1 2013-04-29 00:43:16 <travolta> hey sipa, you know the importaddress function on CodeShark's branch, where it says "label" as one of the params is that synomymus for "account"?
2 2013-04-29 01:07:11 <travolta> Can someone be so kind to explain to be what importprivkey [label] is on that api call?
3 2013-04-29 01:07:17 <travolta> *to me
4 2013-04-29 01:15:33 <EvilPete> gosh damn it.
5 2013-04-29 01:15:46 <EvilPete> "Failed to write undo data" bitcoin-qt
6 2013-04-29 01:16:24 <EvilPete> LevelDB write failure: IO error: /Users/peter/Library/Application Support/Bitcoin/chainstate/067308.log: Too many open files
7 2013-04-29 01:21:55 <denisx> EvilPete: https://github.com/bitcoin/bitcoin/commit/fe15aa335c223a37a1196d5f509fbe8059b733bb
8 2013-04-29 01:39:39 <rfish> Is it possible that certain elements inside the hashPrevBlock make it impossible to find a hash at a certain difficulty regardless of Nonce and hashMerkleRoot?
9 2013-04-29 01:40:12 <Luke-Jr> yes
10 2013-04-29 01:40:27 <rfish> so then that can kill the block chain no?
11 2013-04-29 01:40:32 <Luke-Jr> rfish: yes
12 2013-04-29 01:40:37 <sirdefaced> wow
13 2013-04-29 01:40:50 <Luke-Jr> rfish: the chance of it happening are far far far less likely than the Sun exploded and killing everyone, though
14 2013-04-29 01:40:57 <Luke-Jr> exploding*
15 2013-04-29 01:41:21 <rfish> oh ok
16 2013-04-29 01:41:30 <Luke-Jr> rfish: and if it ever seemed like it did, the community *could* hard fork to resolve it
17 2013-04-29 01:41:31 <rfish> thanks
18 2013-04-29 01:41:52 <rfish> that makes sense
19 2013-04-29 01:46:08 <gonffen> Luke-Jr: I assume there is no way to determine if that has occurred versus say, incredibly unlikely bad luck?
20 2013-04-29 01:46:33 <Luke-Jr> gonffen: there is no difference..
21 2013-04-29 01:47:55 <gonffen> do you mean literally or practically?
22 2013-04-29 01:48:11 <Luke-Jr> literally
23 2013-04-29 01:53:23 <rfish> And another question is what incentive do miners have in propigating transactions throughout the network
24 2013-04-29 01:53:38 <sivu> fees
25 2013-04-29 01:53:50 <rfish> no, that is for including them in the next block
26 2013-04-29 01:53:55 <Luke-Jr> fees, and also a healthy network means Bitcoins have value
27 2013-04-29 01:54:04 <Luke-Jr> rfish: eh, that's all miners do
28 2013-04-29 01:54:10 <rfish> not for propigating the transaciton, that costs bandwidth with no reward
29 2013-04-29 01:54:26 <Luke-Jr> rfish: miners are not unique there
30 2013-04-29 01:54:34 <Luke-Jr> all nodes, mining or not, need to relay transactions
31 2013-04-29 01:54:55 <syskk> hi all. what does vMasterKey represent in wallet.cpp?
32 2013-04-29 01:55:00 <syskk> is it the decryption key?
33 2013-04-29 01:55:09 <rfish> all they need is their coinbase theoritically to generate a new block
34 2013-04-29 01:55:38 <Luke-Jr> rfish: you're confusing something, but I'm not sure what
35 2013-04-29 01:55:43 <Luke-Jr> there are two things:
36 2013-04-29 01:56:04 <Luke-Jr> 1) relaying transactions p2p; all nodes (not miners) must do this for the network to function; the incentive is to keep the network functioning so their own bitcoins have value
37 2013-04-29 01:56:18 <Luke-Jr> 2) including transactions in blocks; only miners do this, and they get the transaction fees for doing it
38 2013-04-29 01:57:33 <rfish> ok, most people don't care about that, they just assume other people will do it
39 2013-04-29 01:57:36 <rfish> take torrents for example
40 2013-04-29 01:57:49 <rfish> people choose not to upload
41 2013-04-29 01:57:54 <rfish> and to only download
42 2013-04-29 01:58:12 <rfish> and when it is required there are ways to upload to yourself to fake out the system
43 2013-04-29 02:00:23 <rfish> I'm sure someone will create a wallet that claims to comsume less bandwidth and it could catch on
44 2013-04-29 02:13:48 <syskk> how can i get a list of private keys?
45 2013-04-29 02:14:32 <rfish> whos private keys?
46 2013-04-29 02:14:44 <Luke-Jr> syskk: I don't think there is a way to do that.
47 2013-04-29 02:15:05 <Luke-Jr> short of manually dumping the database
48 2013-04-29 02:15:10 <syskk> Luke-Jr: I just need some help to get started.. I want to implement this feature
49 2013-04-29 02:15:36 <Luke-Jr> syskk: what feature?
50 2013-04-29 02:15:49 <syskk> i want to implement a way to dump all private keys
51 2013-04-29 02:16:01 <syskk> ive taken a look at dumpprivkey() in rpcdump.cpp
52 2013-04-29 02:16:14 <syskk> but im not sure how to get a list of addresses
53 2013-04-29 02:18:02 <sivu> getaddressbyaccount
54 2013-04-29 02:18:10 <Luke-Jr> sivu: won't get all of them
55 2013-04-29 02:18:14 <Luke-Jr> syskk: why?
56 2013-04-29 02:18:14 <sivu> sorry, getaddressesbyaccount
57 2013-04-29 02:18:50 <syskk> Luke-Jr: to make it easier exporting a wallet
58 2013-04-29 02:20:25 <Luke-Jr> syskk: I believe there's a pullreq for an exportwallet RPC somewhere
59 2013-04-29 02:21:50 <syskk> Luke-Jr: thanks
60 2013-04-29 03:10:49 <ryanender_> looks like the mincoin exchange went live (finally)
61 2013-04-29 03:16:15 <dansmith_btc> Hello, does bitcoind have a built-in events mechanism which notifies when a new block arrives? I know that bitcoinjs does.
62 2013-04-29 03:17:49 <vazakl-> mincoin is cool
63 2013-04-29 03:23:31 <jgarzik> mincoin is from realsolid? ah hum :)
64 2013-04-29 03:23:40 <jgarzik> dansmith_btc: yes
65 2013-04-29 03:23:56 <ryanender_> think he built the exchange
66 2013-04-29 03:25:08 <dansmith_btc> jgarzik, what is the mechanism called?
67 2013-04-29 03:25:33 <jgarzik> dansmith_btc: -blocknotify, which gives a command to execute once per block, on the command line
68 2013-04-29 03:28:53 <dansmith_btc> jgarzik, how can I access the new block programmatically after I'm notified of it?
69 2013-04-29 03:32:04 <jgarzik> dansmith_btc: depends on what you need
70 2013-04-29 03:32:16 <jgarzik> dansmith_btc: getblock <hash> returns everything except the transaction data itself
71 2013-04-29 03:32:24 <dansmith_btc> jgarzik, I need to access block's internals
72 2013-04-29 03:32:46 <dansmith_btc> jgarzik, and the transactions as well
73 2013-04-29 03:32:48 <jgarzik> dansmith_btc: unspent transactions are 'gettransaction', otherwise you will need to get down-n-dirty with the raw block serialization
74 2013-04-29 03:32:54 <jgarzik> or gettxout
75 2013-04-29 03:33:28 <Luke-Jr> or txindex=1 <.<
76 2013-04-29 03:34:08 <jgarzik> dansmith_btc: or txindex, as luke indicated
77 2013-04-29 06:05:25 <jojo_> Just a question. Is it 6 confirmation i should wait for until i can transfer money from an adress if i make an automatic service that makes an adress and then get the money and then it should trasfer that money to another adress..
78 2013-04-29 06:09:06 <weex> jojo_: will the "another address" be yours?
79 2013-04-29 06:09:49 <weex> if so you can do it immediately but if it might be someone else's then it's safer to wait
80 2013-04-29 06:10:25 <jojo_> Ok why is it ok if it is my adress?
81 2013-04-29 06:10:40 <weex> the other thing you can do is make sure the output of the transaction that went to you is used as an input to the payment that sends to another address
82 2013-04-29 06:10:57 <weex> because the whole reason for waiting for confirmations is to protect yourself against double spending
83 2013-04-29 06:11:17 <weex> the risk is higher the fewer confirmations you wait for
84 2013-04-29 06:12:34 <jojo_> ok but if i have a database that keeps track of all the coins will that work to or should i still wait for the confirmationes?
85 2013-04-29 06:13:19 <weex> i'd wait
86 2013-04-29 06:13:36 <weex> a db doesn't much matter if you get scammed
87 2013-04-29 06:13:50 <weex> unless you're going to have some kind of deposit
88 2013-04-29 06:15:22 <jojo_> Ok so then wait to be certain then? How can u get scammed with doublespend?
89 2013-04-29 06:15:43 <jojo_> Sorry if the questions are noobish but im kinda noob toi :)
90 2013-04-29 06:17:08 <weex> https://en.bitcoin.it/wiki/Double-spending
91 2013-04-29 06:17:26 <weex> the wiki is quite good but feel free to ask after reading that
92 2013-04-29 06:29:28 <jojo_> Read a bit now :) Thanks. One question comes to mind, with all these asic power that will be releasd, will the risk of attacks go down then. Just curious cus all the attacks seems to be done by being the biggest hascher in the world. Am i right or am totally of now? :)
93 2013-04-29 06:30:45 <jojo_> on 0 conf it could be a bit unsecure either way right?
94 2013-04-29 06:31:23 <jojo_> Just asking cus im curios now.
95 2013-04-29 06:31:29 <weex> a higher network hashrate makes it harder to do an attack if you don't have an asic
96 2013-04-29 06:31:38 <weex> even if you do have asics
97 2013-04-29 06:32:18 <weex> the defense for 0 confirmation is not to accept incoming connections and try to connect to well connected nodes
98 2013-04-29 06:33:15 <jojo_> As i read there are about 47000 orders on bitforce and that hashrate should be pretty hard to get 10% of right?
99 2013-04-29 06:34:18 <jojo_> what meens by incomming connections?
100 2013-04-29 06:34:25 <weex> well it's better to just look at the network hashrate
101 2013-04-29 06:34:42 <weex> it's 70th now? so 7th is yeah...pretty hard to get
102 2013-04-29 06:35:35 <weex> by default a node will connect with just about any other node but you can set a config variable to disallow connections from other than a certain set of nodes
103 2013-04-29 06:36:16 <jojo_> you meen like ipadresses in the blockchain wallet settings?
104 2013-04-29 06:36:23 <jojo_> OR am i totally of now?
105 2013-04-29 06:39:21 <weex> in bitcoin.conf if that's what you mean
106 2013-04-29 06:42:33 <jojo_> I was thinking of blockchain.info
107 2013-04-29 06:42:43 <jojo_> Or should i not use that?
108 2013-04-29 06:43:11 <jojo_> I dont want to reli on my computer at home for this service
109 2013-04-29 06:44:27 <weex> back up, what are you really trying to do?
110 2013-04-29 06:46:14 <jojo_> im thinging of makiong a app and home page to recieve and send money i have an idea in my head i just have to learn more to get it into a service :)
111 2013-04-29 06:50:32 <weex> jojo_: you can use blockchain.info's api for some things but jsonrpc to bitcoind is more flexible i think
112 2013-04-29 06:55:00 <jojo_> The problem with the wallet at home is that if i turn of the computer the service will be down :)
113 2013-04-29 07:11:27 <jojo_> if i use blockchain.info wallet with api do i have to send an end or something efter i used the api cus when i used the api i can no longer make changes on the homepage i get this all the time "Error Saving Wallet: Error Decrypting Previously encrypted JSON. Not Saving Wallet."
114 2013-04-29 07:34:58 <sipa> jgarzik: for blockchain transactions, use getrawtransaction; gettransaction is for querying the wallet
115 2013-04-29 07:35:17 <jgarzik> sipa: I wish we had getrawblock
116 2013-04-29 07:35:27 <Diablo-D3> why dont we?
117 2013-04-29 07:35:46 <jgarzik> <shrug>
118 2013-04-29 07:35:56 <Diablo-D3> we should have four methods
119 2013-04-29 07:36:11 <Diablo-D3> getrawtransaction, putrawtransaction, getrawblock, putrawblock
120 2013-04-29 07:36:13 <jgarzik> getrawblock is easy and works great
121 2013-04-29 07:36:18 <jgarzik> without needing a txindex
122 2013-04-29 07:36:45 <Diablo-D3> now ask why put
123 2013-04-29 07:36:58 <jgarzik> or maybe getblocktransaction <blockhash> <txid>
124 2013-04-29 07:37:19 <Diablo-D3> p2pool could make use of those to quickly push tx and blocks across the p2pool network
125 2013-04-29 07:37:24 <Diablo-D3> thus giving us a leg up over other pools
126 2013-04-29 07:37:51 <syskk> in which object are private keys stored?
127 2013-04-29 07:37:54 <sipa> travolta: labels and accounts are implementation-wise the same (which just makes it even more confusing), but labels are just for recognizing addresses in your address book, and useful even if you don't use the accounting system
128 2013-04-29 07:43:34 <syskk> is wallet->mapAddressBook guaranteed to conatin all my public keys?
129 2013-04-29 07:45:56 <sipa> syskk: no
130 2013-04-29 07:46:09 <sipa> it doesn't contain change address or pool addresses
131 2013-04-29 07:46:35 <sipa> jgarzik: you mean just the raw bytes of the entire block?
132 2013-04-29 07:47:18 <syskk> sipa: i want to get a list of all my private keys.. where are they stored?
133 2013-04-29 07:47:27 <syskk> sipa: in the code i mean
134 2013-04-29 07:48:14 <sipa> syskk: CKeyStore
135 2013-04-29 07:48:28 <sipa> CWallet derives from it
136 2013-04-29 07:48:50 <syskk> sipa: thanks
137 2013-04-29 07:52:06 <syskk> sipa: how can i retrieve the CKeyStore from wallet.cpp?
138 2013-04-29 07:54:20 <syskk> there is CKeyStore::GetKeys() but it requires a address argument
139 2013-04-29 07:56:12 <sipa> syskk: eh, no
140 2013-04-29 07:56:24 <sipa> it takes a vector to put the KeyID's in
141 2013-04-29 07:56:29 <sipa> *set
142 2013-04-29 07:57:39 <sipa> you can then run GetKey or GetSecret on each
143 2013-04-29 08:00:00 <syskk> got it
144 2013-04-29 08:02:19 <jgarzik> sipa: yes
145 2013-04-29 08:02:39 <sipa> jgarzik: there are several problems with getblock/getransaction/getrawtransaction imho
146 2013-04-29 08:03:01 <jgarzik> sipa: the main use case was mentioned earlier on IRC: watching for new blocks, and scanning their entirety yourself
147 2013-04-29 08:03:41 <sipa> there's no reason why you can't even get parsed raw transactions from the chain, even without txindex
148 2013-04-29 08:04:11 <sipa> and there should be an equivalent to getrawtransaction, but fetching from wallet instead of mempool/chain
149 2013-04-29 08:04:18 <jgarzik> sipa: except the lack of an RPC interface permitting one to do so, that is :)
150 2013-04-29 08:04:24 <sipa> yes, indeed
151 2013-04-29 08:04:39 <jgm> getrawblock would definitely be handy
152 2013-04-29 08:04:41 <buZz> does this still work? https://github.com/sacarlson/MultiCoin
153 2013-04-29 08:04:43 <sipa> there is a pullreq to add a block hash to getrawtransaction
154 2013-04-29 08:04:57 <sipa> which would work, but is still a hack
155 2013-04-29 08:06:26 <jgm> Is there any consideration of breaking bitcoind in to its various components? Daemon/datastore/GUI/wallet is the obvious split, in which case a lot of functionality like this could be added as standalone pieces which wouldn't affect the core bitcoin process
156 2013-04-29 08:06:37 <syskk> sipa: getting this error: error: cannot call member function 'virtual void CKeyStore::GetKeys(std::set<CKeyID, std::less<CKeyID>, std::allocator<CKeyID> >&) const' without object
157 2013-04-29 08:06:41 <jgm> (Once the split has occurred, of course; that itself would be a big job)
158 2013-04-29 08:06:45 <syskk> sorry i am pretty noob with C++
159 2013-04-29 08:06:48 <jgarzik> jgm: yes, old thoughts
160 2013-04-29 08:06:49 <sipa> syskk: you need to call it on the wallet
161 2013-04-29 08:07:40 <sipa> ACTION would first like to see the wallet split off into a SPV client
162 2013-04-29 08:07:51 <syskk> sipa: thanks
163 2013-04-29 08:07:56 <buZz> jgm: i'd like to see the bitcoind to be made multi-coin
164 2013-04-29 08:07:59 <sipa> but that's not compatible with Gavin's idea of monitoring the mempool to make fee decisions
165 2013-04-29 08:08:21 <buZz> jgm: its just so pointless to have 99% the same code running multiple times, imho
166 2013-04-29 08:09:03 <jgm> buZz: no offence meant, but given that a lot of the forks are for code-related reasons as much as philosophical reasons that seems like just degenerating into a binfest
167 2013-04-29 08:09:29 <buZz> hmm yeah i get what you mean
168 2013-04-29 08:09:31 <jgm> jgarzik: have those thoughts been place on the "too scary" pile or are they likely to progress?
169 2013-04-29 08:10:15 <The_Fly> +1 for modularisation of bitcoin
170 2013-04-29 08:10:28 <sipa> there are several changes involved here
171 2013-04-29 08:10:39 <The_Fly> and quite breaking changes
172 2013-04-29 08:10:40 <jgarzik> jgm: It is lots of big changes
173 2013-04-29 08:10:46 <jgarzik> jgm: each with risk
174 2013-04-29 08:10:49 <sipa> first of is modularization of the code - not necessarily even splitting it into several binaries or processes
175 2013-04-29 08:10:56 <syskk> sipa: does that sound right? http://pastebin.com/kdivfqVz
176 2013-04-29 08:11:18 <The_Fly> it is something for a major version number in the milestones perhaps
177 2013-04-29 08:11:41 <sipa> syskk: that'll just get you key ids, but yes
178 2013-04-29 08:12:10 <sipa> not sure if CKeyID has a ToString, but don't expect addresses to show up
179 2013-04-29 08:12:44 <jgm> Yep suspect that the risk is relatively high. Trouble is that at some stage (if not already) it's really going to hold back bitcoin development. If, for example, adding new user-facing features to wallets is considered too dangerous due to potential side-effects to the core daemon that's not a great situation to be in
180 2013-04-29 08:13:07 <sipa> jgm: i'm sure it is holding back development
181 2013-04-29 08:13:29 <jgarzik> indeed
182 2013-04-29 08:13:53 <jgarzik> never enough developer bandwidth to fix immediate things + all the Things That Need To Be Done
183 2013-04-29 08:14:02 <sipa> unfortunately, modularization is not given a high priority (we're too busy with keeping up with other things...)
184 2013-04-29 08:14:10 <The_Fly> ive got the 0mq support up to date with whats in master
185 2013-04-29 08:14:22 <syskk> sipa: it does. its showing me a bunch of strings. what is the difference between Key ids and private keys? how do i get the private key from the key id?
186 2013-04-29 08:14:25 <The_Fly> merge was trivial changes to thread creation/ anagement
187 2013-04-29 08:14:34 <The_Fly> and added walletnotify 0mq message
188 2013-04-29 08:14:36 <sipa> syskk: as i said, use GetKey or GetSecret
189 2013-04-29 08:14:45 <syskk> sipa: ok thanks
190 2013-04-29 08:14:48 <sipa> The_Fly: nice
191 2013-04-29 08:14:54 <The_Fly> anyone able to help with rabbitmq? i found a 0mq/rabbitmq adapter
192 2013-04-29 08:14:56 <jgm> Hmm... what do you reckon would be the lowest-danger/easiest-to-tackle modularization piece? Not sure I have the time, but if I could at least take a look and work out if I have the time to tackle something
193 2013-04-29 08:14:57 <The_Fly> sipa: ty
194 2013-04-29 08:15:16 <sipa> jgm: testing!
195 2013-04-29 08:15:21 <sipa> implementation is not a problem
196 2013-04-29 08:15:27 <The_Fly> should submit a pull request for that?
197 2013-04-29 08:15:32 <The_Fly> not urgent really
198 2013-04-29 08:15:34 <sipa> The_Fly: for what?
199 2013-04-29 08:15:38 <The_Fly> updated 0mq
200 2013-04-29 08:15:49 <sipa> oh, you're not the original author?
201 2013-04-29 08:15:53 <The_Fly> no
202 2013-04-29 08:16:05 <sipa> well, feel free to submit a pullreq with an updated version then
203 2013-04-29 08:16:08 <jgm> sipa: heh, any list out there of what needs testing?
204 2013-04-29 08:16:09 <The_Fly> just his branch had not been merged and was stale
205 2013-04-29 08:16:19 <The_Fly> ok np
206 2013-04-29 08:16:20 <sipa> jgm: any pull request :)
207 2013-04-29 08:17:06 <The_Fly> anyone played with https://github.com/rabbitmq/rmq-0mq/wiki ?
208 2013-04-29 08:17:52 <jgm> sipa: there are a lot of them out there, any prioritization? Apart from any I raise myself, of course
209 2013-04-29 08:19:55 <sipa> jgm: those that are tagged with 0.8.2
210 2013-04-29 08:22:49 <jgm> will take a look and see if any fall anywhere my area of expertise
211 2013-04-29 08:23:18 <syskk> sipa: what is the difference between getkey and getsecret?
212 2013-04-29 08:23:33 <sipa> syskk: one gives you a CKey, the other gives you a CSecret
213 2013-04-29 08:23:52 <sipa> CKey is more powerful, allows conversion/signing/verification, CSecret is just the raw secret
214 2013-04-29 08:23:57 <sipa> CSecret is likely what you want here
215 2013-04-29 08:24:11 <syskk> ok
216 2013-04-29 08:24:21 <syskk> which one does dumpprivkey return?
217 2013-04-29 08:24:32 <sipa> you can't show a CKey
218 2013-04-29 08:24:36 <syskk> CSecret right?
219 2013-04-29 08:24:38 <sipa> yes
220 2013-04-29 08:24:53 <sipa> CKey is a wrapper around some OpenSSL functions that deal with crypto
221 2013-04-29 08:25:12 <sipa> look at the dumpprivkey cod
222 2013-04-29 08:25:30 <sipa> syskk: tbh, i was planning on adding a dumpwallet very soon
223 2013-04-29 08:27:18 <syskk> ok well.. im also trying to learn
224 2013-04-29 08:27:53 <syskk> sipa: why does it show like 20+ secrets but i only actually have 2 receive addresses
225 2013-04-29 08:28:14 <sipa> syskk: it will include 100 key pool addresses, and all your change addresses
226 2013-04-29 08:28:25 <sipa> neither of which are visible in the GUI
227 2013-04-29 08:28:38 <syskk> got it
228 2013-04-29 08:28:59 <syskk> actually there's something weird??? all secrets start with K or L
229 2013-04-29 08:31:16 <syskk> also, my private keys are missing
230 2013-04-29 08:31:17 <syskk> :(
231 2013-04-29 08:32:33 <sipa> why is that weird?
232 2013-04-29 08:32:51 <syskk> http://pastebin.com/wPB1uDkH
233 2013-04-29 08:33:10 <syskk> well, my private keys aren't in there...
234 2013-04-29 08:33:12 <sipa> looks good to me
235 2013-04-29 08:33:28 <syskk> the ones that i can get using dumpprivkey
236 2013-04-29 08:33:52 <sipa> sure?
237 2013-04-29 08:34:53 <syskk> well im not using this wallet so let me pastebin the output
238 2013-04-29 08:35:20 <sipa> try it on testnet
239 2013-04-29 08:36:26 <syskk> http://pastebin.com/YAdcDk7R
240 2013-04-29 08:37:19 <syskk> im probably misunderstanding something
241 2013-04-29 08:37:33 <sipa> they use a different encoding for private keys, it seems
242 2013-04-29 08:37:36 <sipa> no idea which
243 2013-04-29 08:38:00 <syskk> ah ok
244 2013-04-29 08:38:25 <syskk> why are there so many private keys? are they all important?
245 2013-04-29 08:38:38 <SomeoneWeird> keypool?
246 2013-04-29 08:38:40 <syskk> according to blockchain.info those 2 private keys are enough to export my wallet
247 2013-04-29 08:38:41 <sipa> do you understand the purpose of the keypool?
248 2013-04-29 08:38:50 <syskk> no :O
249 2013-04-29 08:39:11 <sipa> bitcoind's wallet uses new addresses continuously for privacy (every change goes to a fresh address)
250 2013-04-29 08:39:24 <sipa> that means that at every transaction, your wallet backups would become outdated
251 2013-04-29 08:39:34 <sipa> so they contain 100 future keys already
252 2013-04-29 08:39:41 <sipa> and new ones are taken from that pool
253 2013-04-29 08:39:42 <syskk> ah ok just read the wiki
254 2013-04-29 08:40:02 <The_Fly> this relates to my question earlier about decentralising the wallets
255 2013-04-29 08:40:14 <The_Fly> now that i have 0mq running i'd like to give it a shot, somehow
256 2013-04-29 08:40:36 <The_Fly> (signing transactions and passing to the daemon/p2p/database process)
257 2013-04-29 08:41:01 <syskk> sipa: how do i find out which keys are from the key pool and which are not?
258 2013-04-29 08:41:03 <The_Fly> i think the stumbling block last time i raised it was how to form the transaction correctly
259 2013-04-29 08:41:44 <The_Fly> which requires you to store locally (with respect to the wallet) some information about the blockchain
260 2013-04-29 08:41:45 <sipa> syskk: public addresses will have a corresponding entry in mapAddressBook
261 2013-04-29 08:42:04 <sipa> The_Fly: not really - you just the need unspent outputs
262 2013-04-29 08:42:09 <syskk> sipa: ok thanks
263 2013-04-29 08:42:10 <sipa> which are maintained by the wallet
264 2013-04-29 08:43:38 <The_Fly> ok, and a keypool local to the wallet right?
265 2013-04-29 08:43:52 <sipa> yes?
266 2013-04-29 08:43:52 <The_Fly> which it can grow as necessary? (or not?)
267 2013-04-29 08:43:57 <sipa> yes, sure
268 2013-04-29 08:44:05 <sipa> but what are you trying to do?
269 2013-04-29 08:45:06 <The_Fly> multiple projects which use bitcoin, so want each to have a separate wallet on a separate box, talking to main bitcoind over 0mq
270 2013-04-29 08:45:33 <sipa> you're implementing a wallet from scratch?
271 2013-04-29 08:46:05 <The_Fly> i might have to
272 2013-04-29 08:46:28 <The_Fly> either that or use codeshark's multiple wallet support
273 2013-04-29 08:46:57 <The_Fly> which may be an option, but is less secure
274 2013-04-29 08:48:08 <The_Fly> having the private keys on very restricted machines (accessible only within the local network) seems desirable
275 2013-04-29 08:50:33 <The_Fly> at least the externalisation of transaction signing would be something i'd like to have
276 2013-04-29 08:51:16 <The_Fly> the main bitcoind process could still respond to wallet specific getbalance/etc queries, just move the signing
277 2013-04-29 08:51:55 <The_Fly> i assume you only need the public keys of your addresses and the blockchain to know a wallet's balance?
278 2013-04-29 08:52:02 <The_Fly> *calculate
279 2013-04-29 08:54:07 <sipa> yes
280 2013-04-29 08:54:27 <sipa> but you don't need the blockchain... just which outputs you can still spend
281 2013-04-29 08:54:46 <The_Fly> i dont need the blockchain to sign, just outputs, yes...
282 2013-04-29 08:54:52 <sipa> obviously you infer that information from the chain by monitoring new blocks, but once you have that, you don't need the chain anymore
283 2013-04-29 08:55:07 <sipa> arguably, you don't need to *have* the chain, you just need to observe it
284 2013-04-29 08:55:16 <The_Fly> and update state
285 2013-04-29 08:55:31 <The_Fly> add db records (where they pertain to your wallet(s))
286 2013-04-29 08:56:25 <The_Fly> i placed the BZmq_SendWalletTransaction within the same function that -walletnotify thread was forked
287 2013-04-29 08:57:12 <The_Fly> so i should see all events
288 2013-04-29 08:57:27 <The_Fly> but say i wanted to import a wallet...
289 2013-04-29 08:57:40 <DiabloD3> hey sipa
290 2013-04-29 08:57:41 <The_Fly> i'd have to scan the blockchain for all outputs
291 2013-04-29 08:57:48 <DiabloD3> whats the maximum number of sends in sendmany?
292 2013-04-29 08:58:33 <sipa> DiabloD3: no specific limit
293 2013-04-29 08:58:41 <sipa> The_Fly: yes
294 2013-04-29 08:59:03 <DiabloD3> sipa: so thousands would be legit?
295 2013-04-29 09:00:27 <sipa> DiabloD3: yes, but the transaction would become very large
296 2013-04-29 09:00:43 <DiabloD3> it'd still be cheaper than doing individual ones, right?
297 2013-04-29 09:07:42 <DiabloD3> sipa: also, how do I unlock the wallet from the rpc to do a tx?
298 2013-04-29 09:08:21 <DiabloD3> oh, walletpassphrase
299 2013-04-29 09:08:24 <DiabloD3> nm
300 2013-04-29 09:08:32 <DiabloD3> I was looking at help and didnt see it at first
301 2013-04-29 09:13:16 <The_Fly> (for now 0mq+multiwallet support will do for me, so im working on the merge for that also)
302 2013-04-29 09:13:44 <The_Fly> then will maybe have a try at building a lightweight wallet/transaction signer
303 2013-04-29 09:17:37 <The_Fly> and rely on a secure/offline wallet to minimise damage in case of intrusion
304 2013-04-29 09:18:06 <buZz> DONT trust ANY 'secure' online wallet!!!
305 2013-04-29 09:18:14 <The_Fly> thats why im doing this
306 2013-04-29 09:18:36 <The_Fly> will keep an offline paper wallet or truecrypt usb key
307 2013-04-29 09:18:42 <The_Fly> or both!
308 2013-04-29 09:18:50 <buZz> nice
309 2013-04-29 09:18:56 <The_Fly> :)
310 2013-04-29 09:19:24 <The_Fly> i think the 0mq+multiwallet will do nicely
311 2013-04-29 09:19:41 <The_Fly> but, im curious as to how much pentesting has been done with bitcoind
312 2013-04-29 09:20:06 <The_Fly> are there any remote code execution, buffer overflow, etc. exploits lurking in the code
313 2013-04-29 09:20:15 <jgm> Eek, which version of boost does bitcoin build with on the test server? Just tried a patch and got a failure due to a missing function which AFAIK has been in there for years...
314 2013-04-29 09:21:39 <The_Fly> i've 1.46 installed here
315 2013-04-29 09:22:13 <DiabloD3> damnit wtf
316 2013-04-29 09:22:18 <DiabloD3> I cannot get sendmany to work
317 2013-04-29 09:22:21 <The_Fly> anyone able to respond to query regarding exploits?
318 2013-04-29 09:22:24 <The_Fly> security etc.
319 2013-04-29 09:22:46 <DiabloD3> The_Fly: most likely not.
320 2013-04-29 09:23:02 <DiabloD3> bitcoind has been heavily tested
321 2013-04-29 09:23:06 <The_Fly> excellent
322 2013-04-29 09:23:13 <The_Fly> thanks
323 2013-04-29 09:23:44 <The_Fly> i'll merge then and submit a pull request for both 0mq and 0mq+multiwallet
324 2013-04-29 09:24:04 <The_Fly> the later needs a tiny bit of work to allow subscribing to individual wallets
325 2013-04-29 09:24:29 <The_Fly> hopefully a one line patch
326 2013-04-29 09:25:38 <alaricsp> In general, only trust open-source wallets, and even then guardedly so
327 2013-04-29 09:25:59 <The_Fly> seems multiwallet is missing a createwallet RPC func
328 2013-04-29 09:26:12 <alaricsp> In theory, the bitcoind devs could try and sneak in a backdoor that, when you enter your wallet passphrase, sends them all your BTC; but being open source, they'd need to be sneakier in doing so.
329 2013-04-29 09:26:13 <The_Fly> im not sure how i'd create a new "empty" wallet.dat
330 2013-04-29 09:26:34 <jgm> Boost 1.46 is the earliest version which defaults to filesystem version being 3, which is what the patch needs. Guess I'll have to ifdef it then, but we must be running a pretty old version of boost
331 2013-04-29 09:26:43 <alaricsp> Prepackaged binaries from distros could have extra patches added beyond what's in the source, too.
332 2013-04-29 09:26:49 <DiabloD3> goddamnit, why the fuck is sendmany so broken
333 2013-04-29 09:26:50 <sipa> jgm: we need to support fs v2 for now
334 2013-04-29 09:26:55 <DiabloD3> no matter what syntax I give it, it bitches
335 2013-04-29 09:26:55 <The_Fly> alaricsp: im building from source
336 2013-04-29 09:27:02 <DiabloD3> error: value is type int, expected obj
337 2013-04-29 09:27:06 <DiabloD3> WHAT DO YOU WANT FROM ME
338 2013-04-29 09:27:17 <sipa> DiabloD3: what do you send?
339 2013-04-29 09:27:18 <The_Fly> it wants to break you
340 2013-04-29 09:27:21 <sivu> alaricsp: so you always go through the whole code of an open source project before building and running the software?
341 2013-04-29 09:27:32 <The_Fly> i suppose you dont have to
342 2013-04-29 09:27:50 <The_Fly> on the assumption that enough eyes are looking to spot something
343 2013-04-29 09:27:59 <DiabloD3> sipa: Ive tried sendmany address:btc, Ive tried sendmany {address:btc}, Ive tried sendmany {"address":btc}
344 2013-04-29 09:28:01 <DiabloD3> sipa: all same error
345 2013-04-29 09:28:03 <alaricsp> sivu: Nope. I just hope that somebody else would have spotted something :-)
346 2013-04-29 09:28:03 <The_Fly> and that the main devs are also trustworthy
347 2013-04-29 09:28:09 <alaricsp> sivu: I'm not saying it would be practical to do so, for me
348 2013-04-29 09:28:31 <sipa> DiabloD3: what's your exact command line (or RPC object you send)?
349 2013-04-29 09:28:34 <alaricsp> sivu: But perhaps a bank or other large organisation using bitcoin might have their own trusted builds, and inspect every patch they pull down from the public repo.
350 2013-04-29 09:28:40 <jgm> sipa: fair enough, will ifdef it and resubmit the patch
351 2013-04-29 09:28:43 <The_Fly> surely some distro package repo has been hacked in the past to spread malicious binaries
352 2013-04-29 09:28:46 <DiabloD3> sipa: bitcoind sendmany
353 2013-04-29 09:28:51 <The_Fly> it could happy with bitcoind
354 2013-04-29 09:28:57 <sipa> DiabloD3: you're not answering my question
355 2013-04-29 09:29:02 <jgm> DiabloD3: Try '{"address1":btc,"address2":btc}'
356 2013-04-29 09:29:07 <DiabloD3> sipa: I dont understand your question then
357 2013-04-29 09:29:18 <DiabloD3> sipa: I wrote it on the command line verbatim
358 2013-04-29 09:29:32 <DiabloD3> jgm: with the single quotes?
359 2013-04-29 09:29:39 <jgm> yep
360 2013-04-29 09:29:50 <jgm> you need the single quotes to stop the shell from parsing it, and the double quotes to make it valid JSON
361 2013-04-29 09:30:06 <The_Fly> any way to create a blank wallet.dat with bitcoind?
362 2013-04-29 09:30:18 <sipa> DiabloD3: 'bitcoind sendmany' will not do anything (it will show you the help message), and '{"address":btc}' is not a full command
363 2013-04-29 09:30:36 <The_Fly> i suppose i could just copy from a seed wallet.dat
364 2013-04-29 09:30:40 <DiabloD3> sipa: er, I clearly said sendmany first.
365 2013-04-29 09:30:41 <sipa> DiabloD3: so i'll try to help you, but i'd like to see exactly what you're doing
366 2013-04-29 09:30:48 <DiabloD3> [07:26:45] <DiabloD3> sipa: Ive tried sendmany address:btc, Ive tried sendmany {address:btc}, Ive tried sendmany {"address":btc}
367 2013-04-29 09:30:52 <DiabloD3> thats the three I tried
368 2013-04-29 09:31:11 <DiabloD3> sendmany '{"address":btc}' just tells me the sendmany help as an error
369 2013-04-29 09:31:16 <sipa> DiabloD3: i'm NOT going to try to infer what you're doing from the bits and pieces you've given m
370 2013-04-29 09:31:29 <DiabloD3> sipa: I dont understand your question then, this is what Ive wrote verbatim
371 2013-04-29 09:31:55 <DiabloD3> that is the command I gave bitcoind
372 2013-04-29 09:32:46 <sipa> DiabloD3: ok, first of all, you need the account to debit
373 2013-04-29 09:33:00 <sipa> the command is sendmany <fromaccount> <amounts>
374 2013-04-29 09:33:03 <DiabloD3> er, why isnt that optional?
375 2013-04-29 09:33:11 <DiabloD3> I dont use accounts
376 2013-04-29 09:33:11 <sipa> is that relevant?
377 2013-04-29 09:33:21 <DiabloD3> yes, its relevant because sendtransaction doesnt require it
378 2013-04-29 09:33:25 <sipa> yes, it could be optional, but it clearly isn't
379 2013-04-29 09:33:27 <DiabloD3> which makes the API confusing
380 2013-04-29 09:33:36 <sipa> and the help message clearly indicates this
381 2013-04-29 09:33:57 <DiabloD3> if it clearly indicated this, I wouldnt be having this problem
382 2013-04-29 09:34:18 <sipa> do you want to bitch about the API and its documentation, or do you want to learn to use sendmany?
383 2013-04-29 09:34:27 <DiabloD3> OMFG.
384 2013-04-29 09:34:55 <DiabloD3> sendmany is useless to me
385 2013-04-29 09:35:07 <sipa> why?
386 2013-04-29 09:35:14 <DiabloD3> error: {"code":-8,"message":"Invalid parameter, duplicated address: address"}
387 2013-04-29 09:35:16 <buZz> http://static.seekingalpha.com/uploads/2012/4/18/saupload_corruptiondemotivator.jpg
388 2013-04-29 09:35:19 <buZz> :DDD
389 2013-04-29 09:35:40 <sipa> DiabloD3: known problem, and i'm in favor of fixing it
390 2013-04-29 09:35:48 <sipa> DiabloD3: but it requires changing the API unfortunately
391 2013-04-29 09:36:05 <sipa> (as JSON doesn't allow duplicate keys in objects)
392 2013-04-29 09:36:16 <DiabloD3> why is it a json array anyhow
393 2013-04-29 09:36:26 <sipa> it isn't
394 2013-04-29 09:36:31 <sipa> it's a json object
395 2013-04-29 09:36:35 <DiabloD3> whatever
396 2013-04-29 09:36:39 <DiabloD3> its one of those
397 2013-04-29 09:36:39 <sirdefaced> lol
398 2013-04-29 09:36:44 <DiabloD3> why cant it just be a list of text
399 2013-04-29 09:36:58 <sipa> it could be [[address,amount],[address,amount],...] for example, that would work
400 2013-04-29 09:37:32 <DiabloD3> yes, so break the api and make it useful
401 2013-04-29 09:38:00 <jgm> The JSON API is... nonstandard, shall we say
402 2013-04-29 09:38:09 <DiabloD3> sipa: okay so wait
403 2013-04-29 09:38:09 <sipa> yeah, it's a mess
404 2013-04-29 09:38:18 <DiabloD3> why is this giving me that error anyhow?
405 2013-04-29 09:38:27 <DiabloD3> bitcoind as an rpc client should be repeating what I say
406 2013-04-29 09:38:34 <DiabloD3> and the server should just accept it as is
407 2013-04-29 09:38:47 <jgm> DiabloD3: so did you get it working then? I just tried ./bitcoind --test --conf=bitcoin-testnet.conf sendmany source '{"mwyMcBpFifgJVvtP2natD12STwRX1px2bK":0.001,"mz1TenD3ZSyiEZtTCLRHhBRXe7KcTUahLx":0.001}' and had no problems
408 2013-04-29 09:38:59 <sipa> jgm: he wants to specify the same address twice
409 2013-04-29 09:39:00 <DiabloD3> jgm: yes, I got it working
410 2013-04-29 09:39:05 <DiabloD3> yes what sipa said
411 2013-04-29 09:39:14 <alaricsp> ACTION wonders why you'd want to do that
412 2013-04-29 09:39:22 <sipa> DiabloD3: the person who implemented it, thought that it was a bad idea to support duplicate addresses
413 2013-04-29 09:39:27 <DiabloD3> alaricsp: because of broken things like SD
414 2013-04-29 09:39:30 <alaricsp> Perhaps if you were building something like satoshidice?
415 2013-04-29 09:39:32 <alaricsp> Ah-hah
416 2013-04-29 09:40:00 <DiabloD3> so wait, why the hell does SD support sendmany
417 2013-04-29 09:40:02 <DiabloD3> if I cant sendmany
418 2013-04-29 09:40:07 <jgm> Yep no way of doing that as valid JSON given the current format. Why not merge the outputs prior to calling sendmany?
419 2013-04-29 09:40:18 <DiabloD3> jgm: because of SD
420 2013-04-29 09:41:27 <sipa> well, you can always use the raw transaction API :p
421 2013-04-29 09:41:30 <alaricsp> SD may be running a patched bitcoind
422 2013-04-29 09:41:37 <DiabloD3> alaricsp: that doesnt help me, now does it
423 2013-04-29 09:41:41 <DiabloD3> I cant _send_ them a sendmany
424 2013-04-29 09:41:41 <sipa> SD uses a heavily-patched BitcoinJ
425 2013-04-29 09:41:47 <jgm> Okay I get it. Although if you want to burn all your money it's friendlier to the network if you do it all with one big roll of the dice rather than lots of little ones.
426 2013-04-29 09:41:49 <alaricsp> DiabloD3: Sadly, no...
427 2013-04-29 09:42:01 <DiabloD3> sipa: meeeeehh, that confuses the hell out of me
428 2013-04-29 09:42:10 <sipa> DiabloD3: understandable
429 2013-04-29 09:42:26 <DiabloD3> all I want to do is make SD pay me lots of money
430 2013-04-29 09:42:33 <DiabloD3> because I think I can consistently win against it
431 2013-04-29 09:42:44 <sipa> then perhaps you need a course in statistics
432 2013-04-29 09:42:52 <DiabloD3> no, I think the people who run SD do
433 2013-04-29 09:43:12 <DiabloD3> Ive already been doing it, but tx fees are eating my lunch
434 2013-04-29 09:44:01 <jgm> Patch sendmany locally then. There's no reason that it isn't allowed in a raw transaction
435 2013-04-29 09:44:19 <jgm> I can do it if you cut me in on the profits :)
436 2013-04-29 09:44:34 <sipa> jgm: also on the losses?
437 2013-04-29 09:45:21 <jgm> Nah, I just want a cut of the winning rolls. The losing rolls are DiabloD3's problem
438 2013-04-29 09:45:35 <DiabloD3> hah.]
439 2013-04-29 09:47:56 <Scrat> sipa +1, wtf you on DiabloD3
440 2013-04-29 09:48:14 <DiabloD3> Scrat: what, its not unusual for people to get statistics wrong
441 2013-04-29 09:48:19 <DiabloD3> I think SD did.
442 2013-04-29 09:48:59 <Scrat> you can prove it wrong on my service :p
443 2013-04-29 09:49:08 <Scrat> s/it/me/
444 2013-04-29 09:49:09 <The_Fly> it doesn't look like it from his blog post about roll distribution
445 2013-04-29 09:50:36 <The_Fly> why not change to [{address:abc, ammount:123.0}, ...]
446 2013-04-29 09:50:45 <Scrat> sipa: how many sipa points do I get for not using the "from address" -> https://coinroll.it
447 2013-04-29 09:52:34 <DiabloD3> Scrat: you probably do things correctly
448 2013-04-29 09:52:41 <The_Fly> where is CodeShark
449 2013-04-29 09:52:42 <The_Fly> hm
450 2013-04-29 09:53:22 <sipa> Scrat: 42, obviously
451 2013-04-29 09:54:55 <The_Fly> hm, i have bugs in his multiwallet patch
452 2013-04-29 09:54:56 <drizztbsd> what is coinrool?
453 2013-04-29 09:55:10 <sipa> The_Fly: feel free to report them
454 2013-04-29 09:55:15 <The_Fly> CDB() : can't open database file wallet.dat, error -30974
455 2013-04-29 09:55:38 <The_Fly> sipa: create issue on main bitcoin githubs?
456 2013-04-29 09:55:55 <The_Fly> also the behaviour of loadwallet is not as described
457 2013-04-29 09:56:02 <The_Fly> does not create new wallet
458 2013-04-29 09:56:09 <sipa> The_Fly: no, on the pullreq page
459 2013-04-29 09:56:14 <The_Fly> ok ty
460 2013-04-29 09:56:31 <The_Fly> (ofc, sorry, this has not been merged yet)
461 2013-04-29 09:56:48 <The_Fly> ACTION is still waking up
462 2013-04-29 09:58:39 <The_Fly> i can unload wallets also but not load them, will try this on a blank ~/.bitcoind and see if i can recreate
463 2013-04-29 10:02:56 <The_Fly> ok, seems that nuking .bitcoind saved the day
464 2013-04-29 10:03:05 <sipa> still shouldn't happen
465 2013-04-29 10:03:13 <The_Fly> i know, but im hopping between branches
466 2013-04-29 10:03:25 <The_Fly> yes, you're right though...
467 2013-04-29 10:03:54 <The_Fly> im not sure what was going on there, all i have is my .bitcoin folder...
468 2013-04-29 10:04:02 <The_Fly> not much of a useful bug report
469 2013-04-29 10:04:29 <The_Fly> i think happened when i moved wallet.dat to another name, because i wanted bitcoind to create a new wallet for me
470 2013-04-29 10:08:10 <The_Fly> ok cool, ive isolated it to that case
471 2013-04-29 10:08:22 <The_Fly> bug only happens when bitcoind created a new wallet.dat on spinup
472 2013-04-29 10:15:31 <The_Fly> is there any convenient way to see if a transaction was failed to be confirmed at a certain depth?
473 2013-04-29 10:16:21 <The_Fly> as far as i can tell i'd have to check that confirmations did not increase upon recepit of a new best block
474 2013-04-29 10:17:08 <The_Fly> (or would #confirmations return to 0 when the chain is simplified?)
475 2013-04-29 10:27:31 <sipa> 'simplified' ?
476 2013-04-29 10:37:10 <The_Fly> sipa: merged, i meant
477 2013-04-29 10:38:36 <sipa> chains aren't merged
478 2013-04-29 10:41:05 <The_Fly> unless there's a fork
479 2013-04-29 10:42:49 <gmaxwell> No, not ever.
480 2013-04-29 10:43:44 <The_Fly> ah i see... "When a block becomes an orphan block, all of its valid transactions are re-added to the pool of queued transactions and will be included in another block"
481 2013-04-29 10:44:54 <The_Fly> im wanting to handle fraudulent transactions
482 2013-04-29 10:45:42 <sipa> transactions will return to confirmations=0 if reorganized
483 2013-04-29 10:46:56 <The_Fly> thanks
484 2013-04-29 10:47:47 <The_Fly> and this has not, or is unlikely to happen, once confirmations >= 6?
485 2013-04-29 10:51:48 <The_Fly> related question, if orphan blocks are still stored is it possible to find all attempts at double-spending in the history of bitcoin?
486 2013-04-29 10:56:56 <sipa> The_Fly: you're not guaranteed to see every stale block
487 2013-04-29 10:57:29 <sipa> as they are not relayed
488 2013-04-29 11:09:20 <The_Fly> i see, thanks, was just curious, would be interesting to get stats on how often people attempt this
489 2013-04-29 11:12:26 <sipa> attempt?
490 2013-04-29 11:12:34 <sipa> they happen inevitably
491 2013-04-29 11:12:57 <The_Fly> sure, orphan blocks do
492 2013-04-29 11:13:01 <The_Fly> but double-spending attempts
493 2013-04-29 11:13:36 <jouke> Can I make a copy of a running .8 node data-dir to bootstrap a new node?
494 2013-04-29 11:14:05 <sipa> jouke: in theory, yes - just copy chainstate and blocks
495 2013-04-29 11:14:31 <sipa> the chainstate cannot be newer than the blocks dir, but they may be out of sync otherwise
496 2013-04-29 11:16:57 <jouke> I would like to have a node with a full transaction index.
497 2013-04-29 11:17:46 <sipa> ok, so set txindex=1 in your config file
498 2013-04-29 11:18:21 <jouke> I don't want to shut down the running node :)
499 2013-04-29 11:18:32 <jouke> So thats why I want to make a copy.
500 2013-04-29 11:18:54 <sipa> just copy the blocks directory then, set txindex on the new node, and do a reindex
501 2013-04-29 11:19:09 <jouke> ok thanks
502 2013-04-29 11:19:13 <sipa> actually, only the blocks/blk* files
503 2013-04-29 11:19:18 <sipa> no need to copy the undo files or the index
504 2013-04-29 11:25:56 <shesek> using bitcoind, can I get transactions made to a multisig address that's in my wallet, but I don't have all the keys for?
505 2013-04-29 11:29:02 <sipa> not yet
506 2013-04-29 11:29:14 <shesek> is there any other tool that allows you to do that?
507 2013-04-29 11:30:13 <shesek> and is it planned for the official client?
508 2013-04-29 11:31:42 <kopy> fontas strategy http://www.flickr.com/photos/95336976@N04/8692748538/
509 2013-04-29 11:52:59 <joe_k> dude that's everybody's strategy
510 2013-04-29 12:20:58 <The_Fly> lol
511 2013-04-29 12:38:36 <sipa> opened my old wallet file again... why do people send me donations of 1 uBTC ?
512 2013-04-29 12:39:08 <Scrat> I do that to piss people off
513 2013-04-29 12:42:04 <tgs3> sipa: for all your great work for community, I will be happy to consolidate your wallet into new (not dust) coins, just send it, I can do it for free :}
514 2013-04-29 12:43:41 <sipa> tgs3: haha
515 2013-04-29 12:43:45 <Belxjander> Scrat: and how do I earn your ire to get 1uBTC out of you ? :P
516 2013-04-29 12:44:27 <tgs3> freewalletfixup dot seriousbusiness at bescammed dot gov dot ng
517 2013-04-29 12:44:46 <_dr> honest.ru
518 2013-04-29 12:44:54 <tgs3> freedom.us
519 2013-04-29 12:44:59 <Scrat> bestwallet.cn
520 2013-04-29 12:45:07 <_dr> serious.biz
521 2013-04-29 12:45:19 <Scrat> Belxjander: quite hard to pull that off, I'm very easy going
522 2013-04-29 12:46:43 <Sebo> Hi, do you know any bitcoind patch that would allow to add bitcoin addresses to be watched by the wallet? i.e. to add only the address but not the privete key?
523 2013-04-29 12:49:25 <travolta> Yes Sebo
524 2013-04-29 12:49:42 <travolta> look for CodeShark's git pull
525 2013-04-29 12:49:43 <Belxjander> Scrat: yeah... I noticed that in the other channel :P
526 2013-04-29 12:50:09 <travolta> his importaddress branch
527 2013-04-29 12:51:19 <travolta> https://github.com/bitcoin/bitcoin/pull/2121
528 2013-04-29 12:51:59 <travolta> is the one
529 2013-04-29 12:51:59 <travolta> that one scrat ^^
530 2013-04-29 12:52:12 <sipa> you mean Sebo i think
531 2013-04-29 12:52:21 <Sebo> travolta: I'll see
532 2013-04-29 12:52:52 <travolta> sebo: ok
533 2013-04-29 12:53:11 <travolta> I'm going to have lunch ttyl!
534 2013-04-29 13:13:28 <shesek> I'm trying to use a two-of-three multisign, and can't get it working. can someone take a look and tell me what I'm doing wrong? the exact steps I took are here: http://pastie.org/private/k8brhriyhosl56iyg1fphw
535 2013-04-29 13:14:16 <shesek> I'm on v0.8.1.0-g34d62a8-beta
536 2013-04-29 13:26:22 <slavik0329> Is there any way to get unspent outputs of an address you do not know using bitcoind RPC?
537 2013-04-29 13:28:17 <nsh> slavik0329, can you rephrase that / expand?
538 2013-04-29 13:30:31 <slavik0329> nsh: The same info i can get here from any address: http://blockchain.info/unspent?address=1BZ2a763GLLBkpKsi94BMkmRfKGY5qcMQ3
539 2013-04-29 13:30:39 <slavik0329> nsh: I want to be able to get this info myself
540 2013-04-29 13:31:38 <shesek> slavik0329, bitcoind only keeps track of addresses in your wallet that you have the private key for
541 2013-04-29 13:31:42 <shesek> so you can't
542 2013-04-29 13:31:50 <slavik0329> shesek: is there any other way?
543 2013-04-29 13:31:51 <shesek> but I understand that its planned
544 2013-04-29 13:32:48 <nsh> it's possible with a little bit of scripting, sl1982-Desktop
545 2013-04-29 13:32:53 <nsh> * slavik0329
546 2013-04-29 13:33:02 <nsh> see: https://bitcointalk.org/index.php?topic=117866.0
547 2013-04-29 13:39:46 <slavik0329> nsh: that still does not provide a method for getting unspent the unspent outputs of an address not owned by me
548 2013-04-29 13:41:27 <nsh> slavik0329, my understanding of that post was the script makes a table of all unspent outputs of transactions, regardless of address, but i only read diagonally
549 2013-04-29 13:41:40 <nsh> i don't feel qualified to advise further and defer to someone with better knowledge
550 2013-04-29 13:43:06 <sipa> slavik0329: see pull request 2121
551 2013-04-29 13:43:57 <nsh> ( https://github.com/bitcoin/bitcoin/pull/2121 )
552 2013-04-29 13:45:03 <slavik0329> sipa: oooo, very nice!
553 2013-04-29 13:47:23 <shesek> any idea about my multisign issue?
554 2013-04-29 13:47:45 <shesek> I really want to get it working
555 2013-04-29 13:51:01 <The_Fly> do you have to manually enter the merge messages e.g. "Merge pull request #2564 from Diapolo/walletview" when working on the repo?
556 2013-04-29 13:51:33 <sipa> no
557 2013-04-29 13:51:41 <The_Fly> oh that's something github do
558 2013-04-29 13:52:03 <The_Fly> right...
559 2013-04-29 13:52:52 <The_Fly> nice githubs
560 2013-04-29 13:58:27 <wumpus> it's something that git does, not so much github
561 2013-04-29 14:02:11 <ColinT> Is the UI and Bitcoin source all mashed together still?
562 2013-04-29 14:04:41 <wumpus> nope; the ui source is in src/qt, the core source in just src
563 2013-04-29 14:04:57 <ColinT> cool - this "json" folder - do I need that?
564 2013-04-29 14:05:03 <wumpus> yes
565 2013-04-29 14:05:12 <ColinT> (if I don't want the UI?)
566 2013-04-29 14:05:26 <wumpus> especially if you don't want the ui
567 2013-04-29 14:05:36 <ColinT> hehe
568 2013-04-29 14:05:39 <wumpus> I suppose you want some kind of entry point? :P
569 2013-04-29 14:06:07 <wumpus> if you strip the json stuff as well as the ui, you have an interesting node
570 2013-04-29 14:06:27 <nsh> headless bitcoin
571 2013-04-29 14:06:34 <sipa> wumpus: a block relay daemon
572 2013-04-29 14:06:40 <sipa> not completely useless
573 2013-04-29 14:06:42 <wumpus> block chain helper
574 2013-04-29 14:06:45 <ColinT> The bitcoin-qt client is fully useable from the command line correct?
575 2013-04-29 14:07:03 <sipa> it has the rpc server built-in
576 2013-04-29 14:07:33 <The_Fly> wumpus: yes but merge messages locally are not what i see in github, so github must detect that you merged a branch in a fork (when you merge with a branch from another remote)
577 2013-04-29 14:07:55 <sipa> github has lots of magic
578 2013-04-29 14:08:03 <sipa> and cats
579 2013-04-29 14:08:10 <The_Fly> that clever octopus
580 2013-04-29 14:08:38 <wumpus> The_Fly: it depends on the options that you provide; by default, in the case of a "fast forward" merge you don't get a merge message, but otherwise it will generate one
581 2013-04-29 14:09:00 <wumpus> and if you have conflicts you even get to edit it yourself :)
582 2013-04-29 14:09:41 <wumpus> github always generates a merge message, even on a fastforward/trivial merge
583 2013-04-29 14:10:02 <The_Fly> ok, im merging locally
584 2013-04-29 14:10:13 <The_Fly> and in the bitcoin repo i see all the merge messages for pull request merges
585 2013-04-29 14:10:44 <wumpus> --no-ff Create a merge commit even when the merge resolves as a fast-forward.
586 2013-04-29 14:10:46 <The_Fly> but here im just merging master into some outdated feature branches of other forks (e.g. fredan/0mq and codeshark/multiwallet)
587 2013-04-29 14:10:59 <The_Fly> i am using --no-ff at work
588 2013-04-29 14:11:11 <The_Fly> to force a merge message
589 2013-04-29 14:11:16 <wumpus> yes
590 2013-04-29 14:11:27 <The_Fly> confusion is that githubs merge message for pull requests is different than that of gits
591 2013-04-29 14:11:43 <The_Fly> but that makes sense because they'll be doing the merge
592 2013-04-29 14:11:56 <The_Fly> when done by the web interface
593 2013-04-29 14:12:13 <wumpus> indeed, the web interface includes the pull request #
594 2013-04-29 14:12:27 <The_Fly> in my case (fixing up these stale branches) am i best to merge them master, or merge master into them, or does it not matter
595 2013-04-29 14:13:49 <wumpus> that only depends on where you want the changes
596 2013-04-29 14:14:00 <The_Fly> i suppose it probably makes sense to merge master into, as i'll be submitting another pull request
597 2013-04-29 14:14:23 <wumpus> may be better to use rebase in that case
598 2013-04-29 14:14:40 <The_Fly> have been doing that so far
599 2013-04-29 14:14:53 <wumpus> it allows you to avoid a labyrinth of merge commits
600 2013-04-29 14:15:29 <wumpus> though it's a matter of personal taste, some people like it that way, but with bitcoin we prefer pull requests that don't contain merges themselves
601 2013-04-29 14:16:13 <The_Fly> and what about the case where master diverges from the pull request
602 2013-04-29 14:16:24 <The_Fly> (by a lot)
603 2013-04-29 14:16:53 <wumpus> then you get conflicts while rebasing, which you have to resolve before you can continue
604 2013-04-29 14:17:11 <The_Fly> sure, i know that lol
605 2013-04-29 14:17:34 <The_Fly> the conflicts in this case were minimal
606 2013-04-29 14:17:46 <The_Fly> but many files changed in master
607 2013-04-29 14:17:54 <The_Fly> only conflict was in init.cpp
608 2013-04-29 14:18:12 <The_Fly> (with 0mq)
609 2013-04-29 14:18:43 <The_Fly> however i can see codeshark's multiwallet having more conflicts
610 2013-04-29 14:18:50 <wumpus> ok
611 2013-04-29 14:19:08 <The_Fly> tbh when i tried rebase vs merge there was no difference
612 2013-04-29 14:19:17 <The_Fly> so im just wondering what the preference/convention is her
613 2013-04-29 14:19:19 <tonikt> hi guys. the private key - it cannot by any random 256-bit number, can it?
614 2013-04-29 14:19:36 <wumpus> yes you'll have exactly the same conflicts
615 2013-04-29 14:19:55 <The_Fly> but history will be different?
616 2013-04-29 14:20:01 <Diapolo> wumpus: any idea what was updated with our en-master file? I got 2 mails over the last 2 days, but no new strings were available to translate. tcatm is mentioned as last one who edited the file...
617 2013-04-29 14:20:22 <wumpus> yes, the history will be different, in the case of a rebase it's easier to preserve a linear-ish history
618 2013-04-29 14:20:33 <wumpus> Diapolo: no idea I got the mails too
619 2013-04-29 14:20:44 <The_Fly> wumpus: ok, im sold, i'll go with that
620 2013-04-29 14:20:47 <The_Fly> thanks
621 2013-04-29 14:21:30 <Diapolo> tcatm: any information what you updated in bitcoin_en.ts?
622 2013-04-29 14:31:18 <Diapolo> wumpus: seems bitcoin_se.ts is missing now on TX but we now have bitcoin_ar.ts ;)... didn't yet diff the en file
623 2013-04-29 14:31:50 <wumpus> hm
624 2013-04-29 14:33:16 <sipa> tonikt: almost any 256-bit number
625 2013-04-29 14:33:29 <Diapolo> wumpus: TX header: <?xml version="1.0" ?><!DOCTYPE TS><TS language="en" version="2.0"> Github header: <?xml version="1.0" encoding="utf-8"?><!DOCTYPE TS><TS version="2.0" language="en">
626 2013-04-29 14:33:33 <Diapolo> that is the only difference
627 2013-04-29 14:33:42 <Diapolo> but could aswell been modified by TX
628 2013-04-29 14:34:18 <sipa> tonikt: http://bitcoin.stackexchange.com/questions/1389/how-are-public-private-keys-in-an-address-created/1715#1715
629 2013-04-29 14:34:57 <tonikt> thx, @sipa
630 2013-04-29 14:35:26 <tonikt> I am writing a key generator and the last thing I need is to generate a key that would not be able to sign, later on
631 2013-04-29 14:36:19 <tonikt> .. so after I generate a key I do a single sign + verify with it. But I do not know if one successful sign+verify is enough to be sure that the key will always work fine
632 2013-04-29 14:41:51 <tonikt> .. anyway, now I know what range to check for - that should do it. :)
633 2013-04-29 14:43:11 <sipa> tonikt: why would it not always work fine?
634 2013-04-29 14:48:30 <topace> is there any way of knowing how often the keypool gets refilled? or does it generate a new address each time it uses one ?
635 2013-04-29 14:48:49 <gmaxwell> sipa: excellent find on the FILE_SHARE_READ
636 2013-04-29 14:48:54 <topace> basically i want to make sure my backups are frequent enough that i have all keeys in the keypool before they are used.
637 2013-04-29 14:49:08 <Scrat> topace: does exactly what you said. getinfo gives you the pool size
638 2013-04-29 14:49:39 <tonikt> sipa: ... yeah, I though there was something wrong with the value, but there was a bug in my verify function. :P
639 2013-04-29 14:50:23 <topace> keypoololdest is the timestamp of the oldest unused address in the keypool?
640 2013-04-29 14:50:46 <tonikt> it would indeed be very unlikely to generate a value form outside the range.
641 2013-04-29 14:54:03 <Diablo-D3> hey gmaxwell
642 2013-04-29 14:54:10 <Diablo-D3> #rust isnt being useful
643 2013-04-29 14:54:27 <Diablo-D3> can I do the "async pipe feeds dequeue and loop eats dequeue" pattern in rust?
644 2013-04-29 14:58:05 <nsh> ACTION wishes he knew what Diablo-D3 is talking about
645 2013-04-29 14:58:31 <helo> select loop?
646 2013-04-29 14:58:51 <Diablo-D3> nsh: its a common large scale programming technique to use async as much as possible and shove it into a dequeue on the other end
647 2013-04-29 14:59:07 <Diablo-D3> helo: basically, if you're going to code it that way
648 2013-04-29 14:59:14 <nsh> Diablo-D3, makes some sense
649 2013-04-29 14:59:48 <nsh> but why deny yourself the educational experience that is threading...
650 2013-04-29 14:59:49 <Diablo-D3> its how you're supposed to program concurrent stuff
651 2013-04-29 14:59:54 <Diablo-D3> nsh: HELL NO
652 2013-04-29 14:59:56 <Diablo-D3> HATE THREADS
653 2013-04-29 15:00:06 <nsh> i figured :)
654 2013-04-29 15:00:08 <Diablo-D3> hell, I gave the entire computer science community the benefit of the doubt
655 2013-04-29 15:00:15 <Diablo-D3> and tried, for 3 years, to make threads efficiently work
656 2013-04-29 15:00:18 <Diablo-D3> IT DOES NOT WORK
657 2013-04-29 15:00:33 <Diablo-D3> I researched every method that exists on how to synchronize things
658 2013-04-29 15:01:06 <nsh> everything old is new again. there are some cogent arguments i've read recently that agree with your position
659 2013-04-29 15:01:14 <grau> Diablo-D3 : try not syncronizing but use immutable
660 2013-04-29 15:01:28 <Diablo-D3> grau: I was already exploiting that actually
661 2013-04-29 15:01:44 <Diablo-D3> I ended up coding a large scale STM implementation that automatically retires attempts using longjmp
662 2013-04-29 15:01:53 <Diablo-D3> at that point I gave up and rejected threads
663 2013-04-29 15:02:08 <Diablo-D3> (best part is, it did read only transactions concurrently, so no locking bullshit)
664 2013-04-29 15:03:06 <The_Fly> oh, switch back from CodeShark/multiwallet and i get a "ERROR: VerifyDB() : *** block.ReadFromDisk failed at 69375, hash=00000000013e96e52b1f014c8e5eec24e419a80b5ab010494c4b6e10cf222e7c"
665 2013-04-29 15:03:47 <grau> My experience is not as bad with threads. In fact I would not know how to solve complex things without them.
666 2013-04-29 15:04:21 <The_Fly> salvagewallet?
667 2013-04-29 15:04:30 <The_Fly> i think i'll backup first...
668 2013-04-29 15:04:44 <Diablo-D3> grau: threads means explicit parallelism
669 2013-04-29 15:04:48 <Diablo-D3> which is usually NOT what you want
670 2013-04-29 15:04:57 <Diablo-D3> you want concurrent tasks that do not inherently block each other
671 2013-04-29 15:05:13 <Diablo-D3> what people do is try to emulate that on top of threads and dont realize that they're doing that
672 2013-04-29 15:05:17 <Diablo-D3> they think thats how its meant to be done
673 2013-04-29 15:06:19 <grau> threads are great for concurrent tasks.
674 2013-04-29 15:07:03 <grau> I used parallel calculation too a lot.
675 2013-04-29 15:07:09 <The_Fly> damn, -salvagewallet not helping
676 2013-04-29 15:07:13 <Diablo-D3> threads are _bad_ for concurrent tasks
677 2013-04-29 15:07:37 <Diablo-D3> they force you to have a unified address space
678 2013-04-29 15:07:37 <sipa> grau: don't try arguing with Diablo-D3, it leads nowhere
679 2013-04-29 15:07:47 <grau> Diablo-D3 : I recon now.
680 2013-04-29 15:07:58 <Diablo-D3> they also force you to use synchronization techniques that ultimately just slow your program down
681 2013-04-29 15:08:01 <grau> Diablo-D3 : How could I miss this.
682 2013-04-29 15:08:06 <Diablo-D3> theres also an upper limit of supporting threads
683 2013-04-29 15:08:24 <Diablo-D3> once 16 and 24 and 32 core CPUs become the norm, threads wont be able to survive on a lot of use cases
684 2013-04-29 15:08:55 <Diablo-D3> you don't have this problem if you use a consumer/producer type of pipeline done asynchronously
685 2013-04-29 15:09:22 <Diablo-D3> sipa: no one is really arguing here
686 2013-04-29 15:09:42 <Diablo-D3> grau: plus, look how NUMA machines are becoming common
687 2013-04-29 15:09:50 <Diablo-D3> threads will destroy performance on those
688 2013-04-29 15:09:50 <helo> having dealt with someone using threads as you describe, i like this kind of thinking
689 2013-04-29 15:10:57 <Diablo-D3> in C depending on your problem, it could be as simple as fork() and pipes
690 2013-04-29 15:24:35 <The_Fly> hm, reindex seemed to fix that problem
691 2013-04-29 15:24:54 <The_Fly> "ERROR: VerifyDB() : *** block.ReadFromDisk failed at 69375, hash=00000000013e96e52b1f014c8e5eec24e419a80b5ab010494c4b6e10cf222e7c"
692 2013-04-29 15:25:20 <The_Fly> appearing when switching between orgin/master and codeshark/multiwallet
693 2013-04-29 15:25:32 <The_Fly> some unclean shutdown somewhere perhaps?
694 2013-04-29 15:53:01 <helo> ACTION wonders how to tell the difference between unclean shutdown effects and disk corruption
695 2013-04-29 15:53:28 <helo> i think the aim is to make unclean shutdown effects nearly impossible, and blame everything on disk corruption
696 2013-04-29 15:54:29 <The_Fly> heh
697 2013-04-29 15:54:39 <The_Fly> idk, it fixed itself *shrugs*
698 2013-04-29 15:55:43 <helo> i'd fear for your hard disk 0_o
699 2013-04-29 15:56:38 <The_Fly> ran a scan recently, no read/write errors
700 2013-04-29 15:56:43 <The_Fly> bad sectors, etc.
701 2013-04-29 15:58:11 <astrolabe> I had an idea for do-it-yourself escrow with bitcoins. Would it be on topic to discuss it here?
702 2013-04-29 15:58:18 <helo> hmmm... seems to be a choice between non-deterministic hashing of block headers by bitcoind, or hard disk problems
703 2013-04-29 15:58:24 <Diablo-D3> half hour left on asicminer blade auction https://bitcointalk.org/index.php?topic=189248.660
704 2013-04-29 16:01:01 <nsh> astrolabe, might be better to write up clearly and post somewhere then link