1 2012-02-08 00:01:48 <Raccoon> is there no Import option for the standard bitcoin client wallet?
  2 2012-02-08 00:01:58 <sipa> in 0.6 there will be
  3 2012-02-08 00:02:13 <Raccoon> thanks.
  4 2012-02-08 00:02:47 <Raccoon> what about an option to cull a wallet of "old empties"
  5 2012-02-08 00:03:48 <riush> Raccoon: not advised. someone might still send money there
  6 2012-02-08 00:04:01 <riush> it would be nice to move old addresses to a 'trash' or something though
  7 2012-02-08 00:04:17 <cjd> probably the best balance is to have an old-address-heap where you can dump them
  8 2012-02-08 00:04:25 <Raccoon> riush: useful if you're utilizing the daemon for API services
  9 2012-02-08 00:04:39 <Raccoon> generating unique addresses for customers or merely potential customers
 10 2012-02-08 00:04:42 <cjd> but /me doesn't have time to write it so shuts up
 11 2012-02-08 00:08:45 <Raccoon> cjd: very hard to set a display filter of addressess with contents, and ability to only export those?
 12 2012-02-08 00:09:17 <Raccoon> multi-wallet support would headway what you suggested
 13 2012-02-08 00:09:47 <Raccoon> removable devices
 14 2012-02-08 00:14:36 <Raccoon> in the API, what does the parameter [account] refer to?
 15 2012-02-08 00:15:24 <Raccoon> a bitcoin address label? a wallet? or some sub structure in between (Wallet -> Accounts -> Addresses) ?
 16 2012-02-08 00:16:48 <gmaxwell> Raccoon: https://en.bitcoin.it/wiki/Accounts_explained
 17 2012-02-08 00:17:09 <Raccoon> thanks
 18 2012-02-08 00:47:00 <Ferroh> Does the blockchain.info guy hang out in here ever?
 19 2012-02-08 02:52:12 <CRichard> hi
 20 2012-02-08 02:54:27 <CRichard> I have an idea w/r/t wallet security. Basically the user optionally sets 4 secret question/answers, any three of which can recover the wallet if the passphrase or secret is lost
 21 2012-02-08 02:55:16 <CRichard> if users are encouraged to use numbers from their documents (i.e. passport etc.) in these questions, it will virtually eliminate wallet loss
 22 2012-02-08 02:55:53 <splatster> I think that this could be applied to 2 factor authentication of sort.
 23 2012-02-08 02:56:10 <splatster> Something you know: Your existing encryption key
 24 2012-02-08 02:56:18 <splatster> Something you have: Your ID
 25 2012-02-08 02:56:18 <theymos> Not much entropy in the answers. Maybe if the answer was hashed billions of times to get the key.
 26 2012-02-08 02:56:36 <splatster> theymos: Billions takes time
 27 2012-02-08 02:56:48 <theymos> Yeah, you'd want each guess to take days.
 28 2012-02-08 02:57:29 <CRichard> hm
 29 2012-02-08 02:57:39 <theymos> I'm not a fan of question/answer schemes because the answers are usually super easy to get with a little research.
 30 2012-02-08 02:57:45 <splatster> It would have to be -the most inefficient- hashing algo
 31 2012-02-08 02:57:59 <splatster> That wallet cracking rigs are out of the question
 32 2012-02-08 02:58:18 <onelineproof> keep paper backups
 33 2012-02-08 02:58:26 <splatster> or that ^
 34 2012-02-08 02:58:45 <CRichard> the way I was thinking of doing it was to simply break the user's wallet key into three parts, with some padding on them, then encrypt each part twice, once with one secret answer and once with another
 35 2012-02-08 02:58:52 <gmaxwell> or you could just have the user record 12 normal randomly english words, and can have a good 128 bits of entropy, have them write it down a few times and put it someplace safe and secure, and not have to worry about all that.
 36 2012-02-08 02:58:58 <CRichard> such that any three valid answers to the four questions unlocks the wallet key
 37 2012-02-08 02:59:01 <onelineproof> also, memorize one complicated password, and use that to store encrypted backups in various locations in the world
 38 2012-02-08 02:59:07 <BlueMatt> I really prefer printable deterministic wallets
 39 2012-02-08 02:59:12 <onelineproof> repeat that password every day
 40 2012-02-08 02:59:18 <onelineproof> that's my plan
 41 2012-02-08 02:59:19 <gmaxwell> And what BlueMatt says.
 42 2012-02-08 02:59:30 <gmaxwell> onelineproof: hope you don't get hit on the head. :)
 43 2012-02-08 02:59:59 <onelineproof> well if my brain fails, then I have more important things to worry about than money
 44 2012-02-08 03:00:03 <gmaxwell> (yes, sounds like a movie in reality people sometime forget things they use every day even without being hit on the head!)
 45 2012-02-08 03:00:07 <splatster> I think you could use the static password profile of a Yubikey to use a super complex randomly generated password for your wallet
 46 2012-02-08 03:00:09 <onelineproof> anyway, theres still the paper backup...
 47 2012-02-08 03:00:33 <gmaxwell> onelineproof: ah, wasn't clear to me that your paper backups weren't encrypted with the same key. Fine enough.
 48 2012-02-08 03:00:41 <splatster> BlueMatt: Armory does just that
 49 2012-02-08 03:00:57 <BlueMatt> yep
 50 2012-02-08 03:01:07 <BlueMatt> though I believe he is changing up the format
 51 2012-02-08 03:01:09 <onelineproof> paper backups for individual bitcoin private keys as well as a paper backup for the master key to my encrypted stuff
 52 2012-02-08 03:01:28 <BlueMatt> also, sipa has a working compatible version of armory's format
 53 2012-02-08 03:01:32 <splatster> BlueMatt: But the old formats should still be usable
 54 2012-02-08 03:01:40 <BlueMatt> (on bitcoind)
 55 2012-02-08 03:01:48 <CRichard> talking to people who have heard about bitcoin, but don't use it, their main complaint is the ease of which they can lose their bitcoins by forgetting their password
 56 2012-02-08 03:01:58 <CRichard> most people are used to the secret question/answer system
 57 2012-02-08 03:02:05 <splatster> CRichard: Don't forget your password
 58 2012-02-08 03:02:09 <NxTitle> >.<
 59 2012-02-08 03:02:17 <CRichard> well yes, thats well and good but people are quite imperfect
 60 2012-02-08 03:02:21 <gmaxwell> CRichard: those systems are horribly insecure, for the most part.
 61 2012-02-08 03:02:29 <NxTitle> I had a 6 year old gmail account that had a strong password (afterchanging it)
 62 2012-02-08 03:02:30 <CRichard> rules of interface design #1: make it forgiving to mistakes
 63 2012-02-08 03:02:34 <gmaxwell> CRichard: then simply write down the secret. There is nothing wrong with that.
 64 2012-02-08 03:02:42 <splatster> I could look at your facebook for such questions
 65 2012-02-08 03:02:46 <smickles> if you think you might forget your password, write it down and put that in a safe deposit box :|
 66 2012-02-08 03:02:50 <splatster> Where were you born?
 67 2012-02-08 03:02:51 <splatster> easy
 68 2012-02-08 03:02:52 <gmaxwell> CRichard: or don't use encryption with bitcoin. It's not a requirement. Has anyone really told you that?
 69 2012-02-08 03:02:52 <NxTitle> and the security question on it was "what's your frequent flyer number" and my friend managed to hack into it because my answer was apparently "1"
 70 2012-02-08 03:03:11 <NxTitle> ^ then hackers steal your coinz
 71 2012-02-08 03:03:15 <CRichard> I know its not a requirement
 72 2012-02-08 03:03:20 <CRichard> I'm not talking about me
 73 2012-02-08 03:03:21 <NxTitle> because people can't help but download bucketloads of porn
 74 2012-02-08 03:03:23 <gmaxwell> CRichard: people losing their passphrase was a major concern of mine when the functionality was added, but I never actually expected a user to be concerned about it (until it was too late)
 75 2012-02-08 03:03:32 <CRichard> developers have no problems with keeping their encryption strong and their passwords safe
 76 2012-02-08 03:03:38 <CRichard> I'm talking about the public in general
 77 2012-02-08 03:03:49 <splatster> Write that shit down and put it a fucking safe
 78 2012-02-08 03:03:52 <BlueMatt> for that we have printable wallets
 79 2012-02-08 03:03:53 <splatster> problem solved
 80 2012-02-08 03:03:57 <gmaxwell> NxTitle: wallet encryption provides very little protection (though not none) against hackers.
 81 2012-02-08 03:04:02 <splatster> or use printable wallet backups
 82 2012-02-08 03:04:22 <NxTitle> keylogger will steal it anyways
 83 2012-02-08 03:04:22 <splatster> gmaxwell: Whaaaaaat?
 84 2012-02-08 03:04:30 <CRichard> no, not at all
 85 2012-02-08 03:04:36 <splatster> </sarcasm>
 86 2012-02-08 03:04:37 <NxTitle> everyone should just have bitcoin wallets on dedicated devices
 87 2012-02-08 03:04:40 <CRichard> take a hint from games like runescape
 88 2012-02-08 03:04:41 <gmaxwell> CRichard: I know the general public will not remember keys. I was quite vocal that I expected more coins to be lost due to forgetting than theft but if people are actually worried, this is a sign I might be wrong.
 89 2012-02-08 03:04:45 <CRichard> you can have a mouse entered pin
 90 2012-02-08 03:04:50 <CRichard> where the locations of the digits move
 91 2012-02-08 03:04:56 <gmaxwell> CRichard: not helpful at all.
 92 2012-02-08 03:05:00 <splatster> Moral of the story: entirely paper wallets
 93 2012-02-08 03:05:06 <gmaxwell> CRichard: the trojan will simply yank it right out of memory.
 94 2012-02-08 03:05:09 <gmaxwell> Alas.
 95 2012-02-08 03:05:09 <smickles> 'cause screen readers don't exist
 96 2012-02-08 03:05:25 <NxTitle> gmaxwell: it'll stop generic keyloggers from getting it, at the very least
 97 2012-02-08 03:05:31 <NxTitle> so it does help, but doesn't solve it
 98 2012-02-08 03:05:38 <CRichard> that is true, a very specialized trojan could
 99 2012-02-08 03:05:44 <gmaxwell> (worse a pin pad thing would severely discourage passwords strong enough to withstand even the weakest search)
100 2012-02-08 03:05:55 <splatster> 0000
101 2012-02-08 03:05:57 <splatster> 0001
102 2012-02-08 03:05:59 <splatster> 0002
103 2012-02-08 03:06:02 <splatster> ...
104 2012-02-08 03:06:04 <splatster> not hard
105 2012-02-08 03:06:05 <gmaxwell> NxTitle: no, it's actually counterproductive due to the above. And there is _already_ specialized wallet stealing stoftware.
106 2012-02-08 03:06:11 <NxTitle> we should store all of our money on MyBitcoin
107 2012-02-08 03:06:14 <NxTitle> there, problem solved
108 2012-02-08 03:06:25 <CRichard> trusted third party
109 2012-02-08 03:06:28 <gmaxwell> Having no money does prevent people from stealing it.
110 2012-02-08 03:06:29 <splatster> NxTitle: That worked out
111 2012-02-08 03:06:32 <NxTitle> :P
112 2012-02-08 03:06:37 <splatster> gmaxwell: Exactly
113 2012-02-08 03:06:54 <splatster> Use a paper wallet.
114 2012-02-08 03:06:59 <smickles> buy that casascius 1oz gold coin
115 2012-02-08 03:07:01 <smickles> ^
116 2012-02-08 03:07:04 <splatster> ya
117 2012-02-08 03:07:14 <NxTitle> and throw it over the internet
118 2012-02-08 03:07:20 <gmaxwell> CRichard: are there any posts, messages or anything you can link me to where people are expressing concern about the risk of forgetting passwords?
119 2012-02-08 03:07:22 <splatster> But paper wallets don't come at a premium
120 2012-02-08 03:07:42 <CRichard> gmaxwell, I heard on a radio station in the US a few weeks ago when they were discussing it
121 2012-02-08 03:07:45 <CRichard> the host brought it up
122 2012-02-08 03:07:46 <smickles> splatster: but i drool for that coin
123 2012-02-08 03:07:58 <CRichard> marc stevens show
124 2012-02-08 03:08:02 <splatster> smickles: What of his coins do you already have?
125 2012-02-08 03:08:09 <gmaxwell> CRichard: Any idea what station? okay. I'll see if I can find a copy of it. Thanks!
126 2012-02-08 03:08:16 <CRichard> np
127 2012-02-08 03:08:20 <smickles> splatster: series 1 ;)
128 2012-02-08 03:08:56 <splatster> I have a couple series one and two 1 BTC coins and a 25 BTC coin with a series two hologram
129 2012-02-08 03:09:19 <CRichard> so just how much leeway does a trojan have thesedays... can it arbitrarily read the memory of another process in win7?
130 2012-02-08 03:09:22 <splatster> I think the series 2 actually looks nicer than the series one
131 2012-02-08 03:09:23 <CRichard> or only the screen?
132 2012-02-08 03:09:37 <splatster> If it's windows, it can get at anything
133 2012-02-08 03:09:39 <smickles> ooh, nifty splatster, i'll catch up to you one day
134 2012-02-08 03:10:15 <CRichard> @splatster: do you know much about win7 security layer?
135 2012-02-08 03:10:15 <splatster> I'm gonna build a display case for that shit
136 2012-02-08 03:10:17 <gmaxwell> CRichard: Yes.
137 2012-02-08 03:10:33 <CRichard> doesn't it need admin rights to do that now?
138 2012-02-08 03:10:37 <gmaxwell> CRichard: because we have to assume the trojan has managed to get itself superuser rights. Alas.
139 2012-02-08 03:10:42 <CRichard> ah
140 2012-02-08 03:10:43 <splatster> windows doesn't do admin rights
141 2012-02-08 03:11:02 <splatster> At least not really
142 2012-02-08 03:11:15 <gmaxwell> splatster: they've worked really hard at that but it's like plugging a dam. :)
143 2012-02-08 03:11:39 <splatster> gmaxwell: *it's like plugging a dam with chewing gum
144 2012-02-08 03:11:52 <BlueMatt> windows admin rights arent like plugging a dam, its more like plugging a river
145 2012-02-08 03:11:57 <CRichard> well if we assume the trojan is either in the kernel or has special rights then I don't think there's a solution
146 2012-02-08 03:12:00 <gmaxwell> Hell, try plugging a dam with 50 people and a backhoe. It's still not easy. :)
147 2012-02-08 03:12:11 <gmaxwell> CRichard: there is. Tada. :)
148 2012-02-08 03:12:21 <gmaxwell> CRichard: Two-factor wallets.
149 2012-02-08 03:12:24 <CRichard> I mean once you get to the point where you can see and edit any memory in any process you can just change the cpu instructions and do as you please
150 2012-02-08 03:12:29 <gmaxwell> Which we're in the slow process of deploying.
151 2012-02-08 03:12:39 <BlueMatt> (bip16 was a part of that)
152 2012-02-08 03:12:54 <CRichard> how does two factor encryption solve the problem?
153 2012-02-08 03:13:19 <FROTUSCI> no wallet, n oproblem
154 2012-02-08 03:13:25 <gmaxwell> With a two-factor wallet you can have a wallet which requires multiple signatures  e.g. one from your desktop one from your smart phone,  ... or one from your desktop, one from a service that SMSes you first.
155 2012-02-08 03:13:48 <CRichard> ah I see
156 2012-02-08 03:14:13 <BlueMatt> we need split-able deterministic wallets
157 2012-02-08 03:14:16 <CRichard> thats quite clever
158 2012-02-08 03:14:17 <gmaxwell> (the system is more general and can support things like two of three, or (A and B) or C.
159 2012-02-08 03:14:21 <gmaxwell> )
160 2012-02-08 03:15:03 <gmaxwell> CRichard: so even if both your desktop and phone are compromised if the same badguy doesn't control both, you should be okay. :)
161 2012-02-08 03:15:16 <splatster> *it's like plugging a dam filled with liquid helium at superfluid temperature
162 2012-02-08 03:15:46 <CRichard> it's a good solution
163 2012-02-08 03:16:24 <BlueMatt> splatster: thats about right
164 2012-02-08 03:17:13 <gmaxwell> CRichard: it'll be a while (6-9 months probably) before it's really ready for most users.. but the basic functionality is now available as a CLI only feature in the 0.6 release candidate stuff.
165 2012-02-08 03:17:24 <FROTUSCI> nice
166 2012-02-08 03:17:32 <splatster> 0.6 is already RC?
167 2012-02-08 03:17:37 <BlueMatt> yea
168 2012-02-08 03:17:40 <BlueMatt> though no builds yet
169 2012-02-08 03:17:46 <splatster> awww
170 2012-02-08 03:17:48 <splatster> damn
171 2012-02-08 03:18:11 <CRichard> would you be introducing a two wallet system? Like a safe and a purse?
172 2012-02-08 03:18:11 <gmaxwell> BlueMatt: http://sourceforge.net/projects/bitcoin/files/Bitcoin/bitcoin-0.6.0/test/
173 2012-02-08 03:18:18 <CRichard> so you can still spend without two factor, for small amounts
174 2012-02-08 03:18:21 <splatster> Are there -any- UI changes in 0.6
175 2012-02-08 03:18:26 <BlueMatt> oh, damn did gavin do gitian builds?
176 2012-02-08 03:18:32 <gmaxwell> splatster: some UI fixes.
177 2012-02-08 03:18:46 <splatster> gmaxwell: But does it look the same?
178 2012-02-08 03:18:51 <gmaxwell> BlueMatt: it's still not official yet, gavin was asking for any comments on the readme.
179 2012-02-08 03:19:02 <BlueMatt> well first it says "NEW FEATURES SINCE BITCOIN VERSION 0.5"
180 2012-02-08 03:19:34 <gmaxwell> CRichard: Yes, you'd be able to have both at once. I don't think anyone has talked about what they're thinking the UI would be like.
181 2012-02-08 03:20:13 <etotheipi_> splatster, did you see on the forum that someone added extra detail to Armory compiling on OSX?
182 2012-02-08 03:20:24 <splatster> etotheipi_: No...
183 2012-02-08 03:20:25 <BlueMatt> oh, also did the cmd parsing changes change bitcoind's upnp default to on?
184 2012-02-08 03:20:28 <splatster> I will go check it out
185 2012-02-08 03:20:32 <etotheipi_> https://bitcointalk.org/index.php?topic=56424.msg735527#msg735527
186 2012-02-08 03:21:05 <etotheipi_> I updated the code based on Joric's findings, then this guy modified the instructions with his own experience
187 2012-02-08 03:21:15 <gmaxwell> BlueMatt: we really should document all these defaults someplace other than the code, I'd totally forgotten UPNP was only default for the GUI.
188 2012-02-08 03:21:26 <BlueMatt> heh
189 2012-02-08 03:21:37 <BlueMatt> yea, help output needs some help
190 2012-02-08 03:22:07 <BlueMatt> we still need a big long comment with all the undocumented options for developer reference
191 2012-02-08 03:22:09 <gmaxwell> A lot of people have never seen help on bitcoind due to the cli multiplexing.
192 2012-02-08 03:22:22 <gmaxwell> yea. Logtimestamps <3
193 2012-02-08 03:22:42 <BlueMatt> why isnt logtimestamps default?
194 2012-02-08 03:22:42 <CRichard> @gmaxell, thanks for the info
195 2012-02-08 03:22:45 <etotheipi_> gmaxwell, I have more information on that tx that the Satoshi client produced that was invalid
196 2012-02-08 03:22:53 <BlueMatt> (other than threading issues)
197 2012-02-08 03:23:06 <etotheipi_> I figured out why it's invalid, and it definitely looks like a Satoshi-client bug
198 2012-02-08 03:23:15 <gmaxwell> BlueMatt: makes it profitable for logfile stealing ninjas to gather logs to deanonymize people
199 2012-02-08 03:23:18 <FROTUSCI> peniz?
200 2012-02-08 03:23:26 <BlueMatt> gmaxwell: oh, ofc
201 2012-02-08 03:23:40 <splatster> "Warning: It appears you have MacPorts or Fink installed."
202 2012-02-08 03:24:05 <splatster> I am going to just format this drive and start over at some point
203 2012-02-08 03:24:49 <BlueMatt> thats why you do everything in a vm  -snapshots :)
204 2012-02-08 03:25:01 <FROTUSCI> im not gay man
205 2012-02-08 03:25:07 <FROTUSCI> stop msg'ing me
206 2012-02-08 03:25:44 <theymos> etotheipi_: Was it a double-spend?
207 2012-02-08 03:26:28 <splatster> You know when you see "ChanServ sets mode +o ____" someone is about to get kicked
208 2012-02-08 03:26:42 <BlueMatt> what other use does anyone use +o for?
209 2012-02-08 03:26:44 <etotheipi_> theymos, http://pastebin.com/2tTbEsit
210 2012-02-08 03:26:47 <BlueMatt> (aside from topics)
211 2012-02-08 03:27:14 <splatster> BlueMatt: for +b
212 2012-02-08 03:27:17 <splatster> and -b
213 2012-02-08 03:27:23 <etotheipi_> theymos, this transaction was zero-fee, spending every last Satoshi from my wallet
214 2012-02-08 03:27:38 <BlueMatt> splatster: meh, I prefer to not ban
215 2012-02-08 03:27:51 <etotheipi_> it looks like it somehow duplicated the top address, which doesn't appear to have ever been seen by the network
216 2012-02-08 03:27:58 <nanotube> CRichard: btw, about that 'security questions' bit... that's not really two factor. it's called 'wish it was two factor' [but really it isn't]. http://thedailywtf.com/Articles/WishItWas-TwoFactor-.aspx
217 2012-02-08 03:28:16 <etotheipi_> theymos, 1GJKutD has never been seen by the network, but 1KWffnjs has the amount it claims in the screenshot
218 2012-02-08 03:28:30 <gmaxwell> :-/
219 2012-02-08 03:28:47 <theymos> Pretty strange. I've never heard of that before.
220 2012-02-08 03:28:54 <etotheipi_> oh, I just realized that it's not a duplicate
221 2012-02-08 03:28:59 <etotheipi_> it's very close, though
222 2012-02-08 03:29:33 <etotheipi_> That screenshot is from Armory, which keeps ending up with the tx everytime localhost rebroadcasts the tx...
223 2012-02-08 03:30:04 <etotheipi_> I suppose there could be a problem with Armory (though I haven't seen such a problem before), so you could take the raw hex from the pastebin and examine the tx yourself
224 2012-02-08 03:30:05 <gmaxwell> etotheipi_: I expect that you sent it some junk transaction while developing that it now has in the wallet.
225 2012-02-08 03:30:24 <etotheipi_> gmaxwell, this has nothing to do with Armory
226 2012-02-08 03:30:32 <gmaxwell> And if it passed ismine on the front side it would try to do spend it with zero confirms.
227 2012-02-08 03:30:45 <splatster> etotheipi_: homebrew is making cryptopp
228 2012-02-08 03:30:47 <etotheipi_> it was only when I started using Armory (never imported any shared addresses), that I tried to empty my wallet into armory
229 2012-02-08 03:31:20 <gmaxwell> etotheipi_: please don't be defensive. I wasn't blaming armory there.
230 2012-02-08 03:31:36 <etotheipi_> gmaxwell, I didn't mean to be defensive, I just wanted to be clear that there was no interaction between programs here
231 2012-02-08 03:31:43 <etotheipi_> at least not as far as I can tell...
232 2012-02-08 03:31:49 <gmaxwell> I'm just suggesting a rational reason why your client would try to spend a non-existing transaction.
233 2012-02-08 03:32:16 <splatster> dun dun dun
234 2012-02-08 03:32:22 <etotheipi_> I don't know what you mean by "junk transaction?"
235 2012-02-08 03:32:37 <gmaxwell> etotheipi_: who knows, something which could never confirm for whatever reason.
236 2012-02-08 03:32:51 <etotheipi_> I definitely sent transactions from Armory to my Satoshi wallet...
237 2012-02-08 03:33:18 <gmaxwell> The client will attempt to spend zero-confirmed inputs which it believes it made "itself", if it needs to do so to get the amount requested.
238 2012-02-08 03:33:19 <etotheipi_> but how did the Satoshi client decide it was isMine?
239 2012-02-08 03:33:51 <etotheipi_> I don't think I ever imported any of my satoshi addresses to Armory
240 2012-02-08 03:34:04 <etotheipi_> I wouldn't because I know that stupid shit can happen...
241 2012-02-08 03:34:23 <etotheipi_> (perhaps I did and simply forgot...)
242 2012-02-08 03:34:59 <gmaxwell> (well really, IsFromMe() + IsMine() in this case...)
243 2012-02-08 03:35:44 <gmaxwell> etotheipi_: so the criteria appears to be that the prev input passed IsMine.
244 2012-02-08 03:35:50 <BlueMatt> luke-jr: did you ever figure out what the indeterminisms were from?
245 2012-02-08 03:36:22 <gmaxwell> Ha. This actually constitutes a bit of a self inflicted trouble making vulnerability with the import function in bitcoin.
246 2012-02-08 03:36:52 <gmaxwell> If I can convince people to import some key that I have.. I can send them crazy transaction and gum up their wallets.
247 2012-02-08 03:37:10 <gmaxwell> (well, mildly crazy)
248 2012-02-08 03:38:02 <gmaxwell> etotheipi_: in any case, if this is the cause then at the time your stuck txn was generated your reference client wallet would have had to had the input there.
249 2012-02-08 03:38:20 <gmaxwell> etotheipi_: can you see if there is more than the one transaction in listtransactions with zero confirmations?
250 2012-02-08 03:39:13 <etotheipi_> actually, if that 1GJKutDW address was mine, it would show up with a walletID
251 2012-02-08 03:39:34 <etotheipi_> okay, let me see if I remember how to use the RPC interface...
252 2012-02-08 03:39:43 <theymos> Can you dump b3dfaa0?
253 2012-02-08 03:40:15 <gmaxwell> etotheipi_: ./bitcoind listtransactions     (perhaps with a "" 1000 tossed on for fun)
254 2012-02-08 03:40:35 <splatster> etotheipi_: homebrew is installing qt
255 2012-02-08 03:40:42 <etotheipi_> splatster, that's good news
256 2012-02-08 03:40:48 <splatster> s/installing/downloading/
257 2012-02-08 03:41:03 <splatster> after that it's just pyqt
258 2012-02-08 03:41:22 <luke-jr> BlueMatt: probably the time. I don't know anything about your faketime lib
259 2012-02-08 03:41:24 <splatster> then some other steps and then actually building armory
260 2012-02-08 03:41:43 <BlueMatt> luke-jr: did you diff the files?
261 2012-02-08 03:41:53 <splatster> 25% done with the download! FUCK YEA!
262 2012-02-08 03:42:00 <luke-jr> BlueMatt: they were significantly different. the first few were clearly timestamps though
263 2012-02-08 03:42:07 <splatster> This is gonna take forever
264 2012-02-08 03:42:17 <BlueMatt> luke-jr: were any of the contained files different?
265 2012-02-08 03:42:18 <splatster> and ever
266 2012-02-08 03:42:27 <luke-jr> BlueMatt: yes of course
267 2012-02-08 03:42:30 <splatster> all the way to
268 2012-02-08 03:42:34 <BlueMatt> luke-jr: oh, damn
269 2012-02-08 03:42:40 <BlueMatt> luke-jr: can you upload yours?
270 2012-02-08 03:42:44 <luke-jr> sure
271 2012-02-08 03:43:04 <etotheipi_> gmaxwell, there's only a few transactions that show up with "listtransactions"
272 2012-02-08 03:43:17 <luke-jr> http://luke.dashjr.org/tmp/code/bitcoin-deps-0.0.2.tbz2 in 30 seconds
273 2012-02-08 03:44:39 <etotheipi_> how do I get the zero-conf tx out of it?
274 2012-02-08 03:47:27 <gmaxwell> etotheipi_: is there more than one?  (even if you give a "" 1000 argument to listtransactions)
275 2012-02-08 03:47:38 <CRichard> for someone writing their own client, what would be the best avenue for research
276 2012-02-08 03:47:41 <etotheipi_> gmaxwell, I don't see any zero-conf tx
277 2012-02-08 03:47:43 <gmaxwell> etotheipi_: and to get it out there are "instructions" on the forum. It's mostly horrible. :)
278 2012-02-08 03:47:46 <splatster> twoot 75%
279 2012-02-08 03:47:47 <CRichard> should I just grab the official code you guys are working on?
280 2012-02-08 03:47:50 <splatster> woot*
281 2012-02-08 03:48:02 <gmaxwell> etotheipi_: none 0_o not even the one you keep rebroadcasting?
282 2012-02-08 03:48:43 <etotheipi_> listtransactions gives me a short list, with the lowest number of confirmations being 378
283 2012-02-08 03:48:46 <gmaxwell> CRichard: the reference client itself and the bitcoin wiki are the best information currently available..
284 2012-02-08 03:49:09 <gmaxwell> etotheipi_: if you give it an argument you can ask for more, it only gives 10 by default IIRC.
285 2012-02-08 03:49:13 <CRichard> does the reference client refer to the original or http://gitorious.org/bitcoin/bitcoind-stable/archive-tarball/v0.5.2#.tar.gz ?
286 2012-02-08 03:49:29 <XMPPwocky> o/ now Bitkit relies on the actual payload to determine how long it is (not just the payload_length in the message header)
287 2012-02-08 03:49:48 <etotheipi_> if I add "" 1000 it only gives me tx with 1000 conf or more
288 2012-02-08 03:49:51 <gmaxwell> CRichard: or really, the source hosted here: https://github.com/bitcoin/bitcoin
289 2012-02-08 03:50:07 <CRichard> ok thanks
290 2012-02-08 03:50:54 <gmaxwell> etotheipi_: er, what version is that? "works for me"
291 2012-02-08 03:50:55 <etotheipi_> I have a tx with 50386 confirmations... I wonder if I can trust it
292 2012-02-08 03:51:09 <BlueMatt> probably not
293 2012-02-08 03:51:11 <gmaxwell> I don't remember that changing rhough.
294 2012-02-08 03:51:19 <theymos> Try gettransaction b3dfaa03828eb9cc698edc41f29544653662ae268de33331639c15499884b338
295 2012-02-08 03:52:49 <etotheipi_> version 0.5.0rc7-linux
296 2012-02-08 03:53:57 <etotheipi_> wtf?  I tried that txid, both endians... I get nothing
297 2012-02-08 03:54:20 <etotheipi_> I'm really confused now
298 2012-02-08 03:54:58 <theymos> Do you have 6acfd032e78a94d7611b5a2b00f190bc0c970224b39817562f3b1f1a7de6c9ee?
299 2012-02-08 03:55:08 <gmaxwell> etotheipi_: listtransactions count argument works fine for me even back in 0.3.19.
300 2012-02-08 03:56:29 <splatster> etotheipi_: Building pyqt
301 2012-02-08 03:57:12 <etotheipi_> I wonder if Armory somehow miscalculated the tx
302 2012-02-08 03:57:24 <etotheipi_> because that hash isn't showing up
303 2012-02-08 03:57:36 <etotheipi_> but I"m not seeing it with listtransactions either
304 2012-02-08 03:57:41 <etotheipi_> is there a way to look up by address?
305 2012-02-08 03:59:32 <etotheipi_> okay, even this tx itself is not showing up under listtransactions...
306 2012-02-08 03:59:49 <etotheipi_> so neither of them are showing up
307 2012-02-08 03:59:52 <gmaxwell> if you could get the count parameter working you could just look at all of them.
308 2012-02-08 04:00:17 <theymos> Grep wallet.dat for the hash. (Not sure which endianness it uses.)
309 2012-02-08 04:01:08 <etotheipi_> I just tried grepping both endiannesses... no dice
310 2012-02-08 04:01:21 <etotheipi_> and the tx that showed up are definitely main-network
311 2012-02-08 04:01:31 <Diablo-D3> kernel hacking: its similar to doing drugs
312 2012-02-08 04:01:33 <etotheipi_> (other tx)
313 2012-02-08 04:03:40 <etotheipi_> so not only is this invalid tx a mystery... but my client seems to be f***ed up
314 2012-02-08 04:03:53 <etotheipi_> but it clearly shows the tx on the UI
315 2012-02-08 04:04:06 <etotheipi_> but the UI gives me no way to examine the txid
316 2012-02-08 04:04:10 <theymos> Run bitcoin with -debug and double-click the tx in the UI.
317 2012-02-08 04:05:03 <BlueMatt> 0.6 ui does
318 2012-02-08 04:05:58 <etotheipi_> okay, this is getting weirder, I think
319 2012-02-08 04:07:06 <etotheipi_> I pasted the debug to the end of this pastebin: http://pastebin.com/6aD2MY0M
320 2012-02-08 04:07:22 <splatster> Holy SHIT!
321 2012-02-08 04:07:24 <etotheipi_> that 1GJKutDWn doesn't appear
322 2012-02-08 04:07:53 <splatster> damn
323 2012-02-08 04:08:04 <splatster> etotheipi_: "Fatal Python error: Interpreter not initialized (version mismatch?)"
324 2012-02-08 04:08:06 <etotheipi_> 8 debits, but only 7 addresses shown at the bottom
325 2012-02-08 04:08:15 <etotheipi_> splatster, what version of python?
326 2012-02-08 04:08:20 <splatster> etotheipi_: 2.7
327 2012-02-08 04:08:38 <splatster> etotheipi_: Should I try restarting or something?
328 2012-02-08 04:08:46 <etotheipi_> splatster, I guess it couldn't hurt
329 2012-02-08 04:08:51 <etotheipi_> but that's probably not it
330 2012-02-08 04:09:05 <splatster> What if my uptime is 22d 10h?
331 2012-02-08 04:09:24 <etotheipi_> I don't know...
332 2012-02-08 04:09:33 <splatster> ok be back in a few
333 2012-02-08 04:10:04 <etotheipi_> theymos, so why does the debug show 8 debits and 8 CTxIn's, but only 7 addresses under "Inputs"?
334 2012-02-08 04:12:06 <theymos> I don't know. I don't think wxbitcoin would have done this. Maybe things changed in qt.
335 2012-02-08 04:13:08 <gmaxwell> presumably because you don't have one of the inputs.
336 2012-02-08 04:13:25 <gmaxwell> so it can't resolve it to print things about it.
337 2012-02-08 04:13:55 <gmaxwell> etotheipi_: you've never removed a key from this wallet, have you?
338 2012-02-08 04:14:35 <etotheipi_> gmaxwell, I don't know how
339 2012-02-08 04:14:51 <etotheipi_> I avoid messing with Satoshi wallets at all costs
340 2012-02-08 04:15:09 <gmaxwell> You can't except via voodoo. Okay.
341 2012-02-08 04:15:22 <gmaxwell> etotheipi_: woah wait.
342 2012-02-08 04:15:41 <gmaxwell> how is armory giving a 'sender' for that mystery input. You said that transaction didn't exist?
343 2012-02-08 04:15:51 <gmaxwell> You can't know a sender for a non-existing transaction!
344 2012-02-08 04:16:20 <etotheipi_> it probably picked it up from the signature
345 2012-02-08 04:16:26 <etotheipi_> err.. public key in the sig
346 2012-02-08 04:17:11 <gmaxwell> ah, sender was meaning something else to me there....
347 2012-02-08 04:18:11 <etotheipi_> yeah, I have a getTxInAddressIfAvail method
348 2012-02-08 04:19:33 <BlueMatt> whats with all the net splits tonight?
349 2012-02-08 04:20:22 <etotheipi_> gmaxwell, ... actually that method isn't used here... it looks like it *is* pulling it from a previous output
350 2012-02-08 04:24:57 <etotheipi_> okay, nevermind, I was looking at the wrong code:  it's definitely getting the address from the sigscript field
351 2012-02-08 04:25:07 <gmaxwell> whew okay.
352 2012-02-08 04:25:49 <splatster> etotheipi_: Same error
353 2012-02-08 04:26:27 <etotheipi_> splatster, google it... I can't really help but I'm sure you're not the first person to have this issue
354 2012-02-08 04:27:15 <etotheipi_> gmaxwell, I can't find the address in my "Receive coins" list or my "Address Book"
355 2012-02-08 04:28:00 <etotheipi_> gmaxwell, I don't *think* I did this... but what would happen if I sent money to an address in the memory pool?
356 2012-02-08 04:28:51 <gmaxwell> Sent to? .. then .. you'd send to it. I suspect I'm not following you!
357 2012-02-08 04:29:26 <etotheipi_> I'm pretty sure I've printed out my wallet addresses before (like when I was searching the wallet when finding the 0.4.0 encryption bug)
358 2012-02-08 04:29:48 <etotheipi_> what if I inadvertantly used one of the memory pool addresses to send money to the Satoshi client?
359 2012-02-08 04:30:15 <etotheipi_> I'm really stretching, here
360 2012-02-08 04:30:29 <gmaxwell> How would you have done that? .. oh you mean imported it? and sent from it bogusly?
361 2012-02-08 04:30:42 <etotheipi_> I don't know how my client offered a valid signature (is it valid?) from an address it doesn't have
362 2012-02-08 04:30:56 <etotheipi_> no importing...
363 2012-02-08 04:31:12 <etotheipi_> just printed out the bank of unseen 100 mem pool addresses
364 2012-02-08 04:31:17 <etotheipi_> and then sent money to one of them
365 2012-02-08 04:31:27 <gmaxwell> oh keypool.
366 2012-02-08 04:31:34 <etotheipi_> err.. keypool
367 2012-02-08 04:31:34 <gmaxwell> It's fine to send money to those.
368 2012-02-08 04:32:33 <etotheipi_> is there a way to clear the satoshi client memory pool so I could unlock those txOuts?
369 2012-02-08 04:32:39 <gmaxwell> My best theory was that some transaction went into the client it couldn't spend but it believed (1) it belonged to it, (2) "it" generated it (came from address that passed IsMine), .. then when you emptied it it tried to use it.. since then its decided that it doesn't really have that input (the result of a rescan)?
370 2012-02-08 04:32:51 <etotheipi_> I'm curious, if I try to dump my entire wallet again, if it happens again
371 2012-02-08 04:32:55 <gmaxwell> etotheipi_: it's not in the memory pool. It's in the wallet.
372 2012-02-08 04:32:57 <BlueMatt> sipa: do you have to define  _WIN32_WINNT in netbase.h?
373 2012-02-08 04:33:40 <etotheipi_> so zero-conf tx are saved in the wallet?
374 2012-02-08 04:33:42 <gmaxwell> With the client offline, copy off the wallet.dat (please do keep a copy of the weird one), restore some old one that has the same keys.
375 2012-02-08 04:33:45 <gmaxwell> etotheipi_: yes.
376 2012-02-08 04:33:46 <BlueMatt> (its already predefined in windef.h (from windows.h) in mingw
377 2012-02-08 04:34:08 <gmaxwell> (your own zero-conf tx.. the memory pool is not saved. it's .. in memory!)
378 2012-02-08 04:34:43 <etotheipi_> gah, I'm going to have to deal with this tomorrow
379 2012-02-08 04:35:00 <etotheipi_> I need to get up "early" tomorrow
380 2012-02-08 04:35:25 <gmaxwell> oh, good point. Goodnight!
381 2012-02-08 04:35:44 <etotheipi_> although I am curious how it is a value that so closely resembles that other address, which is legit
382 2012-02-08 04:35:51 <BlueMatt> ;;later tell gavinandresen can you merge https://github.com/TheBlueMatt/bitcoin/commit/490bef8bff62bebf5236ae271c2ac3fb4e849bbb before 0.6, I hate having compile warnings in release builds
383 2012-02-08 04:35:54 <gribble> The operation succeeded.
384 2012-02-08 04:36:13 <gmaxwell> etotheipi_: I don't think it's that close, I think you're seeing patterns. :)
385 2012-02-08 04:38:07 <etotheipi_> is there a way to definitively check whether that address exists in my wallet?
386 2012-02-08 04:38:20 <etotheipi_> or rather, that my wallet has the private key for it?
387 2012-02-08 04:39:18 <splatster> etotheipi_: Please look at the call stack: http://pastebin.com/WiTKtaVG
388 2012-02-08 04:39:42 <gmaxwell> etotheipi_: attempt to dump the private key.
389 2012-02-08 04:39:50 <splatster> CppBlockUtils is in some way related to the crash
390 2012-02-08 04:39:51 <gmaxwell> (dumpprivkey rpc)
391 2012-02-08 04:40:07 <etotheipi_> gmaxwell, that's part of 0.5.0?
392 2012-02-08 04:40:45 <etotheipi_> splatster, none of that makes sense to me... besides the possibility that you're on a 32-bit OS trying to run a 64-bit app
393 2012-02-08 04:40:51 <etotheipi_> or vice versa
394 2012-02-08 04:41:46 <etotheipi_> splatster, did you run the export commands as it said in the post?
395 2012-02-08 04:41:54 <gmaxwell> alas, no I guess not. I'm not actually sure of a way to answer that question in 0.5
396 2012-02-08 04:41:59 <splatster> etotheipi_: Any thougts?
397 2012-02-08 04:42:27 <etotheipi_> splatster, does "uname -a" work in OSX?
398 2012-02-08 04:42:32 <etotheipi_> if so, tell me the output
399 2012-02-08 04:45:16 <etotheipi_> splatster, btw, to get as far as you did, did you need any of Joric's original directions?  are the directions that Torus posted sufficient alone?  or do I need to merge them?
400 2012-02-08 04:45:32 <splatster> Re-re-re-re-retrying to build and run Armory
401 2012-02-08 04:46:02 <etotheipi_> splatster, we'll talk more later, I *really* need to go to sleep
402 2012-02-08 04:46:16 <splatster> This is a broken record-ecord-ecord-ecord
403 2012-02-08 04:47:36 <splatster> etotheipi_: Ya I will compile my own set of foolproof instructions based on my experiences and both of theirs
404 2012-02-08 04:47:57 <splatster> Compiling...
405 2012-02-08 04:48:08 <etotheipi_> okay, I'll check-in tomorrow after work
406 2012-02-08 04:48:31 <splatster> I'll send you a MemoServ message if I get it to work
407 2012-02-08 04:48:38 <etotheipi_> btw, it looks like my memory-reduction activities are going smoothly
408 2012-02-08 04:48:44 <splatster> Cool
409 2012-02-08 04:48:55 <BlueMatt> arg, when people ask questions on the mailing list, I wish they came here instead
410 2012-02-08 04:49:16 <splatster> Why do people ask questions on the mailing list?
411 2012-02-08 04:49:24 <splatster> IRC is where it's at
412 2012-02-08 04:49:30 <etotheipi_> mmap is a lifesaver!  I just don't know for sure if it's doing what I want, because my system has so much RAM, I think it puts it all in RAM anyway
413 2012-02-08 04:49:36 <BlueMatt> no offense to grarpamp, but asking what plans are for future client optimization doesnt really belong on the mailing list
414 2012-02-08 04:50:55 <splatster> etotheipi_: I'm sure you'll figure it out
415 2012-02-08 04:50:58 <etotheipi_> I need to test the solution on a system with much more limited RAM, to see if it lowers the cache appropriately
416 2012-02-08 04:51:06 <splatster> OS specific shit is the toughest to deal with
417 2012-02-08 04:51:37 <splatster> BlameGame
418 2012-02-08 04:51:40 <splatster> oops
419 2012-02-08 04:51:51 <etotheipi_> anyone know if there's anything I need to do explicitly to get MMAP to start freeing data?  Or does it auto-detect available RAM and set cache-size appropriately?
420 2012-02-08 04:51:52 <splatster> sorry bout that
421 2012-02-08 04:52:41 <etotheipi_> it looks like I can mark regions "DONTNEED" and it will mark those for clean up... but if I never mark any of the mmap'd file as such, I assume it will start limiting the amount of cache it uses under memory pressure...?
422 2012-02-08 04:53:18 <splatster> In my experience, never trust things to perform in any sort of logical way
423 2012-02-08 04:53:29 <splatster> It would make too much sense for it to do that
424 2012-02-08 04:53:33 <FROTUSCI-> bitcoin hashdos. discuss
425 2012-02-08 04:53:57 <etotheipi_> splatster, mmap is a very low-level, kernel-integrated method, used by almost every process on your system
426 2012-02-08 04:54:11 <splatster> damn same errors
427 2012-02-08 04:54:17 <etotheipi_> if anything would be super-intelligently optimized, I expect mmap would be
428 2012-02-08 04:54:20 <splatster> gotta make it into a binary
429 2012-02-08 04:54:38 <FROTUSCI-> what do you mean back? ive been here for ages
430 2012-02-08 04:54:55 <BlueMatt> back making random comments?
431 2012-02-08 05:03:39 <gargadons> I heard BlueMatt's mom sucked off a horse
432 2012-02-08 05:03:40 <amiller> who was it that was talking to me about homomorphic encryption and 'private' bitcoin with me yesterday
433 2012-02-08 05:03:57 <amiller> damn
434 2012-02-08 05:04:08 <amiller> oh i have logs nvm
435 2012-02-08 05:04:46 <SomeoneWeird> 0.o
436 2012-02-08 05:04:57 <BlueMatt> and no, I have nfc who that was
437 2012-02-08 05:07:45 <FROTUSCI-> amiller, we were chatting about that w/gmaxwell werent we
438 2012-02-08 05:08:10 <FROTUSCI-> balance = g^a
439 2012-02-08 05:11:19 <SomeoneWeird> g**a
440 2012-02-08 05:11:51 <SomeoneWeird> WEEEEEE
441 2012-02-08 05:11:55 <SomeoneWeird> creation kit came out i think
442 2012-02-08 05:49:43 <is4tomj> Is there a good place to find sponsors for bitcoin-related sites?
443 2012-02-08 05:50:12 <is4tomj> I don't want to use an ad-network b/c I want to find sponsors for individual BTCUSD visualizations on my homepage.
444 2012-02-08 05:55:24 <AAA_awright> is4tomj: Probably advertising on your own website, then you attract sponsors who already use your product.
445 2012-02-08 08:51:57 <gribble> The operation succeeded.
446 2012-02-08 08:51:57 <sipa> ;;later tell BlueMatt i need that macro for getaddrinfo, so it has to be defined somewhere, and because of modularity, i prefer to do so where it is needed
447 2012-02-08 09:52:56 <Blitzboom> whats the status on BIP 16/17? did anything change yet?
448 2012-02-08 09:53:30 <sipa> http://blockchain.info/p2sh
449 2012-02-08 09:54:01 <Blitzboom> oh boy
450 2012-02-08 09:55:35 <Blitzboom> and deepbit only wants to help when he already sees a majority, right
451 2012-02-08 09:56:24 <sipa> indeed...
452 2012-02-08 09:59:00 <Blitzboom> ok, thanks
453 2012-02-08 10:38:43 <sipa> does anyone have any idea as to why StopNode() doesn't check whether ThreadOpenConnections is still running?
454 2012-02-08 10:39:07 <sipa> it's been that way since at least v0.1.5
455 2012-02-08 12:08:56 <upb> sipa: to allow for use after free exploits i guess
456 2012-02-08 12:11:26 <sipa> ?
457 2012-02-08 13:00:01 <amstan> is gavin in here? there's a reddit iama request for him: http://www.reddit.com/r/IAmA/comments/pftz8/iama_request_gavin_andresen_lead_developer_of_the/
458 2012-02-08 13:03:12 <sipa> amstan: he comes here occassionally
459 2012-02-08 13:04:32 <sipa> ;;later tell gavinandresen < amstan> is gavin in here? there's a reddit iama request for him: http://www.reddit.com/r/IAmA/comments/pftz8/iama_request_gavin_andresen_lead_developer_of_the/
460 2012-02-08 13:04:33 <gribble> The operation succeeded.
461 2012-02-08 13:29:21 <cjd> hmm that's no good
462 2012-02-08 13:29:47 <cjd> Moron__ is now Idiot__, that's an IQ drop of 25 points :(
463 2012-02-08 13:32:06 <Joric> you skipped imbecile (iq of 26-50)
464 2012-02-08 13:32:41 <Joric> moron -> imbecile -> idiot each loses ~25 iq on the way
465 2012-02-08 13:35:50 <cjd> oh indeed, it's worse than I thought
466 2012-02-08 13:36:30 <sipa> at least he's honest
467 2012-02-08 13:37:00 <sipa> damn, i was wondering what this BTC-related domain he is on was... BTCentralplus.com
468 2012-02-08 13:37:23 <cjd> heh
469 2012-02-08 14:00:57 <Idiot__> hello cjd :)
470 2012-02-08 14:01:22 <Idiot__> oh i forgot about imbecile, my bad :)
471 2012-02-08 14:12:44 <Idiot__> bitcoins up ;)
472 2012-02-08 14:26:06 <Idiot__> bidcoins down
473 2012-02-08 15:37:29 <k9quaint> Diablo-D3: you to make fancy gui for your slow miner!
474 2012-02-08 15:38:49 <Diablo-D3> k9quaint: Im thinking about adding full fledged cluster support
475 2012-02-08 16:03:11 <BlueMatt> ;;seen sipa
476 2012-02-08 16:03:12 <gribble> sipa was last seen in #bitcoin-dev 2 hours, 26 minutes, and 11 seconds ago: <sipa> damn, i was wondering what this BTC-related domain he is on was... BTCentralplus.com
477 2012-02-08 16:07:09 <BlueMatt_> sipa: mind putting it in a #ifndef after including windows.h then?
478 2012-02-08 16:07:36 <sipa> why?
479 2012-02-08 16:08:17 <sipa> i doubt it will still work then
480 2012-02-08 16:11:52 <BlueMatt_> sorry, hadnt read the code, I just assumed it was defined, not defined to something useful
481 2012-02-08 16:11:53 <BlueMatt_> one sec
482 2012-02-08 16:12:31 <luke-jr> ugh, this DDoS is annoying
483 2012-02-08 16:12:40 <luke-jr> "Jeremy Miller: [12:01] If you are receiving a 20 Gbps attack the price would be $30,000/mo"
484 2012-02-08 16:14:22 <sipa> to do the attack, or to protect against it?
485 2012-02-08 16:14:37 <luke-jr> protect against
486 2012-02-08 16:14:45 <BlueMatt> you are getting a 20 Gbps ddos?
487 2012-02-08 16:14:48 <luke-jr> yes
488 2012-02-08 16:14:56 <BlueMatt> damn, that sucks
489 2012-02-08 16:15:01 <[eval]> ew
490 2012-02-08 16:15:16 <[eval]> eligius is? are other pools?
491 2012-02-08 16:15:52 <luke-jr> just Eligius afaik
492 2012-02-08 16:15:53 <[eval]> i did notice the last couple blocks have come fairly slowly :(
493 2012-02-08 16:16:03 <BlueMatt> thtats probably not why
494 2012-02-08 16:16:06 <luke-jr> ozcoin and BTCServ were wallet-stolen
495 2012-02-08 16:16:14 <[eval]> ozcoin too?!
496 2012-02-08 16:17:43 <BlueMatt> sipa: nvm, really #include "netbase.h" just needs moved up above uint256.h in util.h
497 2012-02-08 16:17:51 <BlueMatt> so that you wont redefine
498 2012-02-08 16:20:08 <BlueMatt> ;;seen gavinandresen
499 2012-02-08 16:20:09 <gribble> gavinandresen was last seen in #bitcoin-dev 18 hours, 14 minutes, and 57 seconds ago: <gavinandresen> I'm out for tonight, email or gribble me if you have comments on the 0.6 release notes
500 2012-02-08 16:39:54 <topi`> luke-jr: maybe you need a custom firewall that would prefer connections from a whitelist (compiled from the best contributors' ips) and then accept other connections only when there are < N connection attempts per second
501 2012-02-08 16:40:12 <topi`> so it would still work for some under ddos
502 2012-02-08 16:40:38 <luke-jr> topi`: 20 Gbit
503 2012-02-08 16:40:57 <BlueMatt> for some reason I doubt luke has a 20 Gbps port to his isps router to be able to even implement such a thing
504 2012-02-08 16:44:56 <gmaxwell> topi`: you could just ask the provider to only permit tcp dport 22 and 80. .. but the provider doesn't want 20gbit/sec of junk traffic either.
505 2012-02-08 16:45:14 <gmaxwell> s/80/whatever port the pool is on/
506 2012-02-08 16:45:24 <luke-jr> gmaxwell: heck, blocking UDP would do it right now
507 2012-02-08 16:45:27 <BlueMatt> you just need to start rotating ips and rotating names pointing to those ips often enough that the ddos wont follow quickly enough
508 2012-02-08 16:45:35 <luke-jr> BlueMatt: they're in #Eligius
509 2012-02-08 16:45:39 <BlueMatt> also, point the dns names the ddos is targeted at at someone exiting
510 2012-02-08 16:45:48 <luke-jr> within minutes of announcing a new URI privately there, it gets attacked
511 2012-02-08 16:46:06 <BlueMatt> yea, but the botnet doesnt shift instantly
512 2012-02-08 16:46:17 <BlueMatt> it takes an hour or two for the ddos to ramp up
513 2012-02-08 16:47:02 <gmaxwell> luke-jr: take the tor bundle out of torchat and let people reach you via tor hidden services?
514 2012-02-08 16:47:24 <gmaxwell> ls
515 2012-02-08 16:47:42 <BlueMatt> ;;seen genjix
516 2012-02-08 16:47:42 <gribble> genjix was last seen in #bitcoin-dev 1 day, 21 hours, 24 minutes, and 40 seconds ago: <genjix> cya
517 2012-02-08 16:49:49 <luke-jr> gmaxwell: torchat?
518 2012-02-08 16:52:55 <gmaxwell> IIRC torchat has packaged up a sort of single use tor implementation.
519 2012-02-08 16:54:23 <sipa> luke-jr: give each a different url :)
520 2012-02-08 16:55:51 <luke-jr> sipa: DDoS is IP-based
521 2012-02-08 16:56:58 <sipa> ok, use 2 IPs, give each a different domain name, vary the A records of those domains over the two IPs, and see which one gets the DDoS :)
522 2012-02-08 16:59:28 <Idiot__> i heard eligius pool is switching over to dos... isnt that a bit of an old o/s?
523 2012-02-08 17:01:39 <sipa> Idiot__: yeah, sure
524 2012-02-08 17:02:01 <midnightmagic> luke-jr: Hrm. It is also possible that someone is running a relay from #eligius to another network via a gateway bot.
525 2012-02-08 17:02:16 <midnightmagic> luke-jr: So the actual person might not be in #eligius at all..
526 2012-02-08 17:05:06 <phantomcircuit> midnightmagic, unlikely
527 2012-02-08 17:05:38 <midnightmagic> phantomcircuit: There are I2P relay bots in #bitcoin, #bitcoin-discussion, #bitcoin-dev, and about four other channels I am also in. It is not unlikely.
528 2012-02-08 17:06:11 <phantomcircuit> no i mean it's unlikely that the person doing it has bothered
529 2012-02-08 17:06:30 <midnightmagic> ah
530 2012-02-08 18:23:59 <Matoking> Git related question, can I/how can I commit changes without adding a new commit? Like, combining the commit with the previous commit?
531 2012-02-08 18:24:30 <Matoking> Sorry if I sound unclear, Git is peeling off the little sanity I have left as we speak
532 2012-02-08 18:24:44 <NxTitle> I don't believe you can
533 2012-02-08 18:24:59 <NxTitle> well, you can but it's stupid complicated
534 2012-02-08 18:25:01 <josephcp> do you mean git "rebase"?
535 2012-02-08 18:25:03 <NxTitle> why do you want to do this?
536 2012-02-08 18:25:04 <sipa> you can rebase them afterwards
537 2012-02-08 18:25:24 <sipa> and rebase -i is quite easy
538 2012-02-08 18:25:40 <NxTitle> o
539 2012-02-08 18:25:43 <NxTitle> disregard me then
540 2012-02-08 18:37:00 <luke-jr> Matoking: git commit --amend
541 2012-02-08 18:37:07 <luke-jr> Matoking: only works for the last commit
542 2012-02-08 18:49:16 <Matoking> git rebase -i has given me merging errors everytime I've tried it
543 2012-02-08 18:49:35 <Matoking> I dunno if I'm just making it overly complicated
544 2012-02-08 18:49:48 <Matoking> Well
545 2012-02-08 18:49:51 <Matoking> Might as well try it again
546 2012-02-08 18:51:05 <Matoking> What the heck
547 2012-02-08 18:51:07 <Matoking> It worked this time
548 2012-02-08 18:53:32 <Matoking> I've got the pull request squashed into one commit
549 2012-02-08 18:53:33 <Matoking> i think
550 2012-02-08 18:53:46 <Matoking> Although it shows I've changed 87 files instead of 2
551 2012-02-08 18:54:16 <Matoking> And those changed files are from commits that have been done on the main branch on the official bitcoin repository
552 2012-02-08 18:55:04 <sipa> think you did something wrong there
553 2012-02-08 18:55:14 <sipa> what did you rebase it against?
554 2012-02-08 18:55:29 <Matoking> https://github.com/bitcoin/bitcoin/pull/795/files#diff-55
555 2012-02-08 18:55:40 <Matoking> It only has two changes to bitcoingui.h and .cpp
556 2012-02-08 18:55:59 <Matoking> I pulled everything from the main bitcoin branch
557 2012-02-08 18:56:04 <Matoking> On top of my own changes
558 2012-02-08 18:56:06 <Matoking> Rebased them together
559 2012-02-08 18:56:39 <Matoking> For example those translation file changes are from commits that are already in the main bitcoin repository
560 2012-02-08 18:56:46 <Matoking> So why it's listing them in the pull request is beyond me
561 2012-02-08 18:57:04 <sipa> seems you started from 0.6 code
562 2012-02-08 18:57:11 <sipa> but rebased againt 0.5.99 code
563 2012-02-08 18:59:49 <Matoking> It already has all the changes that are in the main bitcoin branch at the moment plus the actual change
564 2012-02-08 19:00:11 <Matoking> Either there is some technicality I haven't grasped yet or it's something else
565 2012-02-08 19:00:13 <sipa> which command line did you use?
566 2012-02-08 19:00:33 <Matoking> Git BASH as is
567 2012-02-08 19:00:42 <sipa> ?
568 2012-02-08 19:00:57 <Diablo-D3> git bash is some weird thing git for windows has
569 2012-02-08 19:01:02 <sipa> git rebase -i <something>
570 2012-02-08 19:01:05 <Diablo-D3> it spawns a functional bash prompt only useful for git
571 2012-02-08 19:01:11 <sipa> ah
572 2012-02-08 19:01:49 <sipa> but what rebase command did you use?
573 2012-02-08 19:01:57 <Matoking> git rebase -i HEAD~2
574 2012-02-08 19:02:04 <luke-jr> Matoking: you know, I *did* tell you how&
575 2012-02-08 19:02:13 <Matoking> Didn't work
576 2012-02-08 19:02:23 <luke-jr> https://github.com/bitcoin/bitcoin/pull/795#issuecomment-3829606
577 2012-02-08 19:03:11 <Matoking> Yes
578 2012-02-08 19:03:13 <Matoking> Tried all that
579 2012-02-08 19:03:22 <Matoking> Spawned a bunch of y/n prompts
580 2012-02-08 19:03:58 <Matoking> The point is
581 2012-02-08 19:04:11 <Matoking> The only difference between my branch and the official branch are a few lines of code
582 2012-02-08 19:04:30 <Matoking> So I don't get it why it has a huge list of changes
583 2012-02-08 19:05:47 <luke-jr> Matoking: there is no prompts for the commadns I gave you
584 2012-02-08 19:08:38 <Matoking> Well
585 2012-02-08 19:08:47 <Matoking> The timestamp seems to be 5 days ago instead of 23 hours ago
586 2012-02-08 19:08:56 <Matoking> Although the changes otherwise seem to be the same
587 2012-02-08 19:14:44 <Matoking> Well I'm off for now
588 2012-02-08 19:14:57 <sipa> Matoking: se
589 2012-02-08 19:14:58 <sipa> sec
590 2012-02-08 19:15:07 <Matoking> Huh
591 2012-02-08 19:15:29 <sipa> Matoking: rebase takes all commits along the path from your current point to the latest common ancestor of what you're rebasing against
592 2012-02-08 19:15:40 <sipa> and puts those on top of the point you ask
593 2012-02-08 19:16:06 <gribble> New news from bitcoinrss: Goonie opened issue 809 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/issues/809>
594 2012-02-08 19:16:07 <sipa> so what you've done now is made a patch against a version of the bitcoin repo from a few hours or days ago, with the same code
595 2012-02-08 19:16:30 <sipa> do a git fetch upstream, and git rebase -i upstream/master
596 2012-02-08 19:17:09 <Matoking> I'll just pick the only commit there is, no?
597 2012-02-08 19:17:25 <sipa> yes
598 2012-02-08 19:17:45 <sipa> there should be only one, as everything before is shared history
599 2012-02-08 19:17:51 <Matoking> And now it's warning about fast-forwarding
600 2012-02-08 19:17:56 <Matoking> When I'm trying to pushit
601 2012-02-08 19:18:27 <sipa> push it using -f
602 2012-02-08 19:18:56 <Matoking> Oh sweet
603 2012-02-08 19:19:00 <Matoking> Everything seems to be in order
604 2012-02-08 19:19:02 <sipa> fast-forwarding means only adding commits, not removing/changing older ones
605 2012-02-08 19:19:03 <Matoking> Thanks very much
606 2012-02-08 19:19:13 <sipa> typically when you rebase, you need -f, as you are changing history
607 2012-02-08 19:19:30 <Matoking> Well
608 2012-02-08 19:19:57 <Matoking> I thought with git rebase the starting point was the latest commit and older commits which would be squashed in order to pack them into a single commit
609 2012-02-08 19:20:37 <sipa> well, yes, but it picks its starting point automatically
610 2012-02-08 19:20:47 <sipa> what you specify is where you want your commits to be put on top of
611 2012-02-08 19:21:06 <Matoking> I used git rebase -i HEAD~2
612 2012-02-08 19:21:29 <Matoking> Which I thought meant it picked the latest commit and the previous commit
613 2012-02-08 19:21:34 <sipa> no
614 2012-02-08 19:21:53 <sipa> it means picking all commits since HEAD~2, and putting them on top of HEAD~2
615 2012-02-08 19:22:06 <Matoking> Okay
616 2012-02-08 19:22:15 <sipa> but you want your patch to be against the current head, not HEAD~
617 2012-02-08 19:22:16 <sipa> but you want your patch to be against the current head, not HEAD~2
618 2012-02-08 19:22:36 <Matoking> Well anyway
619 2012-02-08 19:22:44 <Matoking> Thanks again, although it could have been a simple thing
620 2012-02-08 19:22:53 <Matoking> I've been struggling with Git too much
621 2012-02-08 19:42:13 <luke-jr> wee, the FBI is involved now
622 2012-02-08 19:51:07 <tcatm> luke-jr: ?
623 2012-02-08 19:55:14 <gmaxwell> tcatm: persistant 20gbps DOS attack against eligius.
624 2012-02-08 19:55:18 <luke-jr> tcatm: someone's DDoSing Eligius with over 20 Gbit/s
625 2012-02-08 19:55:37 <gmaxwell> (apparently spoofed as coming from deepbit or slush at times)
626 2012-02-08 19:55:41 <luke-jr> apparently started targetting nearby IPs, so the datacenter called the FBI
627 2012-02-08 19:56:13 <phantomcircuit> feds wont care...
628 2012-02-08 19:56:47 <luke-jr> phantomcircuit: apparnetly last time this happened the guy went to jail
629 2012-02-08 19:57:07 <phantomcircuit> you'd have to be monumentally retarded to get put in jail for a ddos
630 2012-02-08 20:06:56 <gribble> New news from bitcoinrss: Mindey opened issue 810 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/issues/810>
631 2012-02-08 20:21:24 <nonverba> quick question, does one mining cycle count as two hashes or one? i ask because it runs SHA-2 twice per mining cycle... thanks :)
632 2012-02-08 20:23:46 <sipa> nonverba: one "hash" as in Mhash/s, is a double-SHA256
633 2012-02-08 20:24:40 <nonverba> gotcha thanks
634 2012-02-08 20:26:59 <phantomcircuit> sipa, it's more like 1.5 sha256 operations
635 2012-02-08 20:27:08 <sipa> phantomcircuit: implementation detail
636 2012-02-08 20:27:15 <phantomcircuit> :)
637 2012-02-08 22:27:42 <ThePiachu> Hello. I was considering helping out a bit with the client code in an effort to understand what is happening in each file (mentioned it already on the forum, but the question didn't get much attention - https://bitcointalk.org/index.php?topic=62555.0 ). The jist is, viewing the source code as is is hard at times without investing a lot of time to understanding it.
638 2012-02-08 22:27:57 <ThePiachu> I was hoping to make it clearer by adding comments, meaningful variable names and so forth - sort of like what was done for BitcoinJ. My question is - would such a help be welcomed by the developers and viewed as helpful, or more like an unnecessary padding? Are there any guidelines as to a coding style one should follow when contributing code?
639 2012-02-08 22:29:14 <BlueMatt> if you are just doing comments, I dont think everyone will much appreciate it.  Same thing for variable names for the most part
640 2012-02-08 22:29:32 <BlueMatt> Id think you may want to leave the variable name prefixes,  but other than that I think everyone will just thank you
641 2012-02-08 22:29:53 <BlueMatt> also, see doc/coding.txt
642 2012-02-08 22:30:30 <ThePiachu> ah, didn't see the coding.txt, thanks!
643 2012-02-08 22:56:43 <XMPPwocky> hey: does an inv packet usually have 500 inv_vect s?
644 2012-02-08 23:01:15 <lianj> no
645 2012-02-08 23:01:41 <XMPPwocky> hmm
646 2012-02-08 23:03:22 <XMPPwocky> just random?
647 2012-02-08 23:04:15 <lianj> random? no. if you have one new tx, for example, you just send a tx inv packet with one hash
648 2012-02-08 23:09:22 <XMPPwocky> lianj: well, at connect
649 2012-02-08 23:11:06 <lianj> you send getblocks and get a packet inv packet with some block hashes back, max 500 iirc. the spec allows a var_int size iirc though
650 2012-02-08 23:13:14 <XMPPwocky> okay, yeah
651 2012-02-08 23:14:37 <Idiot___> what exactly is a block?
652 2012-02-08 23:15:01 <lianj> !give Idiot___ blocks
653 2012-02-08 23:15:04 <pirateat40> lego blocks?
654 2012-02-08 23:15:08 <lianj> aw
655 2012-02-08 23:15:33 <Idiot___> i was called a blockhead once...
656 2012-02-08 23:15:38 <Idiot___> does that have anything to do with bitcoin?
657 2012-02-08 23:15:56 <lianj> https://en.bitcoin.it/wiki/Block
658 2012-02-08 23:16:49 <Idiot___> ive spotted a problem
659 2012-02-08 23:16:57 <Idiot___> there are in theory, going to be an infinite number of blocks
660 2012-02-08 23:17:06 <Idiot___> but we havent yet invented a storage system with an infinite capacity?
661 2012-02-08 23:17:39 <iz> yeah, but the idea is that storage capacity increases over time
662 2012-02-08 23:18:19 <iz> at an exponential rate, which more than keeps up with the linear growth rate of total block chain size
663 2012-02-08 23:18:50 <Idiot___> uh ok
664 2012-02-08 23:19:41 <lianj> not really, 2012 seams like a bad year for disk prices
665 2012-02-08 23:21:30 <iz> lianj: compare the cost of a 1TB drive in 2012 to the cost of a 1TB drive in 2007 (5 years prior) and i think you'll see the effect of the thailand flooding is just a local fluctuation
666 2012-02-08 23:22:12 <lianj> im not sure its only because of the thailand thing
667 2012-02-08 23:22:43 <Idiot___> urgh... can exponential growth really continue forever though?
668 2012-02-08 23:22:57 <iz> also there could be blockchain pruning, as long as all of the transactions are completed in the pruned blocks
669 2012-02-08 23:23:29 <lianj> but true, until the chain is 1TB big it will look different. but processing/walking that chain then will only done by a handful of services then, the rest needs to trust them
670 2012-02-08 23:24:01 <XMPPwocky> http://pastebin.com/gSv4nWwN
671 2012-02-08 23:24:02 <XMPPwocky> woop wop
672 2012-02-08 23:24:06 <XMPPwocky> woop, even
673 2012-02-08 23:25:53 <XMPPwocky> that was from actually sitting between two nodes and proxying their packets
674 2012-02-08 23:26:11 <XMPPwocky> now I just have to stick a UI on it
675 2012-02-08 23:26:16 <Idiot___> is that a man-in-the-middle attack?
676 2012-02-08 23:26:28 <XMPPwocky> Idiot___: could certainly be used for one
677 2012-02-08 23:26:38 <Idiot___> i was thinking, that term is a bit sexist
678 2012-02-08 23:26:46 <Idiot___> shouldnt we call it person-in-the-middle attack?
679 2012-02-08 23:27:24 <lianj> youre in troll-mode, dont you?
680 2012-02-08 23:27:37 <Idiot___> troll mode?
681 2012-02-08 23:28:20 <iz> similarly, linux is sexist by making you type man all the time.. it should be person
682 2012-02-08 23:29:15 <Idiot___> so true iz, i didnt think of that
683 2012-02-08 23:29:20 <mod6> its man for manual
684 2012-02-08 23:29:28 <Idiot___> and theres a cat command but no dog command
685 2012-02-08 23:29:30 <Idiot___> whats with that?
686 2012-02-08 23:29:36 <mod6> that makes perfect sense to me, its not sexist.
687 2012-02-08 23:29:39 <iz> person is short for personalized manual
688 2012-02-08 23:30:00 <iz> lol
689 2012-02-08 23:30:10 <Idiot___> yeh u would say that wouldnt you... you man :P
690 2012-02-08 23:30:26 <mod6> lol, well im like ?!?!
691 2012-02-08 23:30:32 <iz> haha
692 2012-02-08 23:30:35 <mod6> xD
693 2012-02-08 23:32:00 <Idiot___> i also dont like the idea of a control panel... it sounds too domineering... how about a nurturing panel?
694 2012-02-08 23:32:13 <luke-jr> iz: man is inclusive
695 2012-02-08 23:32:31 <luke-jr> it can mean women too
696 2012-02-08 23:33:04 <Idiot___> bitcoin is very sexist... wallets are owned by men not women
697 2012-02-08 23:33:15 <Idiot___> women have purses
698 2012-02-08 23:33:19 <iz> idk, typing man mount sound pretty gay to me..
699 2012-02-08 23:33:25 <iz> sounds
700 2012-02-08 23:33:46 <iz> or you mean man in the general sense
701 2012-02-08 23:34:27 <iz> like how even female actors are called actors.. and how women are still considered part of mankind