1 2012-08-27 00:00:51 Zarutian has quit (Quit: Zarutian)
2 2012-08-27 00:05:16 <MC-Eeepc> you seem rather perturbed at my offhand africa comment
3 2012-08-27 00:05:42 <MC-Eeepc> no ones criticising anyone really, otherwise i wouldnt even be here
4 2012-08-27 00:05:58 <MC-Eeepc> anyway how do i do this ultraprune shit then
5 2012-08-27 00:11:26 Tiggr-Bot has quit (Ping timeout: 260 seconds)
6 2012-08-27 00:12:04 vampireb has joined
7 2012-08-27 00:16:55 RainbowDashh has quit (Ping timeout: 244 seconds)
8 2012-08-27 00:16:55 <sipa> MC-Eeepc: my gitian environment is currently broken; i don't think i can do a windows build now
9 2012-08-27 00:17:33 <MC-Eeepc> oh
10 2012-08-27 00:17:45 <sipa> BlueMatt: can you?
11 2012-08-27 00:17:51 * Luke-Jr wonders if we have 3 people for 0.7 then O.o
12 2012-08-27 00:18:06 <sipa> Luke-Jr: i'll get it up one of the next days, no worries
13 2012-08-27 00:20:05 RainbowDashh has joined
14 2012-08-27 00:22:01 <gmaxwell> sipa: won't the pulltester do it if you update the pull?
15 2012-08-27 00:23:18 <sipa> gmaxwell: yup, doing that; had to fix unit tests
16 2012-08-27 00:29:57 <sipa> gmaxwell: know of any way to speed the pull tester up? :)
17 2012-08-27 00:33:43 oneb has quit (Changing host)
18 2012-08-27 00:33:43 oneb has joined
19 2012-08-27 00:34:44 mmoya_ has quit (Ping timeout: 276 seconds)
20 2012-08-27 00:41:04 Gladamas has quit (Read error: Connection reset by peer)
21 2012-08-27 00:41:08 Gladamas_ has joined
22 2012-08-27 00:42:24 copumpkin has quit (Ping timeout: 260 seconds)
23 2012-08-27 00:45:17 copumpkin has joined
24 2012-08-27 00:56:07 genjix has joined
25 2012-08-27 00:56:12 <genjix> gmaxwell: you're on bash.org http://bash.org/?latest
26 2012-08-27 00:56:20 <genjix> http://bash.org/?952948
27 2012-08-27 00:57:04 <Luke-Jr> sipa: should be good to pull now
28 2012-08-27 00:57:15 <Luke-Jr> #1716*
29 2012-08-27 00:57:39 <sipa> gmaxwell: context...?
30 2012-08-27 01:00:55 torsthaldo has quit (Ping timeout: 256 seconds)
31 2012-08-27 01:01:55 Mad7Scientist is now known as IlIlIlIlIlIl
32 2012-08-27 01:02:29 enquirer has quit (Quit: back soon)
33 2012-08-27 01:02:53 torsthaldo has joined
34 2012-08-27 01:05:24 IlIlIlIlIlIl is now known as Mad7Scientist
35 2012-08-27 01:08:46 torsthaldo has quit (Read error: Connection reset by peer)
36 2012-08-27 01:08:46 paraipan has quit (Quit: Saliendo)
37 2012-08-27 01:10:22 torsthaldo has joined
38 2012-08-27 01:15:27 vampireb has quit (Quit: Lost terminal)
39 2012-08-27 01:16:39 vampireb has joined
40 2012-08-27 01:17:33 PhantomSpark has quit (3!~kvirc@pool-71-251-16-25.nycmny.fios.verizon.net|Read error: Connection reset by peer)
41 2012-08-27 01:17:57 PhantomSpark has joined
42 2012-08-27 01:18:01 <devrandom> sipa: have you tried gitian + lxc? should be easier to work with than kvm
43 2012-08-27 01:18:55 <sipa> devrandom: last time i tried, it didn't work, but that was some time ago
44 2012-08-27 01:19:52 <genjix> ;;seen phantomcircuit
45 2012-08-27 01:19:52 <gribble> phantomcircuit was last seen in #bitcoin-dev 3 days, 17 hours, 4 minutes, and 27 seconds ago: <phantomcircuit> sad panda
46 2012-08-27 01:19:54 Arnavion has quit (Ping timeout: 240 seconds)
47 2012-08-27 01:19:59 BLZbubba_ has quit (Ping timeout: 265 seconds)
48 2012-08-27 01:20:06 Arnavion has joined
49 2012-08-27 01:20:32 <devrandom> sipa: the latest also works around the vmbuilder issue (hang on unmount)
50 2012-08-27 01:21:01 <sipa> devrandom: let me try it
51 2012-08-27 01:22:26 <devrandom> I'll be around for about an hour if you run into issues, or later-tell me. I'd like to make it more pain free to use...
52 2012-08-27 01:23:03 vampireb has quit (Quit: Lost terminal)
53 2012-08-27 01:23:04 <Luke-Jr> devrandom: it'd be pain-free if it worked on other distros, and without Ruby⦠:P
54 2012-08-27 01:23:40 <devrandom> Luke-Jr: do the "other distros" at least have debootstrap?
55 2012-08-27 01:23:49 <Luke-Jr> devrandom: no? :p
56 2012-08-27 01:24:00 <devrandom> what do they have?
57 2012-08-27 01:24:19 <devrandom> (for building an image)
58 2012-08-27 01:24:25 <Luke-Jr> tar? <.<
59 2012-08-27 01:25:20 <devrandom> heh... but we do need to build an image, so need to install packages
60 2012-08-27 01:25:29 <Luke-Jr> or use a template? :P
61 2012-08-27 01:25:47 <devrandom> where would the template come from?
62 2012-08-27 01:26:05 <Luke-Jr> http://en.gentoo-wiki.com/wiki/LXC#Template_scripts
63 2012-08-27 01:26:07 <devrandom> and how would it be verified not to have been tampered with?
64 2012-08-27 01:26:27 <Luke-Jr> well, to do that, you need to build everything from source <.<
65 2012-08-27 01:27:03 <devrandom> Luke-Jr: well, that may be a good medium term goal...
66 2012-08-27 01:27:35 <Luke-Jr> you'd end up with a Gentoo chroot I think lol
67 2012-08-27 01:27:47 <devrandom> that might not be a bad idea
68 2012-08-27 01:28:05 <devrandom> I notice that lxc has a debian template, which calls debootstrap
69 2012-08-27 01:28:26 <Luke-Jr> hmm
70 2012-08-27 01:28:29 <devrandom> Luke-Jr: what distro do you use? gentoo?
71 2012-08-27 01:28:34 * Luke-Jr ponders if Gentoo can be made fully deterministic
72 2012-08-27 01:28:35 <Luke-Jr> yes
73 2012-08-27 01:30:16 <devrandom> Luke-Jr: is there an 'emerge' (or whatever packages are called) for bootstrap in gentoo?
74 2012-08-27 01:30:40 one_zero has joined
75 2012-08-27 01:30:44 <devrandom> oh yes, there is
76 2012-08-27 01:31:24 <Luke-Jr> devrandom: emerge -e world <-- rebuilds everything
77 2012-08-27 01:31:25 <devrandom> so if I create a debootstrap VM build option, it should work?
78 2012-08-27 01:31:36 <Luke-Jr> I suppose
79 2012-08-27 01:31:50 <Luke-Jr> using lxc-create might be better tho
80 2012-08-27 01:32:09 <devrandom> it will just in turn call debootstrap?
81 2012-08-27 01:32:56 <devrandom> but yes, could just create a suitable lxc template that calls debootstrap for now
82 2012-08-27 01:33:42 <devrandom> I do think that gentoo is more suitable in the long term because it builds everything from source
83 2012-08-27 01:34:15 <sipa> well, obviously someone could rig the 1st stage compiler you compile the rest with ;0
84 2012-08-27 01:35:23 <devrandom> sipa: yes, a recursive compiler trojan is a weak point of all this
85 2012-08-27 01:35:56 <devrandom> there was some paper about using two compilers to detect such a thing
86 2012-08-27 01:36:38 <devrandom> there it is: http://www.dwheeler.com/trusting-trust/
87 2012-08-27 01:37:56 RainbowDashh has quit (Quit: Computer has gone to sleep.)
88 2012-08-27 01:38:25 <sipa> http://scienceblogs.com/goodmath/2007/04/15/strange-loops-dennis-ritchie-a/
89 2012-08-27 01:39:23 <devrandom> yup
90 2012-08-27 01:40:23 <devrandom> fun :)
91 2012-08-27 01:42:06 <[Tycho]> Somehow this reminds me of "Primer" movie.
92 2012-08-27 01:45:17 <devrandom> hm.. the obfuscator idea sounds like a good countermeasure
93 2012-08-27 01:54:37 BTCTrader has quit (Quit: BTCTrader)
94 2012-08-27 01:55:07 eoss has quit (Remote host closed the connection)
95 2012-08-27 01:56:40 vampireb has joined
96 2012-08-27 02:00:21 Gladamas_ is now known as Gladamas
97 2012-08-27 02:00:38 <sipa> devrandom: http://pastebin.com/arKA37xr
98 2012-08-27 02:01:14 <sipa> ubuntu 12.04
99 2012-08-27 02:01:18 copumpkin is now known as napumpkin
100 2012-08-27 02:01:32 napumpkin is now known as copumpkin
101 2012-08-27 02:03:06 <sipa> ;;later tell devrandom http://pastebin.com/arKA37xr
102 2012-08-27 02:03:06 <gribble> The operation succeeded.
103 2012-08-27 02:03:55 <ne0futur> http://www.grsecurity.net/~spender/exploits/exploit2.txt
104 2012-08-27 02:04:45 <ne0futur> was a good example of a bug ( backdoor ) appearing only if you use the good compiler + glib +kernel + selinux
105 2012-08-27 02:04:57 CodeInChaos has quit (Ping timeout: 256 seconds)
106 2012-08-27 02:05:15 <ne0futur> well since redhat manages gcc and glib and more . . .
107 2012-08-27 02:05:20 <gmaxwell> ne0futur: meh, it's an example of a bug which is caused by code obviously depending on undefined behavior in C.
108 2012-08-27 02:05:46 torsthaldo has quit (Remote host closed the connection)
109 2012-08-27 02:06:21 <ne0futur> the problem was a NULL tun variable,
110 2012-08-27 02:06:21 <ne0futur> which from the source should have been unexploitable due to the
111 2012-08-27 02:06:23 <ne0futur> immediate check for !tun
112 2012-08-27 02:06:25 <ne0futur> read it all
113 2012-08-27 02:06:32 <ne0futur> un unexploitable bug
114 2012-08-27 02:07:10 <ne0futur> well anyway lets say i m wrong i wont even try to argue
115 2012-08-27 02:07:23 <ne0futur> lets trust redhat and the nsa :p
116 2012-08-27 02:08:03 <ne0futur> A vulnerability which, when viewed at the source level, is unexploitable! But which, thanks to gcc optimizations, becomes exploitable :)
117 2012-08-27 02:08:12 <devrandom> sipa: there is a brctl command in the README.md ... you need br0
118 2012-08-27 02:08:14 <gmaxwell> ne0futur: the check happened _after_ the dereference. This is a clear violation of the C spec, it's undefined behavior. Yes, it's a bit subtle, but there are lots of ways to have subtle bugs.
119 2012-08-27 02:08:45 <gmaxwell> ne0futur: it's only 'unexplotable when viewed at the source level' by someone who does not have a firm grasp of the fine rules of C.
120 2012-08-27 02:08:49 <sipa> devrandom: i should have RTFM
121 2012-08-27 02:09:21 <devrandom> np
122 2012-08-27 02:09:34 <Arnavion> I seem to recall the Clang team complaining about that thing in the kernel
123 2012-08-27 02:09:42 <Arnavion> Using a variable first, checking if it's null later
124 2012-08-27 02:09:57 <Arnavion> and that they had to special-case removing the optimization in their compiler for the kernel
125 2012-08-27 02:10:51 <gmaxwell> Arnavion: The kernel developers aren't actually writing in C. They're writing in kernel-c. It's an imaginary language that is implemented by a compiler which tranliterates C-like code to asm and which exists only in the minds of the kernel develoepers.
126 2012-08-27 02:14:34 <ne0futur> gmaxwell: and the good gcc version
127 2012-08-27 02:14:56 <sipa> yes, it's a rapedly evolving language
128 2012-08-27 02:15:01 <sipa> *rapidly
129 2012-08-27 02:15:19 <gmaxwell> it doesn't really matter 'the gcc version' there are many compilers which will optimize out a redundant null check.
130 2012-08-27 02:15:36 <gmaxwell> E.g. the intel compiler and the power64 compiler will both do that. I think the intel one will even do it at O1.
131 2012-08-27 02:15:45 has_many has quit (Ping timeout: 244 seconds)
132 2012-08-27 02:16:40 pecket has quit (Ping timeout: 260 seconds)
133 2012-08-27 02:17:04 <genjix> ne0futur: sorry but that's undefined behaviour.
134 2012-08-27 02:17:08 <gmaxwell> Many C developers make a mistake of thinking of C like a preprocessor stage on an assemblerâ Which is pretty much what the PDP11 DMR c compiler was, and thats how it's often taught to people. But that isnt how C is specified, and it's not how any modern compiler works.
135 2012-08-27 02:17:22 <Arnavion> http://blog.llvm.org/2011/05/what-every-c-programmer-should-know_14.html There it is
136 2012-08-27 02:17:45 <ne0futur> still a security problem opened by the compiler
137 2012-08-27 02:18:05 [\\\] is now known as odysseus
138 2012-08-27 02:18:14 copumpkin is now known as cyclops
139 2012-08-27 02:18:16 <gmaxwell> ne0futur: no, it's a security problem opened by the incorrect code. In userspace code that code would reliably crash, but the kenel doesn't have the same memory protection as userspace.
140 2012-08-27 02:18:18 odysseus is now known as [\\\]
141 2012-08-27 02:18:53 has_many has joined
142 2012-08-27 02:18:59 cyclops is now known as copumpkin
143 2012-08-27 02:19:07 <genjix> Arnavion's link is good for explaining why.
144 2012-08-27 02:21:04 <genjix> tbh i prefer a compiler that produces binaries that fail hard and fast (or even better: don't run) with code that has undefined behaviour
145 2012-08-27 02:21:50 <genjix> c/c++ make no guarantees about how a binary with UB should behave. it can produce binaries that do absolutely anything.
146 2012-08-27 02:22:05 <kjj> C gets even stranger when you factor in the crazy things CPUs do. compilers are only the first step
147 2012-08-27 02:22:32 <gmaxwell> genjix: varrious bits of static analysis will happily point out that the code there suggests undefined behavior, so thats helpful. Unfortunately the compiler can't usually tell the difference between a boobytrap an an important operation.
148 2012-08-27 02:23:10 <gmaxwell> kjj: er. No. C specifies an abstract machine. If you write valid C your code will operate idetically on all correct implementations; because defined code is require to operate identically to the behavior of the abstract machine.
149 2012-08-27 02:23:35 <sipa> devrandom: sudo: unable to resolve host alduin
150 2012-08-27 02:23:35 cheako has joined
151 2012-08-27 02:23:41 <gmaxwell> (including widly different ones like ones where char is 16 bits or where arithemetic is ones compliment)
152 2012-08-27 02:23:46 <kjj> yeah, but the way a real machine operates is crazy odd compared to how we see that abstract machine in our heads
153 2012-08-27 02:23:46 <sipa> devrandom: that being the name of the host machine
154 2012-08-27 02:23:55 <genjix> gmaxwell: did you see that you're on bash.org?
155 2012-08-27 02:24:15 <sipa> kjj: sure, but at least that's (at least in sane cases) unobservable
156 2012-08-27 02:24:19 <sipa> except by timing
157 2012-08-27 02:24:21 TiggrBot has joined
158 2012-08-27 02:24:22 nathaniel has quit (Read error: Connection reset by peer)
159 2012-08-27 02:24:34 <kjj> sipa: supposed to be, but often the errata prove otherwise
160 2012-08-27 02:24:38 <gmaxwell> genjix: Yep.
161 2012-08-27 02:25:00 <sipa> gmaxwell: any context?
162 2012-08-27 02:25:26 <sipa> kjj: not surprising, given the several layers of processing involved after machine language
163 2012-08-27 02:25:54 <genjix> kjj: tbh cpu differences are minor compared to platform differences for implementors.
164 2012-08-27 02:26:05 <genjix> maybe it mattered more in the time of gossplatt
165 2012-08-27 02:26:07 <gmaxwell> sipa: it was just from some discussion about surveillance and I was just making a wry comment that as a whole the public is massively opting into a level of surveillance which could never have been forced on them.
166 2012-08-27 02:26:10 <kjj> and since we hit the wall of jello just north of 3 GHz, compilers have become really strange
167 2012-08-27 02:26:45 <gmaxwell> kjj: :( Nothing is strange. Compilers that did these sorts of optmizations are really not new.
168 2012-08-27 02:27:28 <kjj> Yeah, optimization has been around forever
169 2012-08-27 02:27:36 <gmaxwell> If you can tell the difference between a correct optimizing compiler and a non-optimizing one; other than performance; then your code is wrong.
170 2012-08-27 02:27:42 <genjix> *greenplatt not gossplat
171 2012-08-27 02:28:36 <kjj> but modern CPUs spend a non-trivial amount of their lives doing work that they need to roll back, and modern compilers are built for THAT world now
172 2012-08-27 02:29:26 [\\\] is now known as escrow
173 2012-08-27 02:29:37 <kjj> the kernel-c thing is dead on, but all of us do it to some extent. we visualize ASM, and then assume a CPU that we can understand is executing that ASM. neither is true any more
174 2012-08-27 02:30:56 <gmaxwell> It wasn't true 20 years ago. It was somewhat true 30 years ago, so long as you used C and not fortran.
175 2012-08-27 02:31:17 <kjj> even debugging is odd now. we imagine the CPU screeching to a dead stop when it hits a breakpoint, but the truth is that it is travelling back in time a few hundred instructions
176 2012-08-27 02:31:42 <gmaxwell> kjj: the observable behavior is still identical.
177 2012-08-27 02:31:50 <genjix> i'm not a c-coder. i write c++ and deal with objects and abstractions.
178 2012-08-27 02:32:02 <genjix> we don't think imperically.
179 2012-08-27 02:32:12 <genjix> *imperatively
180 2012-08-27 02:32:14 <kjj> heh. the quantum physics argument applied to C code.
181 2012-08-27 02:32:31 <sipa> genjix: still C++ is a very imperative language :)
182 2012-08-27 02:32:49 <kjj> the photons leave here and are observed there. nothing else is real
183 2012-08-27 02:33:13 <sipa> genjix: but I see what you mean; you are abstracting from some details, indeed
184 2012-08-27 02:33:20 <genjix> yeah depends on the level. for methods and small pieces of code.
185 2012-08-27 02:33:58 <genjix> right. that's why template metaprogramming is great for algorithms, but not for the higher levels.
186 2012-08-27 02:34:25 <kjj> brb
187 2012-08-27 02:34:40 MC-Eeepc has quit (Ping timeout: 272 seconds)
188 2012-08-27 02:34:42 TiggrBot has quit (Ping timeout: 260 seconds)
189 2012-08-27 02:36:07 <genjix> https://www.youtube.com/watch?v=zh8W4ZglOlw
190 2012-08-27 02:36:52 escrow is now known as TheOfficialEscro
191 2012-08-27 02:37:14 TheOfficialEscro is now known as [\\\]
192 2012-08-27 02:38:28 fiesh has quit (Ping timeout: 272 seconds)
193 2012-08-27 02:40:31 <kjj> anyone know namecoin in detail? I had an idea for an altchain that would actually be useful, but I figured I'd better educate myself before inflicting it on the world
194 2012-08-27 02:40:51 fiesh has joined
195 2012-08-27 02:41:33 toffoo has joined
196 2012-08-27 02:41:51 phantomcircuit_ has joined
197 2012-08-27 02:44:54 vampireb has quit (Quit: Lost terminal)
198 2012-08-27 02:55:44 phantomcircuit_ has quit (Remote host closed the connection)
199 2012-08-27 02:57:18 RainbowDashh has joined
200 2012-08-27 03:06:09 RainbowDashh has quit (Ping timeout: 256 seconds)
201 2012-08-27 03:08:01 Z0rZ0rZ0r_ has joined
202 2012-08-27 03:08:39 <ne0futur> kjj: what kind of details ? try #namecoin anyway
203 2012-08-27 03:08:43 RainbowDashh has joined
204 2012-08-27 03:09:15 <ne0futur> gmaxwell: also if you find a better eample of a backdoor triggered by a compiler, i m interested
205 2012-08-27 03:09:32 <ne0futur> I just havent found better than http://www.grsecurity.net/~spender/exploits/exploit2.txt until now
206 2012-08-27 03:10:01 Z0rZ0rZ0r has quit (Ping timeout: 252 seconds)
207 2012-08-27 03:10:03 <gmaxwell> It's not 'triggered by a compiler' :( It's triggered by incorrect code that dereferences a null pointer.
208 2012-08-27 03:11:53 pecket has joined
209 2012-08-27 03:14:02 <ne0futur> find a better example
210 2012-08-27 03:14:11 vampireb has joined
211 2012-08-27 03:14:30 <ne0futur> Also, bypass of mmap_min_addr via SELinux vulnerability!
212 2012-08-27 03:14:30 <ne0futur> (where having SELinux enabled actually increases your risk against a
213 2012-08-27 03:14:31 <ne0futur> large class of kernel vulnerabilities)
214 2012-08-27 03:15:49 darkskiez has quit (Ping timeout: 245 seconds)
215 2012-08-27 03:16:12 <ne0futur> lets speak of multi level, well hidden backdoor
216 2012-08-27 03:22:38 <gmaxwell> also, It's _not_ a 'backdoor'. It's a rather boring bug. And mmap_min_addr on many systems without selinux (e.g. ubuntu) has historically been set to zero for compatiblity with wine.
217 2012-08-27 03:28:26 darkskiez has joined
218 2012-08-27 03:30:31 RainbowDashh has quit (Quit: Computer has gone to sleep.)
219 2012-08-27 03:36:37 <ne0futur> thanks for the better example
220 2012-08-27 03:39:17 RainbowDashh has joined
221 2012-08-27 03:42:15 minimoose has quit (Quit: minimoose)
222 2012-08-27 03:45:41 minimoose has joined
223 2012-08-27 03:48:08 TiggrBot has joined
224 2012-08-27 03:51:52 [7] has quit (Disconnected by services)
225 2012-08-27 03:52:00 TheSeven has joined
226 2012-08-27 03:58:17 Karmaon1 has quit (Ping timeout: 260 seconds)
227 2012-08-27 03:58:58 RainbowDashh has quit (Ping timeout: 268 seconds)
228 2012-08-27 04:02:58 RainbowDashh has joined
229 2012-08-27 04:03:15 enquirer has joined
230 2012-08-27 04:03:23 enquirer has quit (Client Quit)
231 2012-08-27 04:07:19 vampireb has quit (Quit: leaving)
232 2012-08-27 04:14:23 maaku has quit (Quit: maaku)
233 2012-08-27 04:14:49 maaku has joined
234 2012-08-27 04:18:16 RainbowDashh has quit (Read error: Connection reset by peer)
235 2012-08-27 04:20:22 RainbowDashh has joined
236 2012-08-27 04:46:56 roconnor has quit (Ping timeout: 260 seconds)
237 2012-08-27 04:52:50 maaku has quit (Quit: maaku)
238 2012-08-27 04:54:45 jouke has joined
239 2012-08-27 05:08:50 Diablo-D3 has quit (Remote host closed the connection)
240 2012-08-27 05:09:34 edcba has quit (Ping timeout: 245 seconds)
241 2012-08-27 05:09:50 D34TH has quit (Quit: Leaving)
242 2012-08-27 05:12:16 Diablo-D3 has joined
243 2012-08-27 05:24:55 brwyatt is now known as brwyatt|Away
244 2012-08-27 05:27:26 Joric has quit (Ping timeout: 244 seconds)
245 2012-08-27 05:32:11 ThomasV has joined
246 2012-08-27 05:36:55 Gladamas_ has joined
247 2012-08-27 05:37:13 Gladamas has quit (Read error: Connection reset by peer)
248 2012-08-27 05:38:03 Joric has joined
249 2012-08-27 05:38:03 Joric has quit (Changing host)
250 2012-08-27 05:38:03 Joric has joined
251 2012-08-27 05:38:06 Gladamas_ is now known as Gladamas
252 2012-08-27 05:45:47 setkeh has quit (Read error: No route to host)
253 2012-08-27 05:46:24 agricocb has joined
254 2012-08-27 05:52:00 Diablo-D3 has quit (Remote host closed the connection)
255 2012-08-27 05:54:59 Diablo-D3 has joined
256 2012-08-27 05:59:15 Diablo-D3 has quit (Remote host closed the connection)
257 2012-08-27 06:00:22 Diablo-D3 has joined
258 2012-08-27 06:07:14 <devrandom> sipa: that is a benign warning
259 2012-08-27 06:14:04 agricocb has quit (Remote host closed the connection)
260 2012-08-27 06:14:50 agricocb has joined
261 2012-08-27 06:17:14 Habbie has quit (Ping timeout: 265 seconds)
262 2012-08-27 06:22:43 Diablo-D3 has quit (Ping timeout: 244 seconds)
263 2012-08-27 06:26:35 dust-otc has joined
264 2012-08-27 06:44:45 RainbowD_ has joined
265 2012-08-27 06:45:25 RainbowDashh has quit (Read error: Connection reset by peer)
266 2012-08-27 06:45:26 RainbowD_ is now known as RainbowDashh
267 2012-08-27 06:48:37 darkee has joined
268 2012-08-27 06:49:47 fiesh has quit (Ping timeout: 268 seconds)
269 2012-08-27 06:49:51 aq83 has quit (Read error: Connection reset by peer)
270 2012-08-27 06:50:16 aq83 has joined
271 2012-08-27 06:50:58 fiesh has joined
272 2012-08-27 06:51:23 TheSeven has quit (Ping timeout: 252 seconds)
273 2012-08-27 06:51:44 TheSeven has joined
274 2012-08-27 06:51:45 guruvan- has quit (Ping timeout: 276 seconds)
275 2012-08-27 06:51:45 guruvan has quit (Ping timeout: 276 seconds)
276 2012-08-27 06:51:45 darkee has quit (Ping timeout: 276 seconds)
277 2012-08-27 06:51:48 stalled has quit (Ping timeout: 272 seconds)
278 2012-08-27 06:52:02 Z0rZ0rZ0r_ has quit (Ping timeout: 265 seconds)
279 2012-08-27 06:53:16 enquirer has joined
280 2012-08-27 06:54:59 Z0rZ0rZ0r has joined
281 2012-08-27 06:57:03 guruvan has joined
282 2012-08-27 06:57:40 stalled has joined
283 2012-08-27 06:57:52 guruvan- has joined
284 2012-08-27 07:03:01 random_cat has quit (Remote host closed the connection)
285 2012-08-27 07:04:05 keverw has joined
286 2012-08-27 07:04:45 minimoose has quit (Quit: minimoose)
287 2012-08-27 07:07:07 kreal has joined
288 2012-08-27 07:07:31 <keverw> Hey guys! Does anyone know of a Node.js script that will console.log in real time the address that got paid, address that paid it, how much got paid only after 6 confirms. I have an idea for something that needs to be real time. Just need that basic thing. I seen JS libs where you can check an address balance and stuff but I would like something higher level. I'm a broke student but willing to pay for this if I had to.
289 2012-08-27 07:09:38 <kreal> read about json-rpc
290 2012-08-27 07:09:59 ThomasV has quit (Ping timeout: 245 seconds)
291 2012-08-27 07:10:54 <kreal> I bet you can even make it yourself with enough reading :)
292 2012-08-27 07:11:03 <kreal> but if that fails I can help if you want.
293 2012-08-27 07:11:11 tower has quit (Disconnected by services)
294 2012-08-27 07:11:21 tower has joined
295 2012-08-27 07:11:23 tower has quit (Changing host)
296 2012-08-27 07:11:23 tower has joined
297 2012-08-27 07:12:13 <kreal> Question, can I make symbolic links to the blockchain and dbs and have more bitcoin daemons refer to this?
298 2012-08-27 07:12:39 <kreal> or use -datadir to the same directory.
299 2012-08-27 07:12:47 <keverw> I figured I would have to build something that would get the recent trans with 6 confirms, send the tx data to my function that takes (toAddr, fromAddr, amount), and then mark that TXID as being used already in a MySQL database.. hmm
300 2012-08-27 07:13:04 <keverw> Like some sort of polling I guess. Does that seem like an idea or is there something better?
301 2012-08-27 07:14:12 <kreal> comet
302 2012-08-27 07:14:23 <kreal> http://en.wikipedia.org/wiki/Comet_%28programming%29
303 2012-08-27 07:14:32 <kreal> long pull http requests.
304 2012-08-27 07:14:48 <kreal> donno ?
305 2012-08-27 07:16:08 <keverw> I also know of http://blockchain.info/api/api_websocket but not sure if I should put my whole business on one API...
306 2012-08-27 07:16:52 <keverw> and I doubt that runs on test net. So doubt I could really test...
307 2012-08-27 07:17:43 iocor has joined
308 2012-08-27 07:19:48 random_cat has joined
309 2012-08-27 07:26:45 MC-Eeepc has joined
310 2012-08-27 07:32:19 PK has joined
311 2012-08-27 07:38:04 davout has joined
312 2012-08-27 07:38:05 davout has quit (Changing host)
313 2012-08-27 07:38:05 davout has joined
314 2012-08-27 07:41:48 davout_ has joined
315 2012-08-27 07:43:12 tower has quit (Disconnected by services)
316 2012-08-27 07:43:26 tower has joined
317 2012-08-27 07:45:05 davout has quit (Ping timeout: 245 seconds)
318 2012-08-27 07:48:14 mmoya_ has joined
319 2012-08-27 07:50:16 edcba has joined
320 2012-08-27 07:51:12 tower has quit (Remote host closed the connection)
321 2012-08-27 07:51:39 tower has joined
322 2012-08-27 07:52:56 BTCTrader has joined
323 2012-08-27 07:58:11 davout_ has quit (Remote host closed the connection)
324 2012-08-27 08:07:35 iocor has quit (Quit: Computer has gone to sleep.)
325 2012-08-27 08:32:29 dvide has quit ()
326 2012-08-27 08:33:08 RainbowDashh has left ("Textual IRC Client: www.textualapp.com")
327 2012-08-27 08:35:35 iocor has joined
328 2012-08-27 08:36:36 RainbowDashh has joined
329 2012-08-27 08:38:01 root2 has joined
330 2012-08-27 08:40:46 ThomasV has joined
331 2012-08-27 08:42:20 davout has joined
332 2012-08-27 08:42:21 davout has quit (Changing host)
333 2012-08-27 08:42:21 davout has joined
334 2012-08-27 08:46:48 toffoo has quit ()
335 2012-08-27 09:01:28 ThomasV is now known as ThomasV_afk
336 2012-08-27 09:02:21 CodesInChaos has joined
337 2012-08-27 09:03:44 Habbie has joined
338 2012-08-27 09:08:06 setkeh has joined
339 2012-08-27 09:33:55 ThomasV_afk is now known as ThomasV
340 2012-08-27 09:40:38 ZephyrVoid has quit (Ping timeout: 244 seconds)
341 2012-08-27 09:54:29 ZephyrVoid has joined
342 2012-08-27 10:01:18 RainbowDashh has quit (Ping timeout: 244 seconds)
343 2012-08-27 10:13:19 p0s has joined
344 2012-08-27 10:18:37 someone42 has joined
345 2012-08-27 10:24:17 paraipan has joined
346 2012-08-27 10:32:40 aq has joined
347 2012-08-27 10:33:40 maaku has joined
348 2012-08-27 10:35:46 aq83 has quit (Ping timeout: 246 seconds)
349 2012-08-27 10:39:24 maaku has quit (Quit: maaku)
350 2012-08-27 10:39:24 LuaKT has joined
351 2012-08-27 10:39:24 LuaKT has quit (Changing host)
352 2012-08-27 10:39:24 LuaKT has joined
353 2012-08-27 10:40:54 <sipa> devrandom: still, it failed
354 2012-08-27 10:41:22 <sipa> in var/install.log i get an error about apt-get not being able to connect to 10.0.2.2
355 2012-08-27 10:42:51 Marf has joined
356 2012-08-27 11:07:28 Marf has quit (Ping timeout: 248 seconds)
357 2012-08-27 11:17:23 Motest031 has quit (Ping timeout: 265 seconds)
358 2012-08-27 11:17:24 Motest003 has joined
359 2012-08-27 11:18:31 maaku has joined
360 2012-08-27 11:30:22 cande has joined
361 2012-08-27 11:39:32 robocoin has joined
362 2012-08-27 11:46:38 genjix has quit (Quit: leaving)
363 2012-08-27 11:49:53 mndrix has quit (Quit: Connection closed for inactivity)
364 2012-08-27 11:52:53 [\\\] has quit (Ping timeout: 244 seconds)
365 2012-08-27 11:57:15 PhantomSpark has joined
366 2012-08-27 11:59:46 PhantomSpark has quit (3!~kvirc@pool-71-251-16-25.nycmny.fios.verizon.net|Ping timeout: 246 seconds)
367 2012-08-27 12:00:48 iocor has quit (Quit: Computer has gone to sleep.)
368 2012-08-27 12:03:41 Arnavion has quit (Quit: ==kimi ni nante okureba ii no==)
369 2012-08-27 12:05:27 olp has joined
370 2012-08-27 12:06:27 Arnavion has joined
371 2012-08-27 12:12:28 Marf has joined
372 2012-08-27 12:20:17 iocor has joined
373 2012-08-27 12:22:35 Fanquake has joined
374 2012-08-27 12:23:07 Fanquake has quit (Remote host closed the connection)
375 2012-08-27 12:23:21 Fanquake has joined
376 2012-08-27 12:23:52 Fanquake has quit (Remote host closed the connection)
377 2012-08-27 12:24:04 Fanquake has joined
378 2012-08-27 12:24:22 maaku has quit (Quit: maaku)
379 2012-08-27 12:25:04 Fanquake has quit (Remote host closed the connection)
380 2012-08-27 12:25:18 Fanquake has joined
381 2012-08-27 12:25:33 Fanquake has quit (Remote host closed the connection)
382 2012-08-27 12:25:46 Guest10546 has joined
383 2012-08-27 12:29:37 datagutt has joined
384 2012-08-27 12:35:08 Guest10546 has left ()
385 2012-08-27 12:37:22 _dr has quit (Remote host closed the connection)
386 2012-08-27 12:40:56 nsh has joined
387 2012-08-27 12:47:33 AlexWaters has joined
388 2012-08-27 12:53:57 kakobrekla has joined
389 2012-08-27 12:54:34 <kakobrekla> are there any txes at the moment that will never be included in a block?
390 2012-08-27 12:56:53 Arnavion3 has joined
391 2012-08-27 12:56:53 Arnavion has quit (Disconnected by services)
392 2012-08-27 12:58:23 Arnavion3 has left ()
393 2012-08-27 12:59:09 cande has quit (Ping timeout: 245 seconds)
394 2012-08-27 13:00:05 Arnavion has joined
395 2012-08-27 13:02:17 malaimo has joined
396 2012-08-27 13:17:56 mndrix has joined
397 2012-08-27 13:20:16 keverw has quit (Quit: keverw)
398 2012-08-27 13:21:17 cande has joined
399 2012-08-27 13:22:16 D34TH has joined
400 2012-08-27 13:22:16 D34TH has quit (Changing host)
401 2012-08-27 13:22:16 D34TH has joined
402 2012-08-27 13:26:09 <gmaxwell> kakobrekla: sure
403 2012-08-27 13:26:17 <kakobrekla> yes?
404 2012-08-27 13:26:26 <gmaxwell> kakobrekla: 4005d6bea3a93fb72f006d23e2685b85069d270cb57d15f0c057ef2d5e3f78d2 for example
405 2012-08-27 13:26:57 <kakobrekla> whats wrong with this one
406 2012-08-27 13:27:16 <gmaxwell> There are also many double spends. E.g. coinbase was paying refferal bonuses with them a couple weeks ago. 0_o.
407 2012-08-27 13:27:22 <gmaxwell> It's an invalid P2SH spend.
408 2012-08-27 13:27:30 <copumpkin> was that just a bug in coinbase?
409 2012-08-27 13:27:36 <gmaxwell> copumpkin: I assume.
410 2012-08-27 13:27:39 <copumpkin> also, did they really call themselves coinbase? :(
411 2012-08-27 13:27:40 <copumpkin> boo
412 2012-08-27 13:28:27 <gmaxwell> copumpkin: no worse than 'blockchain'. Pretty crappy all these companies calling themselves components of bitcoin; creates a lot of confusion. At least coinbase is more obscure.
413 2012-08-27 13:28:35 <kakobrekla> so before tx gets 1 confirm, there there a way of seeing that it wont get included
414 2012-08-27 13:29:00 <gmaxwell> kakobrekla: yes, ... dont make it invalid in the first place?
415 2012-08-27 13:29:00 <copumpkin> well, blockchain.info has always used the .info as part of its name, but yeah it's kinda shitty too
416 2012-08-27 13:29:19 <gmaxwell> copumpkin: they might, but their users don't.
417 2012-08-27 13:29:26 <kakobrekla> whats wrong with blockchain.info?
418 2012-08-27 13:29:34 <copumpkin> oh I guess their logo doesn't either
419 2012-08-27 13:29:38 <copumpkin> so nevermind, it's shitty :)
420 2012-08-27 13:29:59 <gmaxwell> (in fact, it's not an unusual happening that people confuse the blockchain with 'blockchain.info' e.g. they think that blockchain.info is the centeral server of bitcoin)
421 2012-08-27 13:30:15 <gmaxwell> (Then again, I don't think I would have forseen that outcome either)
422 2012-08-27 13:31:00 <gmaxwell> kakobrekla: you can always check the blocktemplate call on a full node to see if its attempting to include it. If it is, any other full node with the same version would be too.
423 2012-08-27 13:31:41 <lianj> gmaxwell: it was a locking bug in the coin selection
424 2012-08-27 13:32:10 <kakobrekla> gmaxwell may i pm you
425 2012-08-27 13:33:00 <gmaxwell> kakobrekla: Sure.
426 2012-08-27 13:33:12 <gmaxwell> But if you ever ask me that in public again the answer will be no.
427 2012-08-27 13:33:15 <gmaxwell> :)
428 2012-08-27 13:40:14 cande has quit (Read error: Connection reset by peer)
429 2012-08-27 13:43:20 slush has joined
430 2012-08-27 13:44:04 one_zero has quit ()
431 2012-08-27 13:45:42 Marf has quit (Ping timeout: 260 seconds)
432 2012-08-27 13:46:13 <slush> hi
433 2012-08-27 13:46:43 <gmaxwell> HI
434 2012-08-27 13:46:45 Habbie has quit (Ping timeout: 265 seconds)
435 2012-08-27 13:47:34 <slush> Yesterday I was offline for some time, but I mined few blocks on testnet3. Today I connected my bitcoin to the network and I expected that it will resync with the network. If my blocks win, I'll see them on blockexplorer.com/testnet, otherwise I'll see my blocks as orphaned.
436 2012-08-27 13:48:11 <gmaxwell> Yes. I would expect that, although note that blockexplorer's handling of large reorgs is flaky.
437 2012-08-27 13:48:29 <slush> On startup my bitcoind displays my chain as valid, my peers have 21108 blocks, my height 21215 and nothing happen
438 2012-08-27 13:48:56 <slush> gmaxwell: it's not just about blockexplorer, I expected that my connected peers will start downloading my blocks
439 2012-08-27 13:49:02 setkeh has quit (Quit: Love Linux ?? and Sharing Experience ?? Come Join us on Freenode at #linuxdistrocommunity)
440 2012-08-27 13:49:08 <gmaxwell> slush: a new block is needed to trigger the reorg.
441 2012-08-27 13:49:11 setkeh has joined
442 2012-08-27 13:49:29 <slush> oh
443 2012-08-27 13:49:57 <slush> so just the fact that other peer advertise different blockheight isn't enough?
444 2012-08-27 13:50:16 <gmaxwell> Those height advertisements are basically used for nothing; just display purposes.
445 2012-08-27 13:50:27 <slush> ok, that explains something
446 2012-08-27 13:51:15 <slush> so if some other peer push block from that shorter chain to me, my bitcoind will tell him that he has longer chain, right?
447 2012-08-27 13:54:20 Habbie has joined
448 2012-08-27 13:54:23 <gmaxwell> IIRC, no. The only thing that will trigger it is you announcing a block.
449 2012-08-27 13:54:57 setkeh has quit (Quit: Love Linux ?? and Sharing Experience ?? Come Join us on Freenode at #linuxdistrocommunity)
450 2012-08-27 13:54:59 <slush> and how can I do that using current git version?
451 2012-08-27 13:55:54 <gmaxwell> The most obvious way would be to mine another block. It might be possible to resubmit using submitblock but I've not tried.
452 2012-08-27 13:56:42 <slush> Maybe I'm bitching, but this looks like a bug. There's already possibly longer chain in existence, but bitcoin don't announce it into the network.
453 2012-08-27 13:58:31 <gmaxwell> "Meh"
454 2012-08-27 13:58:55 <sipa> in normal operation, the active chain is always being worked on
455 2012-08-27 13:59:08 <sipa> it's a rather strange situation for a testnet
456 2012-08-27 13:59:13 <gmaxwell> hm. we don't seem to have a call to get the raw block, so I can't see if I can trigger a reannounce with submit block.
457 2012-08-27 13:59:40 <slush> sipa: how is this relevant to testnet? Shouldn't be bitcoin resistant against network split?
458 2012-08-27 13:59:57 <gmaxwell> slush: this behavior arguably increases resistance against network split.
459 2012-08-27 14:00:18 <gmaxwell> Because it means that the network won't reorg to a temporarily longer but dead fork.
460 2012-08-27 14:00:39 <slush> gmaxwell: yes, getrawblock() would be handy for this situation
461 2012-08-27 14:00:45 TD has joined
462 2012-08-27 14:01:09 <slush> hm, it makes sense. but it looks weird at least at first view
463 2012-08-27 14:01:52 <gmaxwell> (It's not attack resistance, of course, because an attacker could just force the announcements.)
464 2012-08-27 14:02:48 * sipa feels like refactoring main entirely
465 2012-08-27 14:03:12 form_ has joined
466 2012-08-27 14:03:28 * slush is trying to mine new block at my netbook @1MHash/s, not so funny
467 2012-08-27 14:03:59 <gmaxwell> slush: hehe. well, diff 1 should only take it about an hour or so.
468 2012-08-27 14:04:01 * sipa believes slush has enough hash power to assign, if he really needs to
469 2012-08-27 14:04:23 gavinandresen has joined
470 2012-08-27 14:05:26 <TD> good afternoon
471 2012-08-27 14:05:34 <sipa> hi there
472 2012-08-27 14:05:45 <slush> not here. and redirecting pool to testnet isn't so good idea
473 2012-08-27 14:05:53 <sipa> slush: i was kidding :)
474 2012-08-27 14:05:57 <slush> :) i know
475 2012-08-27 14:06:15 variousnefarious has quit (Ping timeout: 245 seconds)
476 2012-08-27 14:06:19 setkeh has joined
477 2012-08-27 14:06:40 <justmoon> TD: hey, leveldb native win32 api port compiles and works, but there is a rare semi-random data corruption issue I'm stuck on
478 2012-08-27 14:06:42 <gmaxwell> TD: so, wrt r decompression, what happens if a peer lies about the R value for a transaction? E.g. can I make an invalid signature look valid in the batch validation by providing the wrong y coorrid?
479 2012-08-27 14:06:49 <TD> justmoon: hrm
480 2012-08-27 14:06:54 <TD> justmoon: i'm rebasing that branch at the moment
481 2012-08-27 14:07:19 <TD> gmaxwell: i don't believe so. at least the paper claims not :) the first step of the ecdsa algorithm is basically finding R from r.
482 2012-08-27 14:07:28 <sipa> gmaxwell: i think checking whether R is on the curve is fast
483 2012-08-27 14:08:00 <gmaxwell> TD: sure sure, but if its someone _else_ that does the finding?
484 2012-08-27 14:08:04 <TD> the checking process solves using the value of s and then compares to R. if you can produce the correct R, it means you have the private key. i think.
485 2012-08-27 14:08:19 <TD> yes, that's what i mean
486 2012-08-27 14:08:26 <TD> it's equivalent
487 2012-08-27 14:08:31 variousnefarious has joined
488 2012-08-27 14:08:39 <TD> you can't find an R that will let you forge a signature
489 2012-08-27 14:08:48 <sipa> but if someone sends a Ctransaction2
490 2012-08-27 14:09:02 <sipa> and some peers modifies the y coordinate in transit
491 2012-08-27 14:09:13 <TD> i think it just breaks the signature then
492 2012-08-27 14:09:24 <sipa> well yes, but that will make you think the transaction is invalid
493 2012-08-27 14:09:30 <gmaxwell> Yea, but thats a kind of DOS you can guard against.
494 2012-08-27 14:09:49 <TD> sipa: yes. but you can modify the signatures in transit already
495 2012-08-27 14:09:56 <TD> i don't see how it's any different?
496 2012-08-27 14:09:56 <gmaxwell> if a Ctransaction2 fails you do it the slow way, and if it still fails only then do you remember the result.
497 2012-08-27 14:10:07 <sipa> TD: you can't make them invalid without changing the txid
498 2012-08-27 14:10:08 talpan has joined
499 2012-08-27 14:10:08 talpan has quit (Client Quit)
500 2012-08-27 14:10:21 <TD> SignatureHash does not include the signatures themselves, obviously, so modifying the signature does not change the txid
501 2012-08-27 14:10:37 <sipa> SignatureHash != TransactionHash
502 2012-08-27 14:10:52 <TD> oh, right
503 2012-08-27 14:10:53 <TD> duh
504 2012-08-27 14:10:58 <sipa> and certainly changing the signature changes the txid
505 2012-08-27 14:11:10 <TD> sure, but then so what? what does this let you do?
506 2012-08-27 14:11:14 roconnor has joined
507 2012-08-27 14:11:26 <gmaxwell> td: kill txn on the network, unless it's guarded against.
508 2012-08-27 14:11:47 <gmaxwell> I get a txn from you, I make an invalid version, announce it everywhere. No one will take your valid version.
509 2012-08-27 14:12:04 <sipa> my question is: can you still do batch verification if you do R-reconstruction at verification time in advance?
510 2012-08-27 14:12:11 <sipa> *with speedup
511 2012-08-27 14:12:28 <TD> gmaxwell: you have to calculate "r" during signature checking anyway, so at that point it's trivial to make sure the legacy value matches
512 2012-08-27 14:13:20 <gmaxwell> TD: er. but you don't expect it to match for an invalid signature.
513 2012-08-27 14:13:32 <sipa> r is part of the signature, not calculated during verification afaik?
514 2012-08-27 14:13:59 <sipa> oh, never mind
515 2012-08-27 14:14:39 <TD> gmaxwell: if they don't match you drop the tx. the correct version is still accepted by the network. i guess the DoS blacklists would use a hash of CTx2 or something
516 2012-08-27 14:14:56 <gmaxwell> The question is "is this invalid because it's really invalid, or because we got bad acceleration data" In the latter case you just ban the peer and reinject the txn with fixed acceleration data. In the former case you remember than the txn is bad so you don't request it again.
517 2012-08-27 14:15:01 <gmaxwell> TD: or that.
518 2012-08-27 14:15:13 <gmaxwell> yes, if ctx2 peers used the ctx2 id instead of the txid that would work.
519 2012-08-27 14:15:40 <TD> sipa: in ecdsa* it's calculated. i'm not sure why the papers are based on ecdsa* instead of regular ecdsa
520 2012-08-27 14:16:11 <sipa> R is calculated in regular ECDSA as well, and its x component is compared to the submitted r value
521 2012-08-27 14:16:15 <gmaxwell> In any case, sorry for the rathole. :) This is all only important once someone has actual code that gets the Nx speedup. :)
522 2012-08-27 14:16:19 <TD> sipa: you're right - given that you can convert between R and r with some work (apparently), I don't see why you need to modify the signatures. in fact there's a paper that deals with batch unmodified ECDSA
523 2012-08-27 14:16:26 <TD> but it reports less speedup. perhaps that's why
524 2012-08-27 14:16:33 theorbtwo has quit (Remote host closed the connection)
525 2012-08-27 14:16:43 <sipa> probably, the reconstruction part cannot be batched
526 2012-08-27 14:17:11 <TD> http://cse.iitkgp.ac.in/~abhij/publications/AfricaCrypt12-ECDSA-bv.pdf
527 2012-08-27 14:17:16 ThomasV has quit (Quit: Leaving)
528 2012-08-27 14:17:17 <TD> that's batching for unmodified ECDSA
529 2012-08-27 14:17:29 <TD> "Experiments carried out on NIST prime curves demonstrate a maximum speedup
530 2012-08-27 14:17:29 <TD> of above six over individual veriï¬cation if all the signatures in the batch belong
531 2012-08-27 14:17:30 <TD> to the same signer, and a maximum speedup of about two if the signatures in the
532 2012-08-27 14:17:30 <TD> batch belong to different signers"
533 2012-08-27 14:17:59 <TD> vs 4x for ECDSA*
534 2012-08-27 14:18:03 <TD> in the multi-signer case
535 2012-08-27 14:18:11 <TD> so apparently it's a big difference. that must be why they modified the signatures.
536 2012-08-27 14:23:59 <TD> also the ecdsa* is a lot faster for non-koblitz curves
537 2012-08-27 14:24:00 iocor has quit (Quit: Computer has gone to sleep.)
538 2012-08-27 14:24:22 balrog has quit (Read error: Connection reset by peer)
539 2012-08-27 14:27:57 <gmaxwell> Gitk visualization of the bitcoin git repo looks like an LA freeway map: https://people.xiph.org/~greg/ss-gitk.png
540 2012-08-27 14:28:21 <TD> LA must have some pretty dangerous freeways
541 2012-08-27 14:28:47 pnicholson has joined
542 2012-08-27 14:29:05 setkeh has quit (Ping timeout: 272 seconds)
543 2012-08-27 14:29:38 iocor has joined
544 2012-08-27 14:29:48 robocoin has quit (Read error: Operation timed out)
545 2012-08-27 14:29:59 Marf has joined
546 2012-08-27 14:30:48 theorbtwo has joined
547 2012-08-27 14:31:07 balrog has joined
548 2012-08-27 14:33:29 Zarutian has joined
549 2012-08-27 14:59:54 <gavinandresen> Looking for review/ACK's for the DoS fix that I've been working on: https://github.com/bitcoin/bitcoin/pull/1729
550 2012-08-27 15:00:48 ThomasV has joined
551 2012-08-27 15:03:34 Arnavion has quit (Ping timeout: 252 seconds)
552 2012-08-27 15:04:38 Arnavion has joined
553 2012-08-27 15:09:58 p0s has quit (Read error: Connection reset by peer)
554 2012-08-27 15:11:52 nonick has joined
555 2012-08-27 15:13:33 darkee has quit (!~darkee@gateway/tor-sasl/darkee|Ping timeout: 276 seconds)
556 2012-08-27 15:14:05 <BlueMatt> sipa: re: gitian, no, but I have an equivalent environment...anyway, Id have to suggest sticking with pull-testers build (Im assuming its way too late anyway)
557 2012-08-27 15:17:31 D34TH_ has joined
558 2012-08-27 15:19:11 davout has quit (Remote host closed the connection)
559 2012-08-27 15:20:02 <gavinandresen> BlueMatt: can you summarize the paths to getting a gitian-built 0.7 ?
560 2012-08-27 15:20:03 D34TH has quit (Ping timeout: 276 seconds)
561 2012-08-27 15:21:53 sebicas has joined
562 2012-08-27 15:24:39 <BlueMatt> I believe luke was working with devrandom to get it working in lxc, so in theory, you can do a gitian build in an ubuntu vm, otherwise I believe you still need vt-x hardware running ubuntu (or, just get the actual ubuntu base vm using the make-base-vm script in a physical ubuntu, then use that on any kvm system, which is what I do)
563 2012-08-27 15:24:50 pnicholson has quit (Quit: pnicholson)
564 2012-08-27 15:24:51 <BlueMatt> I dont know much about the status of the lxc stuff, you'd have to ask luke
565 2012-08-27 15:25:23 <BlueMatt> in terms of build right now, need to upgrade to boost 1.50 and merge one of luke's commits before it will build
566 2012-08-27 15:25:23 <gavinandresen> BlueMatt: I meant a summary of the "pull 1716 versus upgrade to Boost 1.50"
567 2012-08-27 15:26:04 <gavinandresen> gitian building doesn't work right now? Or URI support won't work without upgrading...
568 2012-08-27 15:26:11 <BlueMatt> no, it fails
569 2012-08-27 15:26:19 <BlueMatt> pull 1716 moves the boost monkey-patch to the header before including boost, which would work, but its as ugly (or I'd argue uglier) than the orginal patch
570 2012-08-27 15:26:43 <gavinandresen> what made it start failing?
571 2012-08-27 15:26:52 <BlueMatt> I believe me
572 2012-08-27 15:27:06 <BlueMatt> the upgrade to boost 1.49 killed it, but also my patch (apparently)
573 2012-08-27 15:27:47 <BlueMatt> Id like to see us just upgrade to 1.50 to fix the issue (aside from the typo in 1.49)
574 2012-08-27 15:28:12 <gavinandresen> ok... backing out whatever made things break seems like it would be the safest, but I agree that moving forward is better than back.
575 2012-08-27 15:28:28 <BlueMatt> but, apparently, to get it to build as a part of gentoo's stuff, we'd still need the patch
576 2012-08-27 15:28:41 <BlueMatt> Im torn as to whether or not we should keep it, or make gentoo put it in their patch list
577 2012-08-27 15:29:06 <gavinandresen> what's market share of gentoo?
578 2012-08-27 15:29:30 <BlueMatt> its not just gentoo, its gentoo compiling for win32, so...none?
579 2012-08-27 15:31:06 <BlueMatt> we still have the full rc phase, so I just figure any bugs boost 1.50 will introduce will get found (they are usually quite obvious) and we can revert if we have to
580 2012-08-27 15:31:30 <gavinandresen> ok, so what needs to get done to upgrade to 1.50. And is 1.50 the latest?
581 2012-08-27 15:31:49 <BlueMatt> 1.51 is, but it introduces bugs, I forget which, but laanjw linked it in the pull
582 2012-08-27 15:31:50 <gavinandresen> why not 1.51?
583 2012-08-27 15:32:13 <BlueMatt> #1719
584 2012-08-27 15:32:46 <gavinandresen> mmm... another 'causes issues on a platform we don't care about' issue....
585 2012-08-27 15:32:51 <BlueMatt> I figure upgrading to a fixed version of boost would be better than fixing it ourselves (as it is now, as I did it to fix...whatever it was)
586 2012-08-27 15:33:08 <gavinandresen> (yeah, yeah, I know Wine is useful for wumpus to test stuff)
587 2012-08-27 15:33:30 <BlueMatt> well, itl kill pull-tester
588 2012-08-27 15:33:40 <gavinandresen> pull-tester uses Wine?
589 2012-08-27 15:33:43 <BlueMatt> yea
590 2012-08-27 15:33:47 <gavinandresen> sigh. OK
591 2012-08-27 15:33:56 <BlueMatt> (to run the test cases compiled for win32)
592 2012-08-27 15:34:02 theorbtwo has quit (Read error: Connection reset by peer)
593 2012-08-27 15:34:32 theorbtwo has joined
594 2012-08-27 15:34:47 <gavinandresen> All this mess is triggered by the URI boost:interprocess code, yes?
595 2012-08-27 15:35:03 <BlueMatt> dunno 'bout 1719, but the monkey-patch, yes
596 2012-08-27 15:35:20 <gavinandresen> 1719 is boost::interprocess
597 2012-08-27 15:35:27 <BlueMatt> must be then
598 2012-08-27 15:35:33 <sipa> what about a DISABLE_FANCY_WIN32_PROTECTION_STUFF in the makefiles that disables that stack protection, and using that in the build tester?
599 2012-08-27 15:36:01 <gavinandresen> what does stack protection have to do with any of this?
600 2012-08-27 15:36:05 <BlueMatt> wfm, though I do prefer to keep it as close as possible to the gitian ouptuts
601 2012-08-27 15:36:29 <BlueMatt> I dont think sipa is on the same topic
602 2012-08-27 15:36:37 <BlueMatt> (or is 1719 stack-protector related?)
603 2012-08-27 15:37:13 <gavinandresen> 1719 looks like a purely boost::interprocess issue, ticking a "wine doesn't support the right kind of semaphores" issue
604 2012-08-27 15:38:17 vigilyn has joined
605 2012-08-27 15:38:49 <sipa> BlueMatt: sorry, i just read something about wine compatibility, which that stack protection stuff broke, iirc?
606 2012-08-27 15:38:57 <sipa> but apparently you're talking about something else
607 2012-08-27 15:39:10 <BlueMatt> sipa: yes, adding -fstack-protector kills wine
608 2012-08-27 15:39:26 <BlueMatt> but we're on the gitian booost 1.50/monkey-patch stuff atm
609 2012-08-27 15:39:33 <BlueMatt> (#1716)
610 2012-08-27 15:40:29 <gavinandresen> Ok: seems like the right thing to do is: pull 1716. Monkey-patch or #define before #include are equally ugly in my humble opinion.
611 2012-08-27 15:40:52 <BlueMatt> agreed, but can we please also do 1.50 before we start rc?
612 2012-08-27 15:41:04 <gavinandresen> what needs to be done to "do 1.50" ?
613 2012-08-27 15:41:25 <BlueMatt> s/49/50/ and bump boost-gitian yml version
614 2012-08-27 15:41:40 <gavinandresen> s/49/50/ where ?
615 2012-08-27 15:41:51 <BlueMatt> makefiles and gitian files should be it afaik
616 2012-08-27 15:42:02 <BlueMatt> just the mingw/linux-mingw and qt .pro ones i believe
617 2012-08-27 15:42:15 <TD> how does it kill wine, btw?
618 2012-08-27 15:42:24 <TD> i may be able to fix that. i spent a few years as a codeweavers employee
619 2012-08-27 15:42:34 <TD> so i know the wine codebase better than any healthy person should
620 2012-08-27 15:42:37 <BlueMatt> TD: https://github.com/bitcoin/bitcoin/pull/1674
621 2012-08-27 15:42:51 <gavinandresen> TD: that's the type of thing I would keep a deep, dark secret
622 2012-08-27 15:42:53 <BlueMatt> may want to talk to gmaxwell/Diapolo
623 2012-08-27 15:43:10 D34TH has joined
624 2012-08-27 15:43:10 D34TH has quit (Changing host)
625 2012-08-27 15:43:10 D34TH has joined
626 2012-08-27 15:44:18 <gavinandresen> BlueMatt: if you want boost 1.50, create a pull request that does the right things....
627 2012-08-27 15:44:27 <TD> thanks
628 2012-08-27 15:44:42 <BlueMatt> gavinandresen: ack, give me a few hours
629 2012-08-27 15:45:21 <TD> gavinandresen: it gets worse. my job was to reimplement DCOM
630 2012-08-27 15:45:32 <gavinandresen> TD: oh my.
631 2012-08-27 15:45:39 <BlueMatt> ouch...
632 2012-08-27 15:46:05 <gavinandresen> TD: you must have all sorts of nasty technical details filling up valuable brainspace
633 2012-08-27 15:46:42 D34TH_ has quit (Ping timeout: 276 seconds)
634 2012-08-27 15:46:57 <TD> the sort of thing that will come out in therapy one day. i'll just blurt out CoMarshalInterThreadInterfaceInStream(pUNK) and end up being committed
635 2012-08-27 15:47:12 <gavinandresen> BlueMatt: besides warm fuzzies, do we gain anything by switching from boost 1.49 to boost 1.50 ?
636 2012-08-27 15:47:28 <TD> gavinandresen: why make the string part of the alert protocol? that just means it can't be internationalized
637 2012-08-27 15:47:53 <gavinandresen> TD: because otherwise an attacker with the key could send out a blank message which couldn't be overridden.
638 2012-08-27 15:48:02 <BlueMatt> gavinandresen: nope, just warm fuzzies and less chance of something 'splodin later because we forgot to throw in the patch before #including <boost/foreach.hpp> which for some godawful reason includes <boost/interprocess/*>
639 2012-08-27 15:48:18 <sipa> wait, what?
640 2012-08-27 15:48:19 <TD> i mean, in the maxID case, just ignore the message entirely and then substitute in a hard-coded one at the gui level
641 2012-08-27 15:48:35 <TD> BlueMatt: boost foreach macro includes IPC code, what?!
642 2012-08-27 15:48:44 <gavinandresen> TD: sure, that could be done...
643 2012-08-27 15:48:46 <BlueMatt> TD: no, its an example
644 2012-08-27 15:48:48 <TD> oh
645 2012-08-27 15:48:49 <TD> haha
646 2012-08-27 15:48:50 <TD> phew
647 2012-08-27 15:48:53 <BlueMatt> yea
648 2012-08-27 15:48:57 <BlueMatt> that would be "godawful"
649 2012-08-27 15:48:59 D34TH_ has joined
650 2012-08-27 15:49:18 D34TH has quit (Ping timeout: 276 seconds)
651 2012-08-27 15:49:52 <gavinandresen> TD: hard-coded string appears in RPC output, too, which doesn't go through the i18n stuff
652 2012-08-27 15:49:59 <sipa> TD: I hoped they paid you to compensate for psychological and probably physical damages resulting from that?
653 2012-08-27 15:50:05 <TD> i was paid, yes
654 2012-08-27 15:50:15 <TD> not much but for a student it didn't seem so bad
655 2012-08-27 15:56:57 davout has joined
656 2012-08-27 15:56:57 davout has quit (Changing host)
657 2012-08-27 15:56:57 davout has joined
658 2012-08-27 15:58:02 theorbtwo has quit (Remote host closed the connection)
659 2012-08-27 15:58:48 <BlueMatt> wait...we dont already assign a DoS value to alerts with invalid sigs...
660 2012-08-27 15:59:58 jgarzik has joined
661 2012-08-27 16:00:07 <gavinandresen> nope, we don't.
662 2012-08-27 16:00:24 <gavinandresen> jgarzik: I thought you were at a nice beach somewhere.
663 2012-08-27 16:00:27 <BlueMatt> well, good thing someone's fixing that...
664 2012-08-27 16:01:06 <jgarzik> gavinandresen: I am! :)
665 2012-08-27 16:01:22 <jgarzik> the only way I could do a whole week at the beach was to work Monday & Tuesday :)
666 2012-08-27 16:01:42 * jgarzik turns left, and looks at the Atlantic Ocean from his beach condo
667 2012-08-27 16:01:51 * gavinandresen is jealous
668 2012-08-27 16:01:52 <TD> nice!
669 2012-08-27 16:01:56 <BlueMatt> nice
670 2012-08-27 16:01:57 <jgarzik> great place for coding NFSv4.1 services
671 2012-08-27 16:02:10 <jgarzik> possibly more complex than a bitcoin-validating python node
672 2012-08-27 16:02:50 <BlueMatt> well, get off bitcoin and finish up to look at the beach more ;)
673 2012-08-27 16:03:50 <TD> BlueMatt: is your dns seed coming back at any point?
674 2012-08-27 16:03:55 <BlueMatt> its up
675 2012-08-27 16:03:58 <BlueMatt> is it now?
676 2012-08-27 16:04:19 <TD> dnsseed.bluematt.me ?
677 2012-08-27 16:04:21 <BlueMatt> no, looks up to me
678 2012-08-27 16:04:22 <BlueMatt> yea
679 2012-08-27 16:04:36 <TD> odd
680 2012-08-27 16:04:42 <TD> i'm seeing (from bitcoinj): 05:58:07 11 DnsDiscovery.getPeers: DNS lookup for dnsseed.bluematt.me failed.
681 2012-08-27 16:04:47 <TD> also when I do a dig on it, I get no A records
682 2012-08-27 16:04:55 <sipa> looks up from here
683 2012-08-27 16:04:55 <BlueMatt> what does dig +trace show?
684 2012-08-27 16:05:41 <jgarzik> ;; Got answer:
685 2012-08-27 16:05:41 <jgarzik> ;; ->>HEADER<<- opcode: QUERY, status: SERVFAIL, id: 15400
686 2012-08-27 16:05:41 <gribble> Error: "Got" is not a valid command.
687 2012-08-27 16:05:42 <gribble> Error: "->>HEADER<<-" is not a valid command.
688 2012-08-27 16:05:42 <BlueMatt> and/or do you have an old ip for dnsseedns.bluematt.me
689 2012-08-27 16:05:48 <jgarzik> BlueMatt: ^^
690 2012-08-27 16:05:56 <BlueMatt> :(
691 2012-08-27 16:06:06 <TD> http://pastebin.com/UhiZeLTW
692 2012-08-27 16:06:07 <jgarzik> ;; SERVER: 10.11.255.156#53(10.11.255.156)
693 2012-08-27 16:06:08 <BlueMatt> dns should have propagated by now...
694 2012-08-27 16:06:09 <TD> timeout
695 2012-08-27 16:06:18 <jgarzik> BlueMatt: is ^^ the correct IP?
696 2012-08-27 16:06:23 <TD> btw i've probably got ipv6 here
697 2012-08-27 16:06:41 <TD> ;; ->>HEADER<<- opcode: QUERY, status: SERVFAIL, id: 54035
698 2012-08-27 16:06:41 <BlueMatt> dnsseedns.bluematt.me CNAME jenkins.bluematt.me -> 204.236.249.183
699 2012-08-27 16:06:41 <gribble> Error: "->>HEADER<<-" is not a valid command.
700 2012-08-27 16:06:51 D34TH_ has quit (Ping timeout: 276 seconds)
701 2012-08-27 16:07:54 <TD> ;; Query time: 14 msec
702 2012-08-27 16:07:54 <TD> ;; SERVER: 172.16.255.1#53(172.16.255.1)
703 2012-08-27 16:07:54 <gribble> Error: "Query" is not a valid command.
704 2012-08-27 16:07:55 <gribble> Error: "SERVER:" is not a valid command.
705 2012-08-27 16:08:05 <TD> so it appears i have the wrong ip. when did you change it?
706 2012-08-27 16:08:20 <BlueMatt> I thought it was > 1 week ago, but maybe Im wrong
707 2012-08-27 16:09:08 <TD> is ns1-auth.rollernet.us. correct?
708 2012-08-27 16:09:17 <BlueMatt> yea, thats right
709 2012-08-27 16:09:23 dust-otc has quit (Remote host closed the connection)
710 2012-08-27 16:09:33 <BlueMatt> thats why Im confused, i dont see them down
711 2012-08-27 16:09:34 kakobrekla has left ()
712 2012-08-27 16:09:48 <TD> ok
713 2012-08-27 16:09:52 <TD> so dnsseedns.bluematt.me resolves for me
714 2012-08-27 16:09:55 <TD> dnsseedns.bluematt.me. 3276 IN CNAME jenkins.bluematt.me.
715 2012-08-27 16:09:56 <TD> jenkins.bluematt.me. 86400 IN A 204.236.249.183
716 2012-08-27 16:10:05 <BlueMatt> thats right
717 2012-08-27 16:10:12 <TD> but dnsseed.bluematt.met -> SERVFAIL
718 2012-08-27 16:10:17 maaku has joined
719 2012-08-27 16:10:24 <TD> 172.16.255.1#53
720 2012-08-27 16:10:54 <TD> so for some reason, i'm not seeing the IP match
721 2012-08-27 16:10:55 Arnavion has quit (Ping timeout: 252 seconds)
722 2012-08-27 16:11:02 <jgarzik> $ dig @dnsseedns.bluematt.me. dnsseed.bluematt.me. any
723 2012-08-27 16:11:03 <BlueMatt> somehow that looks familiar, maybe thats the old one
724 2012-08-27 16:11:04 <jgarzik> ;; ->>HEADER<<- opcode: QUERY, status: FORMERR, id: 61891
725 2012-08-27 16:11:04 <jgarzik> ;; flags: rd ad; QUERY: 0, ANSWER: 0, AUTHORITY: 0, ADDITIONAL: 0
726 2012-08-27 16:11:05 <gribble> Error: "->>HEADER<<-" is not a valid command.
727 2012-08-27 16:11:05 <gribble> Error: "flags:" is not a valid command.
728 2012-08-27 16:11:10 <jgarzik> ditto for s/any/a/
729 2012-08-27 16:11:19 <BlueMatt> jgarzik: whats dnsseedns.bluematt.me A?
730 2012-08-27 16:11:40 <jgarzik> ;; SERVER: 204.236.249.183#53(204.236.249.183)
731 2012-08-27 16:12:57 kjj has left ()
732 2012-08-27 16:13:14 * TD wonders why bitcoinj block sync has got so damn slow
733 2012-08-27 16:13:26 <TD> or maybe it's just the remote peers that got slow
734 2012-08-27 16:13:29 <TD> 5 blocks per second is pathetic
735 2012-08-27 16:13:40 kjj_ has joined
736 2012-08-27 16:13:46 <BlueMatt> yea, need to look at pulling blocks from multiple nodes ;)
737 2012-08-27 16:13:56 t7_ has joined
738 2012-08-27 16:13:58 skeledrew has quit (Ping timeout: 252 seconds)
739 2012-08-27 16:14:15 Arnavion has joined
740 2012-08-27 16:14:46 PK has quit (Quit: Leaving)
741 2012-08-27 16:15:17 <TD> yeah but why would they have slowed down so much. these blocks aren't necessarily huge.
742 2012-08-27 16:15:28 <TD> i guess they're just so bogged down processing newly broadcast transactions
743 2012-08-27 16:16:07 buttpirate has joined
744 2012-08-27 16:16:27 <jgarzik> TD: newly broadcast transactions don't bog the network as much as blocks, by far
745 2012-08-27 16:16:44 Diapolo has joined
746 2012-08-27 16:17:02 <jgarzik> TD: block serving involves a ton of seeking. block reception involves much more seeking and writing. TX processing, OTOH, involves a tiny bit of disk seeking and that's it.
747 2012-08-27 16:17:22 <jgarzik> from the standpoint of a public node, block serving probably causes the most disk traffic by far
748 2012-08-27 16:17:29 <TD> block serving is just one seek, no? i mean once you have fetched the block offsets, they're all laid out sequentially on disk
749 2012-08-27 16:17:36 <TD> so it should be one seek+readahead to load all the requested blocks
750 2012-08-27 16:18:27 <jgarzik> TD: if disk cache < sizeof(blk*.dat), then it's all disk, all the time. Remember that a public node is probably serving N clients M different blocks, at any given moment.
751 2012-08-27 16:18:48 buttpirate has quit (Client Quit)
752 2012-08-27 16:19:30 <jgarzik> ...and those clients are all at different levels of IBD or network sync...
753 2012-08-27 16:19:37 <TD> most clients, hopefully not ...
754 2012-08-27 16:19:44 <TD> but yeah. i guess it's possible nodes are just really bogged down
755 2012-08-27 16:19:48 <midnightmagic> Are they sequential on disk? bdb is btree isn't it?
756 2012-08-27 16:19:52 <jgarzik> TD: that's definitely what I see on my 3 public nodes
757 2012-08-27 16:19:57 <TD> hmm
758 2012-08-27 16:19:58 skeledrew has joined
759 2012-08-27 16:20:01 <jgarzik> midnightmagic: block data is sequential on disk
760 2012-08-27 16:20:03 <TD> midnightmagic: blk000?.dat is sequential
761 2012-08-27 16:20:05 ThomasV has quit (Ping timeout: 252 seconds)
762 2012-08-27 16:20:19 <BlueMatt> midnightmagic: looking up each txin for each tx in a block certainly isnt sequential
763 2012-08-27 16:21:07 <jgarzik> TD: You're welcome to request debug.log from my public nodes at any time, if you are curious. That offer is open to all devs... these nodes are wallet-free, so nothing would be compromised.
764 2012-08-27 16:21:31 <TD> ok
765 2012-08-27 16:23:41 Diapolo has quit (Quit: Leaving.)
766 2012-08-27 16:23:51 Diapolo has joined
767 2012-08-27 16:23:58 rcorreia has quit (Read error: Connection reset by peer)
768 2012-08-27 16:24:10 <midnightmagic> jgarzik, TD: page splitting and the build of btree nodes puts it at stepped sequential, assuming internal nodes are caches doesn't it? how is the block data fully sequential? (curious now, since my employer uses a modified bdb and our sequential data is close, but readahead is a big tunable for us.)
769 2012-08-27 16:24:28 <midnightmagic> BlueMatt: Ah, yes of course..
770 2012-08-27 16:24:35 slush has quit (Ping timeout: 245 seconds)
771 2012-08-27 16:25:25 kjj_ has left ()
772 2012-08-27 16:25:29 kjj_ has joined
773 2012-08-27 16:26:15 <kjj_> odd DNS problems are often caused by incorrect glue
774 2012-08-27 16:26:48 * midnightmagic wants to strace a sequential block read now
775 2012-08-27 16:28:29 rcorreia has joined
776 2012-08-27 16:28:36 <jgarzik> midnightmagic: well for bdb, bitcoin updates records, not just creating new ones (it must mark older txout's as spent)
777 2012-08-27 16:28:45 <jgarzik> midnightmagic: the block data is stored outside bdb, in sequential binary files
778 2012-08-27 16:30:41 aq has quit (2!~kvirc@chello080109197142.2.graz.surfer.at|Ping timeout: 246 seconds)
779 2012-08-27 16:30:49 <sipa> i have no numbers but checking a tx for acceptance to the block chain can be a lot more expensive than serving a block
780 2012-08-27 16:30:51 <midnightmagic> jgarzik: oh. outside? wait are you guys talking about the new format or the one we're all using right now?
781 2012-08-27 16:31:03 <TD> w.t.f.
782 2012-08-27 16:31:06 <sipa> if the inputs are in several blocks
783 2012-08-27 16:31:07 <TD> i wonder if it's a timing issue
784 2012-08-27 16:31:20 <TD> on my workstation "dig dnsseed.bluematt.me" fails immediately, but dig +trace dnsseed.bluematt.me works
785 2012-08-27 16:31:24 <TD> what's the difference?
786 2012-08-27 16:31:27 <midnightmagic> blk0001.dat: data <-- oh!
787 2012-08-27 16:31:29 <TD> some cache?
788 2012-08-27 16:31:38 <midnightmagic> thanks jgarzik. I didn't realise. I thought they were all bdb
789 2012-08-27 16:31:55 * TD tries to remember the spell for clearing the dns cache
790 2012-08-27 16:32:09 rcorreia has quit (Remote host closed the connection)
791 2012-08-27 16:32:13 <TD> it seems the server line when i do dig is 172.16.255.1 but when I do +trace it gets the right ip
792 2012-08-27 16:32:14 <midnightmagic> rndc flush {view}
793 2012-08-27 16:32:15 <sipa> midnightmagic: ultraprune still has the blk data files, but doesn't need it for validation of txn
794 2012-08-27 16:32:25 <midnightmagic> sipa: nice
795 2012-08-27 16:33:27 <midnightmagic> sipa: Ultraprune has the utxo monolithic set?
796 2012-08-27 16:33:33 <midnightmagic> hrm.. why aren't I running that..
797 2012-08-27 16:33:47 <sipa> midnightmagic: indeed
798 2012-08-27 16:33:57 <kjj_> TD: ditto. 208.79.240.10 reports that dnsseed.bluematt.me is the NS for dnsseed.bluematt.me, but provides no glue
799 2012-08-27 16:34:09 <jgarzik> sipa: recent blocks are more likely to be cached, as they are more likely to be served. Incoming TX's are more likely to require checking recent blocks, than older blocks.
800 2012-08-27 16:34:20 <jgarzik> This is very apparent in pynode, which is slower than bitcoind
801 2012-08-27 16:34:20 <TD> i don't have rndc
802 2012-08-27 16:34:23 * TD stabs linux
803 2012-08-27 16:34:28 <devrandom> sipa: perhaps apt-cacher-ng is not running on the host?
804 2012-08-27 16:34:29 <sipa> jgarzik: sure
805 2012-08-27 16:34:32 <TD> why is every instruction i find for this out of date?
806 2012-08-27 16:34:32 <TD> urgh
807 2012-08-27 16:34:33 Diapolo has left ()
808 2012-08-27 16:34:54 <jgarzik> TD: probably in a bind-server pkg of some sort
809 2012-08-27 16:35:02 <sipa> devrandom: that was earlier, and then it gave a different error, iirc
810 2012-08-27 16:35:16 <jgarzik> sipa: therefore... serving older blocks punishes public nodes more than incoming TX's
811 2012-08-27 16:35:24 Diapolo has joined
812 2012-08-27 16:35:29 <jgarzik> in terms of disk seeking
813 2012-08-27 16:35:33 <Diapolo> test
814 2012-08-27 16:35:36 Marf has quit (Ping timeout: 252 seconds)
815 2012-08-27 16:35:38 <sipa> jgarzik: sounds reasonable
816 2012-08-27 16:35:46 <Diapolo> sorry, it seems I had to register my nick to be able to speak ^^
817 2012-08-27 16:35:55 <kjj_> TD: are you running your own local resolver? what is in your named.conf?
818 2012-08-27 16:35:55 <sipa> really?
819 2012-08-27 16:36:00 <kjj_> TD: er, resolv.conf rather
820 2012-08-27 16:36:21 <devrandom> sipa: ok, let me know what the other error was
821 2012-08-27 16:36:23 <kjj_> yeah, I had to re-register mine after someone jacked it
822 2012-08-27 16:37:02 <TD> i'm at work. resolv.conf says
823 2012-08-27 16:37:15 <TD> search corp.google.com. prod.google.com. (etc)
824 2012-08-27 16:37:19 <TD> nameserver <ip>
825 2012-08-27 16:37:24 <TD> for a few IPs
826 2012-08-27 16:37:27 <TD> options ndots:2
827 2012-08-27 16:37:34 <kjj_> TD: then you won't have a rndc because you aren't running bind
828 2012-08-27 16:37:39 <TD> right
829 2012-08-27 16:38:03 <kjj_> TD: so, no cache to clear either. I'm still thinking glue
830 2012-08-27 16:38:06 <TD> flushing the cache on my macos laptop didn't seem to make any difference, mind you. i wonder if the bad data has been cached at the google level
831 2012-08-27 16:38:37 <TD> ugh
832 2012-08-27 16:38:38 <jgarzik> TD: on Fedora, rndc is in the "bind" pkg with the server itself. rndc is basically the command line server control util.
833 2012-08-27 16:38:40 <TD> yep
834 2012-08-27 16:38:49 <devrandom> libresolve is a pure library. there's no daemon
835 2012-08-27 16:38:53 <devrandom> so there's nothing to clear
836 2012-08-27 16:39:02 <TD> looks like google corp DNS has bad data :( why didn't it expire by now?
837 2012-08-27 16:39:02 <jgarzik> TD: that's why god invented "dig @server"... eliminates cache issues
838 2012-08-27 16:39:07 <jgarzik> TD: just walk the dns chain manually
839 2012-08-27 16:39:09 <TD> dig @8.8.8.8 dnsseed.bluematt.me works
840 2012-08-27 16:39:20 <TD> dig @172.16.255.1 dnsseed.bluematt.me does not
841 2012-08-27 16:39:33 <devrandom> what does the TTL show?
842 2012-08-27 16:39:39 <kjj_> the problem is that when you try to resolve it normally, it isn't doing the same stuff that dig +trace would do
843 2012-08-27 16:40:22 <TD> the max TTLs i see in the successful +trace are 1 day
844 2012-08-27 16:40:22 <devrandom> TD: the TTL is the number before the "IN"
845 2012-08-27 16:40:25 <TD> i know
846 2012-08-27 16:40:30 <TD> matt said he changed the config a week ago
847 2012-08-27 16:41:02 <TD> maybe our corp DNS is hosed
848 2012-08-27 16:41:13 AlexWaters has quit (Quit: Leaving.)
849 2012-08-27 16:41:25 <devrandom> I get SERVFAIL trying to dig dnsseed.bluematt
850 2012-08-27 16:41:57 <TD> yes
851 2012-08-27 16:41:59 <Diapolo> @sipa If you've got the time, could you help me out with https://github.com/bitcoin/bitcoin/issues/1721?
852 2012-08-27 16:42:04 <TD> but try dig @8.8.8.8 dnsseed.bluematt.me
853 2012-08-27 16:42:35 <devrandom> I get results
854 2012-08-27 16:42:48 torsthaldo has joined
855 2012-08-27 16:42:55 <devrandom> but something is wrong with the authoritative servers for the domain
856 2012-08-27 16:43:13 <devrandom> and probably 8.8.8.8 is caching some old stuff before they got borked
857 2012-08-27 16:43:50 <devrandom> this gives no A records: dig dnsseed.bluematt.me @ns1-auth.rollernet.us.
858 2012-08-27 16:44:09 <kreal> seems like people are alive in here now, so I'll try and ask my question again, if you don't mind.
859 2012-08-27 16:44:14 <kreal> Question, can I make symbolic links to the blockchain and dbs and have more bitcoin daemons refer to this?
860 2012-08-27 16:44:16 <devrandom> same for ns2-auth
861 2012-08-27 16:44:21 <kjj_> ok, whois lists ns1-auth.rollernet.us and ns2-auth. if you dig @ them directly, they respond to your A query with the NS dnsseed.bluematt.me
862 2012-08-27 16:44:32 Arnavion3 has joined
863 2012-08-27 16:44:32 Arnavion has quit (Disconnected by services)
864 2012-08-27 16:45:01 <devrandom> oh
865 2012-08-27 16:45:15 <devrandom> they actually say dnsseedNS.bluematt.me
866 2012-08-27 16:45:24 <kjj_> they are delegating, but not providing glue. every query dead ends with a NS pointer
867 2012-08-27 16:45:49 <Diapolo> kreal: I'm sure some have done this at least for blk000x.dat, but I never used that setup.
868 2012-08-27 16:46:08 <TD> so how comes it works some places and not others?
869 2012-08-27 16:46:14 <TD> if the config didn't change in the last day?
870 2012-08-27 16:46:35 <kreal> Diapolo, ok well guess I'll just have to try.
871 2012-08-27 16:47:33 <Diapolo> kreal: I'm just asking myself, what happens, if more than 2 daemon is writing to it... well sorry can't really help out technical wise here.
872 2012-08-27 16:47:53 <kreal> guess that would mess something up yes.
873 2012-08-27 16:48:17 <kreal> but I had an idea that it only reads from the blockchain. and writes to wallet.daty
874 2012-08-27 16:48:21 <kreal> wallet.dat*
875 2012-08-27 16:48:27 <kreal> but don't know.
876 2012-08-27 16:48:41 <kreal> I'll do study the code again.
877 2012-08-27 16:48:44 <kreal> go*
878 2012-08-27 16:49:14 <Diapolo> new blocks are appended to the blk000x.dat file
879 2012-08-27 16:50:01 <kreal> hmm ok awesome.
880 2012-08-27 16:50:20 <devrandom> TD: maybe 8.8.8.8 happened to cache stuff, so it succeeds from the cache...
881 2012-08-27 16:50:40 <TD> ok, so it seems that people agree BlueMatts setup is borked, somehow
882 2012-08-27 16:50:44 <kreal> My quest is to move the wallet.dat from Berkeley DB to MySQLi, just for fun.
883 2012-08-27 16:51:26 <TD> devrandom: http://code.google.com/r/hearn-bitcoinj/source/detail?r=108c2349942eec0047024f615ea15173d407d105
884 2012-08-27 16:51:32 <Diapolo> you've got much time :)?
885 2012-08-27 16:51:34 dvide has joined
886 2012-08-27 16:51:51 <kreal> Diapolo, plenty.
887 2012-08-27 16:51:52 <kjj_> TD: he needs to register his nameservers with his registrar, and switch the authority from rollernet.us to his own
888 2012-08-27 16:52:50 <kjj_> TD: as to why it sometimes works, I have no idea.
889 2012-08-27 16:54:59 <devrandom> TD: will look at autosave a bit later
890 2012-08-27 16:55:12 <TD> thanks
891 2012-08-27 16:55:15 <TD> it's a pain
892 2012-08-27 16:55:30 <TD> at some point i need to change the code to hold the wallet lock for the minimal time as well, for android responsiveness
893 2012-08-27 16:56:21 darkee has joined
894 2012-08-27 16:57:37 <kreal> Diapolo, I had an idea that I could make bitcoind multi wallet compatible by having the wallet file in database, but hey lets see how far I get.
895 2012-08-27 16:59:30 nonick has quit (Ping timeout: 276 seconds)
896 2012-08-27 17:00:50 <Luke-Jr> gavinandresen: Can I suggest not DoS-banning on "I don't trust this key" for alerts? It would be nice not to punish other clients that might use alerts themselves in the future (or key changes, if it were compromised, etc).
897 2012-08-27 17:00:57 D34TH has joined
898 2012-08-27 17:00:58 D34TH has quit (Changing host)
899 2012-08-27 17:00:58 D34TH has joined
900 2012-08-27 17:01:40 Bachfischer has quit (Quit: ZNC - http://znc.in)
901 2012-08-27 17:04:47 Joric_ has joined
902 2012-08-27 17:04:47 Joric_ has quit (Changing host)
903 2012-08-27 17:04:47 Joric_ has joined
904 2012-08-27 17:05:15 Diapolo1 has joined
905 2012-08-27 17:05:33 <BlueMatt> kjj_: yea, I think its glue, sorry, just got back to irc...Im gonna ping my upstream dns servers and see if they changed/broke something recently
906 2012-08-27 17:05:42 <gmaxwell> Luke-Jr: it doesn't just ban on it, it's effectively rate limited.
907 2012-08-27 17:06:44 Arnavion3 has quit (Quit: ==kimi ni nante okureba ii no==)
908 2012-08-27 17:07:01 Arnavion has joined
909 2012-08-27 17:07:06 <kjj_> BlueMatt: I usually tell people to find out what their registrar needs to register and list a private nameserver, rather than muck around with delegation lower in the tree
910 2012-08-27 17:07:10 aq83 has joined
911 2012-08-27 17:07:53 pjorrit_ has joined
912 2012-08-27 17:08:07 nhodges has quit (Remote host closed the connection)
913 2012-08-27 17:08:07 mndrix has quit (Remote host closed the connection)
914 2012-08-27 17:08:08 Faraday__ has quit (Remote host closed the connection)
915 2012-08-27 17:08:13 <BlueMatt> its not a registrar glue issue, the ns there is under a different tld, and that should always work, the issue is at the next level
916 2012-08-27 17:08:22 topi`_ has joined
917 2012-08-27 17:08:32 TD has quit (Quit: TD)
918 2012-08-27 17:08:39 <sipa> BlueMatt: could it be related to bitcoin-seeder not implementing DNS correctly?
919 2012-08-27 17:08:49 <BlueMatt> sipa: I dont think so, or your seed would be broken too
920 2012-08-27 17:08:52 <kjj_> BlueMatt: the key word is "should"
921 2012-08-27 17:09:13 <BlueMatt> kjj_: ive had that setup forever, and its never broken before that ive seen ;)
922 2012-08-27 17:09:27 <BlueMatt> (and its always been a dnsseed on that setup and it worked, so...)
923 2012-08-27 17:09:44 <BlueMatt> anyway, looks like they updated their ns software and broke it somewhere, Ill ping them
924 2012-08-27 17:09:47 <kjj_> I used to work phone support for a hosting company. Got calls like this nearly every day, the solution was always to either jump right to delegation from the TLD, or use a DNS service
925 2012-08-27 17:10:03 <BlueMatt> this _is_ a dns service ;)
926 2012-08-27 17:10:18 <kjj_> heh. oops
927 2012-08-27 17:10:42 Faraday__ has joined
928 2012-08-27 17:10:46 t7 has joined
929 2012-08-27 17:11:14 nhodges has joined
930 2012-08-27 17:11:21 coblee_ has joined
931 2012-08-27 17:12:33 coblee_ is now known as coblee
932 2012-08-27 17:12:40 <jgarzik> BlueMatt: maybe its a DNSSEC issue
933 2012-08-27 17:12:56 darkee has quit (!~darkee@gateway/tor-sasl/darkee|Remote host closed the connection)
934 2012-08-27 17:12:56 <jgarzik> BlueMatt: maybe the registrar wants a valid NS chain
935 2012-08-27 17:15:16 MC-Eeepc has joined
936 2012-08-27 17:15:16 `2Fast2BCn has joined
937 2012-08-27 17:15:16 Wack0 has joined
938 2012-08-27 17:15:16 freewil has joined
939 2012-08-27 17:15:56 darkee has joined
940 2012-08-27 17:18:26 <BlueMatt> jgarzik: nope, its a glue issue, problem being the ns for dnsseed.bluematt.me is dnsseedns.... which is a cname, thus the server isnt providing it as glue
941 2012-08-27 17:18:26 <BlueMatt> changed it to a duplicate a and it should now be working, but...still a bug that should be fixed
942 2012-08-27 17:18:26 <BlueMatt> TD, jgarzik wait, IIRC an hour and ping me if its still not working
943 2012-08-27 17:18:38 darkee has joined
944 2012-08-27 17:19:36 <kjj_> ahh, CNAME. Wish those had never been invented
945 2012-08-27 17:22:15 darkee has quit (Ping timeout: 276 seconds)
946 2012-08-27 17:22:21 davout has quit (Remote host closed the connection)
947 2012-08-27 17:22:34 <gavinandresen> Luke-Jr: RE: not banning unrecognized keys: we have to. If we don't, then an attacker just pre-generates a few thousand "other" keys and then floods a victim with them.
948 2012-08-27 17:23:05 nonick has joined
949 2012-08-27 17:23:07 <gavinandresen> Luke-Jr: if you have an alternative implementation with its own alert system, then you should only send alerts to peers that are of that implementation
950 2012-08-27 17:23:13 maaku has quit (Quit: maaku)
951 2012-08-27 17:23:22 <kjj_> gavinandresen: could always sign their keys with your keys, at least for now
952 2012-08-27 17:23:29 <gmaxwell> 0_o
953 2012-08-27 17:23:34 <gmaxwell> "Thats not how it works"
954 2012-08-27 17:24:05 <Luke-Jr> gavinandresen: I suppose that makes sense.
955 2012-08-27 17:24:15 <gavinandresen> yeah, we're not going to implement a certificate authority system for an alert system we use maybe once a year
956 2012-08-27 17:24:31 <Luke-Jr> wish there was a way to get foreign alerts relayed too, but I see that'd be a bigger problem
957 2012-08-27 17:24:37 <kjj_> heh. I never said it was a good idea to do that, at least not in the long run
958 2012-08-27 17:24:55 <Eliel> how about banning if there's more than N unrecognized keys within a time limit?
959 2012-08-27 17:25:01 <gavinandresen> yes, relaying foreign alerts would be a quick way to lots of really nasty DoS attacks....
960 2012-08-27 17:25:13 <gmaxwell> Eliel: thats what it does with the pull request, more or less.
961 2012-08-27 17:25:18 <gavinandresen> Eliel: that's what the pull does, with N=10 but no timelimit
962 2012-08-27 17:25:24 <gmaxwell> (the time limit is infinite)
963 2012-08-27 17:25:37 <Eliel> then I suggest a time limit.
964 2012-08-27 17:25:57 <Eliel> it really can't be considered a DoS if they're not coming in too fast.
965 2012-08-27 17:26:03 <form_> hi. someone knows a good website which describes how to mine without a pool? just some information needed, but i find nothing...
966 2012-08-27 17:26:09 darkee has quit (!~darkee@gateway/tor-sasl/darkee|Ping timeout: 276 seconds)
967 2012-08-27 17:26:16 <sipa> form_: what's your hashrate?
968 2012-08-27 17:26:25 <gavinandresen> If there is a time limit, then you open up another possible DoS attack because we have to remember when we've got invalid alerts....
969 2012-08-27 17:26:43 <form_> sipa: 800M, but that doesnt matter here. i just wanna know how it works (which servers he talks to, etc.)
970 2012-08-27 17:26:55 <Eliel> hmm? just increase a counter and decrement it at steady intervals if it's been increased.
971 2012-08-27 17:26:58 <sipa> form_: #bitcoin-mining will tell you more
972 2012-08-27 17:26:58 <Luke-Jr> form_: just point it to bitcoind directly
973 2012-08-27 17:27:33 <form_> sipa: will try // Luke-Jr: point who?
974 2012-08-27 17:27:35 <Eliel> you already need a counter to count to N=10.
975 2012-08-27 17:27:53 <gavinandresen> Eliel: we already had that counter, it is the generic "misbehaving" count.
976 2012-08-27 17:28:04 <gavinandresen> ... and it never goes down.
977 2012-08-27 17:28:14 <Luke-Jr> form_: your miner software
978 2012-08-27 17:28:33 <gmaxwell> Forgetting old misbehaving might be useful, but I see that as something seperate.
979 2012-08-27 17:28:37 theorbtwo has joined
980 2012-08-27 17:28:50 <form_> Luke-Jr: i only know phoenix miner etc. they seem to support pools only
981 2012-08-27 17:29:01 <gmaxwell> e.g. decrease misbehaving by 1 per day for every !=0 nmisbheaving node.
982 2012-08-27 17:29:05 <sipa> form_: point it to 127.0.0.1:8332
983 2012-08-27 17:29:12 <gavinandresen> The penalty for misbehaving isn't perma-ban, so misbehavior IS forgotten eventually.
984 2012-08-27 17:29:16 <Luke-Jr> form_: moving discussion to the mining channel..
985 2012-08-27 17:29:18 <sipa> form_: bitcoind implements the same protocol as pools do (more or less)
986 2012-08-27 17:29:42 D34TH has quit (Read error: Connection reset by peer)
987 2012-08-27 17:29:47 <kjj_> ugh. the alternative seems to be that every other client creates their own parallel p2p structure and relays alerts along THAT. or they just do without
988 2012-08-27 17:29:58 D34TH has joined
989 2012-08-27 17:29:59 D34TH has quit (Changing host)
990 2012-08-27 17:29:59 D34TH has joined
991 2012-08-27 17:30:34 <gavinandresen> gmaxwell: I don't like decreasing misbehaving, because it would make it possible to try to mount very-long-running fill-up-memory-or-disk attacks.
992 2012-08-27 17:31:03 <Eliel> maybe you could fix it with a subscribe mechanism? other nodes can subscribe to receive alerts with a certain key.
993 2012-08-27 17:31:03 <gmaxwell> gavinandresen: so does the current forgetting, of course.
994 2012-08-27 17:32:05 <sipa> Eliel: hey, what about a pubsub system!
995 2012-08-27 17:32:07 dvide has joined
996 2012-08-27 17:32:15 <Eliel> sipa: I'm not familiar with that
997 2012-08-27 17:32:25 Diapolo1 has quit (Quit: Leaving.)
998 2012-08-27 17:32:29 <sipa> Eliel: it existed in half-implemented form in bitcoin for a long time
999 2012-08-27 17:32:33 <sipa> but was never finished
1000 2012-08-27 17:32:35 Diapolo has joined
1001 2012-08-27 17:32:38 <Diapolo> is misbehaviour per IP or based on a bitcoin parameter?
1002 2012-08-27 17:32:40 <sipa> and i'm not sure what satoshi intended it for
1003 2012-08-27 17:32:59 <Eliel> sipa: can you give a short description what it does?
1004 2012-08-27 17:33:35 nonick has quit (Remote host closed the connection)
1005 2012-08-27 17:33:44 <sipa> Eliel: according to the comments in the code back then, s system where nodes could publish information and subscribe to it, using a meet-in-the-middle system
1006 2012-08-27 17:33:47 <sipa> that's all i know
1007 2012-08-27 17:34:06 <sipa> it was removed around 0.5 or so
1008 2012-08-27 17:34:15 nonick has joined
1009 2012-08-27 17:34:31 <Eliel> in a generalized form that could be useful in unexpected ways.
1010 2012-08-27 17:35:53 <kjj_> could use UserAgent in the version message so that nodes only relay to other nodes with the same (or similar) agent and not to others
1011 2012-08-27 17:36:03 <wumpus> gavinandresen, BlueMatt: -fstack-protector-all works fine with wine, it's just that the ancient mingw version that we use now messes it up -- it's not safe to enable it
1012 2012-08-27 17:36:17 <Eliel> I wonder if satoshi intended that to be used for sending messages between users.
1013 2012-08-27 17:36:25 <wumpus> with mingw64 (in 32 bit mode) the resulting executable works fine, both in wine and windows
1014 2012-08-27 17:36:38 andrew_wmf has quit (Quit: andrew_wmf)
1015 2012-08-27 17:36:51 <Diapolo> wumpus: the used version is ancient ^^
1016 2012-08-27 17:37:17 <wumpus> yes
1017 2012-08-27 17:37:34 <wumpus> but I mean it was not wine that caused that problem
1018 2012-08-27 17:37:37 ThomasV has joined
1019 2012-08-27 17:38:00 darkee has joined
1020 2012-08-27 17:38:14 D34TH has quit (Read error: Connection reset by peer)
1021 2012-08-27 17:38:25 <Diapolo> wumpus: I'm glad we got the casue sorted out, as I still think our Win exe needs this.
1022 2012-08-27 17:38:28 D34TH has joined
1023 2012-08-27 17:38:28 D34TH has quit (Changing host)
1024 2012-08-27 17:38:29 D34TH has joined
1025 2012-08-27 17:38:38 <wumpus> with gcc 4.4+ it works, it has nothing to do with wine or (probably) even cross compilation, though I don't have a mingw 4.4 cross compiler to test.. at leat with 4.6.3 it's fine
1026 2012-08-27 17:38:45 <kjj_> the problem is that singly-connected nodes like Armory won't get any messages.
1027 2012-08-27 17:39:51 <sipa> does lucid have a mingw64 compiler?
1028 2012-08-27 17:39:57 <wumpus> anyway, 'it crashes in wine' is not acceptable
1029 2012-08-27 17:40:47 <gmaxwell> I absolutely want -fstack-protector-all for win32... but yea, "'it crashes in wine' is not acceptable"
1030 2012-08-27 17:41:06 nonick has quit (Ping timeout: 276 seconds)
1031 2012-08-27 17:41:10 <sipa> is not acceptable for the binary, or is not acceptable for an excuse for not enabling it?
1032 2012-08-27 17:41:22 <wumpus> it's not acceptable for anything
1033 2012-08-27 17:41:23 Joric_ is now known as Joric
1034 2012-08-27 17:41:25 <wumpus> it points to a deeper issue
1035 2012-08-27 17:41:28 <gmaxwell> Not acceptable for the binary.
1036 2012-08-27 17:41:31 darkee has joined
1037 2012-08-27 17:41:33 <gmaxwell> exactly.
1038 2012-08-27 17:41:39 <gmaxwell> I don't trust that binary.
1039 2012-08-27 17:41:47 <Diapolo> wumpus: I only was thinking about some possibilities and older Wine had problems with it, now it's the compiler who causes this ... well ;).
1040 2012-08-27 17:41:56 <sipa> well, if it's caused by a (known) bug in wine, there's little we can do
1041 2012-08-27 17:42:00 <wumpus> we disabled -fstack-protector on windows already once for corruption reasons, that was with the same old mingw compiler
1042 2012-08-27 17:42:03 <sipa> if it's not, obviously it needs fixing
1043 2012-08-27 17:42:09 <wumpus> no,it's *not* caused by a bug in wine
1044 2012-08-27 17:42:35 <gmaxwell> I assume it's being caused by a compiler bug that is being _exposed_ in wine. (e.g. due to different stack layout)
1045 2012-08-27 17:42:36 <Diapolo> sipa: an old GCC version causes it
1046 2012-08-27 17:42:43 <sipa> sure, i know that now, but just saying "it's not acceptable" without context is wrong
1047 2012-08-27 17:42:50 <wumpus> when you build on windows on 4.4 it works fine, when you cross-compile or build on windows with 4.6.x it's fine
1048 2012-08-27 17:42:56 <wumpus> both in wine and windows
1049 2012-08-27 17:42:58 <sipa> if that statement is made within the context of knowing that it's caused by GCC, duh
1050 2012-08-27 17:43:01 <Diapolo> I never used a GCC or minGW version < 4.4, so my local builds always worked
1051 2012-08-27 17:43:06 <gmaxwell> sipa: fair enough, if we knew what bug it was and it was harmless it would be okay.
1052 2012-08-27 17:43:37 <sipa> anyway
1053 2012-08-27 17:43:42 darkee has quit (!~darkee@gateway/tor-sasl/darkee|Ping timeout: 276 seconds)
1054 2012-08-27 17:43:48 <sipa> is it possible to get gcc4.4 builds in gitian?
1055 2012-08-27 17:44:27 <sipa> gcc-mingw32 has version 4.4.2-3 in lucid
1056 2012-08-27 17:44:46 <sipa> isn't that what we use?
1057 2012-08-27 17:44:59 Cryo has joined
1058 2012-08-27 17:45:22 <wumpus> according to bluematt it was 4.2.x
1059 2012-08-27 17:45:25 <Cryo> anyone have an opinion on the butterfly sc asic box, pm me
1060 2012-08-27 17:45:33 <Luke-Jr> â¦
1061 2012-08-27 17:45:44 <sipa> Cryo: not here
1062 2012-08-27 17:46:00 <gmaxwell> Cryo: try #bitcoin-mining or #bitcoin-mining-vaporware-products-which-people-can-only-speculate-about
1063 2012-08-27 17:46:12 * jgarzik envies anyone who can drink and code. I like beer, but my code is even more crappy after a beer or two.
1064 2012-08-27 17:46:26 <Cryo> you guys totally fail on the pm thing. I wasn't trying to disrupt the channel sheesh
1065 2012-08-27 17:46:27 * jgarzik looks longingly at that Corona-with-lime
1066 2012-08-27 17:46:38 <sipa> jgarzik: xkcd 323
1067 2012-08-27 17:46:49 chrisb__ has joined
1068 2012-08-27 17:47:16 andrew_wmf has joined
1069 2012-08-27 17:47:18 mndrix has joined
1070 2012-08-27 17:47:19 mndrix has quit (Changing host)
1071 2012-08-27 17:47:19 mndrix has joined
1072 2012-08-27 17:47:19 mndrix has quit (Changing host)
1073 2012-08-27 17:47:19 mndrix has joined
1074 2012-08-27 17:47:22 <wumpus> ii mingw32 4.2.1.dfsg-2ubuntu1 Minimalist GNU win32 (cross) compiler that's what pulltester is using, at least
1075 2012-08-27 17:47:28 <jgarzik> heh
1076 2012-08-27 17:47:59 <gmaxwell> 4.2? yuck.
1077 2012-08-27 17:48:29 <Diapolo> right ^^ pretty old
1078 2012-08-27 17:48:32 <Luke-Jr> wumpus: pulltester != gitian
1079 2012-08-27 17:48:48 <wumpus> we understand that Luke-Jr
1080 2012-08-27 17:49:27 copumpkin is now known as pirateat41
1081 2012-08-27 17:50:11 <Diapolo> Luke-Jr: it is a great option to verify if code would at least cross compile and allows me to test things with a nearly Gitian like build process
1082 2012-08-27 17:50:41 <wumpus> btw, any specific reason that gitian is bound to ubuntu lucid?
1083 2012-08-27 17:50:47 <jeremias> hmm, is there any estimates when 0.7 would be ready?
1084 2012-08-27 17:50:51 rdponticelli has quit (Ping timeout: 276 seconds)
1085 2012-08-27 17:50:59 <gmaxwell> wumpus: because if we want binaries that work on older systems we have to build against the older abi.
1086 2012-08-27 17:51:02 <Diapolo> wumups: Am I right, that alert messages are currently untranslated, when not pre-defined somehow in the source?
1087 2012-08-27 17:51:08 D34TH_ has joined
1088 2012-08-27 17:51:25 <Diapolo> gmaxwell: compatibility is nice to a certain extend
1089 2012-08-27 17:51:28 <wumpus> alert messages?
1090 2012-08-27 17:51:41 <gmaxwell> Diapolo: alert messages are text recieved over the network.
1091 2012-08-27 17:51:56 <wumpus> gmaxwell: right, for the linux build that makes sense, for the windows build not at all
1092 2012-08-27 17:52:07 pirateat41 is now known as copumpkin
1093 2012-08-27 17:52:10 <sipa> wumpus: good point
1094 2012-08-27 17:52:11 <Luke-Jr> wumpus: devrandom was talking about using Gentoo the other night, to build everything from source, so I imagine it can change easily
1095 2012-08-27 17:52:21 <Diapolo> gmaxwell: Yes, but I saw that Gavin used a fixed string for his last pulls and I'm asking if that can be translated, it should be IMHO.
1096 2012-08-27 17:52:28 <Luke-Jr> gmaxwell: I don't think that applies to Windows binaries :p
1097 2012-08-27 17:52:30 <wumpus> Luke-Jr: oh please don't make it take even longer :p
1098 2012-08-27 17:52:30 <gmaxwell> Diapolo: that is a special case.
1099 2012-08-27 17:52:45 <gavinandresen> yes, a special "we'
1100 2012-08-27 17:52:54 <gavinandresen> "we're going to work hard so it never happens" case
1101 2012-08-27 17:53:38 <Diapolo> I'm hoping the same btw.
1102 2012-08-27 17:53:39 <wumpus> ehm if you translated that string it'd be a very big problem as it has to exactly match what is received from the network
1103 2012-08-27 17:53:47 Chucknasty has joined
1104 2012-08-27 17:53:53 <Luke-Jr> wumpus: I think if he did that, it'd cache results like it does already. :p
1105 2012-08-27 17:53:55 <Diapolo> wumpus: the string is fixed for that borderline-case now :)
1106 2012-08-27 17:53:57 <wumpus> of course, we could make it german for everyone :p
1107 2012-08-27 17:54:10 * sipa votes Swahili
1108 2012-08-27 17:54:11 <gavinandresen> TD suggested having the GUI code check for alert.nId=max_int, and then showing a transalated string....
1109 2012-08-27 17:54:14 <gmaxwell> Chinese, just to make sure half the world can't google it.
1110 2012-08-27 17:54:21 <Diapolo> wumpus: bad idea ^^ some people don't like us that much
1111 2012-08-27 17:54:38 <wumpus> gavinandresen: yes, that's a good option
1112 2012-08-27 17:54:45 D34TH has quit (Ping timeout: 276 seconds)
1113 2012-08-27 17:54:58 <Diapolo> I just thing the most important string in the Bitcoin code ^^ should be translatable.
1114 2012-08-27 17:55:01 <Diapolo> think
1115 2012-08-27 17:55:09 * Luke-Jr wonders how hard it would be to encode multiple translations in an alert
1116 2012-08-27 17:55:14 <wumpus> I certainly don't hope that's the most important string :p
1117 2012-08-27 17:55:21 <Diapolo> :-D
1118 2012-08-27 17:55:33 <sipa> no problems with adding it as a special case in the GUI code and translating that
1119 2012-08-27 17:55:41 <gmaxwell> Diapolo: there is a tradeoff, if its translated its hard to search for. The nature of the message means people will need to search to find out what is going on.
1120 2012-08-27 17:55:42 <wumpus> well it would not be hard, but it'd result in very large messages
1121 2012-08-27 17:56:51 <kjj_> you could use smilies as universal indicators of severity, followed by a link to a page with translations. silly idea, but not as silly as it sounds the first time you read it
1122 2012-08-27 17:57:18 <gmaxwell> kjj_: normally alerts contain a URL which is translated.
1123 2012-08-27 17:57:25 <gmaxwell> This is a special case that can't contain a url.
1124 2012-08-27 17:57:55 <Diapolo> What about alert IDs that people can google?
1125 2012-08-27 17:58:07 <wumpus> kjj_: yes or like the old atari st errors use a number of bombs to indicate severity :p
1126 2012-08-27 17:58:26 <sipa> devrandom: W: Failed to fetch http://10.0.2.2:3142/archive.ubuntu.com/ubuntu/dists/lucid/main/i18n/Translation-en_US.bz2 Unable to connect to 10.0.2.2:3142:
1127 2012-08-27 17:58:28 <wumpus> I think the message is easy enough to google
1128 2012-08-27 17:58:41 <Diapolo> fine :-D then back to GCC
1129 2012-08-27 17:59:13 <sipa> devrandom: wait, perhaps firewall, let me check
1130 2012-08-27 17:59:50 <Diapolo> wumpus: Btw. did you report that Boost bug upstream?
1131 2012-08-27 17:59:58 <wumpus> 0.7.0rc1 time
1132 2012-08-27 18:00:03 <wumpus> Diapolo: no, not yet
1133 2012-08-27 18:01:02 <Diapolo> OMG no new features until > 0.7 final ... *scared*
1134 2012-08-27 18:01:08 <Diapolo> ^^
1135 2012-08-27 18:01:24 <wumpus> heh
1136 2012-08-27 18:02:27 <Luke-Jr> ping me when 0.7rc1 is tagged and I'll run a gitian build..
1137 2012-08-27 18:02:42 <sipa> devrandom: seems that was it
1138 2012-08-27 18:02:42 <Luke-Jr> I suppose I should do boost now
1139 2012-08-27 18:02:54 darkee has joined
1140 2012-08-27 18:03:25 da2ce7_d has joined
1141 2012-08-27 18:03:42 <Diapolo> Btw. why did no one respond to my question where the ASLR and DEP linker-flags need to be put for bitcoind to take advantage of it?
1142 2012-08-27 18:03:57 <gmaxwell> jeremias: "Two weeks" but every moment we don't have the RC out we're slipping that further.
1143 2012-08-27 18:04:04 rdponticelli has joined
1144 2012-08-27 18:04:19 <wumpus> Diapolo: makefile.mingw and makefile.mingw-linux
1145 2012-08-27 18:04:36 <gavinandresen> RE: 0.7.0rc1 : only thing on my TODO lists are waiting for a pull from Matt to use boost 1.50 for the gitian windows builds. And then commit my "bump version numbers to 0.7.0" branch
1146 2012-08-27 18:05:11 <Diapolo> I'm a little sad that BitcoinPullTester had no exe ready for me to check if the Windows version stuff is working...
1147 2012-08-27 18:05:29 <Diapolo> wumpus: thanks I'll take a look there later
1148 2012-08-27 18:05:41 minimoose has joined
1149 2012-08-27 18:05:46 <wumpus> just be patient, pulltester has a lot to do, it will get there eventually
1150 2012-08-27 18:05:48 darkee has quit (Ping timeout: 276 seconds)
1151 2012-08-27 18:05:48 da2ce7 has quit (Ping timeout: 276 seconds)
1152 2012-08-27 18:06:17 <Diapolo> will come back later to check how far the RC1 work progressed :)
1153 2012-08-27 18:06:23 Diapolo has quit (Quit: Leaving.)
1154 2012-08-27 18:07:26 <Luke-Jr> gavinandresen: btw, if you wanted to pull prioritisetransaction, just let me know and I'll rebase it (and/or whatever other changes you need)
1155 2012-08-27 18:08:26 nsh has quit (Ping timeout: 246 seconds)
1156 2012-08-27 18:08:46 maaku has joined
1157 2012-08-27 18:10:30 nonick has joined
1158 2012-08-27 18:12:05 iocor has quit (Quit: Computer has gone to sleep.)
1159 2012-08-27 18:13:36 darkee has quit (!~darkee@gateway/tor-sasl/darkee|Ping timeout: 276 seconds)
1160 2012-08-27 18:14:34 iocor has joined
1161 2012-08-27 18:14:59 TD has joined
1162 2012-08-27 18:16:30 <BlueMatt> TD: does dnsseed.bluematt.me work now?
1163 2012-08-27 18:16:37 <BlueMatt> or jgarzik
1164 2012-08-27 18:16:58 rdponticelli has quit (Ping timeout: 260 seconds)
1165 2012-08-27 18:17:03 <kjj_> works for me now
1166 2012-08-27 18:17:17 rdponticelli has joined
1167 2012-08-27 18:17:28 <jgarzik> BlueMatt: yes, working. just tried two hosts on two separate networks.
1168 2012-08-27 18:17:28 <BlueMatt> thanks
1169 2012-08-27 18:17:43 <BlueMatt> thanks
1170 2012-08-27 18:19:42 <sipa> works here too
1171 2012-08-27 18:19:52 Erdon has quit (Ping timeout: 244 seconds)
1172 2012-08-27 18:21:38 pnicholson has joined
1173 2012-08-27 18:23:25 molecular has quit (Ping timeout: 252 seconds)
1174 2012-08-27 18:23:49 molecular has joined
1175 2012-08-27 18:24:25 <midnightmagic> jgarzik, TD: thanks for clarifying I appreciate that. it helps a lot.
1176 2012-08-27 18:25:24 <Luke-Jr> BlueMatt: testing 1.50
1177 2012-08-27 18:26:36 Diablo-D3 has joined
1178 2012-08-27 18:29:35 <sipa> ok, what do i need to gitian-(re)build?
1179 2012-08-27 18:29:48 <BlueMatt> boost
1180 2012-08-27 18:29:51 <BlueMatt> I dont think qt
1181 2012-08-27 18:30:15 <sipa> current master is ok, or merge that pull first?
1182 2012-08-27 18:30:32 <BlueMatt> merge first
1183 2012-08-27 18:32:45 nsh has joined
1184 2012-08-27 18:33:35 iocor has quit (Quit: Computer has gone to sleep.)
1185 2012-08-27 18:33:55 <Luke-Jr> wow, gavin and wumpus test fast :o
1186 2012-08-27 18:34:10 <wumpus> I've already tested boost 1.50 + bitcoin yesterday
1187 2012-08-27 18:34:20 nsh has quit (Remote host closed the connection)
1188 2012-08-27 18:34:41 nsh has joined
1189 2012-08-27 18:34:48 nsh has quit (Changing host)
1190 2012-08-27 18:34:48 nsh has joined
1191 2012-08-27 18:34:51 <gavinandresen> ok, now to fetch boost 1.50 and gitian-build it....
1192 2012-08-27 18:35:02 <gavinandresen> (if I can remember how)
1193 2012-08-27 18:35:08 * Luke-Jr facepalms :p
1194 2012-08-27 18:35:19 Detritus has joined
1195 2012-08-27 18:35:39 <sipa> eh, weren't there some +- ready pulls left?
1196 2012-08-27 18:36:02 <sipa> nothing important, i guess
1197 2012-08-27 18:36:13 <wumpus> can wait for rc2 I guess
1198 2012-08-27 18:36:19 <Luke-Jr> doc/release-process.txt is 1.47 still >_<
1199 2012-08-27 18:36:54 <sipa> so, we first gitian-build boost now?
1200 2012-08-27 18:37:11 <wumpus> I'm going to gitian-build everything just in case
1201 2012-08-27 18:37:34 <sipa> i'm using lxc now, i hope that doesn't affect the determinism
1202 2012-08-27 18:38:14 <gavinandresen> boost-win32-1.50.0-gitian.zip is an input to the win32 build, so have to build that first....
1203 2012-08-27 18:38:58 <sipa> $ bin/gbuild ../bitcoin-build/contrib/gitian-descriptors/boost-win32.yml -c bitcoin=03fa53a640
1204 2012-08-27 18:39:01 <sipa> /usr/lib/ruby/1.8/yaml.rb:133:in `load': syntax error on line 21, col 0: `@@ -19,9 +19,9 @@' (ArgumentError)
1205 2012-08-27 18:39:18 <Luke-Jr> sipa: did you pull bitcoin-build?
1206 2012-08-27 18:39:25 <sipa> errrr
1207 2012-08-27 18:39:46 <sipa> i fetched, but didn't switch...
1208 2012-08-27 18:42:04 bonks has joined
1209 2012-08-27 18:42:36 <bonks> Hey are there any tools to that will trace one address to another?
1210 2012-08-27 18:43:08 <bonks> I want to know if coins I have originated from a specific address.
1211 2012-08-27 18:43:12 O2made has joined
1212 2012-08-27 18:43:57 <gmaxwell> bonks: it would be easy to write an external tool for bitcoin 0.7.0+ that used the RPC to give you a list of ancestors to a transaction.
1213 2012-08-27 18:44:18 nsh has quit (Ping timeout: 264 seconds)
1214 2012-08-27 18:44:57 <kjj_> bonks: you can do it the hard way by clicking back on block explorer
1215 2012-08-27 18:44:57 <Luke-Jr> gavinandresen: why did alert.GetHash's implementation change?
1216 2012-08-27 18:44:59 Chucknasty has quit (Read error: Connection reset by peer)
1217 2012-08-27 18:45:01 <bonks> gmaxwell: that would be nice, though I don't install beta
1218 2012-08-27 18:45:29 <jgarzik> gavinandresen: I stuck some test in doc/release-notes.txt, but that wants updating for any changes >= August 1
1219 2012-08-27 18:45:34 <bonks> kjj_: lol I tried and gave up once there were too many addresses. I was hoping someone already created a web tool of some sort
1220 2012-08-27 18:45:59 <gmaxwell> bonks: then you're not running bitcoin at all then?
1221 2012-08-27 18:46:26 <bonks> gmaxwell: I mean unreleased :P
1222 2012-08-27 18:46:46 <gavinandresen> Luke-Jr: to prevent an attacker from keeping your CPU busy checking slightly-different-signature-versions of old alerts.
1223 2012-08-27 18:47:10 slush has joined
1224 2012-08-27 18:48:32 <jgarzik> gavinandresen: just delete doc/release-notes.txt after posting
1225 2012-08-27 18:49:25 <Luke-Jr> gavinandresen: won't this mean, that you need different messages for each alert (eg, same problem affecting multiple version series)? does it affect cancellation relays?
1226 2012-08-27 18:50:08 <gavinandresen> Luke-Jr: no, the hash is computed from all of the data in CUnsignedAlert, so affects-different-versions are different hashes.
1227 2012-08-27 18:50:52 <gavinandresen> Luke-Jr: only the signature isn't part of the hash. I wish transaction IDs had been hashed that way, too, by the way...
1228 2012-08-27 18:50:59 <Luke-Jr> ah, ok
1229 2012-08-27 18:51:19 <Luke-Jr> gavinandresen: so the transaction id is known before signing? yeah, that'd be convenient
1230 2012-08-27 18:51:39 <gavinandresen> yes, and if you signed a transaction twice it'd have the same id....
1231 2012-08-27 18:52:24 <sipa> and you would know the txid while it was only signed partially
1232 2012-08-27 18:52:34 <sipa> and nobody could change the txid by modifying the signature
1233 2012-08-27 18:52:42 iocor has joined
1234 2012-08-27 18:52:44 <Luke-Jr> gavinandresen: sounds like a new DoS attack there ;)
1235 2012-08-27 18:53:26 <gavinandresen> Luke-Jr: should be just the opposite, if you get one validly-signed version of the transaction you could ignore all the rest.
1236 2012-08-27 18:53:31 <sipa> d25ddb68061c11ee0a3fc40d3e6dec2b7b5216ffc3ba0b4de947a7fb04a3900e boost-win32-1.50.0-gitian2.zip
1237 2012-08-27 18:53:34 <sipa> d3f5b2540f7f4b25412a779ca1affc24fd449d9b7dc00444082eb5de90a71301 boost-res.yml
1238 2012-08-27 18:53:40 <wumpus> dd19aff5904280225ca88daf41aeb0c614f131bf1e490078d1f02ef70f0bf606 boost-win32-1.50.0-gitian2.zip
1239 2012-08-27 18:53:56 <Luke-Jr> gavinandresen: but you won't necessarily re-fetch the txn if you've already rejected it as non-standard due to too many sigs
1240 2012-08-27 18:54:19 <sipa> Luke-Jr: nothing says that the number of sigs couldn't be hashed
1241 2012-08-27 18:55:07 iocor has quit (Client Quit)
1242 2012-08-27 18:55:10 <Luke-Jr> sipa: ok, then nastyNode will corrupt your sig to make it invalid⦠he gets DoS-banned, but you lose the txn :P
1243 2012-08-27 18:55:16 slush1 has joined
1244 2012-08-27 18:55:23 <sipa> good point
1245 2012-08-27 18:55:26 <sipa> there are issues involved
1246 2012-08-27 18:55:51 <gavinandresen> yeah, would require lots of Deep Thought.
1247 2012-08-27 18:55:56 <sipa> 42.
1248 2012-08-27 18:55:56 pnicholson has quit (Read error: Connection reset by peer)
1249 2012-08-27 18:56:39 <gavinandresen> Did boost-win32-*gitian.zip build deterministically before? I remember having issues with one of the dependencies....
1250 2012-08-27 18:56:51 pnicholson has joined
1251 2012-08-27 18:57:06 <sipa> gavinandresen: maybe it doesn't, but doesn't influence the rest
1252 2012-08-27 18:57:24 <wumpus> maybe the zip timestamps or so...
1253 2012-08-27 18:57:52 <wumpus> or timestamps within the .a files
1254 2012-08-27 18:57:55 <gavinandresen> bah, I accidently re-built boost-win32-1.47.....
1255 2012-08-27 18:59:18 <sipa> do we need to rebuild win32-deps?
1256 2012-08-27 18:59:33 <jgarzik> gavinandresen: I'll poke at release-notes.txt and add some more stuff
1257 2012-08-27 18:59:40 <gavinandresen> jgarzik: thanks
1258 2012-08-27 19:00:38 <wumpus> b9edb45129d44fb78d34986c4e24965f4fa9d4bb56b96390e045d95d113d59a5 bitcoin-deps-0.0.4.zip
1259 2012-08-27 19:01:17 <BlueMatt> gavinandresen: re: changing hashes to ignore sigs: easily doable for wallets (which is the main dos-ish vector for such attacks now), obviously you would have to take it further and ignore all script data chunks that are ignore is the hash tx for sig stuff
1260 2012-08-27 19:01:45 <BlueMatt> it would have to be wallet-only and done after (and calculated as a part of) script execution
1261 2012-08-27 19:05:01 <Luke-Jr> gavinandresen: re Special-case the last alert for alert-key-compromised case⦠wouldn't it make sense to just trigger that last-alert case if any alert passed sig checks with the id?
1262 2012-08-27 19:05:21 <Luke-Jr> so if someone tried to do something funny, it wouldn't require someone to send a real alert
1263 2012-08-27 19:05:50 <gmaxwell> sipa: do you have a proposed patch for the signature non-standard stuff yet; I know we can't deploy it until at least one high profile user removes their cranial rectal inversion, but it would probably be good to get it in the pipeline.
1264 2012-08-27 19:06:25 <sipa> gmaxwell: yes, i have two implementations even (one openssl, one custom-written, and they agree entirely)
1265 2012-08-27 19:06:39 davout has joined
1266 2012-08-27 19:07:16 rcorreia has joined
1267 2012-08-27 19:07:23 nonick has quit (Remote host closed the connection)
1268 2012-08-27 19:07:30 <gavinandresen> Luke-Jr: you mean: if nId == max then create a hardcoded, unsigned alert and pass that up to higher levels to process? Yeah, but those higher levels might re-check signatures, etc. It was easier to require that the hardcoded alert be the one sent.
1269 2012-08-27 19:07:57 <wumpus> now I'm sure, gitian uses i586-mingw32msvc-g++ (GCC) 4.2.1-sjlj (mingw32-2)
1270 2012-08-27 19:08:44 nonick has joined
1271 2012-08-27 19:09:08 <sipa> wumpus: indeed
1272 2012-08-27 19:09:49 oneb has quit (Read error: Connection reset by peer)
1273 2012-08-27 19:10:18 <gavinandresen> fricking fracking.... my gitian machine just decided to die.
1274 2012-08-27 19:10:36 <sipa> LXC seems to work fine here now
1275 2012-08-27 19:10:47 <sipa> almost said LHC
1276 2012-08-27 19:10:50 <wumpus> eek
1277 2012-08-27 19:10:55 nonick has quit (Remote host closed the connection)
1278 2012-08-27 19:11:43 nonick has joined
1279 2012-08-27 19:12:49 Joric has quit ()
1280 2012-08-27 19:13:17 oneb has joined
1281 2012-08-27 19:13:59 <jgarzik> gavinandresen: updated. Qt still has a big "???" in the UI section of doc/release-notes.txt
1282 2012-08-27 19:14:12 <jgarzik> wumpus: can you update ^^ ?
1283 2012-08-27 19:14:17 <sipa> when building deps:
1284 2012-08-27 19:14:18 <slush> gmaxwell: seems that blockexplorer handled that reorg correctly. You were right, reorg triggered after mining of next block
1285 2012-08-27 19:14:21 <sipa> ./wingenminiupnpcstrings miniupnpcstrings.h.in miniupnpcstrings.h
1286 2012-08-27 19:14:21 <sipa> make: ./wingenminiupnpcstrings: Command not found
1287 2012-08-27 19:15:01 <BlueMatt> you dont have wine or thats missing a .exe, not sure which
1288 2012-08-27 19:15:09 <BlueMatt> or when you installed wine, you didnt get binfmt-support
1289 2012-08-27 19:16:16 <sipa> i don't install anything
1290 2012-08-27 19:16:18 <sipa> this is gitian
1291 2012-08-27 19:16:47 <wumpus> /home/ubuntu/build/boost_1_50_0/stage/lib/libboost_thread_win32-mt-s.a(thread.o):thread.cpp:(.text+0x2f7e): undefined reference to `boost::chrono::system_clock::now()'
1292 2012-08-27 19:16:47 <wumpus> collect2: ld returned 1 exit status
1293 2012-08-27 19:17:17 <sipa> wumpus: where's that?
1294 2012-08-27 19:17:18 nonick has quit (Ping timeout: 276 seconds)
1295 2012-08-27 19:17:46 <Luke-Jr> sigh
1296 2012-08-27 19:18:01 <wumpus> it needs LIBS += -lboost_chrono$$BOOST_LIB_SUFFIX
1297 2012-08-27 19:18:13 <sipa> BlueMatt: seems that it builds wingenminiupnpcstrings.exe
1298 2012-08-27 19:18:24 <sipa> even when the host env is ubuntu
1299 2012-08-27 19:18:28 <wumpus> sipa: at the end of building bitcoin-qt :/
1300 2012-08-27 19:19:14 <BlueMatt> sipa: sounds right, not sure why its missing the exe, is that broken for anyone else? also, why are you re-building the deps?
1301 2012-08-27 19:19:36 <wumpus> deps built fine for me
1302 2012-08-27 19:20:02 datagutt has quit (Quit: Computer has gone to sleep.)
1303 2012-08-27 19:20:14 <jgarzik> There are two types of people in this world: those who can extrapolate from incomplete data
1304 2012-08-27 19:20:18 <sipa> BlueMatt: it builds a windows executable
1305 2012-08-27 19:20:26 <sipa> and then tries to run it as a linux on
1306 2012-08-27 19:20:27 <sipa> e
1307 2012-08-27 19:20:44 <wumpus> but bitcoin-qt and bitcoind build fails
1308 2012-08-27 19:20:51 <BlueMatt> sipa: its building for windows, no?
1309 2012-08-27 19:21:09 <sipa> BlueMatt: yes
1310 2012-08-27 19:21:30 <BlueMatt> so it makes some sense
1311 2012-08-27 19:21:35 <sipa> i586-mingw32msvc-gcc wingenminiupnpcstrings.o -lws2_32 -liphlpapi -o wingenminiupnpcstrings
1312 2012-08-27 19:21:38 <sipa> ./wingenminiupnpcstrings miniupnpcstrings.h.in miniupnpcstrings.h
1313 2012-08-27 19:21:40 <sipa> make: ./wingenminiupnpcstrings: Command not found
1314 2012-08-27 19:21:58 <BlueMatt> sipa: I thought some patch to makefile or something fixed that, dunno, cant remember exactly
1315 2012-08-27 19:22:28 <sipa> anyway, if i don't need to rebuild deps, i won't :)
1316 2012-08-27 19:23:23 BurtyBB has left ()
1317 2012-08-27 19:23:33 BurtyB has joined
1318 2012-08-27 19:25:05 <wumpus> but we really need to add the boost_chrono to the windows builds
1319 2012-08-27 19:25:29 mtve has quit (Ping timeout: 244 seconds)
1320 2012-08-27 19:25:39 <sipa> linux builds work fine, i think
1321 2012-08-27 19:25:47 <wumpus> yes
1322 2012-08-27 19:25:52 mtve has joined
1323 2012-08-27 19:26:08 <wumpus> linux build didn't change boost version afaik
1324 2012-08-27 19:27:51 <sipa> indeed
1325 2012-08-27 19:28:16 <sipa> same error at end of windows qt build
1326 2012-08-27 19:28:21 * sipa -> food
1327 2012-08-27 19:28:23 * BlueMatt nominates the two how can push without having to pull-request
1328 2012-08-27 19:28:37 BitByBit has joined
1329 2012-08-27 19:28:48 <wumpus> yes but what about the version tag?
1330 2012-08-27 19:29:00 <sipa> gavinandresen: can you re-tag 0.7.0rc1, or do we immediately do a rc2? :)
1331 2012-08-27 19:29:02 <BlueMatt> meh
1332 2012-08-27 19:29:15 <sipa> (actually, first fix, then build, and THEN tag should be safer...)
1333 2012-08-27 19:29:17 BitByBit has quit (Client Quit)
1334 2012-08-27 19:29:21 <gmaxwell> Tags can be deleted.
1335 2012-08-27 19:29:25 <Luke-Jr> gmaxwell: barely
1336 2012-08-27 19:29:31 <jgarzik> tags can be updated
1337 2012-08-27 19:29:36 <jgarzik> -f
1338 2012-08-27 19:29:43 <sipa> sure they can
1339 2012-08-27 19:29:48 <sipa> the question is whether they should
1340 2012-08-27 19:29:49 <Luke-Jr> jgarzik: but I don't think git will pull updated tags without manual intervention?
1341 2012-08-27 19:30:10 <jgarzik> if your pull normal fetches tags, yes, it will.
1342 2012-08-27 19:30:19 <jgarzik> an updated tag does not change pull behavior.
1343 2012-08-27 19:30:30 <BlueMatt> in other words do it quick so that no one has to pull -f
1344 2012-08-27 19:30:55 <jgarzik> no one needs pull -f
1345 2012-08-27 19:30:56 mmoya_ has quit (Ping timeout: 246 seconds)
1346 2012-08-27 19:31:00 <jgarzik> the push needs -f
1347 2012-08-27 19:31:54 BitByBit has joined
1348 2012-08-27 19:31:58 davout has quit (Remote host closed the connection)
1349 2012-08-27 19:32:22 <wumpus> I made the changes, but rather first test
1350 2012-08-27 19:33:26 <jgarzik> maybe standard practice should be tagging 0.foo-rc1-test1, then test builds, then tag -rc1 ;p
1351 2012-08-27 19:33:43 <wumpus> https://github.com/laanwj/bitcoin master for anyone that wants to try
1352 2012-08-27 19:34:19 <wumpus> linux gitian build is still running here...
1353 2012-08-27 19:34:25 <Luke-Jr> jgarzik: meh, RC of a RC is silly
1354 2012-08-27 19:34:55 <jgarzik> but apparently necessary
1355 2012-08-27 19:35:00 <jgarzik> if tags want updating
1356 2012-08-27 19:35:04 <sipa> jgarzik: or just try building before tagging...
1357 2012-08-27 19:35:12 <Luke-Jr> ^
1358 2012-08-27 19:35:32 <Luke-Jr> or better yet, before pulling ;)
1359 2012-08-27 19:35:42 <sipa> haha, true
1360 2012-08-27 19:36:29 <BlueMatt> or...even better, let someone who has been paying more attention than me do the last-minute pulls
1361 2012-08-27 19:37:24 mcorlette is now known as mcorlett
1362 2012-08-27 19:38:19 <Luke-Jr> BlueMatt: not sure anyone else had the motivation to try bumping to 1.50 though
1363 2012-08-27 19:41:00 <gavinandresen> well, my gitian machine's hard disk decided to die... let me know when it is safe and I will re-tag.
1364 2012-08-27 19:41:31 <sipa> wumpus: if you give me half an hour, i'll test your branch
1365 2012-08-27 19:41:41 <wumpus> I'm currently building it too
1366 2012-08-27 19:42:20 <gavinandresen> I deleted the v0.7.0rc1 tag at github
1367 2012-08-27 19:42:27 <sipa> good
1368 2012-08-27 19:42:29 drazak_ has quit (Ping timeout: 260 seconds)
1369 2012-08-27 19:42:30 <Luke-Jr> wumpus: I thought chrono was only needed on Windows (it's general in .pro)
1370 2012-08-27 19:42:38 <wumpus> somehow my linux build broke off (with a timeout, must be a local problem)
1371 2012-08-27 19:43:17 <wumpus> aww crap
1372 2012-08-27 19:43:26 <wumpus> nah this is testing only anyway
1373 2012-08-27 19:43:51 <sipa> gavinandresen: also, before we used the 4th digit in the version jumber for rc's; not strictly necessary but maybe nice for distingishing?
1374 2012-08-27 19:43:55 <wumpus> seems my linux boost doesn't even have chrono
1375 2012-08-27 19:44:31 <wumpus> seems it fixed the windows build at least...
1376 2012-08-27 19:44:59 <wumpus> next to make it windows-specific...
1377 2012-08-27 19:45:18 <sipa> i haven't started building
1378 2012-08-27 19:45:32 <Luke-Jr> sipa: only when there were security issues to distinguish IIRC?
1379 2012-08-27 19:45:42 <sipa> Luke-Jr: true
1380 2012-08-27 19:46:13 <sipa> i wondering whether it was considered unnecessary or just forgotten
1381 2012-08-27 19:46:22 <sipa> i don't care either way
1382 2012-08-27 19:47:08 paul_nicholson has joined
1383 2012-08-27 19:47:32 mmoya has joined
1384 2012-08-27 19:47:46 <Luke-Jr> sipa: well, it bumps the 4th-digit revisions if those are necessary, so I'd prefer avoiding it without good reason
1385 2012-08-27 19:47:57 <Luke-Jr> (eg, the first 0.6.0.x was 0.6.0.7)
1386 2012-08-27 19:48:13 pnicholson has quit (Read error: Connection reset by peer)
1387 2012-08-27 19:48:18 <wumpus> ok pushed a new version, now adds chrono only on windows
1388 2012-08-27 19:50:51 Diapolo has joined
1389 2012-08-27 19:51:30 <sipa> wumpus: gitian building fd0460b for windows & linux
1390 2012-08-27 19:51:38 Marf has joined
1391 2012-08-27 19:51:46 <wumpus> yes that's the right one, thanks
1392 2012-08-27 19:54:35 <wumpus> shit, I've used WINDOWS:LIBS instead of windows:LIBS
1393 2012-08-27 19:54:38 <wumpus> seems not to work :(
1394 2012-08-27 19:54:54 <Luke-Jr> case sensitivity fail :/
1395 2012-08-27 19:55:14 <wumpus> pushed 74fb2e0
1396 2012-08-27 19:55:37 <wumpus> @sipa
1397 2012-08-27 19:56:40 <sipa> wumpus: restarted
1398 2012-08-27 19:57:53 rdponticelli has quit (Read error: Connection reset by peer)
1399 2012-08-27 19:58:06 rdponticelli has joined
1400 2012-08-27 19:58:07 tower has quit (Disconnected by services)
1401 2012-08-27 19:58:21 tower has joined
1402 2012-08-27 19:58:55 <Diapolo> hey guys, what'u up here?
1403 2012-08-27 19:59:18 <Luke-Jr> Diapolo: [19:27:54] <wumpus> https://github.com/laanwj/bitcoin master for anyone that wants to try
1404 2012-08-27 19:59:23 <wumpus> Diapolo: boost 1.50 upgrade was a bit more work than expected :p
1405 2012-08-27 19:59:38 <wumpus> we forgot to add the chrono lib
1406 2012-08-27 20:00:35 <BlueMatt> s/we/BlueMatt/
1407 2012-08-27 20:00:41 <Diapolo> LIBS += -lboost_system$$BOOST_LIB_SUFFIX -lboost_filesystem$$BOOST_LIB_SUFFIX -lboost_program_options$$BOOST_LIB_SUFFIX -lboost_thread$$BOOST_THREAD_LIB_SUFFIX -lboost_chrono$$BOOST_LIB_SUFFIX
1408 2012-08-27 20:01:00 <Luke-Jr> my boost 1.50 gitian build finished: f76cd68e5
1409 2012-08-27 20:01:10 <Luke-Jr> Diapolo: there is no chrono on Linux :P
1410 2012-08-27 20:01:17 <Diapolo> ahhhh
1411 2012-08-27 20:01:35 <wumpus> well, I didn't think about it either BlueMatt, even though I added it many times already for my own tests with boost 1.51
1412 2012-08-27 20:02:04 * Luke-Jr somehow got confused and assumed it was only needed in 1.51 <.<
1413 2012-08-27 20:02:10 <wumpus> yes me too
1414 2012-08-27 20:02:45 <Diapolo> I know I had to add it with 1.5 on Windows.
1415 2012-08-27 20:02:47 <Diablo-D3> wumpus: in during boost hate ;)
1416 2012-08-27 20:03:22 <Diapolo> wumpus: Why did you add another line for Chrono? Above we have a windows:LIBS?
1417 2012-08-27 20:03:37 <wumpus> because boost-thread needs chrono
1418 2012-08-27 20:03:43 <wumpus> so if you put it above it won't work
1419 2012-08-27 20:04:28 <Diapolo> does a linker work from last to first?
1420 2012-08-27 20:04:40 <wumpus> ok. build was succesful now
1421 2012-08-27 20:05:08 <wumpus> no, it builds a list of needed symbols from left to right
1422 2012-08-27 20:06:51 <Diapolo> so if a lib needs relies on another it has to come AFTERWARDS in order? sorry if that is a dumb question
1423 2012-08-27 20:07:05 t7 has quit (Remote host closed the connection)
1424 2012-08-27 20:07:43 <Diablo-D3> erm
1425 2012-08-27 20:07:47 <Diablo-D3> linkers dont work from first to last
1426 2012-08-27 20:07:56 <Diablo-D3> they search all libs for symbol resolving
1427 2012-08-27 20:08:02 <wumpus> boost-thread needs boost::chrono::system_clock::now() .. the linker remembers that, and checks for the symbol in the libraries after that
1428 2012-08-27 20:08:08 <Diablo-D3> if boost has managed to do that, they found a bug in ld
1429 2012-08-27 20:08:42 <wumpus> only when dynamic linking...
1430 2012-08-27 20:08:51 <wumpus> when static linking the order is really important
1431 2012-08-27 20:08:56 <Diablo-D3> wumpus: nope.
1432 2012-08-27 20:09:01 <Diablo-D3> ld still solves it correctly
1433 2012-08-27 20:09:09 <Diablo-D3> you might be talking about non-gnu compilers
1434 2012-08-27 20:09:17 [\\\] has joined
1435 2012-08-27 20:09:17 <Diablo-D3> which I couldnt give a shit about
1436 2012-08-27 20:09:23 <sipa> i've ceertainly seen cases where the order of arguments to the linker mattered
1437 2012-08-27 20:09:32 <sipa> and that was gcc
1438 2012-08-27 20:09:33 <Diablo-D3> order of arguments yes
1439 2012-08-27 20:09:37 <Diablo-D3> order of libraries no
1440 2012-08-27 20:09:43 <wumpus> yeah, I've noticed that many times, with gcc
1441 2012-08-27 20:09:54 <Diablo-D3> my own library lugh does order specific shit
1442 2012-08-27 20:09:57 <Diapolo> I don'
1443 2012-08-27 20:09:58 <Diablo-D3> but its with non-library arguments
1444 2012-08-27 20:10:05 <Diablo-D3> it switches static on then back off
1445 2012-08-27 20:10:08 <Diapolo> t get it ^^ will try with my local build to understand
1446 2012-08-27 20:10:08 BitByBit has quit (Ping timeout: 260 seconds)
1447 2012-08-27 20:10:13 * Luke-Jr notes all that matters is that wumpus's fix works.
1448 2012-08-27 20:10:14 <wumpus> even with dynamic linking the order can matter, when --as-needed is used
1449 2012-08-27 20:10:24 <Diablo-D3> Luke-Jr: fucked up makefiles are bad.
1450 2012-08-27 20:10:32 <Diablo-D3> wumpus: if boost is using as needed, its broken
1451 2012-08-27 20:10:38 <Diablo-D3> then again, its boost
1452 2012-08-27 20:10:40 <Diablo-D3> boost IS broken
1453 2012-08-27 20:10:43 <wumpus> ubuntu defaults to as-needed
1454 2012-08-27 20:10:46 <Luke-Jr> everything should use --as-needed
1455 2012-08-27 20:10:47 <wumpus> has nothing to do with boost
1456 2012-08-27 20:10:48 <Diablo-D3> ubunti is broken
1457 2012-08-27 20:10:57 <Diablo-D3> people need to stop using that distro
1458 2012-08-27 20:11:00 <wumpus> sigh
1459 2012-08-27 20:11:04 <Diablo-D3> its anti-user friendly, its bloated, its microsoft funded
1460 2012-08-27 20:11:06 <Luke-Jr> every major distro is using --as-needed
1461 2012-08-27 20:11:12 <wumpus> ok let's go back to the practical world
1462 2012-08-27 20:11:17 <wumpus> seems we have a working fix now
1463 2012-08-27 20:11:21 <Diablo-D3> and any bugs on my software filed from ubuntu users are closed immediately as wontfix.
1464 2012-08-27 20:11:27 <wumpus> how to get it to upstream and re-tag?
1465 2012-08-27 20:11:29 <Luke-Jr> http://fedoraproject.org/wiki/UnderstandingDSOLinkChange
1466 2012-08-27 20:11:30 <Luke-Jr> http://wiki.debian.org/ToolChain/DSOLinking
1467 2012-08-27 20:11:41 <Luke-Jr> http://www.gentoo.org/proj/en/qa/asneeded.xml
1468 2012-08-27 20:11:48 <Diablo-D3> Luke-Jr: your trolling doesnt work on me
1469 2012-08-27 20:11:50 <Diapolo> How got the tabs into the contains(USE_IPV6, -) { part in the project-file?
1470 2012-08-27 20:12:05 <Diablo-D3> actually, how the fuck did you get off my ignore
1471 2012-08-27 20:12:17 <c_k> Diablo-D3: try to be constructive ;)
1472 2012-08-27 20:12:20 <Diablo-D3> stop ignore evading or Ill have jgarzik and gmaxwell ban you from all the relevant channels
1473 2012-08-27 20:12:24 <c_k> of course people are going to use Ubuntu
1474 2012-08-27 20:12:31 <Luke-Jr> lol
1475 2012-08-27 20:12:33 <wumpus> Diapolo: try git blame
1476 2012-08-27 20:12:38 <Diablo-D3> c_k: yes, but I dont have to waste my time supporting them
1477 2012-08-27 20:12:47 <wumpus> you don't have t support anyone
1478 2012-08-27 20:12:55 <c_k> Diablo-D3: for your miner?
1479 2012-08-27 20:13:02 <Diablo-D3> c_k: I write other software you know
1480 2012-08-27 20:13:03 <sipa> c_k: you're trying to use reason against Diablo-D3?
1481 2012-08-27 20:13:22 <c_k> Diablo-D3: no, I didn't
1482 2012-08-27 20:13:33 <Diablo-D3> _heh_
1483 2012-08-27 20:13:41 <Diablo-D3> you know whats going to be on my tombstone?
1484 2012-08-27 20:13:48 <Luke-Jr> on Diapolo's complaint, would anyone object to a pullreq adding a .kateconfig to the git repo to make Kate use spaces instead of tabs? :p
1485 2012-08-27 20:13:55 <Diablo-D3> "author of diablominer" and in smaller print below it "first man to walk on mars"
1486 2012-08-27 20:14:21 <sipa> "and master of wishful thinking"
1487 2012-08-27 20:14:34 <Diablo-D3> sipa: phd.
1488 2012-08-27 20:14:45 <sipa> deal
1489 2012-08-27 20:14:54 <wumpus> so, when are we going to launch him to mars, single way trip :-)
1490 2012-08-27 20:14:59 <Diapolo> wumpus: your patch for Boost Chrono is correct even on Windows, when I put it before Thread it can't resolve the dependecies
1491 2012-08-27 20:15:01 <Luke-Jr> wumpus: lol
1492 2012-08-27 20:15:08 BitByBit has joined
1493 2012-08-27 20:15:20 <Diablo-D3> wumpus: that didnt mean what you thought it did
1494 2012-08-27 20:15:22 <wumpus> Diapolo: thanks for testing
1495 2012-08-27 20:15:25 <Diapolo> Luke-Jr: What the f... is Kate?
1496 2012-08-27 20:15:33 <Diablo-D3> a _lot_ of people would volunteer for a one way mission
1497 2012-08-27 20:15:52 <wumpus> you didn't say the tombstone had to be on earth
1498 2012-08-27 20:15:54 <Luke-Jr> Diapolo: code editor ftw
1499 2012-08-27 20:16:06 <sipa> Diablo-D3: a text editor
1500 2012-08-27 20:16:09 <sipa> eh
1501 2012-08-27 20:16:11 <sipa> Diapolo:
1502 2012-08-27 20:16:13 <Luke-Jr> http://lmgtfy.com/?q=kate
1503 2012-08-27 20:16:13 <Diablo-D3> tabfail
1504 2012-08-27 20:16:16 <Diablo-D3> and who the fuck uses kate
1505 2012-08-27 20:16:17 <Diapolo> I'm that one Windows guy here ^^ sorry.
1506 2012-08-27 20:16:20 <Diablo-D3> VIM NIGGAHS
1507 2012-08-27 20:16:26 <Luke-Jr> Diapolo: it supports Windows
1508 2012-08-27 20:16:30 <gjs278> kate
1509 2012-08-27 20:16:31 <gjs278> lol
1510 2012-08-27 20:16:34 <Diablo-D3> I'd flash gang signs, but thats more of a emacs thing
1511 2012-08-27 20:16:34 <Diapolo> I use Qt Creator and love it
1512 2012-08-27 20:16:38 <BitByBit> VIM all the way
1513 2012-08-27 20:16:45 <Diablo-D3> Diapolo: kate is kde notepad
1514 2012-08-27 20:16:53 <gjs278> leafpad is gnome notepad
1515 2012-08-27 20:16:55 <wumpus> yes +1 for Qt Creator
1516 2012-08-27 20:16:57 <Luke-Jr> no, that's KWrite
1517 2012-08-27 20:17:00 <Diablo-D3> gjs278: they renamed it?!
1518 2012-08-27 20:17:01 <Diablo-D3> wtf
1519 2012-08-27 20:17:04 <Diablo-D3> leafpad
1520 2012-08-27 20:17:05 <Diapolo> Diablo-D3: thanks that's and answer I can work with
1521 2012-08-27 20:17:08 <gjs278> well unless I'm using something different
1522 2012-08-27 20:17:12 <gjs278> it's gnome as hell though
1523 2012-08-27 20:17:16 <Diablo-D3> it used to be gnotepad
1524 2012-08-27 20:17:26 <Diapolo> wumpus: The GUI guys need to stay side by side together ^^.
1525 2012-08-27 20:17:27 <gjs278> hmm
1526 2012-08-27 20:17:30 <wumpus> :D
1527 2012-08-27 20:17:35 <gjs278> maybe this is just another generic gtk notepad then
1528 2012-08-27 20:17:45 * Diablo-D3 asks apt-get
1529 2012-08-27 20:17:50 <Luke-Jr> Qt Creator is a bit too GUI for my tastes ;p
1530 2012-08-27 20:17:57 * sipa uses mcedit
1531 2012-08-27 20:17:58 <Diablo-D3> gjs278: its not gnome more
1532 2012-08-27 20:18:01 * sipa *ducks*
1533 2012-08-27 20:18:02 <Diablo-D3> and I was wrong, it was gedit not gnotepad
1534 2012-08-27 20:18:07 <gjs278> yeah ok
1535 2012-08-27 20:18:17 * Luke-Jr notes WINE bundles a notepad port
1536 2012-08-27 20:18:20 <Diablo-D3> gjs278: it says it aims to be lighter than gedit, so thats fine with me
1537 2012-08-27 20:18:24 <Diablo-D3> its still no vim
1538 2012-08-27 20:18:26 <Diablo-D3> which is lighter yet
1539 2012-08-27 20:18:35 <Diapolo> Guys, the cause for all core devs (expect wumpus) to add silly spaces is because you don't use Qt Creator ^^.
1540 2012-08-27 20:18:36 <Luke-Jr> Kate supports VIM-like key bindings <.<
1541 2012-08-27 20:18:49 <wumpus> so does qt creator
1542 2012-08-27 20:18:58 <Diablo-D3> Diapolo: see the problem with that is
1543 2012-08-27 20:19:07 <wumpus> anyway
1544 2012-08-27 20:19:08 <Diablo-D3> I write my UIs in HTML.
1545 2012-08-27 20:19:09 <Diapolo> you can instruct it to clean up such stuff when saving ^^
1546 2012-08-27 20:19:11 <Luke-Jr> Diapolo: wait, if we get Gavin using Qt Creator, we can use tabs? :D
1547 2012-08-27 20:19:15 <Diablo-D3> cross platform, cross device, and works without installing more software.
1548 2012-08-27 20:19:38 <Diapolo> Qts XML files are not that bad IMHO
1549 2012-08-27 20:19:50 <Diapolo> and cross-platform afaik
1550 2012-08-27 20:20:02 <Diablo-D3> its not cross platform in the way you hope
1551 2012-08-27 20:20:11 <Diablo-D3> qt apps still look like shit on linux unless you run kde
1552 2012-08-27 20:20:21 <Diapolo> Luke-Jr: If he then only uses Bitcoin-Qt I'll allow you to add tabs :-P.
1553 2012-08-27 20:20:24 <sipa> wumpus: build complete here
1554 2012-08-27 20:20:36 <Diablo-D3> the qt-using-gtk-widgets qt theme doesnt quite work right
1555 2012-08-27 20:20:37 <wumpus> just use the gtk theme and it looks just like gtk...
1556 2012-08-27 20:20:48 <wumpus> sipa: ok
1557 2012-08-27 20:20:58 <Diablo-D3> its close but theres still problems with it
1558 2012-08-27 20:21:14 <wumpus> i don't care about 3 pixels difference
1559 2012-08-27 20:21:33 <Diablo-D3> I care about color differences
1560 2012-08-27 20:21:57 <Diablo-D3> qt apps on launch bitch in xterm they cant get certain color properties
1561 2012-08-27 20:22:14 <Diablo-D3> the theme has them, qt just isnt fetching them right (or using the wrong property name)
1562 2012-08-27 20:22:28 <Diablo-D3> and its still not compatible with gtk3
1563 2012-08-27 20:22:34 vampireb has joined
1564 2012-08-27 20:22:42 <wumpus> sigh...
1565 2012-08-27 20:22:46 sebicas has quit (Quit: sebicas)
1566 2012-08-27 20:22:54 <Diablo-D3> not that Im using gtk3 mind you, but its still a problem for those that are
1567 2012-08-27 20:24:01 <wumpus> rebasing my fix to upstream and pushing..
1568 2012-08-27 20:24:02 <sipa> where did mc's sshfs support go? :(
1569 2012-08-27 20:25:19 <Luke-Jr> Diapolo: https://github.com/bitcoin/bitcoin/pull/1731
1570 2012-08-27 20:25:19 <Diablo-D3> <3 sshfs
1571 2012-08-27 20:25:23 <Diablo-D3> best network file system ever
1572 2012-08-27 20:25:49 <sipa> ah, the url change
1573 2012-08-27 20:25:52 <wumpus> ok, fix is pushed, how to re-tag 0.6.3rc1 (or add a new tag, whatever is best)?
1574 2012-08-27 20:26:02 <sipa> /sh:// instead of /#sh:
1575 2012-08-27 20:26:04 <Diapolo> luke-jr: I did the same, but my commit message is better ^^.
1576 2012-08-27 20:26:12 <sipa> wumpus: 0.7.0rc1 !!!
1577 2012-08-27 20:26:17 <wumpus> ehh sorry
1578 2012-08-27 20:26:24 <sipa> wumpus: poke gavin
1579 2012-08-27 20:26:26 <wumpus> yeah
1580 2012-08-27 20:26:36 <wumpus> gavinandresen: can you re-tag?
1581 2012-08-27 20:26:44 <gavinandresen> will do, one sec
1582 2012-08-27 20:27:27 <gjs278> nfs's caching won't cache application/x-program I had to mess with my mime types and make everything text/x-program
1583 2012-08-27 20:27:55 <gjs278> it was missing about 90% of the stuff I needed cached in the first place
1584 2012-08-27 20:28:17 <sipa> wait, nfs cache works based on mime type? :S
1585 2012-08-27 20:28:24 <gavinandresen> wumpus: re-tagged
1586 2012-08-27 20:29:12 <Luke-Jr> jgarzik: git pull didn't get the updated tag..
1587 2012-08-27 20:29:18 <sipa> i needed git fetch --tags
1588 2012-08-27 20:29:26 <wumpus> nfs? nfs is from far before mime types.. I seriously doubt it uses them
1589 2012-08-27 20:29:34 <Luke-Jr> yep, --tags did it
1590 2012-08-27 20:29:42 * sipa builds
1591 2012-08-27 20:29:51 <makomk> Yeah, git pull doesn't replace existing tags by default.
1592 2012-08-27 20:32:50 torsthaldo has quit (Remote host closed the connection)
1593 2012-08-27 20:32:56 <Diapolo> Why is Matts Boost 1.5 pull still in the open pull list?
1594 2012-08-27 20:33:15 <sipa> Diapolo: the commit was pushed, but not from the pullreq
1595 2012-08-27 20:33:26 midnightmagic has quit (Ping timeout: 245 seconds)
1596 2012-08-27 20:34:33 <Diapolo> sipa: understood
1597 2012-08-27 20:37:43 <Diapolo> sipa: can you take a look at my Qt extend proxy options branch in the next time? I can't do that by myself and wouldn't love to only fiddle around to get it working ^^.
1598 2012-08-27 20:38:19 <wumpus> 343f712af8eeac00f49b5cec92f6bf14321f9b5e961719ef4643ebb3c7df2df3 bitcoin-0.7.0-win32-setup.exe
1599 2012-08-27 20:38:19 <wumpus> ec0e1047770dfccc3babf878b6a9d409acf6520f61353d31d53b3f4acbc6c436 bitcoin-qt.exe
1600 2012-08-27 20:38:23 midnightmagic has joined
1601 2012-08-27 20:38:23 <sipa> Diapolo: well the screenshot looked nice
1602 2012-08-27 20:39:27 vampireb has quit (Quit: leaving)
1603 2012-08-27 20:39:44 <gmaxwell> where is the screenshot?
1604 2012-08-27 20:40:06 <Diapolo> gmaxwell: there https://github.com/bitcoin/bitcoin/issues/1721
1605 2012-08-27 20:40:52 <gmaxwell> needs externalip settings.
1606 2012-08-27 20:41:48 <gmaxwell> perhaps a listen checkbox too.
1607 2012-08-27 20:42:15 <Diapolo> I can for sure extend the GUI, but need help with the networking part in optionsmodel.cpp
1608 2012-08-27 20:42:38 CodeLion has joined
1609 2012-08-27 20:42:48 <CodeLion> Hey noob question but I can't figure it out from googling. What determines the txid of a transaction?
1610 2012-08-27 20:42:54 <sipa> CodeLion: its hash
1611 2012-08-27 20:43:26 <sipa> double SHA256 of its binary serialization, specifically
1612 2012-08-27 20:44:48 <Diapolo> Are we using / including pthread.h somewhere?
1613 2012-08-27 20:45:13 tempmidnight has joined
1614 2012-08-27 20:45:19 <sipa> Diapolo: pthread isn't used on every platform, is it?
1615 2012-08-27 20:45:25 tempmidnight has left ()
1616 2012-08-27 20:45:27 <Diapolo> forget the question it's in db.h ...
1617 2012-08-27 20:46:01 <Diapolo> it causes some trouble with Boost and Windows at least for my local build, which I fixed but don't understand why that happens with cross-compilation
1618 2012-08-27 20:46:15 <Diapolo> DOES NOT is missing sorry
1619 2012-08-27 20:46:24 <wumpus> unsure why we include pthread directly at all
1620 2012-08-27 20:46:40 <Diapolo> no now, BDB does it I guess
1621 2012-08-27 20:46:43 <Diapolo> -w
1622 2012-08-27 20:46:54 <wumpus> ah, right
1623 2012-08-27 20:47:01 <Diapolo> db.h line 30
1624 2012-08-27 20:47:11 <wumpus> that's out of our control
1625 2012-08-27 20:47:36 <Diapolo> and that conflicts with util.h, where we do:
1626 2012-08-27 20:47:36 <Diapolo> #ifdef WIN32 typedef HANDLE pthread_t;
1627 2012-08-27 20:47:48 <Diapolo> when I remove that typedef and just use HANDLE it's fine
1628 2012-08-27 20:47:50 <wumpus> hehe
1629 2012-08-27 20:48:08 <Diapolo> otherwise I get: C:\Users\Diapolo\bitcoin.Qt\src\util.h:553: Fehler:'pthread_t' has a previous declaration as 'typedef void* pthread_t'
1630 2012-08-27 20:48:19 <Diapolo> (from db.h) ... sigh
1631 2012-08-27 20:48:35 <wumpus> yes, redefining pthread_t is probably not the nicest
1632 2012-08-27 20:48:48 roconnor has quit (Remote host closed the connection)
1633 2012-08-27 20:48:57 <Diapolo> I'm not sure why that is in anyway ... will git blame ^^.
1634 2012-08-27 20:49:18 <wumpus> it's on the same page as redefining printf :-)
1635 2012-08-27 20:49:31 <Diapolo> wumpus: you did that ^^
1636 2012-08-27 20:49:41 sirk390 has joined
1637 2012-08-27 20:49:49 <wumpus> I know
1638 2012-08-27 20:49:50 <wumpus> :p
1639 2012-08-27 20:49:58 <Diapolo> do you remember why?
1640 2012-08-27 20:50:09 <wumpus> yes, to get it to compile on windows with vc
1641 2012-08-27 20:50:38 <wumpus> it was a quick hack... we shouldn't really be using pthread_t at all
1642 2012-08-27 20:50:40 <Diapolo> that is a dirty hack, which removes the possibility for Windows users to use minGW :D
1643 2012-08-27 20:50:56 <Diapolo> ... without removing that
1644 2012-08-27 20:52:00 <wumpus> it will all go away if we switch to boost::thread everywhere
1645 2012-08-27 20:52:24 <Diapolo> See: http://msdn.microsoft.com/en-us/library/windows/desktop/ms682453%28v=vs.85%29.aspx they use HANDLE there?
1646 2012-08-27 20:52:48 davout has joined
1647 2012-08-27 20:52:48 davout has quit (Changing host)
1648 2012-08-27 20:52:48 davout has joined
1649 2012-08-27 20:53:12 <sipa> uploading build
1650 2012-08-27 20:54:25 <Diapolo> oh wait ... we have an own CreateThread ... and re-define it ...
1651 2012-08-27 20:54:35 <wumpus> but the thing is, we never use the return value of CreateThread anyway :-)
1652 2012-08-27 20:55:07 <wumpus> yes
1653 2012-08-27 20:55:23 <sipa> http://bitcoin.sipa.be/build/0.7.0rc1
1654 2012-08-27 20:55:36 <wumpus> CreateThread is our own function
1655 2012-08-27 20:55:47 <Diapolo> and MS has one that's called as ours
1656 2012-08-27 20:55:47 <sipa> http://bitcoin.sipa.be/builds/0.7.0rc1
1657 2012-08-27 20:55:53 <wumpus> cool sipa
1658 2012-08-27 20:56:07 <wumpus> Diapolo: yes, as I said, it's on the same page as redefining printf
1659 2012-08-27 20:56:33 <wumpus> those are traps to catch new developers off guard
1660 2012-08-27 20:57:01 robocoin has joined
1661 2012-08-27 20:57:10 <jgarzik> wumpus, Diapolo: any notable UI changes in 0.7 ?
1662 2012-08-27 20:57:30 <Diapolo> indeed
1663 2012-08-27 20:57:38 <sipa> sig verification?
1664 2012-08-27 20:57:46 <sipa> better progress bars
1665 2012-08-27 20:57:57 <sipa> ... hopefully :)
1666 2012-08-27 20:58:53 <Luke-Jr> Diapolo: I think jgarzik wants details :p
1667 2012-08-27 20:59:00 <wumpus> wasn't the rp console new in 0.7 as well?
1668 2012-08-27 20:59:04 <Diapolo> Debug window, re-design and extension of the options dialog and the QR-Code dialog
1669 2012-08-27 20:59:07 <sipa> wumpus: yes
1670 2012-08-27 20:59:30 <wumpus> it's been so long since last release that I kind of lost track
1671 2012-08-27 20:59:34 <Luke-Jr> lol
1672 2012-08-27 20:59:40 <Diapolo> sign/verify message dialog overhaul
1673 2012-08-27 20:59:48 <Diapolo> now matches the RPC calls
1674 2012-08-27 21:00:19 <Diapolo> unit change is reflected directly on button press on all places
1675 2012-08-27 21:00:29 <jgarzik> wumpus: git shortlog --no-merges v0.6.3..
1676 2012-08-27 21:00:29 <gmaxwell> BEST RELEASE EVER.
1677 2012-08-27 21:00:31 <Diapolo> better language selection
1678 2012-08-27 21:00:48 <jgarzik> Diapolo: ^^
1679 2012-08-27 21:00:48 <Luke-Jr> gmaxwell: haha, yeah just say that âº
1680 2012-08-27 21:00:52 <Diapolo> much better translation quality ^^
1681 2012-08-27 21:01:03 <wumpus> Diapolo: it has language selection in the UI *at all*
1682 2012-08-27 21:01:09 <Diapolo> I consider that a small child of mine btw ^^
1683 2012-08-27 21:01:19 <jgarzik> Diapolo: may I nominate you to create a quick pull request updating doc/release-notes.txt ? Sounds like you have a good idea...
1684 2012-08-27 21:01:23 <Diapolo> wumpus: indeed you are right
1685 2012-08-27 21:01:26 <wumpus> haha yes you contributed alot Diapolo
1686 2012-08-27 21:01:30 <sipa> wumpus: any build result for 0.7.0rc1?
1687 2012-08-27 21:01:40 <wumpus> yes sipa, just pushed to gitian-sigs
1688 2012-08-27 21:02:06 bakh has joined
1689 2012-08-27 21:02:10 <sipa> wumpus: ehm... 0.7.1rc1 already? :o
1690 2012-08-27 21:02:14 <sipa> geez you're going fast
1691 2012-08-27 21:02:17 <Diapolo> jgarzik: I'll do my best just a few minutes...
1692 2012-08-27 21:02:19 <jgarzik> heh
1693 2012-08-27 21:02:19 <wumpus> uhm
1694 2012-08-27 21:02:35 davout_ has joined
1695 2012-08-27 21:02:35 <sipa> wumpus: i need to clean my eyes
1696 2012-08-27 21:02:38 <wumpus> I really need to go to sleep
1697 2012-08-27 21:02:43 <sipa> no, i don't!
1698 2012-08-27 21:03:27 <sipa> wumpus: linux build doesn't match
1699 2012-08-27 21:03:47 <Diapolo> jgarzik: shall I do this by hand or via that git shortlog --no-merges v0.6.3 thing?
1700 2012-08-27 21:03:51 <wumpus> re-pushed with 0.7.0rc1
1701 2012-08-27 21:04:11 <sipa> windows build matches
1702 2012-08-27 21:04:14 <gmaxwell> Diapolo: you should do a git shortlog and use the result to make a list by hand.
1703 2012-08-27 21:04:27 <gmaxwell> Diapolo: and ask in here for things that you don't know about but sound worth mentioning.
1704 2012-08-27 21:04:36 <wumpus> sipa: curious
1705 2012-08-27 21:04:45 <Diapolo> I guess that list will critically hit me ^^
1706 2012-08-27 21:04:54 <gmaxwell> The raw txn functionality probably deserves a long description copied from gavin's posts.
1707 2012-08-27 21:05:01 <sipa> wumpus: my linux build didn't match gavin's either last time
1708 2012-08-27 21:05:35 ThomasV has quit (Quit: Quitte)
1709 2012-08-27 21:05:44 <wumpus> sipa: the strange thing is that the linux build uses no manually built deps, just ubuntu lucid, right? I don't understand what the difference can come from
1710 2012-08-27 21:06:10 <Diapolo> gmaxwell: that commands prints commits that are pretty old ... can't use that
1711 2012-08-27 21:06:13 davout has quit (Ping timeout: 252 seconds)
1712 2012-08-27 21:06:14 sirk390 has quit (Quit: Leaving.)
1713 2012-08-27 21:06:26 <wumpus> the windows build is so much more involved... still, that matches
1714 2012-08-27 21:06:28 <sipa> wumpus: when were your base images built?
1715 2012-08-27 21:06:42 <gmaxwell> Diapolo: are they just old because they were pending for a long time?
1716 2012-08-27 21:06:43 <wumpus> -rw-r--r-- 1 root root 510459904 Oct 24 2011 base-lucid-amd64.qcow2
1717 2012-08-27 21:06:43 <wumpus> -rw-r--r-- 1 orion orion 494993408 Oct 24 2011 base-lucid-i386.qcow2
1718 2012-08-27 21:07:09 <sipa> wumpus: mine were built today, maybe it matters...
1719 2012-08-27 21:07:11 <Diapolo> it starts with commits from Satoshi?
1720 2012-08-27 21:07:22 <gmaxwell> Diapolo: 0_o
1721 2012-08-27 21:07:39 <wumpus> can I just remove them and they will be rebuilt?
1722 2012-08-27 21:07:39 <sipa> Diapolo: git shortlog --no-merges v0.6.3..v0.7.0rc1
1723 2012-08-27 21:08:00 torsthaldo has joined
1724 2012-08-27 21:08:05 <Diapolo> sipa: I guess I was missing the end I never used that
1725 2012-08-27 21:08:08 <gmaxwell> because 0.6.3 was on a branch it may be a bit weird.
1726 2012-08-27 21:08:13 <wumpus> shouldn't you use 0.6.0 instead?
1727 2012-08-27 21:08:15 <wumpus> exactly
1728 2012-08-27 21:08:17 <sipa> yes, start from 0.6.1
1729 2012-08-27 21:08:23 <gmaxwell> you may need to do 0.6.1 and ignore the ones that were in 0.6.3 manually.
1730 2012-08-27 21:08:25 <sipa> that's an ancestor of 0.7.0rc1
1731 2012-08-27 21:08:59 CodeLion has left ("Cya")
1732 2012-08-27 21:09:17 <Diapolo> can I export that list directly into a file > list.txt?
1733 2012-08-27 21:09:21 <sipa> yes
1734 2012-08-27 21:10:27 <Diapolo> alright and now I just need to get the ones who are Qt related
1735 2012-08-27 21:12:20 <sipa> wumpus: sigs pushed, by the way
1736 2012-08-27 21:14:59 minimoose has quit (Quit: minimoose)
1737 2012-08-27 21:15:35 aq83 has quit (Ping timeout: 246 seconds)
1738 2012-08-27 21:16:51 <wumpus> some package differences (for example, freetype) may explain it
1739 2012-08-27 21:17:02 aq83 has joined
1740 2012-08-27 21:17:27 <wumpus> now building new qcow2 packages
1741 2012-08-27 21:17:36 * sipa is a fan of lxc
1742 2012-08-27 21:17:53 <sipa> seems to work fine after a few configuration things, and faster too
1743 2012-08-27 21:18:13 <wumpus> yeah, sounds nice, but I don't feel like setting up all kinds of new stuff right now
1744 2012-08-27 21:18:16 <sipa> wumpus: i doubt freetype has an effect on bitcoind
1745 2012-08-27 21:18:39 <wumpus> well it's used through qt
1746 2012-08-27 21:18:50 <wumpus> ah, right
1747 2012-08-27 21:18:53 <wumpus> no, not bitcoind
1748 2012-08-27 21:19:36 <jgarzik> Diapolo: maybe 'git shortlog --no-merges <version range> src/qt/'
1749 2012-08-27 21:19:42 iocor has joined
1750 2012-08-27 21:19:42 <jgarzik> that's _only_ the changes in src/qt/
1751 2012-08-27 21:20:31 <jgarzik> 0.6.3 was on a branch, but git is generally smart about such things
1752 2012-08-27 21:23:35 <Diapolo> jgarzik: I'm just wondering how big a fix or a commit needs to be to make it into such a list ... Qt wise there were plenty of small fixes ^^.
1753 2012-08-27 21:23:49 <jgarzik> Diapolo: ignore all those
1754 2012-08-27 21:23:51 <wumpus> just list the major things
1755 2012-08-27 21:23:57 <Diapolo> okay
1756 2012-08-27 21:23:58 <jgarzik> Diapolo: just list the major things users will see / should know about
1757 2012-08-27 21:24:05 <wumpus> (I think we already covered most in the discussion above)
1758 2012-08-27 21:24:12 <Diapolo> user notable changes
1759 2012-08-27 21:24:16 <jgarzik> yes
1760 2012-08-27 21:24:25 <Luke-Jr> jgarzik: git isn't smart when there's cherry-picking
1761 2012-08-27 21:25:07 <jgarzik> For example, there were many source code cleanups and movements in src/ but they barely got a single line mention at the end
1762 2012-08-27 21:28:19 slush1 has quit (Read error: Operation timed out)
1763 2012-08-27 21:30:41 paul_nicholson has left ()
1764 2012-08-27 21:30:57 <jgarzik> heh, wow. the first bitcoin real estate purchase, for 100,000 BTC.
1765 2012-08-27 21:31:15 <Luke-Jr> o.o
1766 2012-08-27 21:31:26 <sipa> jgarzik: now change the 'e' in caymen to cayman
1767 2012-08-27 21:31:43 <Diapolo> (hypo) real estate was a bank who had a bailout here in DE ^^
1768 2012-08-27 21:33:09 bakh has quit (Quit: Ex-Chat)
1769 2012-08-27 21:34:01 <Diapolo> damn my own commit messages sucked bad a few weeks ago -_-
1770 2012-08-27 21:34:02 ovidiusoft has quit (Ping timeout: 260 seconds)
1771 2012-08-27 21:37:43 robocoin has left ("Verlassend")
1772 2012-08-27 21:38:45 ThomasV has joined
1773 2012-08-27 21:39:06 pnicholson has joined
1774 2012-08-27 21:40:40 <TD> jgarzik: if you want one that's apparently real: http://www.bitmit.net/en/trade/i/4552-3-zi-wohnung-3-room-apartment-61m-in-kornwestheim-d/description
1775 2012-08-27 21:40:49 maaku has left ()
1776 2012-08-27 21:43:53 <TD> BlueMatt: i'm reviewing some more of your fullverif change snow
1777 2012-08-27 21:44:12 <TD> just a few minor things that are quick fixes - there seem to be a mix of tabs/spaces or screwy indenting in places?
1778 2012-08-27 21:44:34 <TD> some of the more critical methods you're adding like Transaction.verify don't seem to have any unit tests
1779 2012-08-27 21:44:48 <TD> i think the way i'll merge these changes is to do them one at a time and then go add my own changes on top.
1780 2012-08-27 21:45:02 <TD> i think that's easier than describing what i want and round-tripping all the time.
1781 2012-08-27 21:46:11 <jgarzik> Diapolo: just skim for major changes... the list probably shouldn't be longer than 10-20 lines
1782 2012-08-27 21:46:20 <jgarzik> (though shorter is just fine, too)
1783 2012-08-27 21:46:49 <Diapolo> can I re-word a little or does it need to be the commit messages always?
1784 2012-08-27 21:46:59 <Diapolo> nearly finished the
1785 2012-08-27 21:47:04 <sipa> Diapolo: it doesn't need to be commit messages at all
1786 2012-08-27 21:47:21 <sipa> it just needs to be a nicely human readable piece of text that lists the changes
1787 2012-08-27 21:47:34 <Diapolo> sipa: wow that is hard ... tell a german use git -shortlog and he will try to make the very best of it ^^ LOL
1788 2012-08-27 21:48:05 <sipa> Diapolo: the shortlog is to help you remember :)
1789 2012-08-27 21:48:11 <jgarzik> exactly
1790 2012-08-27 21:49:46 darkee has joined
1791 2012-08-27 21:52:06 tower has quit (Ping timeout: 264 seconds)
1792 2012-08-27 21:53:19 juchebot has joined
1793 2012-08-27 21:53:48 tower has joined
1794 2012-08-27 21:55:44 <Diapolo> got it, now creating the pull ... fell free to remove things that are to much
1795 2012-08-27 21:55:50 juchebot has quit (K-Lined)
1796 2012-08-27 21:55:54 <sipa> pull?
1797 2012-08-27 21:57:53 * jgarzik turns down a request from Huffington Post to do a live webcam chat about bitcoin
1798 2012-08-27 21:57:59 <jgarzik> "I'm on vacation dammit"
1799 2012-08-27 21:58:18 * jgarzik was nicer, but that was the gist ;p
1800 2012-08-27 21:58:39 setkeh has joined
1801 2012-08-27 21:59:20 <jgarzik> sipa: I asked him to create a pull req updating doc/release-notes.txt with UI stuff in it... had no idea it would get so involved ;p
1802 2012-08-27 21:59:24 <Diapolo> pull is ready
1803 2012-08-27 21:59:36 <Diapolo> I'm able to do it faster next time I promise :).
1804 2012-08-27 22:02:28 <jgarzik> gavinandresen, gmaxwell, sipa: ready for release?
1805 2012-08-27 22:02:29 ThomasV has quit (Ping timeout: 260 seconds)
1806 2012-08-27 22:02:53 <sipa> jgarzik: no matching builds yet
1807 2012-08-27 22:04:02 <gavinandresen> jgarzik: my gitian-building machine is dead, trying to resurrect... (but happy to upload/sign binaries if we get reproducable builds from others)
1808 2012-08-27 22:04:42 <sipa> wumpus: did you rebuild already?
1809 2012-08-27 22:05:12 <Diapolo> wumpus fell asleep no?
1810 2012-08-27 22:05:30 <gavinandresen> afk dinner
1811 2012-08-27 22:06:03 Marf has quit (Ping timeout: 246 seconds)
1812 2012-08-27 22:06:08 TD has quit (Quit: TD)
1813 2012-08-27 22:06:24 <sipa> Diapolo: either I missed that, or you're clairvoyant?
1814 2012-08-27 22:07:25 <Diapolo> I'll be surpriesed what clairvoyant means, need a translator ^^... just a sec.
1815 2012-08-27 22:08:16 [Tycho] has quit (Ping timeout: 276 seconds)
1816 2012-08-27 22:08:22 <Diapolo> I consider that a praise :-D
1817 2012-08-27 22:08:45 <Diapolo> ...until he is writing again
1818 2012-08-27 22:10:49 Marf has joined
1819 2012-08-27 22:11:00 <Diapolo> Btw. sipa: Did you consider using memory-mapped files for the block-chain files? I mean your ultraprune patch creates smaller files and a 128MB file could be entirely in RAM. Or do you consider that unsave?
1820 2012-08-27 22:13:13 tower has quit (Disconnected by services)
1821 2012-08-27 22:13:17 * jgarzik has considering mmap for blk*.dat files
1822 2012-08-27 22:13:21 <jgarzik> they are perfect for it
1823 2012-08-27 22:13:26 tower has joined
1824 2012-08-27 22:14:26 <Diapolo> I wrote a small wrapper around Boost IPC file_mapping and mapped_region for the block-files which works and is OS independent
1825 2012-08-27 22:14:32 <gmaxwell> 'meh'
1826 2012-08-27 22:14:53 <Diapolo> I'm just using it for my local setup and consider that a play-ground.
1827 2012-08-27 22:14:54 <gmaxwell> we already need more total blk.dat than 32bit systems have VM space.
1828 2012-08-27 22:15:36 <Diapolo> On Windows you can use larga address aware flag, which allowed me to request a 2GB chain-file as mem mapped file.
1829 2012-08-27 22:15:41 <Diapolo> large
1830 2012-08-27 22:15:49 <midnightmagic> gmaxwell: is blk0*.dat compressed?
1831 2012-08-27 22:15:52 * Eliel doesn't think Diapolo was suggesting mmapping the blk.dat files.
1832 2012-08-27 22:15:54 <gmaxwell> Diapolo: sure, now map two of them.
1833 2012-08-27 22:16:13 <Diapolo> relase one, map the other ... but no sipas patch has 128 MiB files
1834 2012-08-27 22:16:27 <gmaxwell> (I think windows has a 3+1 split)
1835 2012-08-27 22:16:41 <Diapolo> <- x64
1836 2012-08-27 22:17:08 <gmaxwell> Diapolo: what is constantly mmaping and unmapping going to do for them? We need random access to the files.
1837 2012-08-27 22:17:22 <gmaxwell> midnightmagic: No. It's just seralized blocks.
1838 2012-08-27 22:17:31 <gmaxwell> midnightmagic: conventional compression doesn't do much with it.
1839 2012-08-27 22:17:36 <Diapolo> I guess at least a few could stay in RAM, which gives really fast access times
1840 2012-08-27 22:17:38 ThomasV has joined
1841 2012-08-27 22:17:54 <Diapolo> gmaxwell: I trie 7-Zip on them, gave ~20% savings
1842 2012-08-27 22:18:03 <Diablo-D3> try lrzip
1843 2012-08-27 22:18:14 * gmaxwell rolls eyes.
1844 2012-08-27 22:18:45 <Diablo-D3> nothing beats lrzip
1845 2012-08-27 22:18:59 <gmaxwell> Diapolo: memmap doesn't put things in ram. (good thing it doesn't, otherwise your memmap/unmap would be quite slow)
1846 2012-08-27 22:19:13 Zarutian has quit (Quit: Zarutian)
1847 2012-08-27 22:19:27 <Diablo-D3> memmap maps a foreign memory address
1848 2012-08-27 22:19:31 <gmaxwell> Diapolo: it lets you access the files with demand paging, but the data isn't in ram until you access it (+ some readhead)
1849 2012-08-27 22:19:31 <Diablo-D3> which includes files
1850 2012-08-27 22:20:21 <Diapolo> the time needed for writing reading a block for a mapped files is 0ms with our timing function ...
1851 2012-08-27 22:20:37 <Diapolo> which made me think that happens in RAM
1852 2012-08-27 22:20:49 <Diapolo> switching them takes time of course
1853 2012-08-27 22:21:29 <gmaxwell> Diapolo: it should also take 0ms to read from the file, in cases where cache readahead would have put the data in memory already.
1854 2012-08-27 22:22:08 <Diapolo> but that doesn't happen with our current code
1855 2012-08-27 22:22:42 <Diapolo> it always takes a few ms for every block that's what I observed
1856 2012-08-27 22:23:57 <Diapolo> that was no scientific setup and even if my technical understanding is not that deep, memory mapped files are there to speedup I/O at least that should be true?
1857 2012-08-27 22:27:53 <Diapolo> ... bed is calling
1858 2012-08-27 22:28:15 Diapolo has quit (Quit: Leaving.)
1859 2012-08-27 22:28:34 <gmaxwell> No. Memorymapped files are not inherently faster. (in fact because of the extra TLB hits they can be slower)
1860 2012-08-27 22:28:58 rdponticelli has quit (Ping timeout: 260 seconds)
1861 2012-08-27 22:28:58 <gmaxwell> oh! he's gone. goodnight diapolo! :)
1862 2012-08-27 22:30:29 pnicholson has quit (Ping timeout: 260 seconds)
1863 2012-08-27 22:35:34 pickett_ has quit (Remote host closed the connection)
1864 2012-08-27 22:36:10 ThomasV has quit (Quit: Quitte)
1865 2012-08-27 22:36:35 pickett_ has joined
1866 2012-08-27 22:41:57 <Luke-Jr> Win32 0.7.0rc1 343f712af
1867 2012-08-27 22:42:05 <Luke-Jr> matches wumpus
1868 2012-08-27 22:43:09 <Luke-Jr> 1 moreâ¦
1869 2012-08-27 22:43:18 <Luke-Jr> also, 0.4.8rc1 tagged if anyone can build it..
1870 2012-08-27 22:45:39 D34TH__ has joined
1871 2012-08-27 22:45:57 MobiusL has quit (Remote host closed the connection)
1872 2012-08-27 22:45:57 darkee has quit (Remote host closed the connection)
1873 2012-08-27 22:45:58 da2ce7_d has quit (Write error: Broken pipe)
1874 2012-08-27 22:46:46 darkee has joined
1875 2012-08-27 22:47:38 D34TH has joined
1876 2012-08-27 22:47:38 D34TH has quit (Changing host)
1877 2012-08-27 22:47:38 D34TH has joined
1878 2012-08-27 22:49:13 D34TH_ has quit (Ping timeout: 276 seconds)
1879 2012-08-27 22:51:10 D34TH__ has quit (Ping timeout: 276 seconds)
1880 2012-08-27 22:51:19 p0s has joined
1881 2012-08-27 22:51:25 p0s has quit (Changing host)
1882 2012-08-27 22:51:25 p0s has joined
1883 2012-08-27 22:52:40 drazak_ has joined
1884 2012-08-27 22:52:45 <sipa> Luke-Jr: so mine too?
1885 2012-08-27 22:52:57 <Luke-Jr> sipa: I didn't see yours it the sig repo :p
1886 2012-08-27 23:01:24 pnicholson has joined
1887 2012-08-27 23:07:42 drazak__ has joined
1888 2012-08-27 23:07:42 drazak__ has quit (Client Quit)
1889 2012-08-27 23:09:30 p0s has quit (Read error: Connection reset by peer)
1890 2012-08-27 23:11:34 <sipa> Luke-Jr: still?
1891 2012-08-27 23:12:13 <Luke-Jr> I see it now
1892 2012-08-27 23:15:09 <sipa> hm ok
1893 2012-08-27 23:17:10 <gmaxwell> I'm pondering if furture transaction drafting code shouldn't automatically split enormous change outputs.
1894 2012-08-27 23:17:27 Motest031 has joined
1895 2012-08-27 23:17:42 pnicholson has quit (Quit: pnicholson)
1896 2012-08-27 23:17:46 <gmaxwell> e.g. split change up to 4 ways if the change is over 1000 BTC.
1897 2012-08-27 23:18:03 <gmaxwell> These single addresses with hundreds of thousands of bitcoins give me the heebie jeebies.
1898 2012-08-27 23:18:34 <gmaxwell> The fools are just hoping a cosmic ray doesn't flip a bit and cause them to send all their coin to the great unknown.
1899 2012-08-27 23:18:41 Motest003 has quit (Ping timeout: 248 seconds)
1900 2012-08-27 23:18:45 <sipa> gmaxwell: it's been a while, but i once thought about some algorithm that would always create a few outputs (at least two, at most 5 or so), of exponentially distributed size, but similar to the real output (same number of significants digits, ...)
1901 2012-08-27 23:18:46 pnicholson has joined
1902 2012-08-27 23:18:52 LuaKT has quit ()
1903 2012-08-27 23:18:59 <copumpkin> it'd also be nice if it attempted to hide which transactions are change and which are the intended transaction
1904 2012-08-27 23:19:12 <copumpkin> by making nice round numbers of change in roughly the same magnitude of the main transaction
1905 2012-08-27 23:19:14 <copumpkin> or something like that
1906 2012-08-27 23:19:24 <sipa> gmaxwell: splitting change output was actually suggested by satoshi once
1907 2012-08-27 23:19:53 <sipa> copumpkin: indeed
1908 2012-08-27 23:20:11 <gmaxwell> copumpkin: what I'd like to do is have autosweeping that adds aditional inputs, if there are many available inputs it would do so heedlessly of how round the change is, but if you have few available inputs it would only add more if it made the output rounder.
1909 2012-08-27 23:20:18 theymos has joined
1910 2012-08-27 23:20:31 <sipa> if you now see a 1.34571254 input, and a 1.0 and a 0.34571254 output, it's not really a secret what the change is
1911 2012-08-27 23:20:47 <gmaxwell> But even ignoring the privacy implications; single really enormous outputs worry me.
1912 2012-08-27 23:20:51 <sipa> indeed
1913 2012-08-27 23:21:10 <copumpkin> yeah
1914 2012-08-27 23:22:06 <gmaxwell> I also wonder if the txn creation code shouldn't also check that you could write a txn spending that change output(s), just for paranoia sake.
1915 2012-08-27 23:22:29 <sipa> ?
1916 2012-08-27 23:23:04 <gmaxwell> sipa: right now a single bitflip during txn creation could send all your coin to the black neveragain, I think.
1917 2012-08-27 23:23:22 <gmaxwell> e.g. you pull a change address, a bit gets flipped.. you send the coin to some unredeemable address never to be seen again.
1918 2012-08-27 23:23:36 <gmaxwell> we check that the txn is valid, but we don't check that the change address is really yours.
1919 2012-08-27 23:23:41 CodesInChaos has quit (Ping timeout: 244 seconds)
1920 2012-08-27 23:24:20 <gmaxwell> and normally, perhaps it's no big deal, but I'm seeing people link to a txn that seems to be clearly producing some hundreds of thousands of btc in change.
1921 2012-08-27 23:24:24 jouke has quit (Read error: Operation timed out)
1922 2012-08-27 23:24:30 <gmaxwell> "Most expensive cosmic ray ever"
1923 2012-08-27 23:24:56 <BlueMatt> also wanna check the hashed version of the tx message you are about to send also matches the hash of the tx which was verified, then
1924 2012-08-27 23:25:04 <sipa> s/cosmic ray/linode employee/
1925 2012-08-27 23:26:12 davout_ has quit (Remote host closed the connection)
1926 2012-08-27 23:29:17 eoss has joined
1927 2012-08-27 23:29:18 eoss has quit (Changing host)
1928 2012-08-27 23:29:18 eoss has joined
1929 2012-08-27 23:31:04 rbkillea has joined
1930 2012-08-27 23:33:59 iocor has quit (Quit: Computer has gone to sleep.)
1931 2012-08-27 23:35:04 MobiusL has joined
1932 2012-08-27 23:41:45 <weex> wouldn't that bitflip have to pass crc?
1933 2012-08-27 23:43:32 <weex> oh i guess the signature itself would be come impossible to reproduce
1934 2012-08-27 23:53:17 chrisb__ has quit (Quit: Leaving)
1935 2012-08-27 23:53:45 logger_ has quit (Ping timeout: 245 seconds)
1936 2012-08-27 23:55:20 theymos has left ()
1937 2012-08-27 23:55:23 rdponticelli has joined