1 2013-10-26 00:00:26 <TheLordOfTime> gmaxwell: then i'm confused, it happens when i haven't had coffee
   2 2013-10-26 00:00:35 <TheLordOfTime> gmaxwell: explain why they can't find -daemon on their bitcoind then on windows
   3 2013-10-26 00:00:38 <gmaxwell> super3: you're also confused.
   4 2013-10-26 00:00:47 <gmaxwell> TheLordOfTime: because they're confused.
   5 2013-10-26 00:00:56 <TheLordOfTime> gmaxwell: so your solution is that everyone is confused
   6 2013-10-26 00:01:00 <melvster> there you go:
   7 2013-10-26 00:01:02 <melvster> for ((;;)) do BLOCK=`bitcoind getblock $BLOCK | ./jq -r '.nextblockhash'` ; echo $BLOCK ; done
   8 2013-10-26 00:01:07 * super3 is very confused
   9 2013-10-26 00:01:11 * TheLordOfTime gives gmaxwell some tea
  10 2013-10-26 00:01:12 <melvster> set the first block first
  11 2013-10-26 00:01:23 <melvster> then shove in an sql statement in there
  12 2013-10-26 00:01:48 <gmaxwell> TheLordOfTime: they "can't find it" because they're looking at the list of RPC commands (e.g. "bitcoind help")  instead of the list of parameters ("bitcoind --help").
  13 2013-10-26 00:01:56 <TheLordOfTime> gmaxwell: erm...
  14 2013-10-26 00:02:06 <TheLordOfTime> gmaxwell: [13/10/25 19:34:47] <Ascendion> above -testnet I have -paytxfee
  15 2013-10-26 00:02:14 <TheLordOfTime> ... that's... the command line parameters list :/
  16 2013-10-26 00:02:28 <TheLordOfTime> rpc commands don't show with the - next to them
  17 2013-10-26 00:02:42 <gmaxwell> TheLordOfTime: welp then they are confused in some other way.
  18 2013-10-26 00:03:01 <super3> Ascendion, apparently a small portion of Bitcoind does RPC, yet everyone uses it for RPC
  19 2013-10-26 00:03:02 <TheLordOfTime> heh
  20 2013-10-26 00:03:22 <Ascendion> well the friggin port is there and is responsive to connections (telnet, http)
  21 2013-10-26 00:03:43 execut3 has joined
  22 2013-10-26 00:04:17 <gmaxwell> TheLordOfTime: because it certantly can work as a daemon in win32. The command may not be there because there is no parallel to backgrounding.
  23 2013-10-26 00:04:31 crass has quit (Read error: Connection reset by peer)
  24 2013-10-26 00:04:36 <super3> Ascendion, according something i saw from sipa the bitcoin-cli executable is way way smaller and does essentially the same thing
  25 2013-10-26 00:04:58 <gmaxwell> super3: please stop confusing things with random data from future pulls. Its not helping.
  26 2013-10-26 00:05:09 <gmaxwell> Ascendion: are you running bitcoind or bitcoin-qt?
  27 2013-10-26 00:05:15 <Ascendion> no shit -- I need what works NOW
  28 2013-10-26 00:05:17 <melvster> i made a gist https://gist.github.com/melvincarvalho/7163697
  29 2013-10-26 00:05:27 <super3> Ascendion, this is not going to happen anytime soon
  30 2013-10-26 00:05:31 <TheLordOfTime> Ascendion: calming down helps btw
  31 2013-10-26 00:05:40 <Ascendion> bitcoind from bitcoin-qt distro for windows from sourceforge -- just installed 0.8.5
  32 2013-10-26 00:05:40 <super3> Ascendion, bitcoind will still work for a long while
  33 2013-10-26 00:06:00 <gmaxwell> super3: please just stop. What you're talking about isn't even related to his interests.
  34 2013-10-26 00:06:21 <gmaxwell> Ascendion: please answer my question.
  35 2013-10-26 00:06:51 <Ascendion> bitcoind 0.8.5 from the bitcoin-qt distro for windows on sourceforge
  36 2013-10-26 00:07:06 * super3 stops
  37 2013-10-26 00:07:34 <gmaxwell> Ascendion: okay. And when you say "the port" is working, what port are you talking about?
  38 2013-10-26 00:07:36 shesek has quit (Ping timeout: 268 seconds)
  39 2013-10-26 00:07:52 <gmaxwell> e.g. port number.
  40 2013-10-26 00:08:05 <Ascendion> localhost 8333 -- I can connect with telnet and web browser returns "no data" when I try to hit the port
  41 2013-10-26 00:08:31 execut3 has quit (Ping timeout: 252 seconds)
  42 2013-10-26 00:08:32 <gmaxwell> Ascendion: that isn't the rpc port. The rpc port is 8332.
  43 2013-10-26 00:09:02 <Ascendion> thats what I set in the config so as not to conflict with bitcoin-qt when I need to access my personal wallet
  44 2013-10-26 00:10:09 <gmaxwell> you can't do that. Thats the p2p port.
  45 2013-10-26 00:10:25 <Ascendion> ok -- changing to a radically different port number :)
  46 2013-10-26 00:10:33 <gmaxwell> Do you have a seperate data directory for your personal wallet and this other daemon?
  47 2013-10-26 00:10:35 crass has joined
  48 2013-10-26 00:10:39 <Ascendion> yes
  49 2013-10-26 00:10:53 <gmaxwell> You need to change both the p2p and rpc port. And they need to be different from eachother and the other daemon.
  50 2013-10-26 00:11:25 <gmaxwell> in your config:  rpcport=28332  and port=28333 (for example
  51 2013-10-26 00:11:25 <gmaxwell> )
  52 2013-10-26 00:11:49 <gmaxwell> Make sure to put them in the config, not the commandline, or bitcoind (while being used as an rpc) won't find them.
  53 2013-10-26 00:12:18 <gmaxwell> then to connect to your daemon run bitcoind -datadir="other datadir" help  and it will read the port out of the config and use the right one.
  54 2013-10-26 00:13:36 <Ascendion> using the datadir option fixed it -- I got the blockcount using bitcoind on the command line
  55 2013-10-26 00:13:41 <Ascendion> now testing with my app
  56 2013-10-26 00:13:42 Perlboy has quit (Excess Flood)
  57 2013-10-26 00:14:14 Perlboy has joined
  58 2013-10-26 00:14:39 theorbtwo has quit (Remote host closed the connection)
  59 2013-10-26 00:14:48 <gmaxwell> Ascendion: as future advice, when you do something like change a port number, make sure you mention it when asking for help. Even if you don't think its important the person helping you will need a relatively complete and accurate mental picture of your setup.
  60 2013-10-26 00:15:00 bbrian has quit (Ping timeout: 248 seconds)
  61 2013-10-26 00:15:46 <Ascendion> I stuck my conf in pastebin earlier :)
  62 2013-10-26 00:18:01 <TheLordOfTime> http://pastebin.com/Tws7YLqH
  63 2013-10-26 00:18:17 <TheLordOfTime> Ascendion: gmaxwell might have lost it in the scrollback, keeping the link handy is useful
  64 2013-10-26 00:20:01 bitRipperX has joined
  65 2013-10-26 00:22:05 <Ascendion> grrrr had it working off the command line for a second... back to to errors
  66 2013-10-26 00:22:54 Azms has joined
  67 2013-10-26 00:23:11 execut3 has joined
  68 2013-10-26 00:25:30 discrete has joined
  69 2013-10-26 00:26:39 <Ascendion> the port is there -- connected with browser to 9332, it requested auth, I gave it user/pass, and got back a json error -- problem is in my app now :)
  70 2013-10-26 00:28:19 macboz has joined
  71 2013-10-26 00:30:59 theorbtwo has joined
  72 2013-10-26 00:31:17 TD_ has quit (Quit: TD_)
  73 2013-10-26 00:34:20 execut3 has quit (Ping timeout: 260 seconds)
  74 2013-10-26 00:35:04 cysm has quit (Quit: Leaving)
  75 2013-10-26 00:38:08 a_meteor has joined
  76 2013-10-26 00:39:15 skinnkavaj has quit (Read error: Connection reset by peer)
  77 2013-10-26 00:39:59 qbasicer has quit (Read error: Connection reset by peer)
  78 2013-10-26 00:40:12 mappum has joined
  79 2013-10-26 00:41:19 fanquake has joined
  80 2013-10-26 00:41:26 paracyst has quit (Read error: Connection reset by peer)
  81 2013-10-26 00:41:52 arioBarzan has joined
  82 2013-10-26 00:42:09 qbasicer has joined
  83 2013-10-26 00:43:36 skinnkavaj has joined
  84 2013-10-26 00:43:50 <Ascendion> WHEW -- FINALLY :) ty all for the help :) (got my app accessing the rpc interface after launching a private copy of bitcoind if it could not find and already running copy with the correct config)
  85 2013-10-26 00:43:50 arioBarzan has quit (Remote host closed the connection)
  86 2013-10-26 00:44:23 <Ascendion> time to stuff face -- starving :)
  87 2013-10-26 00:46:21 skinnkavaj has quit (Read error: Connection reset by peer)
  88 2013-10-26 00:47:27 mappum has quit (Ping timeout: 245 seconds)
  89 2013-10-26 00:48:06 skinnkavaj has joined
  90 2013-10-26 00:48:06 skinnkavaj has quit (Changing host)
  91 2013-10-26 00:48:06 skinnkavaj has joined
  92 2013-10-26 00:49:23 Anduck has quit (Ping timeout: 245 seconds)
  93 2013-10-26 00:50:55 bbrian has joined
  94 2013-10-26 00:55:42 sustrik has quit (Ping timeout: 268 seconds)
  95 2013-10-26 01:01:01 bbrian has quit (Ping timeout: 245 seconds)
  96 2013-10-26 01:02:40 <melvster> Ascendion: i wrote a quick script to put the headers in a table
  97 2013-10-26 01:08:23 Azms has quit ()
  98 2013-10-26 01:22:48 deepc0re has quit (Quit: deepc0re)
  99 2013-10-26 01:23:09 molec has joined
 100 2013-10-26 01:23:48 shesek has joined
 101 2013-10-26 01:25:15 <Ascendion> melvster -- cool if thats all you need :) I need basically a full abstraction of the blockchain including transactions in the database, with key data exposed so I can reconstruct unspent outputs for any given public address
 102 2013-10-26 01:25:31 <melvster> nice
 103 2013-10-26 01:25:42 <melvster> unfortunately my server doesnt have enough space for that
 104 2013-10-26 01:25:53 <melvster> so im just starting with headers
 105 2013-10-26 01:25:59 gingpark has joined
 106 2013-10-26 01:26:08 <Ascendion> helps alot when you own the server :)
 107 2013-10-26 01:26:42 moleccc has quit (Ping timeout: 272 seconds)
 108 2013-10-26 01:27:21 askmike has quit (Remote host closed the connection)
 109 2013-10-26 01:27:56 askmike has joined
 110 2013-10-26 01:27:58 xiangfu has joined
 111 2013-10-26 01:28:01 deepc0re has joined
 112 2013-10-26 01:28:31 shesek has quit (Read error: Connection reset by peer)
 113 2013-10-26 01:28:39 <Ascendion> the real fun begins when I replicate this code for LTC and a few other popular cryptocoins :)
 114 2013-10-26 01:29:05 deepc0re has quit (Client Quit)
 115 2013-10-26 01:29:57 jegz has quit (Ping timeout: 246 seconds)
 116 2013-10-26 01:30:16 shesek has joined
 117 2013-10-26 01:31:20 <melvster> nice
 118 2013-10-26 01:31:33 <melvster> a lot of data tho
 119 2013-10-26 01:31:48 askmike has quit (Read error: Operation timed out)
 120 2013-10-26 01:31:51 <Ascendion> so ?? thats what 10 terabyte raid arrays are for :)
 121 2013-10-26 01:32:05 eoss has joined
 122 2013-10-26 01:32:22 <melvster> whoa
 123 2013-10-26 01:32:29 <melvster> that's a lotta disk
 124 2013-10-26 01:32:43 roconnor has joined
 125 2013-10-26 01:32:43 xiangfu has quit (Ping timeout: 245 seconds)
 126 2013-10-26 01:35:05 <melvster> Ascendion: how much does that kind of setup cost?
 127 2013-10-26 01:35:23 cads has joined
 128 2013-10-26 01:35:53 <Ascendion> around a grand
 129 2013-10-26 01:36:24 <Ascendion> 8 bay enclosure with integrated raid controller and 6 drives
 130 2013-10-26 01:36:51 [\\\] has joined
 131 2013-10-26 01:37:26 <Ascendion> could actually do 12tb with it -- 6 hot drives, 1 for the raid parity, 1 hot spare
 132 2013-10-26 01:37:31 shesek has quit (Read error: Connection reset by peer)
 133 2013-10-26 01:37:52 <Ascendion> call it raid7+1 :)
 134 2013-10-26 01:38:54 Nesetalis has joined
 135 2013-10-26 01:38:57 <Ascendion> (I dont actually own a setup like that -- I just got 6tb in my home computer.. but that is the type of setup that will go on the server when I build it out)
 136 2013-10-26 01:40:41 shesek has joined
 137 2013-10-26 01:41:32 shesek has quit (Read error: Connection reset by peer)
 138 2013-10-26 01:42:09 apurplehorse has joined
 139 2013-10-26 01:45:31 shesek has joined
 140 2013-10-26 01:46:45 CircusPeanut has quit (Ping timeout: 246 seconds)
 141 2013-10-26 01:51:27 robocoin_ has joined
 142 2013-10-26 01:51:52 mhanne has joined
 143 2013-10-26 01:54:28 robocoin has quit (Ping timeout: 272 seconds)
 144 2013-10-26 01:54:44 neep3r has joined
 145 2013-10-26 01:55:22 discrete has quit (Ping timeout: 245 seconds)
 146 2013-10-26 01:55:31 shesek has quit (Read error: Connection reset by peer)
 147 2013-10-26 01:58:48 shesek has joined
 148 2013-10-26 01:58:50 shesek has quit (Read error: Connection reset by peer)
 149 2013-10-26 02:00:32 shesek has joined
 150 2013-10-26 02:02:31 shesek has quit (Read error: Connection reset by peer)
 151 2013-10-26 02:06:21 shesek has joined
 152 2013-10-26 02:10:54 neep3r has quit (Ping timeout: 246 seconds)
 153 2013-10-26 02:14:11 Polyatomic has quit (Changing host)
 154 2013-10-26 02:14:11 Polyatomic has joined
 155 2013-10-26 02:14:27 shesek has quit (Read error: Connection reset by peer)
 156 2013-10-26 02:15:45 Polyatomic has quit (Quit: Catcha Round)
 157 2013-10-26 02:16:10 Polyatomic has joined
 158 2013-10-26 02:16:10 Polyatomic has quit (Changing host)
 159 2013-10-26 02:16:10 Polyatomic has joined
 160 2013-10-26 02:17:12 <melvster> has anyone ever thought of writing a web based miner where you go to a page and it mines for you, and maybe you get lucky and get a coin, or share of a coin ...
 161 2013-10-26 02:18:25 <copumpkin> lots of people
 162 2013-10-26 02:18:28 <copumpkin> often maliciously
 163 2013-10-26 02:18:43 <melvster> copumpkin: anything that's free software?
 164 2013-10-26 02:18:48 <copumpkin> in the sense that they'll mine with other people's computers. But nowadays difficulty is so high that it's almost impossible
 165 2013-10-26 02:18:50 <copumpkin> so not much point
 166 2013-10-26 02:19:15 <melvster> i understand the economics, but I was more interested in the code ...
 167 2013-10-26 02:19:56 <melvster> it's only uneconomical due to the zero trust + asic landscape ... there are scenarios where it could be made more economical
 168 2013-10-26 02:21:00 <gmaxwell> melvster: no, it wasn't economical at the beginning of 2011.
 169 2013-10-26 02:21:43 <gmaxwell> oneman created a webcl one, which actually had acceptable security, but since webcl isn't part of the standard web (because its nearly impossible to make secure) it wasn't very useful to people.
 170 2013-10-26 02:21:51 <melvster> yes i get that ... but i wanted to experiment with some code, if it exists, so that it could be made economical again, e.g. maybe using colored coins or off block tx
 171 2013-10-26 02:22:13 <gmaxwell> melvster: what you're saying doesn't make a lot of sense to me.
 172 2013-10-26 02:22:23 <gmaxwell> (and leaves me wondering if you have any idea what mining is)
 173 2013-10-26 02:22:26 neep3r has joined
 174 2013-10-26 02:22:38 <melvster> gmaxwell: let me rephrase ... do you know if there is any code available for web mining, im not asking about economics here
 175 2013-10-26 02:22:56 <gmaxwell> melvster: Sorry, I'm can't help you.
 176 2013-10-26 02:23:00 <melvster> np, thanks
 177 2013-10-26 02:23:45 <melvster> gmaxwell: fwiw, im looking at creating an alt coin system that is 100% web based, so I wanted to play around with web mining
 178 2013-10-26 02:24:33 <gmaxwell> melvster: See topic. :)
 179 2013-10-26 02:25:44 <melvster> I think people would still go for web mining even to own like a fraction of a satoshi ...
 180 2013-10-26 02:25:47 AusBitBank has quit (Ping timeout: 245 seconds)
 181 2013-10-26 02:28:08 moleccc has joined
 182 2013-10-26 02:29:33 <helo> unspendable :/
 183 2013-10-26 02:29:34 <melvster> seems abandoned https://github.com/progranism/Bitcoin-JavaScript-Miner
 184 2013-10-26 02:29:53 <warren> for good reason
 185 2013-10-26 02:29:54 <melvster> unspendable only on the block chain
 186 2013-10-26 02:30:07 Goonie_ has joined
 187 2013-10-26 02:30:07 Goonie has quit (Ping timeout: 240 seconds)
 188 2013-10-26 02:30:09 <melvster> spendable off block
 189 2013-10-26 02:30:57 <melvster> and if you build up enough it becomes spendable
 190 2013-10-26 02:31:05 molec has quit (Ping timeout: 240 seconds)
 191 2013-10-26 02:31:35 <melvster> for example one click on a google ad is 'unspendable' but if you have a lot of traffic you  can cash out once a month
 192 2013-10-26 02:31:52 <melvster> or impression ...
 193 2013-10-26 02:31:53 Subo1977_ has joined
 194 2013-10-26 02:34:17 <melvster> and you can also mine other stuff like vanity addresses
 195 2013-10-26 02:35:06 Subo1977 has quit (Ping timeout: 240 seconds)
 196 2013-10-26 02:39:12 <Luke-Jr> melvster: I concur with gmaxwell. You don't understand mining IMO.
 197 2013-10-26 02:40:12 paracyst has joined
 198 2013-10-26 02:40:21 paracyst has quit (Changing host)
 199 2013-10-26 02:40:21 paracyst has joined
 200 2013-10-26 02:41:14 cysm has joined
 201 2013-10-26 02:42:31 <melvster> lol ... im sure you're right ... but there's not much too it imho, just create a nonce/hash pair under the target?
 202 2013-10-26 02:45:11 <Luke-Jr> …
 203 2013-10-26 02:45:19 <Luke-Jr> melvster: it's the *why* part you don't get.
 204 2013-10-26 02:45:22 <Luke-Jr> why does mining exist?
 205 2013-10-26 02:45:28 <Luke-Jr> hint: it isn't to give people money
 206 2013-10-26 02:45:34 <melvster> more than one reason
 207 2013-10-26 02:45:44 <melvster> it's a proof of work solution to byzantine generals problem
 208 2013-10-26 02:45:51 <melvster> to sync up the block chain
 209 2013-10-26 02:45:54 <gmaxwell> can this move out of #bitcoin-dev?
 210 2013-10-26 02:47:08 <melvster> gmaxwell: just responding to your accusations, I didnt really want to have this conversation
 211 2013-10-26 02:47:20 <melvster> we can end here
 212 2013-10-26 02:50:51 <gmaxwell> melvster: You've been talking to luke, not me.
 213 2013-10-26 02:52:14 <melvster> gmaxwell: if you has not accused me of not understanding mining, the conversation with luke would never have taken place, that's the point ... im happy to discontinue this now tho
 214 2013-10-26 02:52:45 _alp_ has joined
 215 2013-10-26 02:52:54 <melvster> if you come across any web mining s/w id be happy to hear, thanks for the pointer to webcl ... i learnt something
 216 2013-10-26 02:55:16 Krellan_ has quit (Ping timeout: 260 seconds)
 217 2013-10-26 02:55:20 apurplehorse has quit (Ping timeout: 268 seconds)
 218 2013-10-26 02:55:31 paracyst has quit ()
 219 2013-10-26 02:59:08 jtimon has quit (Ping timeout: 256 seconds)
 220 2013-10-26 03:00:10 <super3> melvster, you would have to create a new coin to do any kind of web mining
 221 2013-10-26 03:00:33 <TheLordOfTime> in any case that discussion isn't exactly on topic for #bitcoin-dev (because notbitcoin)
 222 2013-10-26 03:00:43 <TheLordOfTime> no?
 223 2013-10-26 03:00:43 <super3> melvster, on the bitcoin network my GPU could probably outmine 10,000 people running a webmining client
 224 2013-10-26 03:01:27 <super3> melvster, DM me its not going to work, but its at least fun to talk about
 225 2013-10-26 03:03:44 comboy has quit (Remote host closed the connection)
 226 2013-10-26 03:11:22 TheSeven has quit (Disconnected by services)
 227 2013-10-26 03:11:33 [7] has joined
 228 2013-10-26 03:13:14 agath has quit (Remote host closed the connection)
 229 2013-10-26 03:13:16 Someguy123 has quit (Excess Flood)
 230 2013-10-26 03:13:32 agath has joined
 231 2013-10-26 03:15:17 Someguy123 has joined
 232 2013-10-26 03:16:58 jegz has joined
 233 2013-10-26 03:20:48 super3_ has joined
 234 2013-10-26 03:20:56 super3 has quit (Ping timeout: 260 seconds)
 235 2013-10-26 03:21:04 super3_ is now known as super3
 236 2013-10-26 03:21:53 paracyst has joined
 237 2013-10-26 03:21:58 <gmaxwell> I wonder if it would be worth supporting "onion keys" in the payment protocol right out the gate.
 238 2013-10-26 03:22:00 paracyst has quit (Changing host)
 239 2013-10-26 03:22:00 paracyst has joined
 240 2013-10-26 03:23:09 Goonie has joined
 241 2013-10-26 03:23:31 gingpark has quit (Ping timeout: 245 seconds)
 242 2013-10-26 03:23:34 <gmaxwell> A tor hidden service (onion site) is an RSA key.  It would be trivial to pack a tor HS pubkey into a x509 certificate, put the cn=foooo.onion   and then set some non-standard flag that says, "no no, you're supposted to authenticate this by checking that the pubkey hashes to the right CN"
 243 2013-10-26 03:23:56 <gmaxwell> and then you could have your payment protocol things come in saying that your payment request was signed by blahblah.onion.
 244 2013-10-26 03:24:18 <gmaxwell> and it's all authenticated and provable to a third party with no ca involved at all— assuming you already had a way to be sure you had the right onion name.
 245 2013-10-26 03:24:20 Goonie_ has quit (Ping timeout: 248 seconds)
 246 2013-10-26 03:24:45 gingpark has joined
 247 2013-10-26 03:25:03 <gmaxwell> also: onion sites can't get x509, so they really have no solution now under the payment protocol.
 248 2013-10-26 03:25:26 <gmaxwell> Also: the people who are unhappy with a CA being involved in any way would want to be using tor anyways.
 249 2013-10-26 03:27:21 * gmaxwell takes it to the list
 250 2013-10-26 03:28:06 Ferroh has quit (Quit: *SMOKE BOMB*)
 251 2013-10-26 03:34:04 K1773R has quit (Ping timeout: 260 seconds)
 252 2013-10-26 03:35:36 chax has quit (Remote host closed the connection)
 253 2013-10-26 03:35:47 chax has joined
 254 2013-10-26 03:35:57 chax has quit (Remote host closed the connection)
 255 2013-10-26 03:36:13 chax has joined
 256 2013-10-26 03:36:53 gulli has quit (Quit: Page closed)
 257 2013-10-26 03:37:31 Coincidental has quit (Remote host closed the connection)
 258 2013-10-26 03:39:03 K1773R has joined
 259 2013-10-26 03:39:08 K1773R has quit (Changing host)
 260 2013-10-26 03:39:08 K1773R has joined
 261 2013-10-26 03:39:37 canoon has joined
 262 2013-10-26 03:40:18 Arnavion has quit (Quit: Arnavion)
 263 2013-10-26 03:41:04 Arnavion has joined
 264 2013-10-26 03:43:32 gingpark has quit (Ping timeout: 245 seconds)
 265 2013-10-26 03:43:49 Arnavion has quit (Client Quit)
 266 2013-10-26 03:44:53 Arnavion has joined
 267 2013-10-26 03:52:24 malaimo has quit (Ping timeout: 256 seconds)
 268 2013-10-26 03:52:52 <gmaxwell> sipa: what do you think about having the inactivity timeout disconnect only fire if we have MAX_OUTBOUND_CONNECTIONS outbound connections up? There is really no point in tearing down a connection that might come back when we're not full.
 269 2013-10-26 03:54:07 malaimo has joined
 270 2013-10-26 03:54:15 <gmaxwell> and avoiding it might make us more robust when there is a DOS attack (for a long time one of the big reasons dos attacks were damaging to the internet was that they delayed BGP keepalives enough that they cause BGP session flaps, this has largely been addressed by prioritizing bgp traffic on routers, less of an option for use because we're not a router :))
 271 2013-10-26 03:55:13 <gmaxwell> (this is all WRT outbound connections, though I'd say the same thing for inbound ones but applying the inbound criteria instead)
 272 2013-10-26 03:55:50 AusBitBank has joined
 273 2013-10-26 03:55:57 roconnor has quit (Remote host closed the connection)
 274 2013-10-26 03:56:20 Subo1977 has joined
 275 2013-10-26 03:56:26 Subo1977_ has quit (Ping timeout: 240 seconds)
 276 2013-10-26 04:08:53 jeewee has quit (Quit: Leaving.)
 277 2013-10-26 04:23:20 <Luke-Jr> gmaxwell: you know Namecoin already supports SSL without CAs?
 278 2013-10-26 04:27:53 paracyst has quit ()
 279 2013-10-26 04:28:41 <Ascendion> ok -- lets say for a moment that I'm ripping through the block chain starting at the genesis block and working my way up... at any time I pull a block via rpc that doesnt have the last block I pulled as its previous -- I've detected a fork and need to roll back ??
 280 2013-10-26 04:29:47 <gmaxwell> Luke-Jr: it doesn't really. You have to run a namecoin node to do it, which is totally useless... because e.g. it needs to be your trezor validating the payment request. The trezor is not going to run a namecoin node.
 281 2013-10-26 04:30:38 <Luke-Jr> gmaxwell: only because normal DNS lacks a way to express "this is the legit key for the host"
 282 2013-10-26 04:31:06 <gmaxwell> Luke-Jr: it's not just that even if DNS did, you could just lie because DNS querying namecoin is not cryptographic.
 283 2013-10-26 04:31:07 shesek has joined
 284 2013-10-26 04:31:16 <gmaxwell> (actually there is a spec for doing that with DNS IIRC)
 285 2013-10-26 04:31:45 <Luke-Jr> gmaxwell: my point is that if you can get the name, you can get the key too
 286 2013-10-26 04:31:46 <gmaxwell> but you have to trust your resolver. DNSSEC addresses this for regular DNS but namecoin can't emit DNSSEC.
 287 2013-10-26 04:31:56 <gmaxwell> Luke-Jr: but can you get the _right_ name?
 288 2013-10-26 04:32:10 <Luke-Jr> that's a problem with namecoin, not the enablement of SSL based on it
 289 2013-10-26 04:32:58 <Luke-Jr> Ascendion: a reorg*
 290 2013-10-26 04:33:01 <gmaxwell> Luke-Jr: say your PC is compromised. You go to order some alpaca socks.  The malware replaces the payment request with one saying to pay the malware author.  Your trezor says the payment request came from alpacasocks.com because your host is lying about the DNS response that its forwarding on the the trezor.
 291 2013-10-26 04:33:12 <gmaxwell> Luke-Jr: yea, it's a problem with namecoin.
 292 2013-10-26 04:33:36 <Luke-Jr> gmaxwell: so you'd give the Trezor CA authority?
 293 2013-10-26 04:33:41 <Luke-Jr> does it even have the storage for that?
 294 2013-10-26 04:34:23 <gmaxwell> Luke-Jr: sure it does. And even if it didn't it would only need to store the root hash of a CA cert hashtree, and the host could give it the right CA cert on demand, and prove that its a real CA cert.
 295 2013-10-26 04:34:51 <Luke-Jr> hm
 296 2013-10-26 04:34:58 <gmaxwell> (by giving the trezor the hashes connecting the ca cert to the root thats already in the trezor)
 297 2013-10-26 04:35:20 <Luke-Jr> and the user still tells Trezor what CAs he trusts somehow?
 298 2013-10-26 04:35:21 <gmaxwell> or by having all the CA keys signed by the makers of the hardware, and just sending the right one. (then it can be updated easily)
 299 2013-10-26 04:35:51 <gmaxwell> Yea, could be done. E.g. some UI on the device.
 300 2013-10-26 04:36:18 <gmaxwell> In any case, I think supporting namecoin would be great, but I think it needs a lot of code and a namecoin protocol change. Not really viable for the initial release.
 301 2013-10-26 04:38:55 contrapumpkin has joined
 302 2013-10-26 04:38:56 copumpkin has quit (Read error: Connection reset by peer)
 303 2013-10-26 04:39:23 <Ascendion> ok -- lets say for a moment that I'm ripping through the block chain starting at the genesis block and working my way up... at any time I pull a block via rpc that doesnt have the last block I pulled as its previous -- I've detected a fork and need to roll back ??
 304 2013-10-26 04:40:01 <Luke-Jr> Ascendion: you already said that.
 305 2013-10-26 04:40:07 <Ascendion> but no one answered :)
 306 2013-10-26 04:40:58 <Ascendion> ahhh you put a colon after my name so my name highlighting didnt pick it up
 307 2013-10-26 04:41:21 <Luke-Jr> that's standard IRC response format.
 308 2013-10-26 04:41:39 <Ascendion> ok -- reorg, fork, whatever -- I need to back out blocks that have been orphaned and start working up through the new chain
 309 2013-10-26 04:42:12 <Luke-Jr> yes. note it may be more than 1.
 310 2013-10-26 04:42:45 <Ascendion> according to some stats I was reading earlier it rarely goes beyond 4 blocks
 311 2013-10-26 04:43:10 <Ascendion> I've though about just delaying what I process into the database and scan for transactions say 8-10 blocks
 312 2013-10-26 04:44:33 <Luke-Jr> it can.
 313 2013-10-26 04:45:16 <Ascendion> yes -- especially if the p2p network gets partitioned.. but that should be rare if I keep a lot of connections active
 314 2013-10-26 04:48:13 happywarrior has joined
 315 2013-10-26 04:48:13 Ascendion has quit (Read error: Connection reset by peer)
 316 2013-10-26 04:48:47 Ascendion has joined
 317 2013-10-26 04:48:58 <Ascendion> ouch :)
 318 2013-10-26 04:49:40 discrete has joined
 319 2013-10-26 04:49:48 happywarrior has quit (Client Quit)
 320 2013-10-26 04:50:03 ajv has joined
 321 2013-10-26 04:51:32 winterblack has joined
 322 2013-10-26 04:52:09 [7] has quit (Disconnected by services)
 323 2013-10-26 04:52:20 TheSeven has joined
 324 2013-10-26 04:53:29 impulse has quit (Quit: leaving)
 325 2013-10-26 04:55:20 nethershaw has quit (Ping timeout: 272 seconds)
 326 2013-10-26 05:02:16 ajv has quit (Quit: Textual IRC Client: www.textualapp.com)
 327 2013-10-26 05:02:29 ajv has joined
 328 2013-10-26 05:06:46 eoss has quit (Remote host closed the connection)
 329 2013-10-26 05:08:03 xiangfu has joined
 330 2013-10-26 05:11:17 <warren> anyone have a working MacOS X build environment?
 331 2013-10-26 05:12:16 <fanquake> warren Last time I checked.
 332 2013-10-26 05:12:55 discrete has quit ()
 333 2013-10-26 05:13:30 <ajv> hello, I'm working on an SPV client, and I'm having a bit of trouble with merkleblock messages on testnet. I have my bloom filter set to full match, and the merkleblock messages I get during the initial chain download have plenty of hashes of matched transactions, however the merkleblock messages are not followed by the tx messages for the block. It always continues to the next merkleblock message without any tx messages at all
 334 2013-10-26 05:15:54 <dobry-den> warren: i use this https://github.com/WyseNynja/homebrew-bitcoin/blob/master/bitcoind.rb
 335 2013-10-26 05:16:20 Application has joined
 336 2013-10-26 05:17:25 <dobry-den> What are main use-cases for the :checksum in an incoming message? What does the client do with it?
 337 2013-10-26 05:18:15 <dobry-den> Does it reject a message if the payload doesnt match the checksum?
 338 2013-10-26 05:18:30 <ajv> checksum is of course useful to tell if the message got corrupted in transmission
 339 2013-10-26 05:18:37 <ajv> yes it'll reject the message
 340 2013-10-26 05:20:46 Someguy123 has quit (Excess Flood)
 341 2013-10-26 05:25:26 Ashaman has joined
 342 2013-10-26 05:25:49 <dobry-den> ajv: if a message is invalid, does the client drop the connection?
 343 2013-10-26 05:26:37 <ajv> it will eventually, I have to check the code, but I don't expect it would for just one garbled message
 344 2013-10-26 05:27:46 Someguy123 has joined
 345 2013-10-26 05:28:09 <dobry-den> ajv: if you lie and say the payload is 6 bytes but really load 7 bytes into the stream. and then send a valid message to the client, how does it skip that extra byte?
 346 2013-10-26 05:28:42 sustrik has joined
 347 2013-10-26 05:28:44 <ajv> there is a magic number that starts each message
 348 2013-10-26 05:28:56 <ajv> it will look for that magic number to start the next message
 349 2013-10-26 05:30:12 <dobry-den> so if you botch the last byte of the magic number, the client continues its magic-number-seeking loop until timeout or whatever?
 350 2013-10-26 05:30:31 <dobry-den> that makes sense
 351 2013-10-26 05:31:01 <ajv> yes, I believe so. there's also a concept of a "misbehaving" score for each peer... if a peer does too many things wrong, it'll get disconnected
 352 2013-10-26 05:31:22 <dobry-den> yeah, that is pretty cool.
 353 2013-10-26 05:36:32 arioBarzan has joined
 354 2013-10-26 05:40:00 Someguy123 has quit (Excess Flood)
 355 2013-10-26 05:40:00 ThomasV has joined
 356 2013-10-26 05:40:06 <ajv> disregard my question... found the problem. client was receiving the tx messages but incorrectly dropping them
 357 2013-10-26 05:40:22 Someguy123 has joined
 358 2013-10-26 05:43:26 * Luke-Jr ponders whether to include PR 3152 in next-test
 359 2013-10-26 05:44:24 Someguy123 has quit (Excess Flood)
 360 2013-10-26 05:45:25 <Ascendion> link PR 3152 ?? :)
 361 2013-10-26 05:46:54 Someguy123 has joined
 362 2013-10-26 05:47:16 <Luke-Jr> https://github.com/bitcoin/bitcoin/pull/3152
 363 2013-10-26 05:49:07 Someguy123 has quit (Excess Flood)
 364 2013-10-26 05:50:51 <Ascendion> hmmm need to make the comparisons case insensitive, and dont forget the CIA -- they need love too :)
 365 2013-10-26 05:52:38 <Ascendion> gmaxwell is right though -- need to add KGB, Mossad, MI6, and any others we can come up with :)
 366 2013-10-26 05:56:09 Someguy123 has joined
 367 2013-10-26 05:56:09 xiangfu has quit (Read error: Connection reset by peer)
 368 2013-10-26 05:58:03 sustrik has quit (Ping timeout: 246 seconds)
 369 2013-10-26 06:00:19 arioBarzan has quit (Remote host closed the connection)
 370 2013-10-26 06:01:17 arioBarzan has joined
 371 2013-10-26 06:04:33 <Ascendion> hmmm would it be worth adding a message to the network protocol so agencies can inject their name and address into P2P network and update all clients, with the data stored in its own DB on each client ?? :)
 372 2013-10-26 06:04:49 ColinT has quit (Quit: Leaving...)
 373 2013-10-26 06:05:12 <Luke-Jr> Ascendion: no
 374 2013-10-26 06:05:14 owowo has quit (Quit: °!°                    ~~ S4n1tY 1S Fut1l3 ~~                        °!°)
 375 2013-10-26 06:05:20 <Luke-Jr> they can just publish it on irs.gov
 376 2013-10-26 06:06:18 <Ascendion> but then we have to manually update the code -- too much work for us :)
 377 2013-10-26 06:06:34 * Luke-Jr ponders if it's possible for the IRS to run a payment protocol proxy that almost-transparently adds sales tax
 378 2013-10-26 06:06:46 <Luke-Jr> Ascendion: nah, code can just verify the SSL cert
 379 2013-10-26 06:06:50 <warren> Luke-Jr: sales tax is a per-state thing ...
 380 2013-10-26 06:06:55 <Luke-Jr> warren: that's not my point
 381 2013-10-26 06:07:07 <warren> Luke-Jr: and how the heck do you prove that a customer is subject to a particular tax jurisdiction
 382 2013-10-26 06:07:19 <Luke-Jr> warren: the customer would configure their client to use the proxy
 383 2013-10-26 06:07:27 <Ascendion> set it up where the name must be prefixed with a country code like US.IRS and the injected name/address pair is only stored if the public ip address of the node is in the same country (using a geoip lookup)
 384 2013-10-26 06:07:53 patcon has joined
 385 2013-10-26 06:08:15 <Luke-Jr> maybe I shouldn't have said "IRS"; I'm talking theoretical fictional world :p
 386 2013-10-26 06:08:28 <Luke-Jr> post-bitcoin taxation methods
 387 2013-10-26 06:08:44 <Ascendion> that entire PR was theoretical fictional -- I was just playing along :)
 388 2013-10-26 06:08:54 <Luke-Jr> exactly
 389 2013-10-26 06:09:26 <Ascendion> nice break from writing code to update my database and roll back the block chain after a detected fork
 390 2013-10-26 06:10:53 <Ascendion> warren -- its worse than per-state -- its per city in most cases -- the state sets the base rate then each city typically tacks on a little more
 391 2013-10-26 06:11:15 <warren> yeah
 392 2013-10-26 06:11:26 <Luke-Jr> at least it's not Europe's VAT.
 393 2013-10-26 06:11:36 <Ascendion> we have rates running from 7 to 8.25% within a 50 mile radius of where I live :)
 394 2013-10-26 06:12:17 patcon has quit (Ping timeout: 245 seconds)
 395 2013-10-26 06:14:28 <Ascendion> grrrrr powering up into a rollback is gonna make this code cumbersome to say the least
 396 2013-10-26 06:15:24 jegz has quit (Ping timeout: 272 seconds)
 397 2013-10-26 06:25:00 <Ascendion> hmmm this api really does return the next block in the chain and doesnt seem to return the previous
 398 2013-10-26 06:26:26 <Ascendion> grrrr the api docs dont document the json message returned
 399 2013-10-26 06:28:42 <Ascendion> so I guess the right way to detect a new block is to poll the current most recent block and see if the nextblockhash is set ??
 400 2013-10-26 06:29:33 arioBarzan has quit (Remote host closed the connection)
 401 2013-10-26 06:30:18 <Ascendion> just seems bass-ackwards to me to have to back up and update a record in the database every time a new record is added
 402 2013-10-26 06:32:23 btcbtc has joined
 403 2013-10-26 06:34:30 jegz has joined
 404 2013-10-26 06:48:20 <Ascendion> just for my edification -- what goes over the network in a block -- the hash of the previous block ?? it cant be the hash of the next block since thats not known
 405 2013-10-26 06:49:59 <Ascendion> ya -- thats what I though -- it sends the hash of the previous block.. but that isnt what gets stored in the client database apparently, or if it does the json-rpc doesnt expose it
 406 2013-10-26 06:51:22 <Ascendion> bass-ackwards to the hilt !!
 407 2013-10-26 06:52:05 jegz has quit (Ping timeout: 240 seconds)
 408 2013-10-26 06:55:47 Ferroh has joined
 409 2013-10-26 06:57:00 <gmaxwell> Ascendion: wtf are you talking about?
 410 2013-10-26 06:57:25 <Ascendion> the data returned by json-rpc for getblock
 411 2013-10-26 06:57:38 <gmaxwell> returns the previous block hash.
 412 2013-10-26 06:57:50 <gmaxwell> it's in the highly cryptic field named "previousblockhash"
 413 2013-10-26 06:57:56 <Ascendion> it doesnt contain the hash of the previous block like the network format does -- it returns the NEXT block
 414 2013-10-26 06:58:06 <gmaxwell> No, it returns the previous block.
 415 2013-10-26 06:58:34 <Ascendion> I just debugged into the json decoding and it clearly said next block, and the values all matched up for the genesis block
 416 2013-10-26 06:59:30 <Ascendion> I'm taking a short break then I'm gonna go look at the bitcoind source code where it creates that json message
 417 2013-10-26 06:59:49 <gmaxwell> $ ./bitcoind  getblock 0000000000000005dbedf196d204e84b88bb54ad7982d89ad273e34199286e8b | egrep '(height)|(previousblockhash)' ; ./bitcoind  getblock  0000000000000009535b1bd3bcac3a16268ac40425070fec24b3300b82c046d5 | egrep '(height)|(previousblockhash)' ; ./bitcoind  getblock 000000000000000712d03f778dc3db4324c094ebba9574065a54e6ace6f31063 | egrep '(height)|(previousblockhash)' "height" : 266146,
 418 2013-10-26 06:59:55 <gmaxwell>  "previousblockhash" : "0000000000000009535b1bd3bcac3a16268ac40425070fec24b3300b82c046d5" "height" : 266145, "previousblockhash" : "000000000000000712d03f778dc3db4324c094ebba9574065a54e6ace6f31063", "height" : 266144, "previousblockhash" : "0000000000000000c308028f5389928ae8b8d37c4b4e4081dfe96ca8d5f7408d",
 419 2013-10-26 07:00:01 <gmaxwell> ^ it returns the previousblockhash.
 420 2013-10-26 07:00:30 <gmaxwell> The genesis block does not return a previousblockhash.
 421 2013-10-26 07:00:55 <gmaxwell> Just like the tip block doesn't return a nextblockhash.
 422 2013-10-26 07:01:22 <Ascendion> gimme a minute -- I'll add code to test both the genesis and the 1st block after genesis and look at the json returned
 423 2013-10-26 07:01:45 <gmaxwell> here is a tip, whatever you're doing, the genesis block is not a good block for testing.
 424 2013-10-26 07:02:44 <Ascendion> the poco that the json message is decoding into doesnt have the previous block hash... I was thinking for the genesis block it would at least pass back an empty string -- but I'll check the height 1 block
 425 2013-10-26 07:03:45 <swulf--> gmaxwell: weird, kinda?  The genesis block's prevblockhash should be 0x00*32, no?
 426 2013-10-26 07:04:24 CryptoBuck has joined
 427 2013-10-26 07:04:34 <Luke-Jr> swulf--: no, that's just how "no value" is encoded in the header
 428 2013-10-26 07:05:02 <swulf--> Those zero bytes are indeed part of the header since they're included in the hash of the genesis block header...no?
 429 2013-10-26 07:05:18 <gmaxwell> swulf--: no,... because that would break the procedure that "getblock <hash> works"
 430 2013-10-26 07:05:27 <swulf--> hmm
 431 2013-10-26 07:05:42 <gmaxwell> swulf--: this is the RPC interface, not "give me some random raw data"
 432 2013-10-26 07:08:25 aceat64 has quit (Ping timeout: 240 seconds)
 433 2013-10-26 07:10:36 paracyst has joined
 434 2013-10-26 07:12:08 <Ascendion> ok -- the library I was using just left that field out... adding it now.. I got the height=1 block and it did indeed have the previous hash in the json response
 435 2013-10-26 07:15:29 grau has joined
 436 2013-10-26 07:15:45 neep3r has quit (Remote host closed the connection)
 437 2013-10-26 07:16:20 neep3r has joined
 438 2013-10-26 07:17:02 <Ascendion> great -- and bitcoind is locking up after I fetch a few blocks with json-rpc
 439 2013-10-26 07:17:47 <Ascendion> or more accuratly -- make a few requests -- doesnt matter what requests
 440 2013-10-26 07:18:09 aceat64 has joined
 441 2013-10-26 07:19:22 ThomasV has quit (Ping timeout: 272 seconds)
 442 2013-10-26 07:20:20 <gmaxwell> Ascendion: you cannot just leave your connections open, whatever library you are using is likely leaking connections and filling up the rpc connection capacity.
 443 2013-10-26 07:21:22 neep3r has quit (Ping timeout: 272 seconds)
 444 2013-10-26 07:24:44 Krellan_ has joined
 445 2013-10-26 07:24:47 dlidstrom has joined
 446 2013-10-26 07:25:12 <warren> https://github.com/bitcoin/bitcoin/issues/3112  how is getblock supposed to behave?  It isn't clear how this worked for anyone?
 447 2013-10-26 07:26:04 super3 has quit (Ping timeout: 268 seconds)
 448 2013-10-26 07:29:27 grau has quit (Remote host closed the connection)
 449 2013-10-26 07:29:43 <gmaxwell> warren: what are you talking about?
 450 2013-10-26 07:30:30 <warren> gmaxwell: AFAICT this script doesn't work for anyone
 451 2013-10-26 07:31:01 dlidstrom has left ()
 452 2013-10-26 07:31:44 <gmaxwell> warren: works perfectly fine for me.
 453 2013-10-26 07:31:51 <gmaxwell> $ python ./linearize.py a.cfg
 454 2013-10-26 07:31:51 <warren> umm...
 455 2013-10-26 07:31:51 <gmaxwell> Wrote block 0
 456 2013-10-26 07:31:52 <gmaxwell> Wrote block 1000
 457 2013-10-26 07:31:52 <gmaxwell> Wrote block 2000
 458 2013-10-26 07:31:52 <gmaxwell> Wrote block 3000
 459 2013-10-26 07:32:18 <warren> I tried it with bitcoin-0.8.5.  does it require something in master?
 460 2013-10-26 07:32:20 daniel has joined
 461 2013-10-26 07:32:34 <Luke-Jr> warren: likely
 462 2013-10-26 07:32:44 daniel is now known as Guest27209
 463 2013-10-26 07:32:46 <Luke-Jr> IIRC getblock only recently added the ability to get it raw
 464 2013-10-26 07:32:46 <warren> ok
 465 2013-10-26 07:32:51 <gmaxwell> yes, it's not part of 0.8.5 ...
 466 2013-10-26 07:33:36 Guest27209 has quit (Client Quit)
 467 2013-10-26 07:36:19 <Ascendion> and there we go -- already processed the 1st 500 blocks :)
 468 2013-10-26 07:37:49 <Ascendion> hmmmm still managed to hit a timeout at block 673
 469 2013-10-26 07:38:31 ajv has quit (Quit: Textual IRC Client: www.textualapp.com)
 470 2013-10-26 07:39:11 dlidstrom has joined
 471 2013-10-26 07:40:15 dlidstrom has left ()
 472 2013-10-26 07:41:38 arioBarzan has joined
 473 2013-10-26 07:42:40 <Ascendion> hmmm gonna have to wrap some retry logic around this crap -- I restarted it after it bombed at 673 and now its ripping through -- over 3500 so far now that bitcoind finished syncing the block chain
 474 2013-10-26 07:42:47 grau has joined
 475 2013-10-26 07:43:14 <Ascendion> 4000
 476 2013-10-26 07:45:10 <Ascendion> 5000
 477 2013-10-26 07:45:21 <Ascendion> not bad -- 2 minutes per 1000 blocks
 478 2013-10-26 07:46:00 <Ascendion> should fully sync in what ?? about 6 hours ??
 479 2013-10-26 07:46:15 <Belxjander> Ascendion: are you downloading the blocks off the blockchain raw ?
 480 2013-10-26 07:46:27 dlidstrom has joined
 481 2013-10-26 07:46:37 <Ascendion> naaa -- using json-rpc to pull from a bitcoind and store to an SQL database
 482 2013-10-26 07:47:17 <Ascendion> not doing transactions yet -- will have to write that code, wipe the database, and start over :) but that's tomorrows work :)
 483 2013-10-26 07:50:10 <Ascendion> I dont really care how long it takes to sync -- once its in sync, the app doing the work will run 24/7 and catch all the changes as they happen
 484 2013-10-26 07:52:24 <Belxjander> okay
 485 2013-10-26 07:52:48 <Belxjander> I'm considering an AmigaOS bitcoind client + library myself...
 486 2013-10-26 07:53:12 <Ascendion> isnt the average rate of creating a new block supposed to be about 1 every 10 minutes ??
 487 2013-10-26 07:53:22 <Belxjander> but as there is basically no support for the opensource template client and that I can't really comprehend the C++ used
 488 2013-10-26 07:53:29 <Belxjander> afaik yes
 489 2013-10-26 07:53:48 <Ascendion> cause I'm seeing new blocks like every 15 seconds or less
 490 2013-10-26 07:54:02 <Ascendion> must be a SHITLOAD of new hashing power come online recently
 491 2013-10-26 07:55:26 <dlidstrom> Since the 80 bytes to be allowed in OP_RETURN transactions isn't enough for a serialized transaction, I'm wondering how the announce/commit sacrifices in the identity protocol might happen.  Is the idea that some miners might take them directly?  Are there any that do now?
 492 2013-10-26 07:56:29 <CodeShark> if you're seeing blocks every 15 seconds regularly then either you're not watching them live or you're hallucinating
 493 2013-10-26 07:56:50 <warren> hmm... I found an oddity
 494 2013-10-26 07:56:54 <warren> src/version.h:static const int MIN_PROTO_VERSION = 209;
 495 2013-10-26 07:56:54 <warren> src/main.cpp:        if (pfrom->nVersion < MIN_PROTO_VERSION)
 496 2013-10-26 07:57:00 <warren> these are fine and good looking
 497 2013-10-26 07:57:05 <warren> src/net.h:    CNode(SOCKET hSocketIn, CAddress addrIn, std::string addrNameIn = "", bool fInboundIn=false) : ssSend(SER_NETWORK, MIN_PROTO_VERSION)
 498 2013-10-26 07:57:05 <warren> src/net.h:        nRecvVersion = MIN_PROTO_VERSION;
 499 2013-10-26 07:57:11 <warren> but why is MIN_PROTO_VERSION used here?
 500 2013-10-26 07:57:31 neep3r has joined
 501 2013-10-26 07:58:06 <Ascendion> hmmm manybe I'm not caught up yet after all -- its processing a load of orphan blocks right now
 502 2013-10-26 07:58:25 <Ascendion> height just passed 235100
 503 2013-10-26 07:58:28 <warren> Ascendion: what version of client?
 504 2013-10-26 07:58:34 <Ascendion> 0.8.5
 505 2013-10-26 07:59:00 <warren> Ascendion: https://github.com/bitcoin/bitcoin/pull/2910
 506 2013-10-26 07:59:02 <warren> Ascendion: that saves time
 507 2013-10-26 08:01:28 arioBarzan has quit (Remote host closed the connection)
 508 2013-10-26 08:03:05 johnsoft has quit (Ping timeout: 240 seconds)
 509 2013-10-26 08:03:37 johnsoft has joined
 510 2013-10-26 08:04:50 deepc0re has joined
 511 2013-10-26 08:05:36 <Ascendion> looks like a nice rework, but I'm not gonna jump on the bleeding edge just yet :)
 512 2013-10-26 08:06:18 <Ascendion> ok -- I'm not caught up yet -- 30k blocks to go in bitcoind -- then I'm realtime
 513 2013-10-26 08:07:44 macboz has quit (Quit: This computer has gone to sleep)
 514 2013-10-26 08:08:02 deepc0re has quit (Client Quit)
 515 2013-10-26 08:09:46 s7r has quit (Ping timeout: 240 seconds)
 516 2013-10-26 08:10:39 s7r has joined
 517 2013-10-26 08:11:29 <warren> hmm.... I don't think MIN_PROTO_VERSION was thought through well.
 518 2013-10-26 08:11:59 neep3r has quit (Ping timeout: 252 seconds)
 519 2013-10-26 08:14:31 grau has quit (Remote host closed the connection)
 520 2013-10-26 08:15:26 grau has joined
 521 2013-10-26 08:18:12 grau_ has joined
 522 2013-10-26 08:18:29 execut3 has joined
 523 2013-10-26 08:20:15 <sipa> warren: because protocols before 209 are no longer supported since februari 12
 524 2013-10-26 08:21:47 <warren> sipa: if you increase that to 70001, it totally breaks all communication
 525 2013-10-26 08:22:04 shesek has quit (Ping timeout: 272 seconds)
 526 2013-10-26 08:22:10 grau has quit (Ping timeout: 272 seconds)
 527 2013-10-26 08:22:28 <warren> it appears MIN_PROTO_VERSION in src/net.h isn't meant to be the same thing as src/main.cpp
 528 2013-10-26 08:23:01 <Ascendion> definitly disk bound at this point -- hammering my main drive to 100% :)
 529 2013-10-26 08:23:03 <Ascendion> might have to move the database off to another drive to speed things up :)
 530 2013-10-26 08:23:20 datagutt has quit (Ping timeout: 272 seconds)
 531 2013-10-26 08:23:39 <sipa> gmaxwell: good suggestion wrt timeout disconnect
 532 2013-10-26 08:26:41 Anduck has joined
 533 2013-10-26 08:26:48 <sipa> warren: heh?
 534 2013-10-26 08:27:06 <sipa> warren: 209 is what the protocol is initialized at
 535 2013-10-26 08:27:26 <sipa> after a succesful version/verack, that is increasesd to whatever is negotiated
 536 2013-10-26 08:27:36 <warren> oh
 537 2013-10-26 08:28:04 <sipa> and initializing at anything above 209 will likely break things yes, because the peer will not expect certain fields
 538 2013-10-26 08:28:16 <Phoebus> Having issues, haven't used bitcoin-qt in a month, upgraded to last, now 250 weeks behind? No visible progress... left it on last night again, nothing.
 539 2013-10-26 08:28:52 mbelshe has joined
 540 2013-10-26 08:29:25 execut3 has quit (Ping timeout: 260 seconds)
 541 2013-10-26 08:29:41 <sipa> Phoebus: that seems bad
 542 2013-10-26 08:30:34 mbelshe has quit (Client Quit)
 543 2013-10-26 08:30:54 <Phoebus> Any ideas on how to fix it?
 544 2013-10-26 08:32:41 <Ascendion> hmmmm bitcoind seems to like to get into states where it doesnt accept rpc requests in a timely fashion... I get random timeouts after running fine for  period of time
 545 2013-10-26 08:35:27 <Ascendion> it happens when its doing setBestChain from what I just saw in the logs
 546 2013-10-26 08:38:04 <Ascendion> yup -- confirmed -- and if it takes more than the 5 seconds my app allows for retries -- BOOM :)
 547 2013-10-26 08:38:10 robocoin_ has quit (Quit: leaving)
 548 2013-10-26 08:39:16 <warren> sipa: I added a print where it checks MIN_PROTO_VERSION in src/main.cpp, it sees 70001 there
 549 2013-10-26 08:39:37 <warren> sipa: leading me to believe MIN_PROTO_VERSION should be split into two separate variables
 550 2013-10-26 08:40:54 _ingsoc has joined
 551 2013-10-26 08:41:41 <warren> INIT_PROTO_VERSION and MIN_PEER_PROTO_VERSION ?
 552 2013-10-26 08:44:12 <Ascendion> seems to also happen when its commiting to the db.. but those are usually short -- a second or 2 and my retry code handles those fine
 553 2013-10-26 08:44:28 saulimus has joined
 554 2013-10-26 08:44:34 execut3 has joined
 555 2013-10-26 08:44:37 a_meteor has quit (Ping timeout: 272 seconds)
 556 2013-10-26 08:45:10 <warren> Ascendion: for initial block sync, add dbcache=1000 (use 1GB of RAM for cache)
 557 2013-10-26 08:47:01 <Ascendion> hmmm so there isnt like a mutex/semaphore thats blocking RPC when its doing setBestChain ??
 558 2013-10-26 08:49:06 neep3r has joined
 559 2013-10-26 08:49:12 shesek has joined
 560 2013-10-26 08:49:46 <Ascendion> not that worried about it since I can bump up the retries or the interval between retries easy enough.. and it shouldnt be a problem at all once bitcoind is synced
 561 2013-10-26 08:52:45 execut3 has quit (Ping timeout: 260 seconds)
 562 2013-10-26 08:54:20 <sipa> Ascendion: yes, most of the validation logic is protected by cs_main, and most RPCs grab that lock too
 563 2013-10-26 08:54:57 <sipa> warren: yes, the variable could be split
 564 2013-10-26 08:55:20 <sipa> that would indeed stress the importance of the initial version
 565 2013-10-26 08:55:26 shesek has quit (Ping timeout: 252 seconds)
 566 2013-10-26 08:55:31 <warren> are those two names fine?
 567 2013-10-26 08:55:43 <warren> sipa: what should MIN_PEER_PROTO_VERSION be set to?
 568 2013-10-26 08:55:45 xiangfu has joined
 569 2013-10-26 08:56:09 <sipa> 209
 570 2013-10-26 08:56:19 <sipa> we support down to 209 :)
 571 2013-10-26 08:56:37 <warren> wow, that is really compatible?
 572 2013-10-26 08:56:41 <sipa> yes
 573 2013-10-26 08:57:04 <sipa> the validation isn't anymore since the 0.7 fork
 574 2013-10-26 08:57:10 <sipa> but the p2p protocol is
 575 2013-10-26 08:57:19 ThomasV has joined
 576 2013-10-26 08:57:25 <warren> what value is there in supporting the older p2p protocol when theycan't validate?
 577 2013-10-26 08:57:38 <sipa> that's a good question
 578 2013-10-26 08:58:07 <Ascendion> I'd say its time to make some of the slackers get a new client :)
 579 2013-10-26 08:58:13 <sipa> we can probably bump it to 70001 (0.8)
 580 2013-10-26 08:58:28 <sipa> hmm, no
 581 2013-10-26 08:58:37 <warren> Ascendion: for Litecoin I'm bumping it to disconnect all old clients, maybe 10-15% of the network that's ignoring months of alerts
 582 2013-10-26 08:58:46 <sipa> 0.7 and older are still compatible if you tweak the db settings
 583 2013-10-26 08:58:51 <warren> hmm
 584 2013-10-26 08:59:04 <warren> might be better if there were an array of specific proto versions you want to disconnect
 585 2013-10-26 08:59:14 <warren> if they want to use 0.7 + db fix they can bump the version
 586 2013-10-26 09:00:06 <warren> shoot, there's no difference between 0.7 and 0.8 huh?
 587 2013-10-26 09:00:20 <Ascendion> warren -- if you are working on litecoind -- gimme a hint how compatible the rpc interface is since I'm gonna be doing an app similar to the one I'm doing now for LTC once this one is working
 588 2013-10-26 09:00:35 <warren> Ascendion: it's exactly the same as bitcoin 0.8.5
 589 2013-10-26 09:00:47 <warren> Ascendion: what are you working on?
 590 2013-10-26 09:01:21 <Ascendion> cryptocoin aware financials website
 591 2013-10-26 09:01:56 <sipa> warren: eh, the p2p protocol indicates which fields and messages are supported
 592 2013-10-26 09:02:06 dswfherg has joined
 593 2013-10-26 09:02:18 <sipa> warren: it doesn't indicate how validation is done, so imho it should be independent
 594 2013-10-26 09:02:30 <sipa> and someone may actually still run a 0.2.9
 595 2013-10-26 09:02:33 <Ascendion> I cant get much more specific than that due to non-disclosure agreements
 596 2013-10-26 09:03:13 <warren> sipa: and how would that be useful for them at all?
 597 2013-10-26 09:03:15 <sipa> we can talk about dropping compatibility with old clients, but that is a different discussion
 598 2013-10-26 09:03:28 <warren> right, for now I just want to split the variable
 599 2013-10-26 09:03:35 <sipa> fair enough
 600 2013-10-26 09:05:23 <warren> 0.7.2 was src/version.h:static const int PROTOCOL_VERSION = 60002;
 601 2013-10-26 09:09:01 <sipa> yes
 602 2013-10-26 09:09:29 arioBarzan has joined
 603 2013-10-26 09:09:30 <sipa> it was an oversight that 0.8 bumped it to 70001 instead of 60003
 604 2013-10-26 09:09:32 <Ascendion> ok -- what other coins can be said to have the same RPC interface ??
 605 2013-10-26 09:09:40 xiangfu has quit (Ping timeout: 272 seconds)
 606 2013-10-26 09:09:50 <sipa> Ascendion: most are based on old forks of bitcoin
 607 2013-10-26 09:10:01 <sipa> some added a few rpcs of their own
 608 2013-10-26 09:11:23 darkFun has quit (Remote host closed the connection)
 609 2013-10-26 09:12:05 MoALTz has joined
 610 2013-10-26 09:13:32 arioBarzan has quit (Client Quit)
 611 2013-10-26 09:14:22 <Ascendion> I'm not doing very much exotic -- getblockcount, getblockhash, getblock, gettransaction
 612 2013-10-26 09:15:48 robocoin has joined
 613 2013-10-26 09:17:08 <warren> sipa: fortunate oversight.  60003 can be 0.7 + db fix
 614 2013-10-26 09:17:34 darkFun has joined
 615 2013-10-26 09:17:47 robocoin has quit (Client Quit)
 616 2013-10-26 09:18:05 shesek has joined
 617 2013-10-26 09:18:44 robocoin has joined
 618 2013-10-26 09:20:28 <warren> /usr/bin/ld: leveldb/libmemenv.a(memenv.o): relocation R_X86_64_32S against `.rodata' can not be used when making a shared object; recompile with -fPIC
 619 2013-10-26 09:20:30 <warren> grrrr
 620 2013-10-26 09:20:38 <warren> I know this is just a one-time problem
 621 2013-10-26 09:20:44 <warren> but I hit the one-time problem multiple times
 622 2013-10-26 09:20:46 <warren> =)
 623 2013-10-26 09:20:57 <warren> we could have automated this away
 624 2013-10-26 09:22:12 <sipa> warren: you don't get it; there is no "db fix"
 625 2013-10-26 09:22:18 <sipa> warren: it's just a configuration parameter
 626 2013-10-26 09:22:24 <sipa> not a change to the binary
 627 2013-10-26 09:22:56 <warren> sipa: right ... ok
 628 2013-10-26 09:23:53 shesek has quit (Ping timeout: 272 seconds)
 629 2013-10-26 09:26:30 <sipa> and it applies to any version, not just 0.7.2
 630 2013-10-26 09:26:44 darkFun has quit (Remote host closed the connection)
 631 2013-10-26 09:27:12 Grouver has joined
 632 2013-10-26 09:30:25 CryptoBuck has quit (Ping timeout: 240 seconds)
 633 2013-10-26 09:32:45 <Ascendion> geeez if you would just make it so setBestChain ran in smaller increments !!!
 634 2013-10-26 09:34:28 <sipa> ?
 635 2013-10-26 09:34:33 <sipa> it runs once per block...
 636 2013-10-26 09:34:38 <sipa> can't go smaller than that
 637 2013-10-26 09:35:11 <Ascendion> huh ?? I'm watching the logs -- it pops off setBestChain for sometimes 100s of blocks in a row and holds the lock that stalls the RPC for that entire time
 638 2013-10-26 09:35:25 ThomasV has quit (Ping timeout: 240 seconds)
 639 2013-10-26 09:35:25 <sipa> ah, that's not setbestchain
 640 2013-10-26 09:35:52 <sipa> if you receive a block that connects multiple orphans, they are all added at once
 641 2013-10-26 09:35:56 <sipa> that could change indeed
 642 2013-10-26 09:36:11 <sipa> but it requires some refactors that aren't yet in 0.8.5
 643 2013-10-26 09:37:03 shesek has joined
 644 2013-10-26 09:38:46 a_meteor has joined
 645 2013-10-26 09:39:14 <Ascendion> I'm watching one chunk right now thats been going over a minute :)
 646 2013-10-26 09:40:09 <feddy3> Could a transaction with 0 inputs ever be considered standard?
 647 2013-10-26 09:40:26 <Ascendion> dont see how
 648 2013-10-26 09:40:41 <Ascendion> no inputs implies no outputs so its not a transaction :)
 649 2013-10-26 09:41:21 <sipa> feddy3: it's not just non-standard, it is outright invalid
 650 2013-10-26 09:41:33 <sipa> and changing that requires a hard forking change
 651 2013-10-26 09:41:40 <feddy3> sipa: ok, cool.
 652 2013-10-26 09:41:48 <feddy3> sipa: seen the latest version of fetchtx.info?  You can paste in your own tx now.
 653 2013-10-26 09:42:01 <sipa> feddy3: haven't tried yet; will do so soon
 654 2013-10-26 09:42:46 <feddy3> cool:)
 655 2013-10-26 09:43:06 <feddy3> It also hands some basic scripts
 656 2013-10-26 09:43:15 <feddy3> handles
 657 2013-10-26 09:44:08 CryptoBuck has joined
 658 2013-10-26 09:45:20 ThomasV has joined
 659 2013-10-26 09:47:26 askmike has joined
 660 2013-10-26 09:49:02 <feddy3> What about a transaction with 0 outputs? Is that valid?
 661 2013-10-26 09:50:04 <Ascendion> only if you like throwing away coins :)
 662 2013-10-26 09:50:04 <sipa> no
 663 2013-10-26 09:50:10 <sipa> no, it is invalid
 664 2013-10-26 09:50:21 <sipa> you _can_ have an output with value 0.0000000 BTC
 665 2013-10-26 09:50:27 <Ascendion> was joking :)
 666 2013-10-26 09:50:29 <sipa> (which is non-standard but allowed)
 667 2013-10-26 09:50:42 <feddy3> right
 668 2013-10-26 09:51:07 <sipa> feddy3: for coinbases there is even the extra constraint that there is exactly 1 input
 669 2013-10-26 09:52:05 <warren> sipa: https://github.com/bitcoin/bitcoin/pull/3155
 670 2013-10-26 09:53:14 Eiii has quit ()
 671 2013-10-26 09:55:08 <Ascendion> crash time -- gotta let me brains unfry before I work on the transaction handling :)
 672 2013-10-26 09:56:23 <feddy3> sipa: right.. as far as my website is concerned, it doesn't know if a tx is meant to be a coinbase or not so I can't warn or error on that
 673 2013-10-26 09:56:49 one_zero has joined
 674 2013-10-26 09:57:09 <sipa> feddy3: actually, you can know that
 675 2013-10-26 09:57:29 <sipa> as only coinbase inputs have a prevout referring to 00000...:0xFFFF
 676 2013-10-26 10:02:15 <feddy3> ahh, good point
 677 2013-10-26 10:03:38 neep3r has quit (Remote host closed the connection)
 678 2013-10-26 10:05:09 Dyaheon- has quit ()
 679 2013-10-26 10:05:10 <sipa> and such a txi  should probably be parsed/shown differently
 680 2013-10-26 10:05:23 <sipa> as its script is not actually a script
 681 2013-10-26 10:07:37 <feddy3> special case that I need to handle:)
 682 2013-10-26 10:08:05 FabianB has joined
 683 2013-10-26 10:08:46 FabianB_ has quit (Ping timeout: 240 seconds)
 684 2013-10-26 10:09:10 mappum has joined
 685 2013-10-26 10:09:48 patcon has joined
 686 2013-10-26 10:10:35 darkFun has joined
 687 2013-10-26 10:14:36 darkFun has quit (Remote host closed the connection)
 688 2013-10-26 10:16:48 mappum has quit (Ping timeout: 272 seconds)
 689 2013-10-26 10:17:36 jtimon has joined
 690 2013-10-26 10:18:58 patcon has quit (Ping timeout: 245 seconds)
 691 2013-10-26 10:20:17 deepc0re has joined
 692 2013-10-26 10:28:34 _ingsoc has quit (Quit: leaving)
 693 2013-10-26 10:32:24 deepc0re has quit (Quit: deepc0re)
 694 2013-10-26 10:33:07 deepc0re has joined
 695 2013-10-26 10:33:23 _ingsoc has joined
 696 2013-10-26 10:35:54 olalonde has joined
 697 2013-10-26 10:35:59 <olalonde> are transactions with 0 fee propagated to the network?
 698 2013-10-26 10:36:25 <sipa> sometimes
 699 2013-10-26 10:37:45 <olalonde> why would nodes refuse to propagate those transactions? they have nothing to gain or lose...
 700 2013-10-26 10:37:53 ThomasV has quit (Quit: Quitte)
 701 2013-10-26 10:39:02 Dyaheon has joined
 702 2013-10-26 10:39:23 Polyatomic has quit (Quit: Catcha Round)
 703 2013-10-26 10:39:27 <sipa> DoS protection
 704 2013-10-26 10:39:36 <olalonde> ok
 705 2013-10-26 10:39:49 patcon has joined
 706 2013-10-26 10:40:36 <olalonde> how does the standard client determines the appropriate fee for a transaction? does it average a fee / transaction size according to most recent blocks?
 707 2013-10-26 10:40:57 <petertodd> olalonde: it's hardcoded/user chosen right now
 708 2013-10-26 10:41:03 <olalonde> ok thanks
 709 2013-10-26 10:41:07 patcon has quit (Remote host closed the connection)
 710 2013-10-26 10:41:24 <petertodd> olalonde: gavin's proposed a fee estimation scheme, which could work in the general case, but it will alwasy have edge cases where it gets the fees wrong
 711 2013-10-26 10:41:43 patcon has joined
 712 2013-10-26 10:42:33 macboz has joined
 713 2013-10-26 10:44:55 patcon_ has joined
 714 2013-10-26 10:45:55 patcon has quit (Read error: Connection reset by peer)
 715 2013-10-26 10:45:55 darkFun has joined
 716 2013-10-26 10:46:20 patcon_ has quit (Remote host closed the connection)
 717 2013-10-26 10:46:54 patcon has joined
 718 2013-10-26 10:51:05 patcon has quit (Ping timeout: 240 seconds)
 719 2013-10-26 10:52:21 ovidiusoft has joined
 720 2013-10-26 10:55:45 rdymac has quit (Excess Flood)
 721 2013-10-26 10:57:29 rdymac has joined
 722 2013-10-26 10:58:55 PhantomSpark has quit (Read error: Connection reset by peer)
 723 2013-10-26 10:59:15 PhantomSpark has joined
 724 2013-10-26 10:59:31 ircuser-6 has quit (Ping timeout: 272 seconds)
 725 2013-10-26 11:00:25 btcbtc has quit (Ping timeout: 272 seconds)
 726 2013-10-26 11:03:45 Neskia has joined
 727 2013-10-26 11:04:25 neep3r has joined
 728 2013-10-26 11:05:18 Nesetalis has quit (Read error: Connection reset by peer)
 729 2013-10-26 11:08:55 ThomasV has joined
 730 2013-10-26 11:13:25 mrkent has quit (Read error: No route to host)
 731 2013-10-26 11:15:37 neep3r has quit (Ping timeout: 245 seconds)
 732 2013-10-26 11:16:01 <olalonde> is it possible to MiTM attack a bitcoin client or would that be easily detected?
 733 2013-10-26 11:16:39 <olalonde> i guess if the client doesnt ~ 6 confirmation / hour  there should be some kind of alert no?
 734 2013-10-26 11:17:26 <olalonde> otherwise the client would be vulnerable to the double spend attack
 735 2013-10-26 11:17:31 <olalonde> isn't it?
 736 2013-10-26 11:21:15 asuk has joined
 737 2013-10-26 11:25:41 <Belxjander> olalonde: it doesn't work that way since the keys are private and never transfer over the network
 738 2013-10-26 11:26:15 <Belxjander> the transactions record "signing" a transfer of value so even if there was an MITM attack... it suffers from nothing of value actually being throughput
 739 2013-10-26 11:27:00 <Belxjander> all it could do is silence "spending" from a particular IP... and the client may dynamically connect to somewhere else before dropping the link rendering the MITM pointless and a waste of time
 740 2013-10-26 11:27:43 <Belxjander> the only possible use of such an attack would be to verify a particular BitCoinAddress to matching a particular IPaddress within the timeframe of the attack
 741 2013-10-26 11:28:14 <olalonde> I don't think u are right
 742 2013-10-26 11:28:48 <olalonde> couldn't u fool a client by sending it a chain u control and double spend this client after 6 confirmations ?
 743 2013-10-26 11:28:58 <olalonde> this would probably take days even with lots of hashing power
 744 2013-10-26 11:29:12 <olalonde> so not a very likely attack
 745 2013-10-26 11:30:05 <olalonde> lets say u can MiTM attack MtGox... u could send it 6 confirmations about a deposit u made and then withdraw the bitcoins
 746 2013-10-26 11:30:45 <olalonde> u can then stop the MiTM attack and MtGox will realize that the 6confirmations are on a shorter block chain and need to be discarded
 747 2013-10-26 11:30:56 <olalonde> MtGox has lost X bitcoins, u win X bitcoins
 748 2013-10-26 11:31:19 <olalonde> of course, it would require enormous hashing power to go undetected
 749 2013-10-26 11:31:50 <olalonde> and it would take a few hours or days before u can generate 6 confirmations
 750 2013-10-26 11:32:17 <olalonde> so the likelihood of going undetected is tiny
 751 2013-10-26 11:40:01 <olalonde> maybe it would be a good idea to have a sort of warning in the client when suddenly blocks are solved much more slowly then expected?
 752 2013-10-26 11:45:02 ircuser-6 has joined
 753 2013-10-26 11:45:09 <a_meteor> olalonde, that's not possible
 754 2013-10-26 11:45:34 <a_meteor> the client connects to more than a single other node
 755 2013-10-26 11:45:52 <olalonde> yes but thats possible to fake it your are a man in the middle
 756 2013-10-26 11:46:01 <olalonde> unless there is SSL going on
 757 2013-10-26 11:47:32 <olalonde> you can filter the traffic so that u discard any block relayed except yours for example
 758 2013-10-26 11:50:16 <sipa> olalonde: a MitM attack doesn't apply, since there is no authentication to fake
 759 2013-10-26 11:50:27 <sipa> clients do not authenticate to eachother in the first place
 760 2013-10-26 11:50:37 <sipa> what you can do, is a sybil attack, and i think that is what you mean
 761 2013-10-26 11:50:47 <sipa> namely surrounding a node with only attacker peers
 762 2013-10-26 11:50:55 <olalonde> yep thats what i mean
 763 2013-10-26 11:50:58 <sipa> to prevent it from seeing the actual best chain
 764 2013-10-26 11:51:02 <olalonde> yep
 765 2013-10-26 11:51:05 hnz has quit (Ping timeout: 272 seconds)
 766 2013-10-26 11:51:15 <sipa> there are several measures to prevent it, but yes, it is a possible attack
 767 2013-10-26 11:51:27 <warren> some of us are in favor of optional authentication, as means to detect that you're surrounded by hostile nodes
 768 2013-10-26 11:51:34 tmsk has joined
 769 2013-10-26 11:51:39 <warren> 1 or 2 authenticated + 6 random might be good
 770 2013-10-26 11:52:30 <sipa> you can do that by -addnode'ing an onion peer :)
 771 2013-10-26 11:52:36 <olalonde> couldnt u just check for the hash rate?
 772 2013-10-26 11:52:44 <olalonde> if it drops down significantly it should raise red flags
 773 2013-10-26 11:52:53 asuk is now known as asuk|afk
 774 2013-10-26 11:53:00 <sipa> olalonde: can you please use proper english?
 775 2013-10-26 11:53:26 <sipa> olalonde: and significiant hash rate dropping should indeed raise red flags
 776 2013-10-26 11:53:44 <sipa> but you it is very hard to measure in a short term
 777 2013-10-26 11:53:54 <sipa> and in the long term, significant variations are expected
 778 2013-10-26 11:54:02 asuk has quit (afk!~asuk@31.129.27.89|)
 779 2013-10-26 11:54:16 <olalonde> regarding my english, it would probably be more productive for you to point out my mistakes ;)
 780 2013-10-26 11:54:26 <sipa> 'u' is not a word
 781 2013-10-26 11:54:37 <sipa> warren: nonetheless, i'm in favor of having authentication in the protocol
 782 2013-10-26 11:54:45 hnz has joined
 783 2013-10-26 11:55:47 <olalonde> ok I will try my best but I'm used to type u instead of you when chatting. bad habit I guess
 784 2013-10-26 11:57:28 <sipa> thanks :)
 785 2013-10-26 11:59:19 asuk has joined
 786 2013-10-26 12:00:31 patcon has joined
 787 2013-10-26 12:01:36 patcon has quit (Remote host closed the connection)
 788 2013-10-26 12:02:11 patcon has joined
 789 2013-10-26 12:03:43 AusBitBank has quit (Ping timeout: 245 seconds)
 790 2013-10-26 12:05:01 dlidstrom has quit (Ping timeout: 272 seconds)
 791 2013-10-26 12:06:39 patcon has quit (Ping timeout: 272 seconds)
 792 2013-10-26 12:09:00 jtimon_ has joined
 793 2013-10-26 12:11:05 jtimon has quit (Ping timeout: 272 seconds)
 794 2013-10-26 12:11:26 neep3r has joined
 795 2013-10-26 12:15:08 * jgarzik wakes up
 796 2013-10-26 12:15:16 <jgarzik> theymos or anyone fixed the forum yet?
 797 2013-10-26 12:15:48 <sipa> is it broken?
 798 2013-10-26 12:16:15 neep3r has quit (Ping timeout: 265 seconds)
 799 2013-10-26 12:16:39 <CodeShark> it never looked fully functional even when it didn't have software or hacking issues
 800 2013-10-26 12:16:50 <sipa> that's another issue :)
 801 2013-10-26 12:17:09 asuk is now known as asuk|afk
 802 2013-10-26 12:18:04 <CodeShark> sipa, would a tool that generates partial merkle tree diagrams from arbitrary inputs be useful, you think?
 803 2013-10-26 12:18:33 <CodeShark> I'm thinking of just outputting Mathematica code
 804 2013-10-26 12:19:39 darkFun has quit (Remote host closed the connection)
 805 2013-10-26 12:19:50 darkFun has joined
 806 2013-10-26 12:21:37 darkFun has quit (Remote host closed the connection)
 807 2013-10-26 12:22:39 <jgarzik> sipa, someone hacked an account -- or the whole forum -- and started replacing existing posts with a image-post about dogs in Vietnam
 808 2013-10-26 12:22:54 <jgarzik> I deleted a few posts before realizing my mistake
 809 2013-10-26 12:23:35 darkFun has joined
 810 2013-10-26 12:23:41 <jgarzik> sipa, So, it was the kind of thing that requires an annoying bit of cleanup, not just a simple troll ban
 811 2013-10-26 12:24:08 <jgarzik> really, the forum needs to store its data in git, to enable reverts :)
 812 2013-10-26 12:25:03 <jgarzik> I also don't like how tons of minor admins on the site can edit any post.  They claim it is so that they can fix broken bbcode, which is claimed a necessity.  Just creates broken situations like this...
 813 2013-10-26 12:29:25 <sipa> jgarzik: i really don't care about the forum
 814 2013-10-26 12:29:46 shesek has quit (Read error: Connection reset by peer)
 815 2013-10-26 12:30:06 <jgarzik> sipa, probably wise :
 816 2013-10-26 12:30:07 <jgarzik> :)
 817 2013-10-26 12:30:18 shesek has joined
 818 2013-10-26 12:31:13 execut3 has joined
 819 2013-10-26 12:32:48 darkFun has quit (Read error: Connection reset by peer)
 820 2013-10-26 12:34:49 shesek has quit (Ping timeout: 245 seconds)
 821 2013-10-26 12:35:31 murkmans has joined
 822 2013-10-26 12:40:12 canoon has quit (Ping timeout: 245 seconds)
 823 2013-10-26 12:42:18 one_zero has quit ()
 824 2013-10-26 12:43:11 chris_l has joined
 825 2013-10-26 12:46:05 Ogig has joined
 826 2013-10-26 12:46:09 grau has joined
 827 2013-10-26 12:46:38 shesek has joined
 828 2013-10-26 12:49:33 grau_ has quit (Ping timeout: 245 seconds)
 829 2013-10-26 12:49:41 execut3 has quit (Ping timeout: 248 seconds)
 830 2013-10-26 12:50:46 asuk has quit (afk!~asuk@31.129.27.89|)
 831 2013-10-26 12:52:39 canoon has joined
 832 2013-10-26 12:53:43 a_meteor has quit (Ping timeout: 245 seconds)
 833 2013-10-26 12:55:44 <sipa> CodeShark: sounds useful... we could probably use some nice example images within the BIP
 834 2013-10-26 12:55:51 <sipa> i don't know any mathematica though
 835 2013-10-26 12:56:30 <CodeShark> more useful might be a javascript-generated graph
 836 2013-10-26 12:57:17 shesek has quit (Ping timeout: 245 seconds)
 837 2013-10-26 12:57:39 <CodeShark> I used mathematica to generate the examples I posted yesterday
 838 2013-10-26 12:58:38 Dyaheon has quit ()
 839 2013-10-26 12:58:51 Grouver has quit (Quit:  I love my HydraIRC -> http://www.hydrairc.com <-)
 840 2013-10-26 12:59:05 mappum has joined
 841 2013-10-26 12:59:16 sustrik has joined
 842 2013-10-26 12:59:58 Dyaheon has joined
 843 2013-10-26 13:02:55 Ashaman has quit (Ping timeout: 262 seconds)
 844 2013-10-26 13:04:12 Grouver has joined
 845 2013-10-26 13:06:17 mapppum has joined
 846 2013-10-26 13:09:25 mappum has quit (Ping timeout: 265 seconds)
 847 2013-10-26 13:09:55 abrkn has joined
 848 2013-10-26 13:10:37 sustrik has quit (Ping timeout: 245 seconds)
 849 2013-10-26 13:12:07 neep3r has joined
 850 2013-10-26 13:13:54 shesek has joined
 851 2013-10-26 13:16:01 elgrecoFL has quit (Excess Flood)
 852 2013-10-26 13:16:21 Goonie has quit (Remote host closed the connection)
 853 2013-10-26 13:16:53 shesek has quit (Read error: No route to host)
 854 2013-10-26 13:17:13 neep3r has quit (Ping timeout: 272 seconds)
 855 2013-10-26 13:17:16 Ashaman has joined
 856 2013-10-26 13:17:38 elgrecoFL has joined
 857 2013-10-26 13:17:58 shesek has joined
 858 2013-10-26 13:19:02 dgolds has joined
 859 2013-10-26 13:19:12 deepc0re has quit (Quit: deepc0re)
 860 2013-10-26 13:20:30 shesek has quit (Max SendQ exceeded)
 861 2013-10-26 13:21:22 shesek has joined
 862 2013-10-26 13:22:54 Grouver has quit (Quit:  HydraIRC -> http://www.hydrairc.com <- In tests, 0x09 out of 0x0A l33t h4x0rz prefer it :))
 863 2013-10-26 13:24:27 shesek has quit (Max SendQ exceeded)
 864 2013-10-26 13:24:54 shesek has joined
 865 2013-10-26 13:24:58 olalonde has quit (Ping timeout: 245 seconds)
 866 2013-10-26 13:26:58 shesek has quit (Max SendQ exceeded)
 867 2013-10-26 13:27:25 shesek has joined
 868 2013-10-26 13:33:10 asuk has joined
 869 2013-10-26 13:33:14 xiangfu has joined
 870 2013-10-26 13:34:18 ValicekB has quit (Remote host closed the connection)
 871 2013-10-26 13:34:38 dgolds has quit (Remote host closed the connection)
 872 2013-10-26 13:36:29 shesek has quit (Ping timeout: 265 seconds)
 873 2013-10-26 13:41:15 dgolds has joined
 874 2013-10-26 13:44:47 rdymac has quit (Excess Flood)
 875 2013-10-26 13:46:15 ValicekB has joined
 876 2013-10-26 13:46:30 rdymac has joined
 877 2013-10-26 13:47:50 xiangfu has quit (Ping timeout: 267 seconds)
 878 2013-10-26 13:47:52 Ashaman has quit (Ping timeout: 267 seconds)
 879 2013-10-26 13:48:09 xiangfu has joined
 880 2013-10-26 13:49:39 <sipa> BlueMatt: the comparison tool is still disabled in pulltester?
 881 2013-10-26 13:51:49 dgolds has quit (Remote host closed the connection)
 882 2013-10-26 13:52:04 saulimus has quit (Quit: saulimus)
 883 2013-10-26 13:52:24 dgolds has joined
 884 2013-10-26 13:54:19 askmike has quit (Remote host closed the connection)
 885 2013-10-26 13:54:47 askmike has joined
 886 2013-10-26 13:55:37 <Anduck> hmm
 887 2013-10-26 13:55:48 dgolds has quit (Read error: No route to host)
 888 2013-10-26 13:55:48 <Anduck> i made a tx with bitcoind, it didnt propogate...
 889 2013-10-26 13:55:54 <Anduck> i pushed it at blockchain.info/pushtx
 890 2013-10-26 13:55:58 <Anduck> it didnt propogate
 891 2013-10-26 13:56:13 <Anduck> propagate*
 892 2013-10-26 13:56:20 dgolds has joined
 893 2013-10-26 13:56:34 _ingsoc has quit (Ping timeout: 272 seconds)
 894 2013-10-26 13:57:33 <Anduck> now i made it again... and it still doesn't propagate
 895 2013-10-26 13:58:20 askmike_ has joined
 896 2013-10-26 13:58:40 asuk is now known as asuk|afk
 897 2013-10-26 13:59:19 askmike has quit (Ping timeout: 246 seconds)
 898 2013-10-26 14:00:11 asuk has quit (afk!~asuk@31.129.27.89|)
 899 2013-10-26 14:01:42 askmike_ has quit (Remote host closed the connection)
 900 2013-10-26 14:02:08 askmike has joined
 901 2013-10-26 14:02:57 <sipa> anything that appears in bitcoind's debug.log when you do sendrawtransaction?
 902 2013-10-26 14:05:26 <Anduck> it's made with sendmany (not raw tx).. but i then took the raw tx hash and tried to send it via sendrawtransaction
 903 2013-10-26 14:05:30 <Anduck> it didnt send it...
 904 2013-10-26 14:05:36 <Anduck> but blockchain.info/pushtx accepted it
 905 2013-10-26 14:05:59 <Anduck> when i tried sendrawtransaction of the hex, it gave me error: {"code":-22,"message":"TX rejected"}
 906 2013-10-26 14:06:05 <sipa> you haven't answered my question :)
 907 2013-10-26 14:06:15 <Anduck> well
 908 2013-10-26 14:06:20 Ashaman has joined
 909 2013-10-26 14:06:26 _ingsoc has joined
 910 2013-10-26 14:06:29 askmike has quit (Ping timeout: 245 seconds)
 911 2013-10-26 14:06:31 <Anduck> i dont see anything special in the debug.log
 912 2013-10-26 14:06:51 <sipa> hmm, i would expect an error message that explains why it is rejected
 913 2013-10-26 14:06:58 dswfherg has quit ()
 914 2013-10-26 14:07:05 <sipa> (and yes, it should be reported via RPC)
 915 2013-10-26 14:07:29 <Anduck> hmm
 916 2013-10-26 14:07:44 <Anduck> there should be something in debug.log when i search it with txid?
 917 2013-10-26 14:07:58 <Anduck> i got ResendWalletTransactions()
 918 2013-10-26 14:07:58 <Anduck> Relaying wtx <txid>
 919 2013-10-26 14:08:12 <sipa> that's only for wallet transactions
 920 2013-10-26 14:08:23 <sipa> ah, if you created it with sendmany, it is
 921 2013-10-26 14:08:38 <sipa> so yes, it is broadcast
 922 2013-10-26 14:08:41 <Anduck> well
 923 2013-10-26 14:08:42 <Anduck> can i pm you?`
 924 2013-10-26 14:08:44 <sipa> yes
 925 2013-10-26 14:12:52 neep3r has joined
 926 2013-10-26 14:14:38 datagutt has joined
 927 2013-10-26 14:14:44 CircusPeanut has joined
 928 2013-10-26 14:15:25 BlackPrapor has joined
 929 2013-10-26 14:17:37 datagutt_ has joined
 930 2013-10-26 14:17:45 neep3r has quit (Ping timeout: 272 seconds)
 931 2013-10-26 14:18:39 BlackPrapor has quit (2!~kvirc@94.41.42.139.dynamic.ufanet.ru|Ping timeout: 272 seconds)
 932 2013-10-26 14:19:09 deepc0re has joined
 933 2013-10-26 14:20:38 datagutt has quit (Ping timeout: 272 seconds)
 934 2013-10-26 14:30:47 rdponticelli has quit (Ping timeout: 240 seconds)
 935 2013-10-26 14:31:06 datagutt_ has quit (Ping timeout: 265 seconds)
 936 2013-10-26 14:32:49 askmike has joined
 937 2013-10-26 14:34:28 askmike_ has joined
 938 2013-10-26 14:37:06 askmike has quit (Ping timeout: 240 seconds)
 939 2013-10-26 14:38:57 askmike_ has quit (Ping timeout: 245 seconds)
 940 2013-10-26 14:45:52 CircusPeanut has quit (Ping timeout: 246 seconds)
 941 2013-10-26 14:54:34 macboz has quit (Quit: This computer has gone to sleep)
 942 2013-10-26 14:58:53 macboz has joined
 943 2013-10-26 14:59:49 ambimorph has joined
 944 2013-10-26 15:01:23 grau has quit (Remote host closed the connection)
 945 2013-10-26 15:03:18 macboz has quit (Ping timeout: 245 seconds)
 946 2013-10-26 15:03:31 Guest5265 has joined
 947 2013-10-26 15:07:56 roconnor has joined
 948 2013-10-26 15:08:08 Guest5265 has quit (Ping timeout: 272 seconds)
 949 2013-10-26 15:13:39 neep3r has joined
 950 2013-10-26 15:17:17 skinnkavaj has quit (Ping timeout: 260 seconds)
 951 2013-10-26 15:18:28 neep3r has quit (Ping timeout: 265 seconds)
 952 2013-10-26 15:18:51 <ThomasV> sipa: https://en.bitcoin.it/wiki/Talk:BIP_0039
 953 2013-10-26 15:19:48 <ThomasV> if you could have a look
 954 2013-10-26 15:22:20 canoon has quit (Ping timeout: 265 seconds)
 955 2013-10-26 15:27:39 skinnkavaj has joined
 956 2013-10-26 15:29:16 CircusPeanut has joined
 957 2013-10-26 15:30:54 patcon has joined
 958 2013-10-26 15:31:10 shesek has joined
 959 2013-10-26 15:31:48 bbrian has joined
 960 2013-10-26 15:32:34 askmike has joined
 961 2013-10-26 15:32:47 <sipa> ThomasV: i'm not sure about putting client-specific information inside
 962 2013-10-26 15:33:23 <ThomasV> sipa: that's not what I mean
 963 2013-10-26 15:33:45 askmike has quit (Read error: Connection reset by peer)
 964 2013-10-26 15:33:58 <melvster> did bitcoin ever settle on a 3 letter short code consensus ie BTC vs XBC ?
 965 2013-10-26 15:34:04 <sipa> also, if we're going to use a one-direction seeding, why drop the ability to have variable attack difficulty?
 966 2013-10-26 15:34:31 <Belxjander> melvster: XBT and BTC?
 967 2013-10-26 15:34:32 askmike has joined
 968 2013-10-26 15:34:52 <melvster> we're going to model all the common 3 letter ISO codes in the web payments group plus the common new crypto currencies like XRP ... im unsure what would be definitive for bitcoin?
 969 2013-10-26 15:35:04 <melvster> there was some threads in the forum, and voting was mixed last i knew
 970 2013-10-26 15:35:10 <sipa> well for ISO there's no way around X<something>
 971 2013-10-26 15:35:21 <melvster> this is not part of ISO
 972 2013-10-26 15:35:29 <sipa> then i don't know
 973 2013-10-26 15:35:33 <melvster> they charge $$$$ for registration
 974 2013-10-26 15:35:35 <ThomasV> sipa: actually I do not know why slush insisted to have a symmetric encryption
 975 2013-10-26 15:35:40 <sipa> ThomasV: neither do i
 976 2013-10-26 15:35:44 <melvster> this is extending what exists in ISO to the web
 977 2013-10-26 15:36:00 <sipa> ThomasV: but i'm saying that if that requirement isn't there, i find my own proposal way better
 978 2013-10-26 15:36:02 <melvster> if the bitcoin community is settled on BTC we can use that
 979 2013-10-26 15:36:14 <sipa> as it automatically becomes stronger over time
 980 2013-10-26 15:36:38 <sipa> (it encodes its own number of iterations)
 981 2013-10-26 15:36:52 <ThomasV> sipa: I probably didn't understand everything about your proposal.. but could you add metadata to it?
 982 2013-10-26 15:36:54 patcon has quit (Ping timeout: 245 seconds)
 983 2013-10-26 15:37:11 <melvster> ok ... im going to go with "BTC" unless there's any objections ...
 984 2013-10-26 15:37:32 <sipa> ThomasV: not without breaking its nice mathematical properties, i think
 985 2013-10-26 15:37:47 <ThomasV> that's the feeling I had
 986 2013-10-26 15:37:50 paraipan has joined
 987 2013-10-26 15:37:52 <sipa> but there's no need
 988 2013-10-26 15:37:58 <sipa> encode the metadata outside
 989 2013-10-26 15:38:06 <ThomasV> where?
 990 2013-10-26 15:38:10 <sipa> i.e., as a property of the unhashed string
 991 2013-10-26 15:38:34 <sipa> that has no computational overhead
 992 2013-10-26 15:38:51 <ThomasV> I see
 993 2013-10-26 15:38:54 <sipa> it does make the string longer, but that is inevitable if you want to add data
 994 2013-10-26 15:40:05 <ThomasV> yes but if we add it to the string, we fall back to the problem of agreeing on a dictionary
 995 2013-10-26 15:40:12 <sipa> not at all
 996 2013-10-26 15:40:22 ambimorph1 has joined
 997 2013-10-26 15:40:23 <sipa> it will be a client-specific property
 998 2013-10-26 15:40:30 <sipa> but that doesn't have to be a dictionary
 999 2013-10-26 15:40:41 Grouver has joined
1000 2013-10-26 15:40:45 <sipa> something like "if the first vowel is an a, use ..." or something
1001 2013-10-26 15:41:05 <sipa> so when generating your seed string, only search with words where the first vowel is an a
1002 2013-10-26 15:41:35 <ThomasV> I wanted to avoid any assumption on the string (eg to have chinese)
1003 2013-10-26 15:41:53 ambimorph has quit (Ping timeout: 252 seconds)
1004 2013-10-26 15:41:55 <sipa> ok, then take some hash over the first N bytes of the string, and use a property of that
1005 2013-10-26 15:42:13 <sipa> or something like that
1006 2013-10-26 15:42:13 <ThomasV> isn't that what I do?
1007 2013-10-26 15:43:13 <sipa> but I is the output, or not?
1008 2013-10-26 15:43:18 Applicat_ has joined
1009 2013-10-26 15:43:26 <sipa> what is the actual masterkey/seed/... that is generated?
1010 2013-10-26 15:44:54 PhantomSpark has quit (Read error: Connection reset by peer)
1011 2013-10-26 15:45:08 <ThomasV> yes, I isthe output
1012 2013-10-26 15:45:16 PhantomSpark has joined
1013 2013-10-26 15:45:22 <ThomasV> it will be rehashed in bip32
1014 2013-10-26 15:45:27 porquilho has joined
1015 2013-10-26 15:45:42 <sipa> right, but you're defining your metadata as part of the output
1016 2013-10-26 15:45:59 <sipa> i'm saying that if you do it as a property of the input, it can be done without computational overhead
1017 2013-10-26 15:46:01 Application has quit (Ping timeout: 265 seconds)
1018 2013-10-26 15:46:18 <sipa> and without reducing the entropy in the actual key used
1019 2013-10-26 15:46:24 <sipa> (which is what you're doing now)
1020 2013-10-26 15:47:04 <ThomasV> it's easy to see how much entropy I reduce, and to offset that
1021 2013-10-26 15:47:05 cris has quit (Ping timeout: 240 seconds)
1022 2013-10-26 15:47:17 <sipa> how do you offset that?
1023 2013-10-26 15:47:42 <ThomasV> by adding an extra word to my string
1024 2013-10-26 15:47:51 <sipa> that doesn't change anything
1025 2013-10-26 15:48:00 <ThomasV> yes it does
1026 2013-10-26 15:48:17 <sipa> an attacker will still know the first bits of the actual key
1027 2013-10-26 15:48:45 <sipa> which probably doesn't matter, as searching over all input strings is probably much faster than searching over the output strings
1028 2013-10-26 15:48:51 <ThomasV> but that I is rehashed
1029 2013-10-26 15:48:54 chris_l has quit (Remote host closed the connection)
1030 2013-10-26 15:48:56 <sipa> so?
1031 2013-10-26 15:49:11 <sipa> it still reduces the number of possibilities
1032 2013-10-26 15:49:12 <ThomasV> which first bits are you talking about?
1033 2013-10-26 15:49:18 <sipa> of I
1034 2013-10-26 15:49:44 <sipa> sorry, i said actual key before, i meant I
1035 2013-10-26 15:49:49 <ThomasV> I see
1036 2013-10-26 15:49:54 <sipa> but that doesn't really matter, as I -> key is fast
1037 2013-10-26 15:50:16 <sipa> the entropy reduction is probably only theoretical
1038 2013-10-26 15:50:26 <sipa> but if you can put a property on the input string to encode the metadata
1039 2013-10-26 15:50:32 <sipa> it doesn't cost you anything at all
1040 2013-10-26 15:50:45 <sipa> as you can just iterate over only inputs that have that property
1041 2013-10-26 15:50:46 <ThomasV> ok, that's why you use the n-1 th hash in your proposal?
1042 2013-10-26 15:51:15 <sipa> yes
1043 2013-10-26 15:51:27 <sipa> to compensate for the lack of entropy with extra iterations
1044 2013-10-26 15:51:29 dgolds has quit (Ping timeout: 245 seconds)
1045 2013-10-26 15:51:49 <porquilho> hey in Clarkmoody site the last thing i see is: '16:41:59 > MtGox: Connected'
1046 2013-10-26 15:51:55 <porquilho> and doesnt show time&Sales
1047 2013-10-26 15:51:57 <porquilho> or updates anything
1048 2013-10-26 15:51:59 <ThomasV> I could just do like you, hash twice in the loop
1049 2013-10-26 15:52:01 <porquilho> it just stay like that
1050 2013-10-26 15:52:13 <sipa> ThomasV: why not use my proposal in the first place? :(
1051 2013-10-26 15:52:25 <sipa> you can put any metadata property on the input
1052 2013-10-26 15:52:28 dgolds has joined
1053 2013-10-26 15:52:47 <ThomasV> sipa: I'll try to understand it
1054 2013-10-26 15:53:11 <sipa> ThomasV: what i mean is, when generating an input string, only iterate over strings that have a prefix X
1055 2013-10-26 15:53:19 <sipa> where X has some property that has meaning for you
1056 2013-10-26 15:53:35 <ThomasV> ok
1057 2013-10-26 15:54:00 <sipa> what you add to X is your entropy
1058 2013-10-26 15:54:19 <sipa> so it makes the string longer exactly by how much your metadata demands
1059 2013-10-26 15:55:25 <ThomasV> yes, but it constraints the type of string that is going to be used
1060 2013-10-26 15:55:39 <sipa> well obviously metadata will add a constraint!
1061 2013-10-26 15:55:48 <ThomasV> sure :)
1062 2013-10-26 15:55:59 <sipa> but it doesn't have to be a constraint that translates to a limiting the dictionary
1063 2013-10-26 15:56:07 <sipa> use the hash of the first word for example
1064 2013-10-26 15:56:12 cris has joined
1065 2013-10-26 15:56:23 grau has joined
1066 2013-10-26 15:57:05 <ThomasV> that's a big constraint, one will have to add words that satisfy the desired hash
1067 2013-10-26 15:57:22 <ThomasV> and when the desired hash changes, one will have to change the dictionary
1068 2013-10-26 15:57:29 tmsk has quit (Quit: tmsk)
1069 2013-10-26 15:57:30 <sipa> i mean just some bits of that hash
1070 2013-10-26 15:57:39 <ThomasV> yes, but still
1071 2013-10-26 15:57:41 paraipan has quit (Remote host closed the connection)
1072 2013-10-26 15:57:53 <ThomasV> sipa: I'll think about it and get back to you
1073 2013-10-26 15:57:53 <sipa> how is it any different than the constraint you're proposing?
1074 2013-10-26 15:58:01 <sipa> except faster
1075 2013-10-26 15:58:21 <porquilho> can someone help with this bitcoin.clarkmoody thing?
1076 2013-10-26 15:58:31 <porquilho> it doesnt show Time&Sales
1077 2013-10-26 15:58:33 <sipa> porquilho: offtopic here, i'm afraid
1078 2013-10-26 15:58:42 <porquilho> i think their dev was here
1079 2013-10-26 15:58:56 <sipa> ok, then ask him
1080 2013-10-26 15:59:03 <porquilho> i dont remember what was his nickname
1081 2013-10-26 15:59:22 <sipa> then send a mail or something
1082 2013-10-26 16:00:06 _Avyd has quit (Ping timeout: 240 seconds)
1083 2013-10-26 16:00:26 cris has quit (Ping timeout: 240 seconds)
1084 2013-10-26 16:00:51 crass has quit (Read error: Connection reset by peer)
1085 2013-10-26 16:01:11 _Avyd has joined
1086 2013-10-26 16:01:56 crass has joined
1087 2013-10-26 16:02:53 bbrian has quit (Ping timeout: 245 seconds)
1088 2013-10-26 16:04:37 cris has joined
1089 2013-10-26 16:06:48 Applicat_ has quit (Ping timeout: 265 seconds)
1090 2013-10-26 16:07:35 PhantomSpark has quit (Ping timeout: 272 seconds)
1091 2013-10-26 16:08:01 ColinT has joined
1092 2013-10-26 16:10:01 abueesp has joined
1093 2013-10-26 16:10:36 Application has joined
1094 2013-10-26 16:12:01 jegz has joined
1095 2013-10-26 16:13:09 Subo1977_ has joined
1096 2013-10-26 16:13:27 Subo1977 has quit (Ping timeout: 240 seconds)
1097 2013-10-26 16:14:17 askmike has quit (Remote host closed the connection)
1098 2013-10-26 16:14:24 neep3r has joined
1099 2013-10-26 16:14:43 askmike has joined
1100 2013-10-26 16:14:57 owowo has joined
1101 2013-10-26 16:17:15 <porquilho> Okay i found replys for my problem
1102 2013-10-26 16:17:21 <porquilho> https://bitcointalk.org/index.php?topic=43514.560
1103 2013-10-26 16:17:22 <porquilho> its there
1104 2013-10-26 16:17:27 justusranvier has quit (Ping timeout: 240 seconds)
1105 2013-10-26 16:17:44 <porquilho> so for anyone who also has problems on clarkmoody not updating book/Time&Sales etc
1106 2013-10-26 16:17:49 <porquilho> go there
1107 2013-10-26 16:18:05 dfsfsd121312 has joined
1108 2013-10-26 16:19:17 neep3r has quit (Ping timeout: 252 seconds)
1109 2013-10-26 16:19:30 askmike has quit (Ping timeout: 268 seconds)
1110 2013-10-26 16:20:43 dfsfsd121312 has quit (Client Quit)
1111 2013-10-26 16:24:06 Belkaar has quit (Ping timeout: 240 seconds)
1112 2013-10-26 16:26:23 Belkaar has joined
1113 2013-10-26 16:27:17 <Ascendion> hmmmm check me on this -- if I want to look at transactions not associated with an address in the wallet -- I need to look at the raw transaction data ??
1114 2013-10-26 16:27:17 _ingsoc has quit (Ping timeout: 260 seconds)
1115 2013-10-26 16:27:19 datagutt has joined
1116 2013-10-26 16:28:11 _ingsoc has joined
1117 2013-10-26 16:29:12 <sipa> Ascendion: use getrawtransaction
1118 2013-10-26 16:29:26 <sipa> (after enabling txindex, which needs a reindex)
1119 2013-10-26 16:30:22 <sipa> for mempool transactions it should always work, actually
1120 2013-10-26 16:31:00 nomailing has joined
1121 2013-10-26 16:31:16 gingpark has joined
1122 2013-10-26 16:31:43 justusranvier has joined
1123 2013-10-26 16:32:59 <Ascendion> I know I'm a newb as far as bitcoin internals go -- so far you've mentioned coinbase and mempool transactions -- what are those and what others might there be ?? :)
1124 2013-10-26 16:33:24 CryptoBuck has quit (Read error: Connection reset by peer)
1125 2013-10-26 16:35:52 <sipa> coinbase tramsactions are the ones that introduce new currency, they are only creatred by miners
1126 2013-10-26 16:36:26 <sipa> mempool transactions are transactions that are in your mempool, i.e. not in any block so not confirmed, but still known to your node
1127 2013-10-26 16:37:23 <Ascendion> ok --- my code is not going to be aware of mempool transactions -- I only pay attention to transactions listed with a block
1128 2013-10-26 16:38:07 <sipa> ok, so enable txindex and use getrawtransaction
1129 2013-10-26 16:38:25 <Ascendion> already done -- 30k into the reindex now
1130 2013-10-26 16:38:30 datagutt has quit (Ping timeout: 260 seconds)
1131 2013-10-26 16:38:41 deepc0re has quit (Quit: deepc0re)
1132 2013-10-26 16:42:09 <Ascendion> can it be said that once a block is available all its transactions are available too or do I have to handle waiting for transactions to show up ??
1133 2013-10-26 16:42:23 johnsoft has quit (Ping timeout: 252 seconds)
1134 2013-10-26 16:42:48 johnsoft has joined
1135 2013-10-26 16:46:04 askmike has joined
1136 2013-10-26 16:46:07 <sipa> all transactions in a block are available
1137 2013-10-26 16:46:20 <sipa> as they have been verified before the block can be accepted
1138 2013-10-26 16:46:48 <Ascendion> cool -- I dont *HAVE* to run a seperate thread for transactions then -- makes things a bit simpler :)
1139 2013-10-26 16:46:57 bitnumus has joined
1140 2013-10-26 16:47:40 <ThomasV> sipa: I do not understand why you need to link space and time in your proposal. could it work with fixed space, checking every 2^N iterations?
1141 2013-10-26 16:49:59 <bitnumus> ah, i was trying to install bitcoin PPA for ubuntu 13.10, but its in the main repos :D
1142 2013-10-26 16:50:43 <sipa> ThomasV: yes, but the counts would have to be recalculated
1143 2013-10-26 16:50:56 <sipa> ThomasV: and it wouldn't give you the exponential scaling
1144 2013-10-26 16:51:15 <ThomasV> what counts?
1145 2013-10-26 16:51:32 <sipa> eh, the values to check
1146 2013-10-26 16:51:39 <sipa> the lower bounds on the next hash
1147 2013-10-26 16:51:51 <ThomasV> 2^N looks exponential to me..
1148 2013-10-26 16:52:12 <sipa> wait, what do you suggest exactly?
1149 2013-10-26 16:52:18 <sipa> checking what and when?
1150 2013-10-26 16:52:45 <ThomasV> every 2^N iterations, check that the k first bits are zero, with k constant
1151 2013-10-26 16:53:00 <sipa> and N constant too?
1152 2013-10-26 16:53:07 <ThomasV> no
1153 2013-10-26 16:53:17 <sipa> well then you lose the compemsation property
1154 2013-10-26 16:53:23 <ThomasV> just as you did, except you don't increase k
1155 2013-10-26 16:53:43 <ThomasV> what is that?
1156 2013-10-26 16:53:48 <sipa> that the entropy reduction is equal to the strengthening
1157 2013-10-26 16:54:08 <ThomasV> oh right :)
1158 2013-10-26 16:54:10 <sipa> which is mostly theoretical
1159 2013-10-26 16:54:17 <sipa> but it's nice to have
1160 2013-10-26 16:54:24 <ThomasV> so that's the  reason why you linked space and time?
1161 2013-10-26 16:54:34 <sipa> all you really need is defining what difficulty scales you want
1162 2013-10-26 16:54:45 <sipa> the iteration counts and hash bounds flow from that
1163 2013-10-26 16:54:54 <sipa> yes, i guess
1164 2013-10-26 16:55:28 <ThomasV> thinking pragmaticlly, I tend to believe that the input string should have enough entropy in the first place
1165 2013-10-26 16:56:09 <sipa> absolutely
1166 2013-10-26 16:56:17 <sipa> but that is somethimg you can't enforce
1167 2013-10-26 16:56:35 <ThomasV> no, but your proposal does not enforce it either
1168 2013-10-26 16:56:40 <sipa> indeed
1169 2013-10-26 16:57:20 <sipa> if all seeds had sufficient entropy in the first place, there is no need for iteration counts or checksums or whatever
1170 2013-10-26 16:57:47 abueesp has quit (Quit: Saliendo)
1171 2013-10-26 16:57:54 <sipa> but as people like to come up with human-rememberable things, i dount every seed will be actually random
1172 2013-10-26 16:58:26 <ThomasV> well, your proposal is mostly focused on strenghtening
1173 2013-10-26 16:58:37 <sipa> yes
1174 2013-10-26 16:58:47 <ThomasV> bur that's not really the point of BIP39
1175 2013-10-26 16:58:51 <ThomasV> *but*
1176 2013-10-26 16:58:52 <sipa> fair enough
1177 2013-10-26 16:58:53 abueesp has joined
1178 2013-10-26 16:59:06 deepc0re has joined
1179 2013-10-26 16:59:16 <ThomasV> although strenghtening cannot hurt
1180 2013-10-26 16:59:38 <ThomasV> my main point is how do we ensure inter client compatibility
1181 2013-10-26 17:00:30 <sipa> well i'm not sure about tryimg to encode client-specific metadata in the seed
1182 2013-10-26 17:00:53 <sipa> i do understand that some will want to deviate from the suggested wallet structure
1183 2013-10-26 17:01:27 <sipa> but that seems like a per-client default, or something to comfigure independently
1184 2013-10-26 17:01:32 <ThomasV> there is no suggested wallet structure, is there?
1185 2013-10-26 17:01:33 askmike has quit (Remote host closed the connection)
1186 2013-10-26 17:01:40 <sipa> bip32 specifies one
1187 2013-10-26 17:01:48 <sipa> root/account/subchain/key
1188 2013-10-26 17:01:49 caloda has joined
1189 2013-10-26 17:01:58 <HM2> i should probably use github one day
1190 2013-10-26 17:01:59 askmike has joined
1191 2013-10-26 17:02:18 <sipa> trying to make clients compatible on that level seems very hard in any case
1192 2013-10-26 17:02:34 <ThomasV> I agree
1193 2013-10-26 17:02:55 <sipa> so just use the seed to transfer actually secret information (the key)
1194 2013-10-26 17:03:29 <sipa> and if two wallets use differemt structure, then you'lk need to manually configure them
1195 2013-10-26 17:04:01 <sipa> or potentially be (unfortunately) incompatible
1196 2013-10-26 17:04:11 <ThomasV> well, if two versions of electrum use different wallet structures, it would be nice to keep compatibility
1197 2013-10-26 17:04:14 <sipa> which is why there is a suggested structure
1198 2013-10-26 17:04:30 <sipa> yeah i understand the problem
1199 2013-10-26 17:04:38 <sipa> but i'm not sure a good solution exists
1200 2013-10-26 17:04:53 <sipa> it's hard enough to make clients agree on what a wallet even is
1201 2013-10-26 17:05:54 <Ascendion> internally in my system I'm not even considering account/subchain -- a wallet is a collection of keys -- thats it
1202 2013-10-26 17:06:23 <sipa> the account part is mostly there for wallets that have some subdivision support
1203 2013-10-26 17:06:29 askmike has quit (Ping timeout: 260 seconds)
1204 2013-10-26 17:06:36 <sipa> if they don't, just use account 0
1205 2013-10-26 17:06:57 <sipa> subchains enable distinguishing change from public addresses
1206 2013-10-26 17:07:00 askmike has joined
1207 2013-10-26 17:07:22 deafboy has joined
1208 2013-10-26 17:07:42 <ThomasV> sipa: for normal accounts, I need to use a deeper level than what's suggested in bip32
1209 2013-10-26 17:08:04 <sipa> ThomasV: i saw a post about it once; can you elaborate?
1210 2013-10-26 17:08:34 askmike has quit (Remote host closed the connection)
1211 2013-10-26 17:08:35 <ThomasV> that's because I want Electrum to detect the existence of accounts, so that different instances can sels synchronize
1212 2013-10-26 17:08:42 <ThomasV> *self*
1213 2013-10-26 17:08:48 <sipa> ok?
1214 2013-10-26 17:08:56 askmike has joined
1215 2013-10-26 17:09:01 <ThomasV> so I use m/0'/i instaed of m/i'
1216 2013-10-26 17:09:20 <ThomasV> because for a private derivation the user needs to type their password
1217 2013-10-26 17:09:44 <sipa> so m/0'/account/subchain/key ?
1218 2013-10-26 17:09:49 <ThomasV> yes
1219 2013-10-26 17:10:05 <sipa> that means that if any key ever leaks, the entire wallet is compromised
1220 2013-10-26 17:10:28 <sipa> the private derivation is there to avoid that leak
1221 2013-10-26 17:10:34 <ThomasV> yes
1222 2013-10-26 17:11:54 <sipa> seems to me "search for extra accounts" button, which requires your password, isn't so bad
1223 2013-10-26 17:12:04 asuk has joined
1224 2013-10-26 17:12:22 <sipa> and limits the impact of a leaked key to one accoumt
1225 2013-10-26 17:12:35 <sipa> and maintaims compatibility
1226 2013-10-26 17:13:09 askmike has quit (Ping timeout: 248 seconds)
1227 2013-10-26 17:15:06 neep3r has joined
1228 2013-10-26 17:15:23 <ThomasV> sipa: what about p2sh accounts? there's no spec about that
1229 2013-10-26 17:15:52 <sipa> there are some suggestions, but nothing really standardized
1230 2013-10-26 17:16:30 <sipa> like using a number of bip32 chains, and sorting the pubkeys lexicographically before turning into a multisig script
1231 2013-10-26 17:16:41 <sipa> so you can't know which one came from where
1232 2013-10-26 17:16:52 neep3r_ has joined
1233 2013-10-26 17:17:57 <ThomasV> I mean, if you use m/i'/ as the root of accounts, which branches of the tree can be specifically reserved for p2sh?
1234 2013-10-26 17:18:11 patcon has joined
1235 2013-10-26 17:19:24 <sipa> you could use i> 0x01000000 for such
1236 2013-10-26 17:19:31 <ThomasV> :)
1237 2013-10-26 17:19:52 <sipa> or somethimg
1238 2013-10-26 17:19:53 neep3r has quit (Ping timeout: 272 seconds)
1239 2013-10-26 17:20:02 <sipa> i don't like fixing that
1240 2013-10-26 17:20:14 <ThomasV> why?
1241 2013-10-26 17:20:17 roconnor has quit (Quit: Konversation terminated!)
1242 2013-10-26 17:20:28 <sipa> as i don't think we've already found the best practices for implementing multisig-compatible wallets
1243 2013-10-26 17:20:34 <sipa> too little experience
1244 2013-10-26 17:20:36 <ThomasV> exactly
1245 2013-10-26 17:20:43 <ThomasV> that's my point about metadata
1246 2013-10-26 17:21:02 neep3r_ has quit (Ping timeout: 245 seconds)
1247 2013-10-26 17:23:19 shesek has quit (Ping timeout: 272 seconds)
1248 2013-10-26 17:23:25 <sipa> ThomasV: but the amount of such metadata is potentially unlimited
1249 2013-10-26 17:23:42 <sipa> at some point, you may want to encode the M and N of multisig transactions in it
1250 2013-10-26 17:23:47 <sipa> but what about other scripts
1251 2013-10-26 17:23:56 <sipa> like A-or-(B-and-C)
1252 2013-10-26 17:24:08 <sipa> maybe you'll want the entire script template in there
1253 2013-10-26 17:24:14 <ThomasV> sipa: the most compact description is probably to have some version numbers
1254 2013-10-26 17:24:30 <ThomasV> it will be limited by the schedule of releases
1255 2013-10-26 17:24:30 <sipa> right, as long as it is limited to some version numbers
1256 2013-10-26 17:24:54 contrapumpkin is now known as copumpkin
1257 2013-10-26 17:24:57 <sipa> but i don't think it's viable to put all information about how to use a particular set of keys in the seed for creating it
1258 2013-10-26 17:25:15 <sipa> for multisig, you won't just want p2sh-only wallets i guess
1259 2013-10-26 17:25:26 <sipa> you may want to use one account for p2sh keys
1260 2013-10-26 17:25:32 <sipa> eh, M-of-N keys
1261 2013-10-26 17:25:35 patcon_ has joined
1262 2013-10-26 17:25:39 <sipa> and another for something else
1263 2013-10-26 17:25:40 <ThomasV> that's what I do
1264 2013-10-26 17:25:49 <sipa> so, that's not metadata that should go in the seed
1265 2013-10-26 17:26:13 <ThomasV> well, it's implicit, it's a convention
1266 2013-10-26 17:26:20 <sipa> i don't understand
1267 2013-10-26 17:26:38 <sipa> the seed cannot possibly tell you which accounts you'll want to use in which way
1268 2013-10-26 17:26:40 caloda has quit (Quit: leaving)
1269 2013-10-26 17:27:00 <sipa> it may be able to tell you what sort of features the wallet has, but that isn't enough to actually describe usage
1270 2013-10-26 17:27:02 <ThomasV> my wallet expects m/0'/i to be normal accounts, m/1' & m/2' for p2sh
1271 2013-10-26 17:27:26 shesek has joined
1272 2013-10-26 17:27:30 <sipa> well, using public derivation for accounts still seems dangerous to me
1273 2013-10-26 17:27:50 <sipa> and what if new script types are added
1274 2013-10-26 17:27:53 <ThomasV> well, that's something I can change
1275 2013-10-26 17:27:57 <sipa> what if those require longer parameters
1276 2013-10-26 17:28:06 <sipa> like a hash-protected key
1277 2013-10-26 17:28:16 <sipa> you can't keep encoding those in the derivations used
1278 2013-10-26 17:28:35 patcon has quit (Ping timeout: 252 seconds)
1279 2013-10-26 17:28:43 <sipa> maybe you can
1280 2013-10-26 17:28:45 <sipa> i don't know
1281 2013-10-26 17:28:47 <ThomasV> ok, but I can for the moment :)
1282 2013-10-26 17:29:18 <ThomasV> and it's nice for users if everything they need to remember is a seed
1283 2013-10-26 17:29:41 <sipa> i fundamentally disagree with that idea, but i know it's how your client works :)
1284 2013-10-26 17:30:18 <ThomasV> you disagree because it means I use the blockchain to code some information?
1285 2013-10-26 17:30:35 <sipa> not really
1286 2013-10-26 17:30:40 <ThomasV> so, why?
1287 2013-10-26 17:30:50 <sipa> i disagree because it relies on indexing the blockchain to retrieve your wallet
1288 2013-10-26 17:30:59 <ThomasV> oh ok
1289 2013-10-26 17:31:03 <sipa> and it cannot deal with annotations on transactions
1290 2013-10-26 17:31:22 <sipa> i disagree with the idea of mapping bitcoin transactions 1-to-1 to payments, basically
1291 2013-10-26 17:31:22 <ThomasV> huh?
1292 2013-10-26 17:31:31 sustrik has joined
1293 2013-10-26 17:31:40 <sipa> a wallet to me is something that maintains information about payments
1294 2013-10-26 17:31:44 <sipa> linked to bitcoin transactions
1295 2013-10-26 17:31:57 <sipa> but bitcoin transactions are not payments
1296 2013-10-26 17:32:02 <sipa> you don't know who they came from
1297 2013-10-26 17:32:09 <sipa> you cannot put a comment on them
1298 2013-10-26 17:32:14 <ThomasV> I don't get the last point
1299 2013-10-26 17:32:33 <sipa> if all you have is the blockchain, how would you associate a comment with a transaction?
1300 2013-10-26 17:32:38 <sipa> or remember whom it's from?
1301 2013-10-26 17:32:42 <ThomasV> oh sure
1302 2013-10-26 17:32:46 <sipa> the server can do that for you of course
1303 2013-10-26 17:32:59 <sipa> but then you're relying (and at least partially trusting) the server
1304 2013-10-26 17:33:04 <ThomasV> there's an electrum plugin that does that
1305 2013-10-26 17:33:09 <ThomasV> with a webservice
1306 2013-10-26 17:33:26 <ThomasV> http://labelectrum.herokuapp.com/
1307 2013-10-26 17:33:34 deepc0re has quit (Quit: deepc0re)
1308 2013-10-26 17:33:57 <sipa> right, so basically you get a service that maintains wallet metadata for you
1309 2013-10-26 17:34:06 <sipa> why can't that service remember the wallet structure as well?
1310 2013-10-26 17:34:15 <ThomasV> I personally don't use it
1311 2013-10-26 17:34:15 <sipa> it's not less secret
1312 2013-10-26 17:34:48 <ThomasV> hmm, wallet structure is more crucial, because if you don't know it you cannot restore your wallet from seed
1313 2013-10-26 17:35:07 qbasicer has quit (Read error: Connection reset by peer)
1314 2013-10-26 17:35:09 <ThomasV> I would not rely on an external provider for that
1315 2013-10-26 17:35:21 <sipa> ok, i agree with that distinction
1316 2013-10-26 17:35:53 tesserajk has joined
1317 2013-10-26 17:35:55 <sipa> for businesses, the money itself may be just as important as the metadata (=bookkeeping), but for end users that probably isn't the case
1318 2013-10-26 17:36:43 Grouver has quit (Quit:  HydraIRC -> http://www.hydrairc.com <- Would you like to know more?)
1319 2013-10-26 17:36:48 <sipa> anyway, i'm off
1320 2013-10-26 17:36:55 <sipa> it's a hard problem :)
1321 2013-10-26 17:37:03 <ThomasV> right, and businesses may use such a remote service. the one I showed you is not really secure, but another user is working on a similar service
1322 2013-10-26 17:37:10 <ThomasV> ok
1323 2013-10-26 17:37:21 <ThomasV> thanks for your fedback
1324 2013-10-26 17:37:23 qbasicer has joined
1325 2013-10-26 17:37:53 sustrik has quit (Ping timeout: 272 seconds)
1326 2013-10-26 17:38:53 ThomasV has quit (Quit: Quitte)
1327 2013-10-26 17:39:30 ThomasV has joined
1328 2013-10-26 17:39:56 patcon_ has quit (Remote host closed the connection)
1329 2013-10-26 17:40:26 patcon has joined
1330 2013-10-26 17:40:27 * Ascendion wishes he understood a fraction of what you are trying to accomplish with all that mess :)
1331 2013-10-26 17:40:34 ThomasV has quit (Client Quit)
1332 2013-10-26 17:40:43 deepc0re has joined
1333 2013-10-26 17:41:11 Delerium has quit ()
1334 2013-10-26 17:44:29 _ingsoc has quit (Ping timeout: 272 seconds)
1335 2013-10-26 17:45:09 patcon has quit (Ping timeout: 248 seconds)
1336 2013-10-26 17:45:49 _ingsoc has joined
1337 2013-10-26 17:47:41 roconnor has joined
1338 2013-10-26 17:49:25 super3 has joined
1339 2013-10-26 17:51:25 jakov has joined
1340 2013-10-26 17:51:38 eoss has joined
1341 2013-10-26 17:53:09 paracyst has quit (Read error: Connection reset by peer)
1342 2013-10-26 17:54:45 bitRipperX has quit (Ping timeout: 248 seconds)
1343 2013-10-26 17:56:53 dgolds has quit (Ping timeout: 248 seconds)
1344 2013-10-26 17:56:57 mbelshe has joined
1345 2013-10-26 17:58:07 dgolds has joined
1346 2013-10-26 17:59:31 cysm has quit (Ping timeout: 245 seconds)
1347 2013-10-26 17:59:42 valparaiso is now known as valparaiso_afk
1348 2013-10-26 18:00:36 cysm has joined
1349 2013-10-26 18:02:21 sustrik has joined
1350 2013-10-26 18:04:00 Zarutian has joined
1351 2013-10-26 18:04:30 hydrogenesis has joined
1352 2013-10-26 18:07:01 cysm has quit (Ping timeout: 245 seconds)
1353 2013-10-26 18:07:39 sustrik has quit (Ping timeout: 272 seconds)
1354 2013-10-26 18:07:40 cysm has joined
1355 2013-10-26 18:07:43 murkmans has quit (Read error: Connection reset by peer)
1356 2013-10-26 18:14:06 cysm has quit (Ping timeout: 245 seconds)
1357 2013-10-26 18:14:34 cysm has joined
1358 2013-10-26 18:15:18 hydrogenesis has quit (Quit: Colloquy for iPad - http://colloquy.mobi)
1359 2013-10-26 18:17:51 neep3r has joined
1360 2013-10-26 18:20:13 skinnkavaj has quit (Ping timeout: 254 seconds)
1361 2013-10-26 18:20:38 <dobry-den> I'm having trouble reading the source, but I would like to superficially replicate some of the satoshi client's tolerance/behavior when it comes to bad data from another client. Could anyone provide a high-level summary?
1362 2013-10-26 18:21:15 patcon has joined
1363 2013-10-26 18:21:29 sustrik has joined
1364 2013-10-26 18:21:51 Anduck has quit (Ping timeout: 272 seconds)
1365 2013-10-26 18:22:19 neep3r has quit (Ping timeout: 245 seconds)
1366 2013-10-26 18:22:20 <sipa> dobry-den: when simething bad is detected, we increase a peer's badness score
1367 2013-10-26 18:22:35 <sipa> if that exceeds a treshold, we ban the IP for a day
1368 2013-10-26 18:22:42 <sipa> and disconnect
1369 2013-10-26 18:23:41 <sipa> actions which are propagated shouldn't cause badness triggers though
1370 2013-10-26 18:24:11 <sipa> for example, if a new rule is introduced, we typically don't give it a ban score immediately
1371 2013-10-26 18:24:27 <sipa> as that could lead to banning all old clients simultaneously
1372 2013-10-26 18:25:35 <dobry-den> ah. how about at the byte level? what's the satoshi client's tolerance if you connect to it and just send 0x42 a bunch of times instead of a real message sequence starting at the magic seq.
1373 2013-10-26 18:26:17 sustrik has quit (Ping timeout: 260 seconds)
1374 2013-10-26 18:26:18 <jgarzik> dobry-den, increases ban score
1375 2013-10-26 18:26:42 <dobry-den> for every byte?
1376 2013-10-26 18:26:51 <jgarzik> no
1377 2013-10-26 18:27:05 <jgarzik> when it sees it is not possible that input data is a valid message
1378 2013-10-26 18:27:22 melvster has quit (Read error: Connection reset by peer)
1379 2013-10-26 18:27:24 <jgarzik> input data is read in chunks
1380 2013-10-26 18:27:26 cysm has quit (Ping timeout: 245 seconds)
1381 2013-10-26 18:27:29 <sipa> iirc it just disconnects if you send an invalod message
1382 2013-10-26 18:27:42 <dobry-den> so, let's say you send it fifty 0x42 bytes and then begin a valid message
1383 2013-10-26 18:27:53 <jgarzik> still invalid
1384 2013-10-26 18:27:56 btcbtc has joined
1385 2013-10-26 18:27:56 <sipa> you'll be disconnected after 24
1386 2013-10-26 18:28:07 <dobry-den> of course, the length of the header
1387 2013-10-26 18:28:09 <dobry-den> thanks
1388 2013-10-26 18:28:15 <sipa> (i think)
1389 2013-10-26 18:28:29 jcorgan has joined
1390 2013-10-26 18:28:35 <dobry-den> how about one 0x42 byte and then a valid message?
1391 2013-10-26 18:28:46 * jgarzik changed it a while ago to be more aggressive about booting garbage-sending clients
1392 2013-10-26 18:28:49 cysm has joined
1393 2013-10-26 18:28:56 <jgarzik> dobry-den, still invalid
1394 2013-10-26 18:29:18 melvster has joined
1395 2013-10-26 18:29:29 <jgarzik> anything less than valid is.. uncivilized
1396 2013-10-26 18:29:47 <dobry-den> so what does it do? does badness++ and then ignore all bytes until the next valid magic sequence in the byte stream?
1397 2013-10-26 18:30:12 grau_ has joined
1398 2013-10-26 18:32:05 skinnkavaj has joined
1399 2013-10-26 18:32:09 <sipa> dobry-den: iirc, if an unparseable messages is received, you just get disconnected
1400 2013-10-26 18:32:16 <dobry-den> sipa: thanks
1401 2013-10-26 18:32:16 <sipa> garbage bytes are not supported
1402 2013-10-26 18:32:37 grau has quit (Ping timeout: 248 seconds)
1403 2013-10-26 18:33:19 jcorgan has quit (Write error: Broken pipe)
1404 2013-10-26 18:33:34 sustrik has joined
1405 2013-10-26 18:33:51 <jgarzik> sipa, just reviewed
1406 2013-10-26 18:34:04 <jgarzik> sipa, any falling outside of message boundaries results in immediate disconnect
1407 2013-10-26 18:34:09 <sipa> jgarzik: i know
1408 2013-10-26 18:34:14 <sipa> you recently changed that
1409 2013-10-26 18:34:23 <sipa> well, recently
1410 2013-10-26 18:34:33 <jgarzik> sipa, garbage with proper msg boundaries is parsed then ignored
1411 2013-10-26 18:34:40 <sipa> together with the message-receive queue refactor
1412 2013-10-26 18:34:47 <jgarzik> i.e. invalid checksum + proper boundaries
1413 2013-10-26 18:34:56 cysm has quit (Ping timeout: 245 seconds)
1414 2013-10-26 18:34:58 <sipa> ah
1415 2013-10-26 18:35:20 <sipa> jgarzik: btw, perfectly fine with not using my script if you want to do things manually :)
1416 2013-10-26 18:35:28 <sipa> jgarzik: i'd just like to see signed commits
1417 2013-10-26 18:35:33 cysm has joined
1418 2013-10-26 18:35:33 cysm has quit (Changing host)
1419 2013-10-26 18:35:33 cysm has joined
1420 2013-10-26 18:35:40 <jgarzik> sipa, +1
1421 2013-10-26 18:35:50 Anduck has joined
1422 2013-10-26 18:36:07 <dobry-den> jgarzik: can you explain what youre referring to as 'msg boundaries'?
1423 2013-10-26 18:36:19 <jgarzik> dobry-den, bitcoin is a binary protocol
1424 2013-10-26 18:36:24 <jgarzik> packetized
1425 2013-10-26 18:36:32 <jgarzik> message-based
1426 2013-10-26 18:37:30 <dobry-den> on testnet3, i sometimes encounter a byte-stream that looks like [version-msg 0x42 0x42 verack]
1427 2013-10-26 18:38:23 <dobry-den> so, if i wanted to replicate satoshiclient behavior, i would parse version-message, then invalidate the rest of the stream at that first 0x42 byte and disconnect
1428 2013-10-26 18:38:41 <dobry-den> is that true?
1429 2013-10-26 18:40:20 <dobry-den> last night i basically went down a rabbit hole of trying to be too robust (i've come to realize) and i'm trying to draw some lines
1430 2013-10-26 18:40:35 MobPhone has quit (Ping timeout: 272 seconds)
1431 2013-10-26 18:40:40 jcorgan has joined
1432 2013-10-26 18:40:54 <dobry-den> for now i'll aggressively disconnect
1433 2013-10-26 18:42:01 cysm has quit (Ping timeout: 245 seconds)
1434 2013-10-26 18:42:37 cysm has joined
1435 2013-10-26 18:44:24 deepc0re has quit (Quit: deepc0re)
1436 2013-10-26 18:49:11 sustrik has quit (Ping timeout: 272 seconds)
1437 2013-10-26 18:49:21 rdymac has quit (Read error: Connection reset by peer)
1438 2013-10-26 18:50:02 a_meteor has joined
1439 2013-10-26 18:50:14 Eiii has joined
1440 2013-10-26 18:50:16 valparaiso_afk is now known as valparaiso
1441 2013-10-26 18:52:59 jcorgan has quit (Quit: jcorgan)
1442 2013-10-26 18:53:18 jcorgan has joined
1443 2013-10-26 18:54:54 a_meteor has quit (Ping timeout: 268 seconds)
1444 2013-10-26 18:54:58 Application has quit (Remote host closed the connection)
1445 2013-10-26 18:55:09 shesek has quit (Ping timeout: 272 seconds)
1446 2013-10-26 18:55:46 cysm has quit (Ping timeout: 245 seconds)
1447 2013-10-26 18:56:48 cysm has joined
1448 2013-10-26 18:57:00 rdymac has joined
1449 2013-10-26 18:57:23 shesek has joined
1450 2013-10-26 18:59:32 shesek has quit (Max SendQ exceeded)
1451 2013-10-26 19:00:00 shesek has joined
1452 2013-10-26 19:04:12 gingpark has quit (Ping timeout: 265 seconds)
1453 2013-10-26 19:05:19 [Author] has quit (Ping timeout: 256 seconds)
1454 2013-10-26 19:06:46 [Author] has joined
1455 2013-10-26 19:08:16 cysm has quit (Ping timeout: 245 seconds)
1456 2013-10-26 19:12:49 cysm has joined
1457 2013-10-26 19:17:04 cysm has quit (Max SendQ exceeded)
1458 2013-10-26 19:17:45 cysm has joined
1459 2013-10-26 19:18:00 sustrik has joined
1460 2013-10-26 19:18:24 neep3r has joined
1461 2013-10-26 19:19:33 cysm has quit (Client Quit)
1462 2013-10-26 19:21:32 patcon has quit (Remote host closed the connection)
1463 2013-10-26 19:22:08 patcon has joined
1464 2013-10-26 19:22:27 jcorgan has quit (Quit: jcorgan)
1465 2013-10-26 19:22:43 neep3r has quit (Ping timeout: 245 seconds)
1466 2013-10-26 19:23:09 daybyter has joined
1467 2013-10-26 19:23:43 valparaiso is now known as valparaiso_afk
1468 2013-10-26 19:24:03 debiantoruser has quit (Ping timeout: 240 seconds)
1469 2013-10-26 19:25:16 BtcBandit has joined
1470 2013-10-26 19:25:30 valparaiso_afk is now known as valparaiso
1471 2013-10-26 19:26:04 BtcBandit has left ()
1472 2013-10-26 19:26:17 debiantoruser has joined
1473 2013-10-26 19:26:46 asuk is now known as asuk|afk
1474 2013-10-26 19:26:58 patcon has quit (Ping timeout: 268 seconds)
1475 2013-10-26 19:34:01 paracyst has joined
1476 2013-10-26 19:34:49 debiantoruser has quit (Ping timeout: 245 seconds)
1477 2013-10-26 19:35:29 cysm has joined
1478 2013-10-26 19:36:52 deepc0re has joined
1479 2013-10-26 19:36:55 debiantoruser has joined
1480 2013-10-26 19:37:48 deepc0re has quit (Client Quit)
1481 2013-10-26 19:40:46 dlidstrom has joined
1482 2013-10-26 19:42:55 <sipa> ;;blocks
1483 2013-10-26 19:42:56 <gribble> 266236
1484 2013-10-26 19:52:52 <melvster> are the fields in the block header UNSIGNED ints ( version / time / target / nonce )
1485 2013-10-26 19:53:34 <sipa> yes
1486 2013-10-26 19:53:38 <melvster> thanks!
1487 2013-10-26 19:53:50 <jakov> none of those really make sense as a negative number
1488 2013-10-26 19:53:55 <sipa> (though for compatibility reasons, version >= 2^31 can't be used)
1489 2013-10-26 19:53:57 <jakov> nonce doesnt even need to be interpreted as a number
1490 2013-10-26 19:54:11 <sipa> and target is not part of the block
1491 2013-10-26 19:54:16 <sipa> it's decoded from nBits
1492 2013-10-26 19:54:22 <melvster> oic thanks
1493 2013-10-26 19:54:27 <sipa> which is hardly a number itself
1494 2013-10-26 19:54:38 <melvster> im using hexBinary for that
1495 2013-10-26 19:54:40 <sipa> but version and time are unsigned
1496 2013-10-26 19:54:57 <helo> minecraft
1497 2013-10-26 19:55:06 <sipa> ...?
1498 2013-10-26 19:55:15 <melvster> nonce is unsigned too, right?
1499 2013-10-26 19:55:17 _ingsoc has quit (Ping timeout: 248 seconds)
1500 2013-10-26 19:55:33 <sipa> as jakov said, it's not really a number
1501 2013-10-26 19:55:49 <sipa> it's used as a number by the miner, but nobody else interprets it
1502 2013-10-26 19:56:05 <melvster> got it
1503 2013-10-26 19:56:13 <sipa> you can model it as byte[4]
1504 2013-10-26 19:56:23 <melvster> unfortunately my documentation needs to have all the tiny details
1505 2013-10-26 19:56:43 <sipa> it is "4 bytes which are freely chosen by the miner"
1506 2013-10-26 19:56:55 <melvster> great, thanks!
1507 2013-10-26 19:57:04 <sipa> while version and time do have an actual interpretation as an integer
1508 2013-10-26 19:57:05 _ingsoc has joined
1509 2013-10-26 19:57:22 <melvster> so i was previously calling it target, "nbits" is a better name, right?
1510 2013-10-26 19:57:49 btcbtc_ has joined
1511 2013-10-26 19:58:01 <sipa> that's how it's called in the reference client source code
1512 2013-10-26 19:58:08 btcbtc has quit (Ping timeout: 245 seconds)
1513 2013-10-26 19:58:12 <sipa> it's not a very informative name
1514 2013-10-26 19:58:15 <melvster> perfect, ill use that then
1515 2013-10-26 19:58:30 <sipa> it is the target, but in a compact and custom 4-byte encoding
1516 2013-10-26 19:58:32 * melvster translating bitcoin block headers to the semantic web
1517 2013-10-26 19:58:44 <sipa> (the actual target is a 256-bit number)
1518 2013-10-26 19:58:51 tesserajk has quit ()
1519 2013-10-26 19:58:56 <melvster> ah right
1520 2013-10-26 19:59:25 <Luke-Jr> [19:53:29] <sipa> it's used as a number by the miner <-- it is? :P
1521 2013-10-26 19:59:40 <sipa> Luke-Jr: miners increment it
1522 2013-10-26 19:59:51 <sipa> incrementing is an operation you do an integer :)
1523 2013-10-26 20:00:22 <Luke-Jr> GPU miners don't increment it
1524 2013-10-26 20:00:25 <sipa> no?
1525 2013-10-26 20:00:25 <Luke-Jr> ASICs might..
1526 2013-10-26 20:00:36 <Luke-Jr> sipa: no, GPUs run the hash algo on each nonce, in no given order
1527 2013-10-26 20:00:41 <sipa> right
1528 2013-10-26 20:00:51 <sipa> good point
1529 2013-10-26 20:01:06 sustrik has quit (Ping timeout: 240 seconds)
1530 2013-10-26 20:01:34 CircusPeanut has quit (Ping timeout: 246 seconds)
1531 2013-10-26 20:05:35 sustrik has joined
1532 2013-10-26 20:06:05 Application has joined
1533 2013-10-26 20:10:27 asuk has quit (afk!~asuk@31.129.27.89|)
1534 2013-10-26 20:12:09 tyn has joined
1535 2013-10-26 20:17:48 CircusPeanut has joined
1536 2013-10-26 20:18:43 digitalmagus has quit (Remote host closed the connection)
1537 2013-10-26 20:19:02 digitalmagus has joined
1538 2013-10-26 20:19:15 digitalmagus has quit (Changing host)
1539 2013-10-26 20:19:15 digitalmagus has joined
1540 2013-10-26 20:19:20 neep3r has joined
1541 2013-10-26 20:24:03 neep3r has quit (Ping timeout: 252 seconds)
1542 2013-10-26 20:24:43 digitalmagus has quit ()
1543 2013-10-26 20:27:18 crass has quit (Remote host closed the connection)
1544 2013-10-26 20:29:44 tyn has quit (Quit: Leaving)
1545 2013-10-26 20:31:50 grau_ has quit (Remote host closed the connection)
1546 2013-10-26 20:32:22 grau has joined
1547 2013-10-26 20:36:06 skinnkavaj has quit (Read error: Connection reset by peer)
1548 2013-10-26 20:37:00 skinnkavaj has joined
1549 2013-10-26 20:37:00 skinnkavaj has quit (Changing host)
1550 2013-10-26 20:37:00 skinnkavaj has joined
1551 2013-10-26 20:37:07 grau has quit (Ping timeout: 272 seconds)
1552 2013-10-26 20:40:46 sustrik has quit (Ping timeout: 240 seconds)
1553 2013-10-26 20:42:03 <Ascendion> *sigh* playing catch up again -- just a few thou blocks to go and I can start testing code again :)
1554 2013-10-26 20:43:11 deepc0re has joined
1555 2013-10-26 20:44:44 <Luke-Jr> Ascendion: are you still planning to do the GamaStack stuff?
1556 2013-10-26 20:45:06 idstam-se has quit (Ping timeout: 240 seconds)
1557 2013-10-26 20:45:51 <Ascendion> yes -- but I'm not real sure of time frame right now -- I need to call the assembly shop and arrange and time to get up there with my last 2 blank boards and the 4 chips I have left and have them do all the soldering
1558 2013-10-26 20:46:20 <Ascendion> remind me to ping wiz about the chips I already paid for :)
1559 2013-10-26 20:47:04 CircusPeanut has quit (Ping timeout: 246 seconds)
1560 2013-10-26 20:47:21 <Ascendion> the funds cleared into my bank yesterday so I got the cash to get up there once I arrange time with them...
1561 2013-10-26 20:48:36 <Ascendion> someone in eligius asking ??
1562 2013-10-26 20:48:38 <Luke-Jr> well, don't feel like you have to prioritise it or anything
1563 2013-10-26 20:48:58 <Luke-Jr> no, I was just releasing BFG 3.4 and was going to remove it from the planned-to-support list if you had dropped the project
1564 2013-10-26 20:49:07 <Luke-Jr> (since BFL chips are unprofitable, I figured I should ask)
1565 2013-10-26 20:49:45 <Ascendion> ya with K&C kicking out those new systems all the older asics are DOA
1566 2013-10-26 20:49:58 <sipa> ;;genrate 4500
1567 2013-10-26 20:49:59 <gribble> The expected generation output, at 4500.0 Mhps, given difficulty of 390928787.638, is 0.00578898099987 BTC per day and 0.000241207541661 BTC per hour.
1568 2013-10-26 20:50:12 <Ascendion> but I might as well get the chips I've already paid for hashing
1569 2013-10-26 20:50:43 <Ascendion> ya -- thats almost as bad as my GPUs were when I first joined eligius :)
1570 2013-10-26 20:50:50 <sipa> haha
1571 2013-10-26 20:51:53 debiantoruser has quit (Ping timeout: 245 seconds)
1572 2013-10-26 20:51:54 <sipa> ;;genrate 70 12000
1573 2013-10-26 20:51:55 <gribble> The expected generation output, at 70.0 Mhps, given difficulty of 12000.0, is 2.93362134585 BTC per day and 0.122234222744 BTC per hour.
1574 2013-10-26 20:52:08 <sipa> ^ when I first GPU mined
1575 2013-10-26 20:52:40 <Ascendion> ya -- I know -- I'm a late joiner :)
1576 2013-10-26 20:53:08 <sipa> ;;calc 2.93*0.27
1577 2013-10-26 20:53:09 <gribble> 0.7911
1578 2013-10-26 20:53:34 <sipa> ;;calc 0.012*190
1579 2013-10-26 20:53:34 <gribble> 2.28
1580 2013-10-26 20:53:46 daybyter has quit (Ping timeout: 240 seconds)
1581 2013-10-26 20:53:48 debiantoruser has joined
1582 2013-10-26 20:56:00 <Ascendion> I'm half way tempted to just do a run of those coinado boards the other guy did (cant recall his nick off hand) since its already got working firmware
1583 2013-10-26 20:57:07 ThomasV has joined
1584 2013-10-26 20:57:44 [\\\] has quit (Ping timeout: 240 seconds)
1585 2013-10-26 21:01:35 datagutt has joined
1586 2013-10-26 21:01:53 Krellan__ has joined
1587 2013-10-26 21:03:58 Krellan_ has quit (Ping timeout: 245 seconds)
1588 2013-10-26 21:04:09 Krellan has quit (Ping timeout: 245 seconds)
1589 2013-10-26 21:04:45 Krellan has joined
1590 2013-10-26 21:07:04 datagutt has quit (Ping timeout: 245 seconds)
1591 2013-10-26 21:08:08 datagutt has joined
1592 2013-10-26 21:08:59 <Ascendion> ;;blocks
1593 2013-10-26 21:08:59 <gribble> 266243
1594 2013-10-26 21:13:36 <Ascendion> grrrrr -- so I cant use gettransaction on just any transaction id ?? I've gotta use getrawtransaction to pull transactions not associated with any address in the wallet ??
1595 2013-10-26 21:15:09 <sipa> yes
1596 2013-10-26 21:15:13 elevatioN has joined
1597 2013-10-26 21:15:14 <sipa> gettransaction is a wallet RPC
1598 2013-10-26 21:15:25 <sipa> getrawtransaction is a blockchain RPC
1599 2013-10-26 21:19:27 <jouke> Ascendion: what is wrong with getrawtransaction?
1600 2013-10-26 21:19:47 <Ascendion> ok -- figured out why I was getting weirdness from the library -- getraw just returns the hex string.. I gotta call another method to decode it
1601 2013-10-26 21:19:55 ambimorph1 has quit (Quit: Leaving.)
1602 2013-10-26 21:20:04 <jouke> Use "getrawtransaction <txit> 1
1603 2013-10-26 21:20:05 <jouke> Q
1604 2013-10-26 21:20:07 <jouke> doh
1605 2013-10-26 21:20:09 <jouke> Use "getrawtransaction <txit> 1"
1606 2013-10-26 21:20:17 <jouke> txid
1607 2013-10-26 21:20:29 <Ascendion> whats the 1 on the end do to it ??
1608 2013-10-26 21:20:36 <jouke> magic
1609 2013-10-26 21:20:46 <sipa> it's an extra parameter to tell it to decode the transaction
1610 2013-10-26 21:20:55 <sipa> you can also use decoderawtransaction separately
1611 2013-10-26 21:23:49 <Ascendion> ok -- I'm looking at the decoded transaction for the genesis block just to get a start understanding this mess :) looking at the trans output there is a chunk of asm that looks like an address public key and an op_checksig
1612 2013-10-26 21:24:18 neep3r has joined
1613 2013-10-26 21:25:12 <Ascendion> should I be expecting that format for the script in pretty much all trans outputs ??
1614 2013-10-26 21:25:27 <sipa> it is decoded several times
1615 2013-10-26 21:25:31 <sipa> with the same information
1616 2013-10-26 21:25:52 <sipa> 'hex' is just the raw string of bytes in the script
1617 2013-10-26 21:26:15 <sipa> asm is the same, but decoded into human-readable operators
1618 2013-10-26 21:26:33 CircusPeanut has joined
1619 2013-10-26 21:26:39 <sipa> and 'addresses' tries to show you which addresses are affected, but this obviously only works for 'normal' scripts
1620 2013-10-26 21:26:46 datagutt has quit (Ping timeout: 240 seconds)
1621 2013-10-26 21:26:54 <sipa> (not all scripts have corresponding addresses)
1622 2013-10-26 21:28:51 <Ascendion> I'm gonna have to rerun the request and pull the json -- you are refering to properties that I dont see -- probably the library missed one
1623 2013-10-26 21:29:30 <sipa> yes, please just use bitcoind directly so you know what's there
1624 2013-10-26 21:29:40 <sipa> if something is missing, it seems trivial to add to your library
1625 2013-10-26 21:29:54 zeddan81 has quit (Quit: Lämnar)
1626 2013-10-26 21:31:14 elevatioN has quit (Ping timeout: 245 seconds)
1627 2013-10-26 21:34:07 <Ascendion> ya -- just gotta fix the poco and the json lib takes it from there
1628 2013-10-26 21:35:00 debiantoruser has quit (Ping timeout: 252 seconds)
1629 2013-10-26 21:35:00 daybyter has joined
1630 2013-10-26 21:36:05 AusBitBank has joined
1631 2013-10-26 21:36:24 sam___ has joined
1632 2013-10-26 21:36:55 sam___ has left ()
1633 2013-10-26 21:37:34 sam___ has joined
1634 2013-10-26 21:37:59 sam___ has left ()
1635 2013-10-26 21:38:22 daybyter has quit (Client Quit)
1636 2013-10-26 21:38:43 daybyter has joined
1637 2013-10-26 21:38:45 Krellan_ has joined
1638 2013-10-26 21:39:44 <Ascendion> need to find me a decend json pretty printer in C# :)
1639 2013-10-26 21:40:43 Krellan__ has quit (Ping timeout: 272 seconds)
1640 2013-10-26 21:40:46 zenflower has joined
1641 2013-10-26 21:45:00 <Luke-Jr> C# = worthless
1642 2013-10-26 21:45:19 <daybyter> :-)
1643 2013-10-26 21:45:30 <daybyter> java rules...
1644 2013-10-26 21:45:48 <Ascendion> {FLAME ON} only if you dont think productivity is important {FLAME OFF}
1645 2013-10-26 21:45:51 <copumpkin> no, Haskell is the clear winner, and I'm sure Luke-Jr would agree
1646 2013-10-26 21:45:55 * copumpkin runs
1647 2013-10-26 21:46:15 deepc0re has quit (Quit: deepc0re)
1648 2013-10-26 21:46:57 * sipa agrees
1649 2013-10-26 21:47:10 <sipa> now, let's keep it flameless
1650 2013-10-26 21:47:21 <Ascendion> sipa -- agree with WHICH statement ?? :)
1651 2013-10-26 21:47:29 <copumpkin> in a pure environment, fire is impossible
1652 2013-10-26 21:47:49 <sipa> Ascendion: i'll leave the interpretation to you, as i said to keep it flameless
1653 2013-10-26 21:48:05 * daybyter wears flameproof underwear...
1654 2013-10-26 21:48:31 * Ascendion cooks daybyter's face
1655 2013-10-26 21:48:44 <sipa> (hint: I discovered Bitcoin because of a discussion on it in #haskell-blah on this server)
1656 2013-10-26 21:49:46 <Ascendion> LOL -- I've been banging out C# for nearly the last 10 years -- I will never go back to c/c++/vb/java -- the world has moved on -- get with the times :)
1657 2013-10-26 21:50:30 <sipa> since I learnt C++, I didn't want to go back to Java either :p
1658 2013-10-26 21:50:42 <Ascendion> though acutally my current project has me interfacing with a lot of existing java code -- THE PAIN !!!!
1659 2013-10-26 21:51:54 <Ascendion> I learned C++ back when it was a preprocessor for the C compiler :) and I was bloodly happy to get away from it when VB and Java came out
1660 2013-10-26 21:52:11 <daybyter> I learned c++ before java...and still think it's a step forward
1661 2013-10-26 21:52:39 <sipa> when I write something, and performance doesn't matter, I'll use python or haskell
1662 2013-10-26 21:52:40 porquilho has quit ()
1663 2013-10-26 21:53:00 <sipa> when I write something, and performance does matter, C or C++
1664 2013-10-26 21:53:21 <Ascendion> my real problem with c/c++ is you are either incredibly careful writing the code, or you spend a lotta time dealing with obscure bugs -- between that and piss poor compiler error messages, I prayed for something better for many years :)
1665 2013-10-26 21:53:34 <sipa> agree about the compiler errors
1666 2013-10-26 21:53:49 <sipa> with C++ it's actually a lot easier to write safe code than with C
1667 2013-10-26 21:53:52 <sipa> RAII does so much
1668 2013-10-26 21:54:12 <daybyter> I don't like all those platform dependencies...
1669 2013-10-26 21:54:24 <daybyter> ./configure ; qmake ; make
1670 2013-10-26 21:54:26 <daybyter> etc...
1671 2013-10-26 21:54:45 <Ascendion> that may be the case now -- but c++ was a bloody mess when I switched -- template hell to say the least :)
1672 2013-10-26 21:55:18 <sipa> i only learnt C++ to understand the bitcoin source code
1673 2013-10-26 21:55:20 a_meteor has joined
1674 2013-10-26 21:55:21 <sipa> i only knew C before that
1675 2013-10-26 21:55:36 <warren> me too!
1676 2013-10-26 21:55:37 <warren> =)
1677 2013-10-26 21:55:43 zenflower has quit (Quit: Page closed)
1678 2013-10-26 21:55:58 <Ascendion> I've programmed in every major language for the last 35+ years -- I'm happy where I'm at :)
1679 2013-10-26 21:56:15 <sipa> i started programming at 11
1680 2013-10-26 21:56:25 <sipa> but that's still only 18 years ago :p
1681 2013-10-26 21:56:38 <Ascendion> 13 -- when the TRS80 mod 1 first showed up at my local radio shack :)
1682 2013-10-26 21:56:48 <daybyter> me at 13...33 years ago...
1683 2013-10-26 21:57:14 <daybyter> I still do some C on the c64, though (vice emulator actually)
1684 2013-10-26 21:57:24 * sipa feels young
1685 2013-10-26 21:57:24 <Ascendion> hey -- another old fart like me -- lets stink up the place :)
1686 2013-10-26 21:57:35 <daybyter> use 50+ languages in projects..
1687 2013-10-26 21:57:37 Anduckkk has joined
1688 2013-10-26 21:57:53 <Anduckkk> the tx i talked about earlier.. still not relayed
1689 2013-10-26 21:58:12 Anduck has quit (Read error: Operation timed out)
1690 2013-10-26 21:58:14 <Anduckkk> it doesnt get sent to other nodes at all
1691 2013-10-26 21:58:15 <sipa> did you find out why your second system didn't accept it?
1692 2013-10-26 21:58:23 <Anduckkk> nope
1693 2013-10-26 21:58:28 <Anduckkk> but it accepted it finally
1694 2013-10-26 21:58:34 <Anduckkk> i didnt change -anything-
1695 2013-10-26 21:58:39 ThomasV has quit (Ping timeout: 252 seconds)
1696 2013-10-26 21:58:52 <Anduckkk> might still been user error (i think so) but no idea what
1697 2013-10-26 21:59:07 <Anduckkk> but the tx still doesnt relay to other nodes which is weird..
1698 2013-10-26 21:59:31 <daybyter> Ascendion: so you do trade stuff in C#, I guess?
1699 2013-10-26 21:59:39 <Ascendion> I had an advantage -- my uncle worked on the team that created the BASIC language at Dartmouth, and my dad did financial programming for the Federal Reserve -- programming is in my blood :)
1700 2013-10-26 22:00:17 <Ascendion> daybyter -- I do anything the customer/employer wants :)
1701 2013-10-26 22:00:29 <Ascendion> right now its heavy on cryptocoin stuff :)
1702 2013-10-26 22:00:36 <Anduckkk> sipa do you know why the tx doesnt relay?
1703 2013-10-26 22:00:42 <Anduckkk> is this regular behaviour?
1704 2013-10-26 22:00:51 <sipa> Anduckkk: no, but i guess the low fee has something to do with it
1705 2013-10-26 22:00:57 KillYourTV has joined
1706 2013-10-26 22:01:02 elevatioN has joined
1707 2013-10-26 22:01:23 <Anduckkk> yup..
1708 2013-10-26 22:01:32 <Anduckkk> but 0fees get relayed too
1709 2013-10-26 22:01:40 <Anduckkk> brb
1710 2013-10-26 22:01:57 <daybyter> Ascendion: me writes java trade stuff...
1711 2013-10-26 22:02:00 <Luke-Jr> Anduckkk: only with high priority
1712 2013-10-26 22:02:33 <Ascendion> daybyter -- when you say "trade stuff" you mean ???
1713 2013-10-26 22:02:46 <gmaxwell> Dunno if people saw, but apparently the FBI (partially?) compromised the SR guy's encrypted wallet. They claim to have gone and pulled keys off freespace from before he encrypted it. Seems kinda odd: many of the keys they moved were first used rather recently. So unless he just recently encrypted it, it's kinda puzzling.
1714 2013-10-26 22:03:03 <daybyter> http://i.imgur.com/MqHfF.jpg
1715 2013-10-26 22:03:19 pooler has quit (Quit: ChatZilla 0.9.90.1 [Firefox 24.0/20130911163500])
1716 2013-10-26 22:03:22 <daybyter> http://i.imgur.com/OoP0aNo.jpg
1717 2013-10-26 22:03:25 <sipa> Anduckkk: wait a second, it has only 0.00001 BTC fee?
1718 2013-10-26 22:03:36 <Luke-Jr> gmaxwell: I'd lol if he setup a killswitch to encrypt..
1719 2013-10-26 22:03:40 <sipa> Anduckkk: i missed a zero
1720 2013-10-26 22:03:46 <Luke-Jr> gmaxwell: or maybe even destroy
1721 2013-10-26 22:04:00 pooler has joined
1722 2013-10-26 22:04:35 <sipa> Anduckkk: 0.8.5 created this?
1723 2013-10-26 22:05:32 <Ascendion> daybyter -- too bad you don't do c# -- when I get the framework for my system done I'm gonna be looking for programmers to work on parts of the project on contract -- maybe I can use you for some of the deep java interfacing thats going to have to be done :)
1724 2013-10-26 22:06:09 <warren> gmaxwell: we're testing externalip.  would an added print statement showing actual advertisements be welcome?  currently there's no visible indication that it's doing anything.
1725 2013-10-26 22:06:12 <daybyter> I'm looking for coders, too...
1726 2013-10-26 22:06:33 <gmaxwell> warren: certantly I did that for testing.
1727 2013-10-26 22:06:43 <daybyter> I have a bot framework running, and have to work on a trading language.
1728 2013-10-26 22:06:48 <warren> would be too noisy if it were left in production?
1729 2013-10-26 22:07:12 <daybyter> already have some rule-based bots running...
1730 2013-10-26 22:07:18 <Ascendion> my setup is 100% web based though -- using c# for the web front and and all the controlling logic --
1731 2013-10-26 22:10:13 <daybyter> pure java here...
1732 2013-10-26 22:10:27 <daybyter> working on an android app to control the bot.
1733 2013-10-26 22:10:41 <daybyter> but I can just log in so far...not enough time...
1734 2013-10-26 22:14:36 <daybyter> maybe we can collaborate on some strategies, or so...
1735 2013-10-26 22:14:53 <daybyter> syntax is close enough, so porting wouldn't be an issue.
1736 2013-10-26 22:15:01 <Ascendion> sorry to say probably not -- what I'm doing is proprietary and closed source
1737 2013-10-26 22:15:32 <daybyter> I released my API implementations. Not the bot stuff yet.
1738 2013-10-26 22:15:49 <daybyter> Once I get drools going, I can release more code.
1739 2013-10-26 22:16:03 <daybyter> so I can keep the strategies private.
1740 2013-10-26 22:20:26 elevatioN has quit (Ping timeout: 240 seconds)
1741 2013-10-26 22:22:34 Coincidental has joined
1742 2013-10-26 22:23:24 zeddan81 has joined
1743 2013-10-26 22:25:56 elevatioN has joined
1744 2013-10-26 22:27:48 wbill has quit (Ping timeout: 252 seconds)
1745 2013-10-26 22:28:15 <Anduckkk> sipa: yes, 0.8.5 created
1746 2013-10-26 22:28:55 <Anduckkk> i did install the dev version first but removed and installed 0.8.5 and sendmany'd
1747 2013-10-26 22:29:08 _ingsoc has quit (Quit: leaving)
1748 2013-10-26 22:33:52 <sipa> that is very strange
1749 2013-10-26 22:34:12 <sipa> the minimum fee is 0.0001 iirc, not 0.00001
1750 2013-10-26 22:34:47 <Apocalyptic> minimum fee is 0
1751 2013-10-26 22:34:48 * Apocalyptic hides
1752 2013-10-26 22:35:34 <jegz> do we know the average time it takes to get 6 confirmations of a transaction?
1753 2013-10-26 22:35:51 <Ascendion> about an hour from first confirm
1754 2013-10-26 22:36:14 Polyatomic has joined
1755 2013-10-26 22:36:14 Polyatomic has quit (Changing host)
1756 2013-10-26 22:36:14 Polyatomic has joined
1757 2013-10-26 22:36:20 <Ascendion> since difficulty is supposed to fine tune the block generation rate to 1 per 10 minutes :)
1758 2013-10-26 22:36:34 <Ferroh> jegz: Right now it is more like 6*7 minutes
1759 2013-10-26 22:36:35 <gmaxwell> Ascendion: 50 minutes after first confirm would be the by design target, you mean.
1760 2013-10-26 22:37:15 <jegz> a full hour? damn
1761 2013-10-26 22:37:38 <Ferroh> jegz: Ask how long to confirm a wire transfer.
1762 2013-10-26 22:37:49 <jegz> how long to confirm a wire transfer?
1763 2013-10-26 22:38:08 <Ferroh> days
1764 2013-10-26 22:38:14 <jegz> shit
1765 2013-10-26 22:38:18 <jegz> that's longer than an hour
1766 2013-10-26 22:39:13 <Ascendion> and some banks charge fees for sending/accepting wire transfers :) and sometimes they are a percentage of the amount transfered, not a fixed amount :)
1767 2013-10-26 22:40:25 <jegz> haha yeah that's fucked up
1768 2013-10-26 22:40:28 <Ascendion> bitcoin would die in a hurry if there were a fixed 0.1% transaction fee built in and transactions that didnt have that much fee were auto-rejected :)
1769 2013-10-26 22:40:33 <jegz> well btc charged fees too
1770 2013-10-26 22:40:41 <jegz> yes
1771 2013-10-26 22:41:13 <jegz> 0.0001 is acceptable
1772 2013-10-26 22:43:55 ielo has joined
1773 2013-10-26 22:44:00 <Ascendion> hmmmm the json rpc shows that it is possible, but do transactions that have multiple addresses for a single transaction output really happen ??
1774 2013-10-26 22:44:18 ielo has quit (Changing host)
1775 2013-10-26 22:44:18 ielo has joined
1776 2013-10-26 22:46:24 <gfawkes_> yes
1777 2013-10-26 22:47:12 <Ascendion> hmmmmm gonna need some explantion of that so I can model the database right :)
1778 2013-10-26 22:47:28 <gfawkes_> the simplest example would be to consolidate dust
1779 2013-10-26 22:48:22 <Ascendion> but thats gonna be multiple inputs to a single output address
1780 2013-10-26 22:48:31 <Ascendion> that my model already handles
1781 2013-10-26 22:49:05 <gfawkes_> well the reverse is acceptable too, theres someone spamming wallets from one of the gambling sites thats doing it currently
1782 2013-10-26 22:49:24 <gfawkes_> single input multiple outputs
1783 2013-10-26 22:49:53 <Ascendion> multiple outputs is fine -- its multiple addresses in a single output that I'm asking about
1784 2013-10-26 22:50:43 wbill has joined
1785 2013-10-26 22:50:49 shesek has quit (Ping timeout: 245 seconds)
1786 2013-10-26 22:50:50 <gfawkes_> why wouldnt it? if i get 0.1 btc to address X and 0.1 from address Y and i sent 0.2 to address Z
1787 2013-10-26 22:51:04 <gfawkes_> err poorly worded =\
1788 2013-10-26 22:51:20 MrDaneelOlivaw has joined
1789 2013-10-26 22:51:25 <gfawkes_> if someone sent me 0.1 to x and 0.1 to y, as long as i have those two keys i could sent 0.2 to z
1790 2013-10-26 22:51:55 <Ascendion> but could you send an amount to address A and address B using only a single output section in the transaction
1791 2013-10-26 22:52:01 ovidiusoft has quit (Quit: Ex-Chat)
1792 2013-10-26 22:52:21 super3 has quit (Ping timeout: 248 seconds)
1793 2013-10-26 22:52:22 <Ascendion> the fact that the Vout subsection of the transaction allows for a list of addresses implies the possibility
1794 2013-10-26 22:52:36 <gfawkes_> right
1795 2013-10-26 22:53:13 <Ascendion> so whats the rule in that case -- the actual amount sent is the output value * the number of addresses listed ??
1796 2013-10-26 22:53:31 <Ascendion> otherwise you are double spending the amount
1797 2013-10-26 22:53:36 <gfawkes_> i think it would be dependent upon the implementation as to whether or not it supported that particular transaction structure, one of the core devs might need to answer this one... ive only wrapped and worked with the json-rpc interface from the client side
1798 2013-10-26 22:54:20 <Ascendion> ya -- thats what I'm doing -- json-rpc looking at the decodedrawtransaction data so I can import it into an SQL database
1799 2013-10-26 22:54:33 <gfawkes_> i wouldnt think it would be *, just a sum
1800 2013-10-26 22:54:52 <gfawkes_> not sure tho
1801 2013-10-26 22:54:56 cysm has quit (Ping timeout: 245 seconds)
1802 2013-10-26 22:54:56 paracyst has quit (Ping timeout: 245 seconds)
1803 2013-10-26 22:55:08 <Ascendion> ?? one output, one value, 2 or more addresses ??? what are you summing in that case ??
1804 2013-10-26 22:55:13 wbill has quit (Quit: Textual IRC Client: www.textualapp.com)
1805 2013-10-26 22:55:38 <Ascendion> gmaxwell -- HELP :)
1806 2013-10-26 22:55:58 <sipa> Ascendion: the addresses list tells you which addresses are recognized to be involved in the output script
1807 2013-10-26 22:56:05 daybyter has quit (Quit: Konversation terminated!)
1808 2013-10-26 22:56:14 <sipa> it may mean that any or all are needed to spend that output
1809 2013-10-26 22:56:34 <Ascendion> so its not that each one gets the value, but that any of them can use the output as an input ??
1810 2013-10-26 22:56:56 <Ascendion> ooooo OUCH
1811 2013-10-26 22:56:57 <sipa> every output can be spent exactly once
1812 2013-10-26 22:57:30 <Ascendion> in your opinion -- how often does this feature get used ??
1813 2013-10-26 22:57:39 <sipa> not
1814 2013-10-26 22:57:49 wbill has joined
1815 2013-10-26 22:57:50 <sipa> at least not in that form
1816 2013-10-26 22:58:03 <skinnkavaj> gmaxwell you are famous! http://www.reddit.com/r/Bitcoin/comments/1p9ikw/why_brainwallets_are_a_bad_idea/
1817 2013-10-26 22:58:06 <sipa> more complex scripts are done through p2sh
1818 2013-10-26 22:58:12 wbill has quit (Remote host closed the connection)
1819 2013-10-26 22:58:24 <sipa> where you get an address that refers to the hash of the actual script
1820 2013-10-26 22:58:39 <Ascendion> ok -- I've seen that p2sh several times -- wtf is it ?? :)
1821 2013-10-26 22:58:43 askmike has joined
1822 2013-10-26 22:58:49 <sipa> but you don't know the actual script until the output is redeemed
1823 2013-10-26 22:59:02 <sipa> p2sh = pay to script hash
1824 2013-10-26 22:59:03 <MrDaneelOlivaw> hello, I'm trying to build a git clone on saucy, some issues with boost-chrono, both using boost 1.49, 1.53, 1.54, both from the ubuntu repo and from sources (stage, BOOST_REPO). Is this someone has seen already ?
1825 2013-10-26 22:59:19 <MrDaneelOlivaw> something*
1826 2013-10-26 23:01:27 <Ascendion> sipa -- so a script hash will appear as an address with no way to identify it as a script hash or to determine who actually gets the output ??
1827 2013-10-26 23:02:35 <MrDaneelOlivaw> sipa: that's not entirely true,  the script could be public long before it is redeemed
1828 2013-10-26 23:03:36 <MrDaneelOlivaw> anyone building master on ubuntu saucy ? link error on configure, boost-chrono
1829 2013-10-26 23:04:18 <MrDaneelOlivaw> has anyone tried btcd yet ? I've started it to see how it works
1830 2013-10-26 23:04:23 <sipa> Ascendion: p2sh addresses start with a 3
1831 2013-10-26 23:04:27 <Ascendion> read the configure script and see what version of boost chrono its looking for :)
1832 2013-10-26 23:05:02 <sipa> Ascendion: and if you ask the question "who gets the output" you are already seeimg it the wrong way
1833 2013-10-26 23:05:22 <MrDaneelOlivaw> Ascendion, I had a look and I figured it just doesn't like some versions like 1.52 because of some bug
1834 2013-10-26 23:05:28 <sipa> Ascendion: transaction outouts are sent to a script; anyone who can satisfy that script can spend it
1835 2013-10-26 23:05:29 <MrDaneelOlivaw> but i'm not using that
1836 2013-10-26 23:05:46 <sipa> Ascendion: you can create an output that everyone can spend even
1837 2013-10-26 23:05:52 <sipa> Ascendion: or nobody
1838 2013-10-26 23:06:17 <sipa> Ascendion: or some weird combination like "2 out of these 3 keys need to sign off on it"
1839 2013-10-26 23:06:25 askmike has quit (Remote host closed the connection)
1840 2013-10-26 23:06:29 <sipa> or much more complicated schemes
1841 2013-10-26 23:06:44 CircusPeanut has quit (Ping timeout: 246 seconds)
1842 2013-10-26 23:06:46 askmike has joined
1843 2013-10-26 23:06:51 askmike has quit (Remote host closed the connection)
1844 2013-10-26 23:07:07 <MrDaneelOlivaw> as far as I know the pay to script is such that the input script can't have operations and that since it's a hash the data can't change sipa
1845 2013-10-26 23:07:17 rdymac has quit (Excess Flood)
1846 2013-10-26 23:07:28 askmike has joined
1847 2013-10-26 23:08:09 abueesp has quit (Ping timeout: 245 seconds)
1848 2013-10-26 23:08:15 <Ascendion> ok -- but I could say "if there are more than one address in the output, they all get the credit until that output is spent"
1849 2013-10-26 23:09:01 rdymac has joined
1850 2013-10-26 23:10:08 <MrDaneelOlivaw> more than one script hash ?
1851 2013-10-26 23:10:19 <Luke-Jr> Ascendion: that's impossiblew
1852 2013-10-26 23:10:23 <Ascendion> would that be reasonably consistent with what the wallet does for transactions where it owns one of the addresses in a multi-address output ??
1853 2013-10-26 23:10:37 <Luke-Jr> addresses aren't in outputs
1854 2013-10-26 23:11:01 <gmaxwell> sipa: what happened to the comparison tool? why isn't it running in the pull tester anymore?
1855 2013-10-26 23:11:10 <Ascendion> the bloody RPC results has a List<string> Addresses property
1856 2013-10-26 23:11:12 <Luke-Jr> gmaxwell: it broke and nobody fixed it yet
1857 2013-10-26 23:11:26 <Luke-Jr> Ascendion: blame me for that I guess
1858 2013-10-26 23:11:32 <sipa> Ascendion: "ownership" is tok simple to represent this
1859 2013-10-26 23:11:38 askmike has quit (Ping timeout: 246 seconds)
1860 2013-10-26 23:11:46 <sipa> Ascendion: you will likely never see multiple addresses in that list
1861 2013-10-26 23:12:02 <gmaxwell> is that list useful for anything at all?
1862 2013-10-26 23:12:03 <MrDaneelOlivaw> hey Luke-Jr gmaxwell , has any of you tried to build on saucy or do you know anyone that does ? I'm trying but I'm having issues. I'm happy to provide a patch but so far I have not figured out what's the issue
1863 2013-10-26 23:12:16 <sipa> gmaxwell: pure multisig
1864 2013-10-26 23:12:25 <gmaxwell> MrDaneelOlivaw: what is saucy?
1865 2013-10-26 23:12:30 <Ascendion> we are gonna find out -- I'm about to rip every transaction in the block chain and process it into my database :)
1866 2013-10-26 23:12:34 <MrDaneelOlivaw> gmaxwell: sorry, the latest ubuntu release
1867 2013-10-26 23:12:38 mapppum has quit (Ping timeout: 265 seconds)
1868 2013-10-26 23:12:44 <MrDaneelOlivaw> I'm trying to build a bit glone
1869 2013-10-26 23:12:53 <MrDaneelOlivaw> s/glone/clone
1870 2013-10-26 23:13:01 <MrDaneelOlivaw> s/bit/git
1871 2013-10-26 23:13:03 <lianj> Ascendion: have fun with 100gb then
1872 2013-10-26 23:13:37 <Ascendion> lianj -- I got at least 3tb free space -- 100gb is nothing around here :)
1873 2013-10-26 23:13:55 <MrDaneelOlivaw> gmaxwell: i get a similar issue with both stage boost and installed from ubuntu repo http://pastebin.com/CMZyFwba
1874 2013-10-26 23:14:16 <Ascendion> and I got space in this machine for at least 3 more drives :)
1875 2013-10-26 23:14:27 <sipa> Ascendion: a database of 100gb is something else than 100gb of stprage
1876 2013-10-26 23:14:30 <gmaxwell> MrDaneelOlivaw: does 0.8.5 build for you?
1877 2013-10-26 23:14:33 <sipa> but feel free, of course
1878 2013-10-26 23:14:34 <lianj> Ascendion: just saying its a lot of db writes, see you in a couple of days ;)
1879 2013-10-26 23:15:15 <Ascendion> yup -- I'm about to move the sql database to a different drive than the one housing the blockchain to speed things up :)
1880 2013-10-26 23:15:16 nomailing has quit (Quit: nomailing)
1881 2013-10-26 23:15:33 <sipa> also, just calling the rpc for every transaction alone will take ages
1882 2013-10-26 23:15:39 a_meteor has quit (Ping timeout: 245 seconds)
1883 2013-10-26 23:15:43 <MrDaneelOlivaw> gmaxwell: I think it did but let me check quickly
1884 2013-10-26 23:16:15 asuk has joined
1885 2013-10-26 23:16:17 <Ascendion> ya well -- I didnt create the requirements that led to this -- I just gotta code it :)
1886 2013-10-26 23:16:47 <sipa> may i ask the requirements?
1887 2013-10-26 23:17:13 <CodeShark> Ascendion: fwiw, I've implemented an SQL blockchain database…and it can take over a week to sync even on a system with relatively fast I/O
1888 2013-10-26 23:17:20 <CodeShark> and I used the p2p protocol directly
1889 2013-10-26 23:17:29 <CodeShark> I can't imagine how long it would take using RPC
1890 2013-10-26 23:17:36 <Ascendion> the ability to quickly get a current balance for any given address (and in aggregate for any collection of addresses)
1891 2013-10-26 23:17:52 <lianj> yea, while everything is bad, best is to read from a giant bootstrap.dat
1892 2013-10-26 23:17:56 <sipa> Ascendion: you better iterate the utxo set for that
1893 2013-10-26 23:18:08 deepc0re has joined
1894 2013-10-26 23:18:16 <sipa> Ascendion: will build you a balance for every address out there in seconds
1895 2013-10-26 23:18:38 <Ascendion> utxo ??
1896 2013-10-26 23:18:51 <sipa> (even though reasoning in term of address balances is quite flawed)
1897 2013-10-26 23:19:04 <sipa> the unspent transaction output set
1898 2013-10-26 23:19:07 <MrDaneelOlivaw> gmaxwell: 0.8.5 works. it doesn't even use configure OR boost-chrono so yes, the issue is only with 0.9
1899 2013-10-26 23:19:20 <sipa> that's a database that bitcoind holds internally
1900 2013-10-26 23:19:33 <sipa> it's very optimized and very small :)
1901 2013-10-26 23:19:33 <lianj> (per default)
1902 2013-10-26 23:20:36 <MrDaneelOlivaw> why not using electrum server
1903 2013-10-26 23:20:46 <sipa> yeah, they do that too i think
1904 2013-10-26 23:20:49 <MrDaneelOlivaw> if you need the current balance quick
1905 2013-10-26 23:21:06 <MrDaneelOlivaw> who else does it _ sorry i seem to have missed that from the converstion
1906 2013-10-26 23:21:26 <Ascendion> when you say "internally" do you mean in memory or is it on disk ??
1907 2013-10-26 23:21:34 hsmiths has quit (Read error: Connection reset by peer)
1908 2013-10-26 23:21:34 <sipa> on disk
1909 2013-10-26 23:21:43 <sipa> though cached in memory for fast access
1910 2013-10-26 23:22:09 <sipa> also, why do you need two question marks for every question? :)
1911 2013-10-26 23:22:24 <MrDaneelOlivaw> gmaxwell: what do you build 0.9.x on ? any hint on how you built boost ?
1912 2013-10-26 23:22:42 <sipa> 0.9 doesn't exist :)
1913 2013-10-26 23:22:52 <MrDaneelOlivaw> sipa: whatever is in git now then.
1914 2013-10-26 23:22:52 <sipa> and sorry, i haven't tried saucy yet
1915 2013-10-26 23:23:06 <gmaxwell> MrDaneelOlivaw: why are you talking about building boost?
1916 2013-10-26 23:23:33 <Ascendion> grrr -- here is the scenario -- a user is gonna give me address, I'm gonna store that in the database, scan the transaction outputs to get all the coins sent to that address, rip through the inputs looking for spent outputs and deduct that amount, then balance updates on that address happen in realtime as blocks are recieved and their transactions imported
1917 2013-10-26 23:23:53 <MrDaneelOlivaw> gmaxwell: well cause when i had the bitcoin ./configure issue with boost-chrono I tried all versions in the ubuntu repo and then i tried without and have my own built from src
1918 2013-10-26 23:24:29 <gmaxwell> MrDaneelOlivaw: you're going to screw up your distribution install randomly overwriting things.
1919 2013-10-26 23:24:30 <MrDaneelOlivaw> gmaxwell: I wanted to rule out it was something in the ubuntu boost package that went wrong
1920 2013-10-26 23:24:44 <MrDaneelOlivaw> gmaxwell: i'm using docker or vms, no issue
1921 2013-10-26 23:24:46 <gmaxwell> ... you should have started at looking at the configure log output to see why it failed.
1922 2013-10-26 23:24:52 <sipa> Ascendion: seems very inefficient
1923 2013-10-26 23:24:57 <MrDaneelOlivaw> I did, link error with boost-chrono
1924 2013-10-26 23:25:01 hsmiths has joined
1925 2013-10-26 23:25:22 <sipa> Ascendion: but i think the concept of them giving you a single address is flawed in the first place
1926 2013-10-26 23:25:41 <MrDaneelOlivaw> boost-chrono is a new addition, maybe boost has changed the signature slightly ?
1927 2013-10-26 23:26:33 <Ascendion> you keep saying flawed and I kinda understand why cause the script system allows something other than one output to one address and it takes the privkey for that address to use the ouput
1928 2013-10-26 23:26:51 <sipa> no, flawed because a wallet shouldn't consist of just an address
1929 2013-10-26 23:26:56 <Ascendion> but I have to understand the non-standard script cases currently in general use to code from them
1930 2013-10-26 23:27:05 <sipa> an address should just be used once to receive a transaction
1931 2013-10-26 23:27:13 <Ascendion> we already agreed that a wallet is a collection of addresses earlier
1932 2013-10-26 23:27:15 <sipa> and a wallet consists of a bunch of them
1933 2013-10-26 23:27:17 <sipa> ok
1934 2013-10-26 23:27:30 <sipa> so, purely technical
1935 2013-10-26 23:27:40 <MrDaneelOlivaw> gmaxwell:  configure log http://pastebin.com/Gug4aDHw
1936 2013-10-26 23:27:46 <sipa> the utxo set is sort of the result of all blocks
1937 2013-10-26 23:27:50 <Ascendion> but as far as the transaction is concerned -- the output goes to an address even it is only used once and never again
1938 2013-10-26 23:27:54 <sipa> it is not indexed by address
1939 2013-10-26 23:28:04 <sipa> but the utxo set is very small
1940 2013-10-26 23:28:09 <sipa> (250 MB)
1941 2013-10-26 23:28:39 <Ascendion> what format is the DB stored in ??
1942 2013-10-26 23:28:49 <sipa> so searching through it to find unspent coins assigned to a particular address is a ton faster than looking through the whole blockchain
1943 2013-10-26 23:28:54 <gmaxwell> MrDaneelOlivaw: you appear to have cut off the actual error.
1944 2013-10-26 23:28:55 <sipa> very custom designed
1945 2013-10-26 23:29:25 <Ascendion> I wont be processing the whole blockchain -- I've got a seperate table for trans outputs in the sql database
1946 2013-10-26 23:29:36 <Ascendion> same for trans inputs
1947 2013-10-26 23:29:47 <sipa> it's a leveldb database, but the encoding used for the outputs is custom
1948 2013-10-26 23:30:00 <sipa> Ascendion: how about only storing unspent outputs? :)
1949 2013-10-26 23:30:53 bitRipperX has joined
1950 2013-10-26 23:31:12 <Ascendion> that breaks things if I have to roll back due to a reorg the way my code is currently set up
1951 2013-10-26 23:31:53 <Ascendion> if the reorg crosses an output and an input I couldnt roll back then roll forward the balance
1952 2013-10-26 23:32:02 <sipa> haha
1953 2013-10-26 23:32:16 <sipa> you're going through the exact same evolution as bitcoind itself
1954 2013-10-26 23:32:28 <sipa> we used to have such a full database for tracking things
1955 2013-10-26 23:32:38 <sipa> but it became too large and too slow
1956 2013-10-26 23:32:58 <sipa> so it was changed to a separate database with just the unspent outputs in it
1957 2013-10-26 23:33:06 <sipa> and rollback data for every block
1958 2013-10-26 23:35:48 <Ascendion> ya unfortunatly I do have to replicate some of the bitcoind logic -- and keeping the spent outputs doesnt slow me down much since its all indexed in the sql database -- dont have to do scans
1959 2013-10-26 23:36:04 ielo has quit (Ping timeout: 245 seconds)
1960 2013-10-26 23:36:09 a_meteor has joined
1961 2013-10-26 23:36:11 <sipa> your database still does :)
1962 2013-10-26 23:36:16 <MrDaneelOlivaw> gmaxwell: https://dpaste.de/Yqdu here's full. Sorry it took me a second to curl the file into a pastebin like site
1963 2013-10-26 23:37:52 <Ascendion> only of the records that have a given address in the output table.. then that can be filtered based on if the output has an input linked to get the unspent outputs
1964 2013-10-26 23:38:51 hsmiths has quit (Quit: Nettalk6 - www.ntalk.de)
1965 2013-10-26 23:38:55 <sipa> Ascendion: sure, but you still have to build and maintain that database
1966 2013-10-26 23:39:07 <Ascendion> thats the code I'm working on right now :)
1967 2013-10-26 23:39:35 <sipa> and for performance, you probably want to read block data in p2p rather than using rpc
1968 2013-10-26 23:39:56 <sipa> otherwise you might just need several weeks to just query the information
1969 2013-10-26 23:39:59 <Ascendion> the only P2P c# implementation I found was pure junk -- out of date
1970 2013-10-26 23:40:35 <sipa> there are certainly language with better support in bitcoin libraries :)
1971 2013-10-26 23:40:53 <sipa> *languages
1972 2013-10-26 23:41:02 <Ascendion> we'll see how it goes -- I've opened up bitcoind to allow 16 rpc calls at a time and set up my code to run as many as 8 in parallel :)
1973 2013-10-26 23:41:09 reneg has quit (Ping timeout: 265 seconds)
1974 2013-10-26 23:42:13 hsmiths has joined
1975 2013-10-26 23:42:24 grau has joined
1976 2013-10-26 23:42:32 <gmaxwell> MrDaneelOlivaw: that output makes no sense, it shows the test passing.
1977 2013-10-26 23:42:36 <sipa> and you'll be iterating over 25 million transactions or so
1978 2013-10-26 23:42:51 <sipa> how many do you do per second?
1979 2013-10-26 23:43:19 <Ascendion> we'll find out when I start the clean pull here in a couple of hours -- still got some code to finish
1980 2013-10-26 23:43:33 <sipa> ok
1981 2013-10-26 23:43:33 <MrDaneelOlivaw> gmaxwell: agreed, it is confusing. It does the same whether i use a staged boost not patched by ubuntu and debian.
1982 2013-10-26 23:44:00 reneg has joined
1983 2013-10-26 23:44:07 <sipa> if you get to 100 per second, it'll take 3 days
1984 2013-10-26 23:44:08 <MrDaneelOlivaw> gmaxwell: what's you configure like ? do you use boost-chrono ? boost version ?
1985 2013-10-26 23:44:11 <gmaxwell> I don't know what you mean by "staged boost"
1986 2013-10-26 23:44:44 <gmaxwell> MrDaneelOlivaw: it compiles on a standard fedora 19 system, other people here use ubuntu but I don't know if they're using "saucy" yet.
1987 2013-10-26 23:44:47 <Ascendion> 3 days is acceptable -- even a week would be acceptable :) cause once its in sync its gonna be fine :)
1988 2013-10-26 23:44:48 Anduckkk has quit (Ping timeout: 252 seconds)
1989 2013-10-26 23:45:27 <sipa> Ascendion: if your database can keep up, yes
1990 2013-10-26 23:45:44 <sipa> it's certainly possible
1991 2013-10-26 23:46:32 McKay has quit (Ping timeout: 240 seconds)
1992 2013-10-26 23:46:35 <Ascendion> my blockchain is synced -- so its all local access -- once I move the sql db off the drive that has the blockchain, I'll be going as fast as I can with 10krpm drives
1993 2013-10-26 23:46:53 <MrDaneelOlivaw> gmaxwell: https://dpaste.de/QC8y grep for BOOST_ROOT
1994 2013-10-26 23:46:53 lclc has joined
1995 2013-10-26 23:47:27 McKay has joined
1996 2013-10-26 23:47:29 <Ascendion> if I had a couple of SSDs then we could see some REAL speed :)
1997 2013-10-26 23:47:42 <MrDaneelOlivaw> or lost of ram :)
1998 2013-10-26 23:47:43 <gmaxwell> MrDaneelOlivaw: you really shouldn't be recompiling boost, actually you've voided any support warranty from me by doing that. Good luck.
1999 2013-10-26 23:47:46 <lclc> hi, getinfos last line is "errors". What can the value of errors be? is it an integer with the number of errors, an array with error description strings?
2000 2013-10-26 23:47:50 <MrDaneelOlivaw> s/lost/lots/
2001 2013-10-26 23:48:05 <MrDaneelOlivaw> gmaxwell: I don't *want* to recompile boost
2002 2013-10-26 23:48:28 <gmaxwell> lclc: its a string.
2003 2013-10-26 23:48:33 <MrDaneelOlivaw> I did to see if the error went away, if it had to do with the package being somehow broke in ubuntu latest
2004 2013-10-26 23:48:44 deafboy has quit (Ping timeout: 245 seconds)
2005 2013-10-26 23:48:53 <lclc> thx
2006 2013-10-26 23:49:34 <MrDaneelOlivaw> gmaxwell: i'm only asking what version of boost do you use so that I can find the same for it
2007 2013-10-26 23:49:59 <MrDaneelOlivaw> I quite like the new flag in getblock, the gVerbose flag
2008 2013-10-26 23:50:03 <gmaxwell> MrDaneelOlivaw: please stop fucking up your distribution.
2009 2013-10-26 23:50:21 <gmaxwell> I will not tell you what version of boost I have. You've probably hosed your distribution seriously now.
2010 2013-10-26 23:50:22 <MrDaneelOlivaw> gmaxwell: i'm not fucking up my distribution, i'm building a docker dev environment
2011 2013-10-26 23:50:29 <MrDaneelOlivaw> docker, container, vm
2012 2013-10-26 23:50:34 <sipa> gmaxwell: ehm, calm down
2013 2013-10-26 23:50:36 <MrDaneelOlivaw> i'm not running that in my distro
2014 2013-10-26 23:51:11 <MrDaneelOlivaw> I'm not even running ubuntu on my box, i'm running arch, i don't get why you get so upset, i'm just testing/devving on latest of the stack
2015 2013-10-26 23:51:17 <gmaxwell> I'm not upset.
2016 2013-10-26 23:51:34 <gmaxwell> MrDaneelOlivaw: you're failing into the standard yakshaving trap. There is some boring simple problem, you're on an incorrect tangent for fixing it, and you're digging yourself in deeper.
2017 2013-10-26 23:51:35 <MrDaneelOlivaw> well I felt you were and it wasn't my intention
2018 2013-10-26 23:52:09 <sipa> i have boost 1.48 here
2019 2013-10-26 23:52:13 <gmaxwell> I'd like to help get it fixed, but I don't think I can now, because you've probably put your system in a state which is screwed up.
2020 2013-10-26 23:52:19 <MrDaneelOlivaw> MrDaneelOlivaw: gmaxwell possibly the tip of bitcoin just doesn't build on latest ubuntu. I was trying to fix it and find what's wrong.
2021 2013-10-26 23:52:38 <MrDaneelOlivaw> gmaxwell:  do you even know what docker or virtual machines are ?
2022 2013-10-26 23:52:48 <sipa> gmaxwell: did you see he's doing that in a vm? :)
2023 2013-10-26 23:52:49 <gmaxwell> MrDaneelOlivaw: yes, and overwriting your system copy of boost is not the way to do that. Some minute change in the configure script probably is.
2024 2013-10-26 23:52:49 <MrDaneelOlivaw> I run a new instance every time
2025 2013-10-26 23:53:04 <MrDaneelOlivaw> what system it's a container ?
2026 2013-10-26 23:53:07 <MrDaneelOlivaw> docker
2027 2013-10-26 23:53:13 <MrDaneelOlivaw> http://www.docker.io/
2028 2013-10-26 23:53:19 <gmaxwell> MrDaneelOlivaw: sure, so go back to a clean system and give me a configure log from that.
2029 2013-10-26 23:53:31 <MrDaneelOlivaw> I just did
2030 2013-10-26 23:53:36 <MrDaneelOlivaw> do you want the docker sources ?
2031 2013-10-26 23:54:02 <gmaxwell> Which one is the clean build?
2032 2013-10-26 23:54:10 <MrDaneelOlivaw> it takes saucy, downloads the dev packages and dependencies, clones git repo and tries to build it with ./autogen, configure and fails there.
2033 2013-10-26 23:55:05 <gmaxwell> as an aside, why are you buiding with --disable-ipv6?
2034 2013-10-26 23:55:30 <gmaxwell> Also, does prefix work without --?
2035 2013-10-26 23:56:03 <MrDaneelOlivaw> i think i tried without, let me remove it and rerunning from scratch
2036 2013-10-26 23:57:13 <gmaxwell> MrDaneelOlivaw: you're currently building with prefix without the --
2037 2013-10-26 23:57:28 <gmaxwell> which I think is incorrect, though it's not obvious to me why it should be harmful.
2038 2013-10-26 23:57:56 <MrDaneelOlivaw> agreed, i'm removing and retrying for the sake of it
2039 2013-10-26 23:58:06 <gmaxwell> Please answer: 16:51 <@gmaxwell> Which one is the clean build?
2040 2013-10-26 23:58:16 <gmaxwell> Which is the configure log from a bog standard system?
2041 2013-10-26 23:58:52 lclc has left ("Konversation terminated!")
2042 2013-10-26 23:59:59 <MrDaneelOlivaw> both were saucy clean system with just build-essential and dev pages libtool etc, the first one with boost dev packages installed and the second one without them