-05-11 03:16:22 <luke-jr> Diablo-D3: it is, if you include the right header
85 2012-05-11 03:17:02 <Diablo-D3> I thought I was including pretty much every c99 header
86 2012-05-11 03:17:23 <Diablo-D3> because theres a few places I'd rather use ssize_t over normal pointer math
87 2012-05-11 03:17:45 <jgarzik> Diablo-D3: ssize_t is of course defined on linux
88 2012-05-11 03:18:04 <jgarzik> it's a bloody return value for read(2) and many other system calls
89 2012-05-11 03:19:32 <Diablo-D3> jgarzik: well wtf
90 2012-05-11 03:19:33 <Diablo-D3> where is it
91 2012-05-11 03:19:42 <jgarzik> Diablo-D3: it's a type. #include <sys/types.h>
92 2012-05-11 03:20:02 <Diablo-D3> well wtf, Im pretty sure Im including that header
93 2012-05-11 03:21:45 <jgarzik> hum, blockexplorer.com is stuck a few blocks back
94 2012-05-11 03:22:38 <Diablo-D3> http://caspar.adterrasperaspera.com/cgit/libugh.git/tree/src/libugh/base/stdinc.h
95 2012-05-11 05:04:29 <gribble> New news from bitcoinrss: Diapolo opened issue 1256 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/issues/1256>
96 2012-05-11 06:59:31 <rebroad> hopefully no more constant disconnects...
97 2012-05-11 07:39:32 <gribble> New news from bitcoinrss: Diapolo opened pull request 1257 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/pull/1257>
98 2012-05-11 07:55:26 <Twitchy> I've been attempting to use the mtgox api to retrieve my mtgox account stats, etc. But for some reason when I'm using the given PHP code here - https://en.bitcoin.it/wiki/MtGox/API/HTTP - I get an error saying im not logged in, can anyone help?
99 2012-05-11 08:03:21 <TuxBlackEdo> Twitchy, #mtgox
100 2012-05-11 08:03:38 <Graet> Twitchy, #mtgox has its own chan, this is the Bitcoin dev chanel :)
101 2012-05-11 08:03:43 <Graet> oops beaten :P
102 2012-05-11 08:06:03 <Twitchy> ta
103 2012-05-11 08:35:44 <gribble> New news from bitcoinrss: Diapolo opened pull request 1258 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/pull/1258>
104 2012-05-11 08:40:47 <gribble> New news from bitcoinrss: Diapolo opened pull request 1259 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/pull/1259>
105 2012-05-11 10:49:05 <sipa> jgarzik: any suggestions wrt IPV6_V6ONLY?
106 2012-05-11 10:50:58 <sipa> from what i read, OSX is always v6only
107 2012-05-11 10:51:12 <sipa> same on windows xp
108 2012-05-11 10:51:27 <sipa> more recents windowses have the option, but default to on
109 2012-05-11 10:51:52 <sipa> linux has the option, and most distro's set v6only to off by default
110 2012-05-11 10:52:12 <drizztbsd> sipa: *BSD and OSX is IPV6 by defaul
111 2012-05-11 10:52:22 <drizztbsd> and you can't clear it
112 2012-05-11 10:52:31 <drizztbsd> 22 (EINVAL)
113 2012-05-11 10:52:48 <drizztbsd> do you need some test under bsd?
114 2012-05-11 10:52:56 <sipa> no, i google; i have no bsd or osx
115 2012-05-11 10:53:31 <sipa> drizztbsd: IPV6 by default, you mean IPV6-only by default?
116 2012-05-11 10:55:13 <drizztbsd> yes, if you use AF_INET6
117 2012-05-11 10:55:28 <sipa> ok, good
118 2012-05-11 10:55:51 <sipa> i'll make it IPV6_V6ONLY on every platform, and listen separately for v4 and v6
119 2012-05-11 10:56:17 <drizztbsd> are you porting bitcoin to ipv6?
120 2012-05-11 10:57:11 <sipa> yes
121 2012-05-11 10:57:16 <sipa> mostly done already
122 2012-05-11 10:57:22 <sipa> it's already partially merged even
123 2012-05-11 11:00:22 <drizztbsd> by the way setsockopt(s, IPPROTO_IPV6, IPV6_V6ONLY, (char *)&on, sizeof(on)); works on BSD
124 2012-05-11 11:00:29 <drizztbsd> but you can't clear it
125 2012-05-11 11:00:35 <sipa> ok, good
126 2012-05-11 11:23:01 <davout> whoops, zhoutong did it again http://is.gd/WbGopV
127 2012-05-11 11:24:21 <sipa> my god...
128 2012-05-11 11:34:16 <SphericalCow> wow
129 2012-05-11 11:35:08 <sipa> drizztbsd: care to try if it builds/works on BSD? https://github.com/bitcoin/bitcoin/pull/1021
130 2012-05-11 11:35:38 <Diablo-D3> not that made up shit again
131 2012-05-11 11:36:02 <Diablo-D3> I should delete that thread and have bitcoinica banned from the forums
132 2012-05-11 11:36:19 <Diablo-D3> I dont believe in any of that made up fairy tale shit
133 2012-05-11 11:36:28 <Diablo-D3> its not even a scam, I dont think its even a real website
134 2012-05-11 11:36:42 <Diablo-D3> Ive never seen this "bitcoinica" site
135 2012-05-11 11:37:00 <Diablo-D3> just more lies and bullshit
136 2012-05-11 11:38:30 <Lexa> Diablo-D3: maybe because you use DNSSEC? they have an invalid DNSSEC config on their site, if it is this one, see http://dnsviz.net/d/bitcoinica.com/dnssec/
137 2012-05-11 11:38:54 <Diapolo> sipa: What triggers a block-chain rescan on startup or better how is decided how many blocks get checked?
138 2012-05-11 11:38:55 <Diablo-D3> Lexa: they dont even have a url listed anywhere
139 2012-05-11 11:39:37 <Diablo-D3> Eliel: the guy who owns bitcoinica is claiming someone stole his money
140 2012-05-11 11:40:03 <Eliel> no, not yet, he isn't.
141 2012-05-11 11:40:05 <sipa> Diapolo: there's a pointer to the last seen best block in the wallet
142 2012-05-11 11:40:16 <Diablo-D3> dont care
143 2012-05-11 11:40:21 <Diablo-D3> Im going to ask theymos to ban them I think
144 2012-05-11 11:40:59 <Diapolo> sipa: So something like Rescanning last 7844 blocks (from block 163193)... is normal?
145 2012-05-11 11:41:07 <sipa> yes
146 2012-05-11 11:42:26 <drizztbsd> sipa: yes but only bitcoind (I have not X on that machine)
147 2012-05-11 11:42:30 <Diapolo> sipa: Well, it seams fstream.read() is a lot slower than fread() or I have a strange bug in my code ... perhaps you can look at it when I updated the branch over the next few days?
148 2012-05-11 11:43:24 <sipa> drizztbsd: sure, i haven't tried compiling the gui even yet
149 2012-05-11 11:43:37 <drizztbsd> Should I use https://github.com/sipa/bitcoin.git branch ipv6 directly?
150 2012-05-11 11:43:55 <sipa> drizztbsd: i don't care how you use git :)
151 2012-05-11 11:46:10 <pierre`> Diablo-D3: bitcoinica seems to be a scam or a company managed by incompetents.
152 2012-05-11 11:48:31 <wumpus> you got to give it to him ,though, zhoutong is very honest ... a lot of sites would have swept this under the rug
153 2012-05-11 11:49:29 <sipa> if he really is 17, he is a genius. the question is whether he is a naive genius, or an evil genius :)
154 2012-05-11 11:58:57 <wumpus> only a very thin line between those :-)
155 2012-05-11 13:02:07 <clr_> Can someone help me debug longpolling code?
156 2012-05-11 13:08:57 <gribble> New news from bitcoinrss: sipa opened pull request 1260 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/pull/1260>
157 2012-05-11 13:30:52 <luke-jr> do we want to keep "/P2SH/" in coinbases still?
158 2012-05-11 13:33:22 <sipa> i'd say keep it as long as there is no 95% support
159 2012-05-11 13:38:05 <luke-jr> fwiw, stable versions never included it
160 2012-05-11 13:50:34 <gavinandresen> sipa: I uploaded a 0.6.2.2 binary just for OSX. I didn't grab your very latest changes because I didn't want to re-test.
161 2012-05-11 13:51:03 <sipa> luke-jr: i don't think there's a functional ipv6 network yet
162 2012-05-11 13:51:10 <sipa> we'd need seednodes, DNS seeds, ...
163 2012-05-11 13:51:44 <luke-jr> sipa: oh :<
164 2012-05-11 13:52:15 <sipa> 0.6.0+ nodes (should) pass on IPv6 addresses, but without critical mass that can find eachother, they would get lost in the abundance of IPv4 addresses
165 2012-05-11 13:53:30 <sipa> gmaxwell: had any chance in verifying unstuck?
166 2012-05-11 13:56:39 <gmaxwell> sipa: still pulling a fork chain (slowly)
167 2012-05-11 14:00:00 <drizztbsd> sipa: it works
168 2012-05-11 14:00:09 <drizztbsd> drizzt bitcoind 13883 14 tcp4 *:8333 *:*
169 2012-05-11 14:00:10 <drizztbsd> drizzt bitcoind 13883 15 tcp6 *:8333 *:*
170 2012-05-11 14:00:21 <sipa> ah, nice!
171 2012-05-11 14:00:22 <sipa> thanks
172 2012-05-11 14:00:32 <drizztbsd> sorry for the delay but I'm at work :)
173 2012-05-11 14:04:12 <jgarzik> sipa: when a new semOutbound is created inside OpenNetworkConnection()... where is it stored, and when is it deallocated?
174 2012-05-11 14:04:23 <jgarzik> sometimes C++ implicit actions are not obvious when reading a patch :)
175 2012-05-11 14:05:31 <jgarzik> sipa: oh, and you forgot to 'git add sync.cpp'
176 2012-05-11 14:05:49 <jgarzik> sipa: so your commit is missing that entire file, AFAICS
177 2012-05-11 14:08:25 <luke-jr> drizztbsd: what's your IP?
178 2012-05-11 14:09:09 <jgarzik> yep
179 2012-05-11 14:09:25 <jgarzik> I can see AAAA to bitseed.xf2.org. Any static IPv6 addresses for bitcoin... let me know.
180 2012-05-11 14:09:36 <jgarzik> s/see/add/
181 2012-05-11 14:09:45 <sipa> i intend to add a seed6.bitcoin.sipa.be
182 2012-05-11 14:10:40 <sipa> jgarzik: it's not created inside OpenNetworkConnection but by the called; it's given to ONC, and if that succeeds, that passes the lock to the created Node
183 2012-05-11 14:11:03 <sipa> afterwards, it is either released by the caller of ONC (when the object goes out of scope), or when the node object is deleted
184 2012-05-11 14:11:12 <sipa> s/called/caller/
185 2012-05-11 14:11:28 <jgarzik> sipa: pointers created by 'new' are not released when scope disappears
186 2012-05-11 14:11:38 <jgarzik> sipa: trying to find deletion point...
187 2012-05-11 14:12:30 <sipa> jgarzik: oh you mean the CSemaphore object, not CSemaphoreGrant ?
188 2012-05-11 14:12:46 <jgarzik> sipa: correct
189 2012-05-11 14:12:46 <sipa> the semaphore is only created in StartNode
190 2012-05-11 14:12:55 <jgarzik> sipa: semOutbound symbol, as I stated
191 2012-05-11 14:12:59 <sipa> right
192 2012-05-11 14:13:09 <sipa> no delete; there should probably be one in StopNode, though
193 2012-05-11 14:13:50 <gavinandresen> sipa: I think you forgot to git add sync.h/sync.cpp
194 2012-05-11 14:14:13 <sipa> gavinandresen: yes, jgarzik already told me; rebuilding and adding
195 2012-05-11 14:14:48 <sipa> updated
196 2012-05-11 14:15:29 <sipa> jgarzik: there's no semOutbound anywhere in ONC, by the way
197 2012-05-11 14:15:47 <jgarzik> sipa: just typing that :) read patch wrong, and thought it was a temp var, rather than a static var
198 2012-05-11 14:16:07 <jgarzik> sipa: all good
199 2012-05-11 14:16:10 <sipa> the variables you're looking for are probably called 'grant'
200 2012-05-11 14:17:11 <jgarzik> sipa: yeah, already saw that part. I didn't see 'StartNode' bit in patch chunk header, and thought that bit of code was part of ONC.
201 2012-05-11 14:17:18 <jgarzik> sipa: as part of StartNode(), it is indeed correct
202 2012-05-11 14:18:26 <sipa> luke-jr: running a torhs node on [2001:0:53aa:64c:2850:600e:af37:e874]:8333, if you like to try
203 2012-05-11 14:18:41 <sipa> (not that the torhs part matters too much, but it includes ipv6)
204 2012-05-11 14:19:37 <jgarzik> sipa: does your ipv6 patch yet cover DNS seeds?
205 2012-05-11 14:19:42 <sipa> no
206 2012-05-11 14:20:05 <sipa> well, it does request V6 addresses from the seeds
207 2012-05-11 14:20:30 <sipa> but i'd prefer having dedicated v6 seeds
208 2012-05-11 14:20:56 <jgarzik> sipa: why?
209 2012-05-11 14:21:16 <jgarzik> sipa: nothing wrong with overloading a name, or having one query return both A and AAAA
210 2012-05-11 14:21:25 <jgarzik> sipa: one query + one response is more efficient
211 2012-05-11 14:21:48 <jgarzik> let the DNS seeds choose
212 2012-05-11 14:21:56 <sipa> if your node supports v6, you definitely want some v6 results
213 2012-05-11 14:22:29 <jgarzik> sipa: that's a temporary policy that will go away once ipv6 reaches a steady state
214 2012-05-11 14:22:43 <sipa> and old nodes may also request with type=ANY, but will discard part of the rest
215 2012-05-11 14:22:46 <sipa> *result
216 2012-05-11 14:23:09 <jgarzik> sipa: that's definitely in the realm of DNS seed policy. current code is just fine, and will work for seeds that return A + AAAA.
217 2012-05-11 14:23:18 <jgarzik> sipa: no objection to adding your v6-only seed though
218 2012-05-11 14:23:25 <sipa> the bitcoin-ipv6 network may well be 100x smaller for the next year or so
219 2012-05-11 14:23:44 <sipa> i don't want to only get an ipv6 response from the seeds once every few attempts
220 2012-05-11 14:24:20 <sipa> but if requested an ANY address, the reasonable policy is for the seed to return results in proportion to what exists
221 2012-05-11 14:25:21 <sipa> i see the ipv6 network really as a separate network from the ipv4 network; it just happens to overlap pretty well with a subset of the old one, and most nodes functioning as proxies between the networks
222 2012-05-11 14:25:37 <sipa> and as a separate network, it needs a separate seeding
223 2012-05-11 14:25:43 <jgarzik> sipa: all of that is in the realm of the seed owner's policy, ultimately. for the bitcoin codebase, existing netbase.cpp code + some v6-specific seeds suffice.
224 2012-05-11 14:26:04 <sipa> some v6-specific seeds is all i'm arguing for :)
225 2012-05-11 14:26:07 <jgarzik> sipa: bitseed.xf2.org will -always- return both A and AAAA, for everybody
226 2012-05-11 14:26:20 <jgarzik> (once I add AAAA records)
227 2012-05-11 14:26:36 <luke-jr> sipa: brackets needed?
228 2012-05-11 14:26:45 <sipa> luke-jr: preferred
229 2012-05-11 14:27:06 <sipa> if it's unambiguous, they are optional
230 2012-05-11 14:28:09 <luke-jr> when is it ambiguous? O.o
231 2012-05-11 14:28:23 <sipa> 2001:0:53aa:64c:2850:600e:af37:e874
232 2012-05-11 14:28:31 <sipa> that :e874 cannot be interpreted as a port number
233 2012-05-11 14:29:23 <sipa> jgarzik: i'm not arguing for separating the seeds entirely, but some v6-specific ones are necessary (at least during initial adoption, but probably useful afterwards as well)
234 2012-05-11 14:31:38 <luke-jr> tcp6 0 0 2001:470:5:265:22:36503 2001:0:53aa:64c:28:8333 ESTABLISHED
235 2012-05-11 14:32:46 <sipa> which code version is that?
236 2012-05-11 14:32:54 <sipa> your node identified itself using its ipv4 address
237 2012-05-11 14:33:04 <sipa> oh, wait, it should!
238 2012-05-11 14:33:16 <sipa> i'm on teredo, so it prefers giving me an ipv4 one
239 2012-05-11 14:33:37 <luke-jr> sipa: next-test + some private stuff
240 2012-05-11 14:34:21 <sipa> Added 1 addresses from 2001:470:5:265:222:4dff:fe50:4c49: 102 tried, 11527 new
241 2012-05-11 14:35:18 <luke-jr> sipa: is it possible to tell my node to only give IPv6 over IPv6 etc?
242 2012-05-11 14:35:30 <luke-jr> I'd rather not let people link my v6 to my v4
243 2012-05-11 14:36:25 <jgarzik> luke-jr: it seems unwise to pretend one node is really two nodes
244 2012-05-11 14:36:39 <jgarzik> lead to redundant traffic
245 2012-05-11 14:36:46 <jgarzik> redundant connections
246 2012-05-11 14:39:57 <jgarzik> sipa: adding some v6-only seeds is fine
247 2012-05-11 14:41:09 <sipa> luke-jr: you can make your node entirely v6-only, but separating the networks is not possible (for now)
248 2012-05-11 14:41:16 <helo> so any ipv6 node could trivially appear to the network as thousands of unique nodes?
249 2012-05-11 14:41:39 <sipa> helo: yes, but nodes will only attempt an outgoing connection to one
250 2012-05-11 14:42:06 <helo> by that you mean "normal nodes will only attempt to connect to one ipv6 node"?
251 2012-05-11 14:42:31 <sipa> it will only connect to one address per /16 on IPv4 and one address per /32 on IPv6
252 2012-05-11 14:42:46 <helo> ahh, good
253 2012-05-11 14:43:26 <sipa> also, addrman first select a number of buckets per /16 or /32, and then randomly chooses one from them to store the actual address
254 2012-05-11 14:43:50 <sipa> this means someone spamming you with addresses all from the same /32, will only ever touch a subset of buckets
255 2012-05-11 14:44:16 <theymos> Is /32 big enough? Everyone gets assigned a /64 AFAIK.
256 2012-05-11 14:44:29 <sipa> theymos: a /32 is significantly larger than a /64 :)
257 2012-05-11 14:44:45 <theymos> oh whoops
258 2012-05-11 14:44:51 <luke-jr> theymos: everyone gets assigned a /48
259 2012-05-11 14:45:15 <luke-jr> so IPv6 /24 would be the logical equivalent to IPv4 /16
260 2012-05-11 14:45:27 <sipa> /24 is too large, i'm afraid
261 2012-05-11 14:45:42 <helo> won't it still be pretty easy for a big organization to have huge numbers of unique /32's?
262 2012-05-11 14:46:09 <luke-jr> helo: just like it's pretty easy for big corps to have large numbers of unique /16s&
263 2012-05-11 14:47:03 <sipa> i think you need about 40 to have a reasonable chance for dirtying an entire addrman table
264 2012-05-11 14:47:23 <luke-jr> sipa: I think the problem is more of all 8 connections
265 2012-05-11 14:48:25 <sipa> ?
266 2012-05-11 14:48:37 <luke-jr> sipa: if they can get all 8 connections, they can do ugly things
267 2012-05-11 14:48:52 <sipa> to get all 8 connections, they need to dirty your address table
268 2012-05-11 14:48:55 <luke-jr> has anyone reviewed #1240, before I rebase it?
269 2012-05-11 14:49:08 <sipa> not yet
270 2012-05-11 14:49:15 <luke-jr> sipa: not all of it, just enough to get 8 connections :p
271 2012-05-11 14:49:17 <helo> luke-jr: are the ratios (centrally controlled ipv4 /16s)/(legit nodes) vs (centrally controlled ipv6 /32s)/(legit nodes) about the same?
272 2012-05-11 14:49:31 <sipa> helo: unsure
273 2012-05-11 14:50:01 <sipa> if there's a good database of independent network, i wouldn't mind using it to decide outgoing network connections
274 2012-05-11 14:50:14 <luke-jr> look up ASNs? :p
275 2012-05-11 14:50:28 <sipa> last time i checked that was hard/large
276 2012-05-11 14:50:55 <luke-jr> DHT! oh wait&
277 2012-05-11 14:51:15 <sipa> hahaha
278 2012-05-11 14:51:18 <helo> so we need to start publishing our ipv6 addresses in the WoT :/
279 2012-05-11 14:51:25 <sipa> helo: ?
280 2012-05-11 14:51:57 <sipa> luke-jr: you're going to run an ipv6 node for the time being now?
281 2012-05-11 14:52:00 <sipa> with a static ip?
282 2012-05-11 14:52:19 <helo> i.e. build node list organically based on trust relationships, instead of randomly
283 2012-05-11 14:52:25 <luke-jr> sipa: no promises it's 24/7; I'll try to set a permanent one up for #Bitcoin-Watch and Eligius sometime
284 2012-05-11 14:53:09 <helo> assuming large orgs interested in controlling/monitoring bitcoin will represent a large proportion of arbitrarily selected nodes
285 2012-05-11 14:53:11 <sipa> i'll switch mine on bitcoin.sipa.be to IPv6 soon; if we have some nodes, i guess we can add a list of static IPv6 seeds in the code
286 2012-05-11 14:53:26 <luke-jr> sipa: to clarify, my home node is *mostly* 24/7, but I do stop/restart it when I'm developing
287 2012-05-11 14:53:32 <sipa> ic
288 2012-05-11 14:54:04 <luke-jr> I could give you future IPv6 addresses for #Bitcoin-Watch and Eligius if that helps
289 2012-05-11 14:55:40 <sipa> if they're not operational, it doesn't make much sense to do so untill we do a release with IPv6 support
290 2012-05-11 15:03:03 <Nachtwind> hi, i have a small question regarding Transactions. When i send n Bitcoins from User A to user B. Will i always end up having two outputs in that transaction with n Bitcoins and one with fee Bitcoins?
291 2012-05-11 15:04:02 <sipa> you'll have one output to B, one output to yourself, and the different between the inputs and the outputs is fee
292 2012-05-11 15:04:26 <Nachtwind> oh, ok
293 2012-05-11 15:04:28 <sipa> unless you have an input of exactly the size you want to send to B, in that case no change output to yourself should be created
294 2012-05-11 15:04:56 <Nachtwind> but i will always end up having one output showing the full amount n?
295 2012-05-11 15:05:35 <sipa> yes, but people will not necessarily be able to know whether that's the payment or the change
296 2012-05-11 15:05:44 <sipa> they are in random order
297 2012-05-11 15:05:57 <Nachtwind> oh, thats ok for me
298 2012-05-11 15:06:15 <Nachtwind> wonderful, makes it a bit more easier for me
299 2012-05-11 15:06:19 <Nachtwind> thanks for the help :0)
300 2012-05-11 15:07:29 <drizztbsd> luke-jr: why?
301 2012-05-11 15:07:52 <luke-jr> drizztbsd: I was going to addnode, but sipa gave me his :p
302 2012-05-11 15:08:09 <drizztbsd> oh ok :P
303 2012-05-11 15:08:21 <drizztbsd> I'm too natted at work
304 2012-05-11 15:08:38 <drizztbsd> (2 nats, without port forwarding or upnp)
305 2012-05-11 15:09:37 <luke-jr> I meant v6
306 2012-05-11 15:11:47 <drizztbsd> ohhh
307 2012-05-11 15:21:57 <drizztbsd> what is the best (cheap) vps in bitcoins?
308 2012-05-11 15:22:38 <drizztbsd> Actually I'm using a dedicated server (kimsufi( at ovh
309 2012-05-11 15:22:42 <drizztbsd> but I want to change :P
310 2012-05-11 15:34:37 <neofutur> drizztbsd: ask on #bitcoin-hosting
311 2012-05-11 15:34:47 <neofutur> I recommend http://gw.gd/bitcoin_vps
312 2012-05-11 15:34:59 <neofutur> good irc support on #bitvps
313 2012-05-11 15:49:25 <Joric> mmm... almost $75, sweet https://bitcointalk.org/index.php?topic=80815.msg892825
314 2012-05-11 15:50:41 <drizztbsd> I prefer chromium extensions
315 2012-05-11 15:50:48 <drizztbsd> more easy to develop and more performance
316 2012-05-11 15:52:02 <Joric> i barely use firefox nowadays
317 2012-05-11 15:53:31 <luke-jr> Joric: you can detect base58 addresses too
318 2012-05-11 15:55:08 <Joric> luke-jr, you mean, via the firefox extension?
319 2012-05-11 15:55:34 <luke-jr> Joric: I presume
320 2012-05-11 15:56:14 <luke-jr> pretty much, just look for words matching [13][[:base58:]]{26,35}
321 2012-05-11 15:56:16 <luke-jr> or whatever
322 2012-05-11 15:56:25 <luke-jr> chances are it's a bitcoin address
323 2012-05-11 15:57:51 <Joric> didn't write a single firefox extension in my life
324 2012-05-11 15:58:06 <Joric> though guess it's a piece of cake
325 2012-05-11 16:02:43 <zooko> sipa: http://bitcoin.sipa.be/ is not answering ping for me, nor HTTP
326 2012-05-11 16:03:12 <zooko> Hm, or is it DNS lookup failure. What's the IP address?
327 2012-05-11 16:04:12 <Joric> note ajax can't open sites without Access-Control-Allow-Origin header, not sure electrum servers have it, i use proxy server, it's not very cool
328 2012-05-11 16:04:48 <zooko> Hm, or is it that the coffeeshop network I'm on just stopped giving me IP service to my DNS server ...
329 2012-05-11 16:04:52 <Joric> if electrum would have Access-Control-Allow-Origin: true it would be possible to fetch data directly
330 2012-05-11 16:06:05 <Joric> erm, '*', not 'true'
331 2012-05-11 16:06:09 <copumpkin> zooko: 178.18.90.41
332 2012-05-11 16:06:11 <copumpkin> not a vhost, either
333 2012-05-11 16:07:27 <zooko> I think it was a problem on my end.
334 2012-05-11 16:11:58 <luke-jr> would anyone object to a -freerelay option, to make GetMinFee return 0 for relaying?
335 2012-05-11 16:27:51 <gribble> New news from bitcoinrss: laanwj opened pull request 1261 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/pull/1261>
336 2012-05-11 17:43:13 <davout> luke-jr: division by zero which instantly summons a black hole into existence
337 2012-05-11 18:17:50 <jgarzik> sipa gmaxwell: sorry to bring this up again, but gmaxwell was a bit vague :) what are the ipv6 blockers at this point?
338 2012-05-11 19:32:54 <helo> are there any plans to make generating and injecting transactions from offline wallets easier with the 'official' client?
339 2012-05-11 19:33:23 <gmaxwell> helo: absolutely. Or at least so much as anything can be called plans when he code doesn't exist yet.
340 2012-05-11 19:33:34 <gmaxwell> s/he/the/
341 2012-05-11 19:34:04 <gmaxwell> Though in principle you can't 'generate' a transaction from a totally disconnected node.
342 2012-05-11 19:34:26 <gmaxwell> At least not a transaction reflecting inputs that it doesn't know about.
343 2012-05-11 19:35:00 <helo> so transporting inputs to the offline machine might be step #1
344 2012-05-11 19:35:27 <gmaxwell> What I expect us to implement is the ability to setup keys as watch-only in an online node. The online node can then craft transactions, which you then carry to the offline node for signing, and then carry the result back for injection.
345 2012-05-11 19:35:28 <helo> well, facilitating such an export
346 2012-05-11 19:35:51 <gmaxwell> (the export for signing would ship out the copies of the inputs so the offline node could correctly display the amounts/fees)
347 2012-05-11 19:36:06 <helo> yeah, that would be pretty nice
348 2012-05-11 19:37:25 <gmaxwell> When combined with the HDwallet proposal it should be possible to have an offline device be some very small/low memory microcontroller system. too..
349 2012-05-11 20:00:19 <xenland> Why is it when i do "listtransactions" I see that I sent 0.00205001 but it says that I receieved 0.00205001 yet the fee was 0.005 ? Why would it report the amount recieved is the same?
350 2012-05-11 20:01:45 <xenland> http://pastebin.com/Pv11iUL1
351 2012-05-11 20:02:43 <sipa> jgarzik: i'm fine with pulling ipv6 now
352 2012-05-11 20:22:02 <gribble> New news from bitcoinrss: Diapolo opened pull request 1262 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/pull/1262>
353 2012-05-11 20:37:12 <gribble> New news from bitcoinrss: Diapolo opened pull request 1263 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/pull/1263>
354 2012-05-11 21:02:29 <luke-jr> MagicalTux: can you give Sgornick delete powers on the wiki? he's trying to clean up spam, but can only rename it :/
355 2012-05-11 21:14:27 <MagicalTux> k
356 2012-05-11 21:17:55 <gmaxwell> sipa: when logtimestamps the timestamps are emitted by printf directly. However, some entries always have timestamps and they get double stamps when logtimestamps is enabled.
357 2012-05-11 21:19:05 <gmaxwell> sipa: Diapolo is trying to fix it by turning off the redundant ones when logtimestamps is set... I think we should just nix the always on messages instead.
358 2012-05-11 21:21:17 <sipa> ooh, now i see
359 2012-05-11 21:21:49 <sipa> i thought his double timestamps were caused by two threads simultaneously trying to produce debug output
360 2012-05-11 22:38:58 <denisx> version right now is 69900?
361 2012-05-11 22:39:08 <denisx> sounds like the next version is 70000
362 2012-05-11 22:40:18 <gmaxwell> denisx: We advance git to next-major-minus-epsilon during development for the next major.
363 2012-05-11 22:42:10 <denisx> when 0.6.1 comes out, which version will it be?
364 2012-05-11 22:42:45 <denisx> ups, 0.6.2 is already out?
365 2012-05-11 22:42:49 <denisx> ok, I missed something ;)
366 2012-05-11 22:43:13 <gmaxwell> denisx: 0.6.1 has been out for a bit actually 0.6.2 has been out for a couple days now.
367 2012-05-11 22:44:50 <denisx> ok, 4.5. and 8.5.
368 2012-05-11 22:45:25 <gmaxwell> ????
369 2012-05-11 22:45:33 <denisx> that are the dates
370 2012-05-11 22:46:06 <gmaxwell> My hovercraft is full of eels.
371 2012-05-11 22:46:51 <denisx> 5/4/12 and 5/8/12
372 2012-05-11 22:46:53 <denisx> better? ;)
373 2012-05-11 22:47:15 <gmaxwell> ah. Yes.
374 2012-05-11 22:47:39 <sipa> and 0.6.2.2 came out today ;)
375 2012-05-11 22:47:51 <denisx> sipa: what about the 100% cpu bug
376 2012-05-11 22:47:52 <denisx> ?
377 2012-05-11 22:47:58 <gmaxwell> Are we even bothering to build that for anything except OSX?
378 2012-05-11 22:48:17 <gmaxwell> denisx: That OSX 100% cpu bug is fixed by 0.6.2.2.
379 2012-05-11 22:48:37 <denisx> gmaxwell: I have that bug on freebsd, so it is maybe a BSD bug
380 2012-05-11 22:49:31 <denisx> I pulled bitcoind some minutes ago and it was not fixed for me
381 2012-05-11 22:49:52 <gmaxwell> Does compiling the 0.6.2 branch solve it for you?
382 2012-05-11 22:49:53 <sipa> denisx: 0.6.2.2 fixes it for OSX
383 2012-05-11 22:50:13 <sipa> but it's in an #ifdef MAC_OSX
384 2012-05-11 22:50:27 <sipa> so maybe you need to add BSD there
385 2012-05-11 22:50:39 <denisx> sipa; ok, I try that
386 2012-05-11 22:50:48 <sipa> well, fixes... works around it
387 2012-05-11 22:51:08 <sipa> it's in util.h, the definition of CSemaphore
388 2012-05-11 23:02:29 <splatster> My bitcoind has 0 blocks but has > 1 GB datadir. "blocks" : 0,
389 2012-05-11 23:02:58 <splatster> Time to get rid of everything besides wallet.dat?
390 2012-05-11 23:03:09 <gmaxwell> splatster: whats large?
391 2012-05-11 23:03:27 <gmaxwell> That sounds like you erased your chain but left a big debug log behind, or erased blkindex without erasing the blocks file.
392 2012-05-11 23:04:14 <splatster> blk0001.dat, 257496 blkindex.dat, 32044 __db.003
393 2012-05-11 23:04:24 <splatster> those are the three bigest from ls -s
394 2012-05-11 23:05:32 <splatster> Maybe a corrupt chain or something?
395 2012-05-11 23:07:00 <splatster> gmaxwell: ^
396 2012-05-11 23:07:30 <gmaxwell> it would be interesting to see what your debug.log says.
397 2012-05-11 23:07:38 <gmaxwell> (esp around startup)
398 2012-05-11 23:08:54 <splatster> 5412 lines
399 2012-05-11 23:09:13 <gmaxwell> pastebin is your friend.
400 2012-05-11 23:09:19 <splatster> ProcessBlock: ORPHAN BLOCK, prev=00000000000000674933
401 2012-05-11 23:09:24 <splatster> and many similar lines
402 2012-05-11 23:09:44 <splatster> Maybe I got a hold of a bad chain to start with?
403 2012-05-11 23:10:02 <gmaxwell> That isn't how it works.
404 2012-05-11 23:10:02 <splatster> I just started up that client last night.
405 2012-05-11 23:10:12 <gmaxwell> What version are you running?
406 2012-05-11 23:10:31 <gmaxwell> in any case, pastebin the whole debug.log please. :)
407 2012-05-11 23:10:53 <splatster> 0.6.2
408 2012-05-11 23:11:18 <splatster> How can I pastebin the log if I'm running this on a remote server?
409 2012-05-11 23:11:28 <sipa> scp?
410 2012-05-11 23:11:29 <jgarzik> http://www.reddit.com/r/IAmA/comments/sq7cy/iama_a_malware_coder_and_botnet_operator_ama/
411 2012-05-11 23:11:46 <jgarzik> interesting botnetter comments on bitcoin
412 2012-05-11 23:12:39 <splatster> I've got to go for a bit.
413 2012-05-11 23:12:53 <splatster> I'll fetch the log and put it on pastebin later
414 2012-05-11 23:13:09 <gmaxwell> ah, the "Wallet just got emptied" thread on bitcoin talk turns how he had an exposed rpc port ... for no particular reason.
415 2012-05-11 23:14:26 <luke-jr> diff of diff of diff is mind boggling
416 2012-05-11 23:15:33 <gmaxwell> luke-jr: hm?
417 2012-05-11 23:15:33 <jgarzik> the Internet constantly impresses me with its... biological nature. Just leave open a port, and eventually _someone_ will probe it, or attempt an infection through it. Given the breadth of the IPv4 address space, that's pretty impressive. Tells you how fast computer and networks are these days, too.
418 2012-05-11 23:15:54 <jgarzik> sipa: let's pull ipv6
419 2012-05-11 23:17:46 <gmaxwell> one of the folks that previously got robbed via an open rpc port went scanning around and found a few hundred of them.
420 2012-05-11 23:22:09 <sipa> gmaxwell: ok for pulling ipv6?
421 2012-05-11 23:22:27 <gmaxwell> Yep.
422 2012-05-11 23:23:36 <luke-jr> gmaxwell: diff a-a-a a-a-b > a-a; diff a-b-a a-b-b > a-b; diff a-a a-b > a; diff b-a-a b-a-b > b-a; diff b-b-a b-b-b > b-b; diff b-a b-b > b; diff a b
423 2012-05-11 23:24:27 <luke-jr> gmaxwell: this is the process I use to audit my own rebases and merges :P
424 2012-05-11 23:24:37 <luke-jr> (though I have it simplified usually)
425 2012-05-11 23:24:42 <gmaxwell> ++--++- ... is this code or nethack?
426 2012-05-11 23:25:16 <sipa> ok, ipv6 pulled
427 2012-05-11 23:25:28 <denisx> sipa: the patch seems to work under FreeBSD too
428 2012-05-11 23:25:51 <sipa> with BSD, i suppose?
429 2012-05-11 23:25:59 <denisx> sipa: no, __FreeBSD__
430 2012-05-11 23:26:03 <denisx> we had that already
431 2012-05-11 23:26:19 <luke-jr> gmaxwell: only 3 levels of [+-] ;)
432 2012-05-11 23:26:22 <sipa> i mean, you modified the macro condutions around CSemaphore or not?
433 2012-05-11 23:27:43 <luke-jr> anyone know if Gavin is done rolling his own stable branch so I can integrate it into the stable repo?
434 2012-05-11 23:28:36 <denisx> sipa: yes, the ifdef before class CSemaphore
435 2012-05-11 23:29:03 <sipa> luke-jr: if a critical issue appeara before 0.7.0 has had enough time to stabilize, there will be another 6.x something
436 2012-05-11 23:29:26 <luke-jr> sipa: yes, but usually that's my dept
437 2012-05-11 23:29:27 <sipa> luke-jr: i'd just start doing backports once the next release is there
438 2012-05-11 23:31:26 <luke-jr> guess I need to redo my gmp_bip branch first anyway
439 2012-05-11 23:39:09 <jgarzik> w00t
440 2012-05-11 23:39:12 <jgarzik> Added 8 addresses from 2001:470:9ff2:2:0:0:0:12: 26 tried, 9319 new
441 2012-05-11 23:39:37 <jgarzik> this is me:
442 2012-05-11 23:39:39 <jgarzik> AddLocal([2001:4830:1603:2:21c:c0ff:fe79:c8c2]:8333,1)
443 2012-05-11 23:42:24 <splatster> gmaxwell: Debug.log pastebin: http://pastebin.com/raw.php?i=Tq5QsdF8
444 2012-05-11 23:45:46 <sipa> jgarzik: connected
445 2012-05-11 23:45:47 <gmaxwell> Bitcoin version vCLIENT_VERSION_MAJOR.CLIENT_VERSION_MINOR.CLIENT_VERSION_REVISION.CLIENT_VERSION_BUILD-g8ff1873-beta ()
446 2012-05-11 23:45:50 <gmaxwell> thats helpful.
447 2012-05-11 23:46:18 <luke-jr> odd, I can't seem to find that branch inserting height into coinbases
448 2012-05-11 23:46:23 <luke-jr> anyone know where it was?
449 2012-05-11 23:47:04 <luke-jr> gmaxwell: 8ff1873 = v0.6.2
450 2012-05-11 23:47:20 <sipa> gmaxwell: pre-1f5a7cef6f2a
451 2012-05-11 23:47:49 <gmaxwell> sipa: I know
452 2012-05-11 23:48:06 <gmaxwell> splatster: very interesting debug.log.
453 2012-05-11 23:48:53 <splatster> gmaxwell: I'm glad to have made this a little more interesting :)
454 2012-05-11 23:49:24 <splatster> So, should I just leave it as is for now? (in case you want me to run some tests or something)
455 2012-05-11 23:50:05 <gmaxwell> can you tar up a copy of your .bitcoin directory for me (except the wallet) I can give you an account to upload it to.
456 2012-05-11 23:50:36 <splatster> The wallet is fresh, so I haven't really used it anyway.
457 2012-05-11 23:50:44 <splatster> I can send you the whole thing.
458 2012-05-11 23:50:49 <gmaxwell> After copying it, shut down cleanly and delete everything except the wallet and config. You're hopelessly corrupted in some crazy unrecoverable way.
459 2012-05-11 23:51:00 <gmaxwell> if it's fresh just delete everything but the config after copying it.
460 2012-05-11 23:55:30 <jgarzik> sipa: is there a way to dump addrman's state to something vaguely readable?
461 2012-05-11 23:56:52 <luke-jr> jgarzik: do you know any way to grep through all git's objects, including ones unreachable from any ref?