1 2013-02-14 00:23:52 <BlueMatt> sipa: ping
2 2013-02-14 04:58:58 <becheesus> hi there
3 2013-02-14 04:59:45 <becheesus> i'm writing a web application that i want to be able to receive incoming bitcoin transactions to an address
4 2013-02-14 05:00:08 <becheesus> what is the best way to be notified on incoming new transactions
5 2013-02-14 05:02:41 <becheesus> trivially it can be done without bitcoind by using the blockchain websockets api
6 2013-02-14 05:02:50 <becheesus> but are these transactions even committed to a block?
7 2013-02-14 05:04:39 <becheesus> or should i be using bitcoind getreceivedbyaddress
8 2013-02-14 05:06:16 <weex> becheesus: if it doesn't get into a block, it's doesn't count longterm
9 2013-02-14 05:06:30 <weex> if you can handle it, i'd go with bitcoind
10 2013-02-14 05:06:50 <becheesus> yep i can handle it. i was just trying to do it as simply as possible really.
11 2013-02-14 05:07:10 <weex> sometimes simple means as a package for others to be able to setup easily and maintain
12 2013-02-14 05:07:25 <weex> for example i made a mybb plugin that uses the blockchain.info api because i wanted it to be easy to setup
13 2013-02-14 05:07:34 <becheesus> for me simple means i wont screw it up and accidently lose money :)
14 2013-02-14 05:08:12 <weex> test using testnet is the recommendation there
15 2013-02-14 05:09:02 <weex> transfer funds off server as soon as practical etc
16 2013-02-14 05:27:00 <Jouke> I installed rc1 on my vps with ubuntu. Now when I look in "top", it shows that rc1 has 46% memory. The blockchain is allready updated.
17 2013-02-14 05:27:37 <Jouke> The 7.2 instance I have running on the same vps uses 3.2% memory
18 2013-02-14 05:28:12 <gmaxwell> I have no idea what that percent memory is. What is the RES size?
19 2013-02-14 05:28:29 <Jouke> What is RES?
20 2013-02-14 05:28:55 <Jouke> Mem: 4049692k total
21 2013-02-14 05:29:00 <gmaxwell> The resident memory, it should be the sixth or seventh column in top.
22 2013-02-14 05:29:46 <Jouke> 1.9g
23 2013-02-14 05:30:32 <gmaxwell> thats interesting. Is there any unusual way that you're using bitcoin (and unusual RPC calls?) how many connections do you have?
24 2013-02-14 05:30:41 <gmaxwell> What is the VIRT? (usually the column before RES)
25 2013-02-14 05:30:51 <Jouke> 25 connections
26 2013-02-14 05:31:49 <Jouke> no rpc-calls (it is an empty wallet in front of a hot (7.1) wallet that is used a bit in the last hours)
27 2013-02-14 05:32:10 <gmaxwell> And you are running the rc1 binaries from bitcoin.org?
28 2013-02-14 05:32:16 <Jouke> VIRT: 3092m
29 2013-02-14 05:32:19 <Jouke> gmaxwell: yes
30 2013-02-14 05:32:27 <Jouke> the one gavin posted
31 2013-02-14 05:32:29 <gmaxwell> has this been restarted since it initially synced?
32 2013-02-14 05:32:34 <Jouke> no
33 2013-02-14 05:33:01 <gmaxwell> are you giving it any interesting (non-default) commandline options or bitcoin.conf parameters?
34 2013-02-14 05:33:57 <Jouke> in bitcoin.conf: rpcuser, rpcpassword,rpcport,maxconnections=1000 . No commandline options
35 2013-02-14 05:35:04 <gmaxwell> maxconnections is inadvisable, you'll likely crash when you run out of file descriptors at 1024 unless you've taken extra measures, or get memory corruption if select ends up with more than 1024.
36 2013-02-14 05:35:11 <gmaxwell> er maxconnections of 1000
37 2013-02-14 05:35:17 <gmaxwell> But I doubt thats at all related.
38 2013-02-14 05:35:29 <Jouke> yeah ok, but I usually only get 80 connections
39 2013-02-14 05:35:47 <Jouke> at the moment it has 24 connections
40 2013-02-14 05:35:51 <gmaxwell> I'm going to guess that if you restart it now the memory usage will go down and stay down... but that won't enlighten us as to the cause of the current bloat.
41 2013-02-14 05:36:05 <gmaxwell> What OS version is this exactly? ubuntu what?
42 2013-02-14 05:39:11 <Jouke> 12.4
43 2013-02-14 05:39:48 <Jouke> 64 bit
44 2013-02-14 05:40:30 <gmaxwell> and this was a fresh install of 0.8?
45 2013-02-14 05:40:44 <Jouke> yes, removed the whole .bitcoin dirc
46 2013-02-14 05:41:16 <gmaxwell> can you send me a copy of your debug.log? it won't have anything private in it except your IP and potentially your txnid if that node has made any. (gmaxwell@gmail.com)
47 2013-02-14 05:41:54 <gmaxwell> otherwise, for now I can just suggest restarting that node??? and I strongly expect the memory usage will be normal and remain that way. Could you do that then watch it and see that I'm correct?
48 2013-02-14 05:51:29 <Jouke> Ok, I restarted the client and sent you the debug.log. Can you confirm you received it?
49 2013-02-14 06:00:56 <gmaxwell> Jouke: thanks. Got it.
50 2013-02-14 07:15:17 <mariusursache> I noticed something strange in the last couple of days. My minings/transactions on testnet have a really high number of confirmations. yesterday I had +100 confirmations in 1-2 hours. Do you think there is ASIC client running on testnet?
51 2013-02-14 07:19:31 <sipa> BlueMatt: pong
52 2013-02-14 07:47:18 <Jouke> gmaxwell: As said I have the 8 rc1 client to connect to other nodes and a 7.2 client that only connects to that 8rc1 client. I see in my 7.2 wallet that a transaction was first seen (created in that client) 4 hours before blockchain.info sees it. I am wondering if maybe the 8rc1-client took its time before sending the transaction onwards?
53 2013-02-14 07:49:59 <Jouke> It was also only included in a block just three minutes after it was seen by blockchain.info, while I really expected it to be included way earlier.
54 2013-02-14 07:50:21 <gmaxwell> Jouke: if your 8 node wasn't current when it first saw it, it might have tossed it.
55 2013-02-14 07:50:46 <Jouke> Ah ok, so it waited before the 7.2 client rebroadcasted it?
56 2013-02-14 08:00:14 <sipa> the 0.8rc1 also isn't aware it is your transaction, so it will not actively broadcast
57 2013-02-14 08:02:44 <Jouke> Hmm, so I can expect this behaviour in the future?
58 2013-02-14 08:03:56 <Jouke> I was trying to fix this: https://bitcointalk.org/index.php?topic=135856.msg1447313#msg1447313
59 2013-02-14 08:06:03 <Jouke> I have reason to believe that bitcoinaddresses may be checked against our wallet.
60 2013-02-14 08:06:52 <sipa> yes, of course?
61 2013-02-14 08:07:57 <Jouke> I was under the impression that a client broadcasts any inconming transactions as fast as possible.
62 2013-02-14 08:08:11 <sipa> yes
63 2013-02-14 08:08:22 <sipa> but only once
64 2013-02-14 08:10:18 <Jouke> yes ok, but if I keep both clients synced up, I can expect my transactions that are only created with the 7.2 client to propagate the network quite fast.
65 2013-02-14 08:17:06 <sipa> sure
66 2013-02-14 11:16:51 <enferex> I am trying to read the blocks in blk00000.dat. How is teh data stored? I did not see it listed in the Protocol spec https://en.bitcoin.it/wiki/Protocol_specification. I am loking at the source code for the client, and I see it locates a magic number first 0xf9, 0xbe, 0xb4, 0xd9. Is this blk00000.dat format written up any where in teh docs?
67 2013-02-14 11:18:59 <sipa> 4 byte magic + 4 byte length descriptor + serialized block in network format
68 2013-02-14 11:19:06 <sipa> repeat
69 2013-02-14 11:19:10 <enferex> sipa: kick ass!
70 2013-02-14 11:19:12 <enferex> you rock
71 2013-02-14 11:24:13 <Luke-Jr> enferex: note the magic codes are documented at https://en.bitcoin.it/wiki/Protocol_specification#Message_structure
72 2013-02-14 11:24:38 <enferex> yup! thanks!
73 2013-02-14 11:34:14 <enferex> sipa: [magic][size][block header][block] ... repeat blockhead + block
74 2013-02-14 11:42:38 <ciphermonk> it's a berkeleydb database
75 2013-02-14 11:42:50 <ciphermonk> ( I think )
76 2013-02-14 11:46:33 <sipa> ciphermonk: it's not
77 2013-02-14 11:46:50 <sipa> ciphermonk: block files are just a concatenation of blocks
78 2013-02-14 11:47:04 <enferex> with block headers?
79 2013-02-14 11:47:04 <sipa> blkindex.dat and wallet.dat are berkeley databases
80 2013-02-14 11:47:08 <ciphermonk> the blk00000.dat is actually raw data?
81 2013-02-14 11:47:12 <sipa> ciphermonk: yes
82 2013-02-14 11:47:28 <sipa> enferex: block headers are considered a part of blocks
83 2013-02-14 11:47:37 <enferex> ok
84 2013-02-14 11:47:39 <sipa> block = block header + transactions
85 2013-02-14 11:47:51 <ciphermonk> oh cool, didn't know that, just assumed it was all berkeleydb
86 2013-02-14 11:48:13 <enferex> I was refering to the blockheaders that are the result of teh getheaders command.
87 2013-02-14 11:48:28 <sipa> enferex: yeah, those are the block headers + 0 byte
88 2013-02-14 11:48:48 <enferex> ok
89 2013-02-14 11:49:02 <enferex> thank you
90 2013-02-14 11:50:31 <sipa> and block files are indeed [magic][size][block][magic][size][block]..., with block consisting of [blockheader][numtx][transaction][transaction]...
91 2013-02-14 11:50:58 <enferex> ok cool
92 2013-02-14 11:51:13 <ciphermonk> so the bitcoin client opens up a filehandle and appends block data directly into it?
93 2013-02-14 11:52:11 <sipa> yes
94 2013-02-14 11:52:40 <ciphermonk> is there a risk of data corruption if the client crashes in the middle of writing a block?
95 2013-02-14 11:52:40 <sipa> as of 0.8, it pre-allocates and writes to specific positions rather than appending though
96 2013-02-14 11:52:56 <ciphermonk> sorry if I'm asking silly questions, I didn't read that part of the code
97 2013-02-14 11:53:00 <sipa> no, as the index entry for the block is only written after writing the block data to disk
98 2013-02-14 11:53:06 <sipa> (and syncing it)
99 2013-02-14 11:53:15 <sipa> except during IBD
100 2013-02-14 11:53:26 <ciphermonk> ok good
101 2013-02-14 13:15:06 <Jouke> I am trying some things with php jsonRPCclient, and my 7.2 client gives this error: "Incorrect response id (request id: 1, response id: )", while the same request om 8rc1 doesn't result in an error.
102 2013-02-14 13:17:59 <jaromil> guys, OT but askin in dev for comments from experienced ppl :^) I'm writing an academic article, non-technical, about the social progression of bitcoin and will have a draft ready in one or two days. I'd appreciate comments, criticism and suggestions from devs here before I send it for publishing. Is there people willing to receive it and comment on it? if yes, plz contact me in priv and send me your email address or stuff like that.
103 2013-02-14 13:30:27 <BlueMatt> sipa: what ever happened to the dnsseed hosted on the bitcoin foundation server?
104 2013-02-14 13:31:06 <slush_web> gavinandresen: sipa: I'm working on BIP for seed mnemonic (compatible with BIP32). Can I ask for new BIP number?
105 2013-02-14 13:32:01 <gavinandresen> slush_web : sure; gmaxwell is BIP editor
106 2013-02-14 13:32:15 <slush_web> gmaxwell: ^
107 2013-02-14 13:33:01 <gavinandresen> (oh, gmaxwell: I edited BIP 0001 to make you official Bitcoin BIP Numbering Authority a while ago)
108 2013-02-14 13:33:09 <slush_web> loi
109 2013-02-14 13:36:46 <sipa> slush_web: did you ever see https://bitcointalk.org/index.php?topic=102349.0 ?
110 2013-02-14 13:37:18 <sipa> slush_web: i never had the time to develop it further, but it resulted from a discussion with gmaxwell and etotheipi, though was decided not to be included in BIP32 itself
111 2013-02-14 13:37:49 <sipa> BlueMatt: never got around to set it up
112 2013-02-14 13:46:45 <slush_web> sipa: Basically we want to standardize "Electrum mnemonic" algorithm, so the way how to convert list of words to 128-bit binary. As I understand your thread, you're solving the next step of processing that binary.
113 2013-02-14 13:48:07 <slush_web> sipa: oh, I now I see that javascript demo. I should re-read the propsal once again ;-)
114 2013-02-14 13:52:03 <eckey> is anyone knowledgeable with bitcoinj?
115 2013-02-14 13:52:25 <slush_web_> sipa: your demo crashed my browser while trying 24bit strength :-)
116 2013-02-14 13:54:53 <sipa> slush_web_: it's a proposal for converting a textual string to a key, without the need for separately encoding a checksum or strength setting
117 2013-02-14 13:55:37 <Scrat> sipa: if I understand this correctly then hashing a cleartext with high strength will take a while
118 2013-02-14 13:56:04 <Scrat> I mean the seed generation, not the verification
119 2013-02-14 13:57:15 <ThomasV> sipa: nice idea
120 2013-02-14 13:57:53 <ThomasV> if I understand it correctly, you're killing two birds with one stone here
121 2013-02-14 13:58:25 <stick`> or rather killing all birds with a massive cannon :)
122 2013-02-14 13:58:30 <sipa> lol
123 2013-02-14 13:58:33 <TD> sipa has a massive cannon?
124 2013-02-14 13:58:38 <TD> my word
125 2013-02-14 13:59:01 <TD> hey there, are you the stick` doing the trezor?
126 2013-02-14 13:59:46 <slush_web_> sipa: ThomasV: Are all that key stretching techniques really necessary? Is bruteforce attack by generating of 128bit seed really possible attack vector?
127 2013-02-14 13:59:46 <stick`> TD: hey, yep
128 2013-02-14 13:59:46 <ThomasV> "Of course, I still hate brainwallets: I don't like the idea of users taking their Bitcoins to their own grave when they get hit by a bus, but so many people are so passionate about it, it's tough to ignore. "
129 2013-02-14 14:00:06 <TD> stick`: awesome. i'm so looking forward to buying my first.
130 2013-02-14 14:00:15 <stick`> :))
131 2013-02-14 14:00:29 <TD> ThomasV: or just forget ???.
132 2013-02-14 14:00:37 <sipa> slush_web_: if the seed really has 128 bits of entropy, it's not feasible
133 2013-02-14 14:00:37 <slush_web_> Even these 100.000 iterations of sha256 in Electrum's seed generation looks like an overkill for me
134 2013-02-14 14:00:46 <TD> brainwallets == sadness. at least until somebody finds a great way for people to memorize private keys. probably by turning them into a story.
135 2013-02-14 14:00:49 <ThomasV> I don't think the situation is different if a non-brainwallet user gets hit by a bus
136 2013-02-14 14:01:19 <TD> ThomasV: that's true if they're encrypted, i guess
137 2013-02-14 14:01:28 <sipa> slush_web_: but people may (and probably will) find ones to use a password to generate the seed
138 2013-02-14 14:01:52 <sipa> so the ambition should be to make the path of least resistance secure enough
139 2013-02-14 14:02:13 <slush_web_> sipa: when the seed is generated by the machine (which is recommended), then all these algorithms are unnecessary. And we cannot protect users from their own stupidity.
140 2013-02-14 14:02:51 <sipa> slush_web_: agree
141 2013-02-14 14:02:51 <TD> we can try :)
142 2013-02-14 14:03:04 <slush_web_> sipa: I'm concerned mostly because these algorithms are too slow for embedded systems...
143 2013-02-14 14:03:32 <Scrat> sipa: nvm. I just realized how stupid my remark was
144 2013-02-14 14:03:34 <eckey> TD: I'm in AbstractWalletEventListener().onCoinsSent(),trying to find the value. I'm iterating through the TransactionInput objects. For each I getOutpoint() I getHash(), which I'm assuming is the original transaction identifier. Do I then go back to that transaction to get the value?
145 2013-02-14 14:04:31 <TD> eckey: hmm? onCoinsSent() already gives you the transaction that moved coins out of your wallet, along with the pre and post move balances.
146 2013-02-14 14:04:37 <TD> eckey: what is it you're trying to do at a higher level?
147 2013-02-14 14:05:21 <ThomasV> slush_web_: the key stretching in Electrum is probably useless for users who follow the normal path and let the machine generate the seed. but I thought it would not hurt to add it
148 2013-02-14 14:06:34 <sipa> i still haven't heard anything from hal regarding bip32
149 2013-02-14 14:06:41 <TD> sipa: you know hal is sick, right?
150 2013-02-14 14:06:45 <sipa> yes
151 2013-02-14 14:06:53 <TD> i worry that he may not be in a good state to respond :(
152 2013-02-14 14:06:59 <ThomasV> slush_web_: note that I also implemented the current version of BIP32, although it is not activated for the moment
153 2013-02-14 14:07:01 <TD> sipa: did you ever talk to emilia?
154 2013-02-14 14:07:17 <sipa> TD: that's why i don't feel like mailing him again
155 2013-02-14 14:07:21 <sipa> TD: no, not yet
156 2013-02-14 14:08:03 <TD> it's definitely worth talking to her and adam langley
157 2013-02-14 14:08:13 <TD> CC me on the thread if/when you do so
158 2013-02-14 14:08:35 <slush_web_> sipa: ThomasV: Current draft of BIP32 is supposed to be final?
159 2013-02-14 14:09:18 <ThomasV> no, it is not final afaik
160 2013-02-14 14:09:37 <ThomasV> they are waiting for feedback from experts
161 2013-02-14 14:09:47 <slush_web_> ThomasV: hal?
162 2013-02-14 14:09:59 <sipa> we should ask some other people too
163 2013-02-14 14:09:59 <ThomasV> idk, ask sipa
164 2013-02-14 14:10:34 <TD> i think asking emilia and adam is enough
165 2013-02-14 14:10:46 <TD> heck, email koblitz himself, why not? :)
166 2013-02-14 14:11:05 <eckey> TD: I'm trying to manage "sub" wallets in the main wallet. I need to follow coins received and coins spent on an individual BTC address basis.
167 2013-02-14 14:12:19 <TD> eckey: so you're trying to handle transactions that could potentially send/receive from multiple sub-wallets in one transaction?
168 2013-02-14 14:12:35 <TD> i know we discussed this before, but what's the reason for not just using multiple Wallet objects again?
169 2013-02-14 14:12:36 <eckey> TD yes
170 2013-02-14 14:12:38 <TD> i forgot
171 2013-02-14 14:13:46 <TD> well, in SPV mode you can't actually know the value of the transactions inputs individually. you don't have that data. in full verification mode, i don't remember if you have access at that point, or if the connected outputs may have already been pruned away. it's a bit dangerous to rely on the ordering there, as it's not guaranteed.
172 2013-02-14 14:15:35 <Jouke> I am testing a server with multiple wallets at the moment :)
173 2013-02-14 14:15:36 <TD> sipa: why not do it now? more input can't possibly hurt
174 2013-02-14 14:16:47 <Jouke> TD: maybe you mistake ekey with me?
175 2013-02-14 14:17:07 <TD> Jouke: no, i was talking to eckey.
176 2013-02-14 14:17:23 <TD> i think he previously explained why he wants to do this sub-wallet thing, but i lost track of it and now can't remember.
177 2013-02-14 14:17:34 <TD> some missing feature in the library, no doubt
178 2013-02-14 14:17:35 <Jouke> (I posted on the bitcoinj mailing list about that)
179 2013-02-14 14:17:47 <Jouke> a while back
180 2013-02-14 14:17:50 <TD> ah, ok
181 2013-02-14 14:18:02 <TD> it's a bit hard for me to keep track of all the users and their IRC nicks these days, which i suppose is a good problem to have :)
182 2013-02-14 14:18:32 <TD> i wish i had time to bring the GCJ/CNI branch up to speed. bitcoinj is turning into quite a widely used API, and [objective] C++ access would go a long way to completing that
183 2013-02-14 15:31:28 <RedNifre> Hey there!
184 2013-02-14 15:33:09 <RedNifre> I haven't looked into the details of the default bitcoin client, but is it possible to run it entirely from the command line? I'm asking because I wonder if it's feasable to have it run on a server (updating the block chain all the time) and simply use ssh from my smartphone whenever I want to deal with bitcoins.
185 2013-02-14 15:33:31 <RedNifre> (Since I don't want to download the block chain on my smartphone and using third party servers kinda defeats the purpose of bitcoin)
186 2013-02-14 15:34:25 <andytoshi> RedNifre: sure, just run bitcoind
187 2013-02-14 15:36:22 <andytoshi> if you download the source tar.gz, or grab the git repo, then compile like normal, that's what you'll wind up with
188 2013-02-14 15:36:23 <RedNifre> THanks, I'll look into that.
189 2013-02-14 15:46:41 <muhoo> TD: pin
190 2013-02-14 15:46:45 <muhoo> heh ping
191 2013-02-14 15:47:30 <TD> hi
192 2013-02-14 15:52:14 <muhoo> TD: hi, i have let my bitoinj project be dormant for a while, but want to pick it up again. i was thinking of subclassing Wallet and using a DB back-end
193 2013-02-14 15:52:25 <muhoo> this is for a store
194 2013-02-14 15:53:14 <muhoo> and i can't see duplicating so much of the functionality of a wallet in callbacks etc, after thinking about it, i realized that the store pretty much is a wallet, and all the wallet callbacks are more or less what i need
195 2013-02-14 15:53:29 <muhoo> TD: my question then would be, would you consider this to be .... evil?
196 2013-02-14 15:58:52 <TD> muhoo: it's open source, do what you want
197 2013-02-14 15:59:07 <TD> muhoo: i never figured out the right way to make wallet support database backends. i think it'll require some pretty fundamental refactorings
198 2013-02-14 15:59:13 <muhoo> heh, that's a very diplomatic answer ;)
199 2013-02-14 15:59:36 <TD> muhoo: the slower but better way would be to figure out a sensible design that makes both file and db backends work cleanly and then submit a series of patches for review that does it
200 2013-02-14 15:59:54 <muhoo> i'm hoping to avoid touching any other class. that may be wishful thinking.
201 2013-02-14 16:00:05 <TD> subclassing the wallet might do what you want, but you'd have to copy/paste some code anyway and then when you rebase you're likely to end up with weird bugs and data corruptions caused by not keeping up. so i'd just fork the library and apply your changes on top using git
202 2013-02-14 16:00:13 <TD> that way you know it's always merged properly
203 2013-02-14 16:00:47 <muhoo> i'd be OK with doing that, but forks make me sad
204 2013-02-14 16:01:08 <TD> so find a way to merge your fork back upstream :)
205 2013-02-14 16:01:38 <TD> you can't really do this without forking. see what satoshidice had to do
206 2013-02-14 16:01:43 <TD> the changes required are just too fundamental
207 2013-02-14 16:01:58 <TD> the existing wallet just maintains in-memory data structures. persistence is literally "read those structures and write them to a file"
208 2013-02-14 16:02:16 <TD> for a database you need to rebuild the wallet to be based on sending transactions to a separated wallet store.
209 2013-02-14 16:02:22 <muhoo> hmm. right it gets protobuffed, IIRC
210 2013-02-14 16:02:45 <TD> right. that store can then maintain the in-memory transactions and save to a file, to match the old code, or it can convert the transactions into SQL transactions or whatever
211 2013-02-14 16:03:37 <muhoo> and then it needs to generate transactions too, in order to send the coins off-server to the merchants
212 2013-02-14 16:03:55 <muhoo> so at that point, i realized what i had was a wallet
213 2013-02-14 16:04:26 <muhoo> i'll have to play with it some more then. i really would like to avoid having to maintain a fork.
214 2013-02-14 16:05:38 <muhoo> TD: thanks for writing bitcoinj, and for your help
215 2013-02-14 16:06:20 <TD> np
216 2013-02-14 16:35:52 <jgarzik> ASICMINER is online, it seems
217 2013-02-14 16:55:20 <muhoo> the 800GH one? yikes.
218 2013-02-14 16:56:38 <TD> muhoo: 1.5THash
219 2013-02-14 16:56:48 <TD> muhoo: look at btcguild.com and click hall of fame
220 2013-02-14 16:56:50 <muhoo> jeezus fuck
221 2013-02-14 16:57:01 <TD> ACTION shrugs
222 2013-02-14 16:57:12 <TD> we're still a long way from the days when ArtForz had a third of the network power
223 2013-02-14 16:58:47 <HM> ACTION pulls the shoebox from under his bed and resumes tinkering with his quantum supercomputer
224 2013-02-14 17:10:09 <TD> gavinandresen: so you're going to be ticking more todo list items off payment protocol in the next couple of weeks?
225 2013-02-14 17:19:58 <Tatsuya> hey - a bit of a noob, here sorry. Is it possible / can someone explain to me how to connect to testnet3 using bitcoin-qt from the command line? I'm having a hard time finding documentation on this
226 2013-02-14 17:22:00 <ProfMac> Sometimes I wonder if many log files in a unix environment could be moved to mySQL. In the bitcoin context, that might mean moving the debug.log entries that contain the line 'version message' to a mySQL intereaction instead. What are some of the reasons to avoid doing this?
227 2013-02-14 17:22:58 <tcatm> ProfMac: You might want to look at systemd's journal.
228 2013-02-14 17:25:32 <ProfMac> tcatm, ok. Google (bitcoin systemd journa) didn't help much. Can you give me a pointer?
229 2013-02-14 17:26:05 <tcatm> It's a syslog replacement with some nice features.
230 2013-02-14 17:29:19 <gmaxwell> Tatsuya: -testnet=1
231 2013-02-14 17:31:40 <Tatsuya> aha, thanks gmaxwell. For future reference, is that flag documented anywhere?
232 2013-02-14 17:33:08 <Scrat> Tatsuya: https://en.bitcoin.it/wiki/Running_Bitcoin#Command-line_arguments
233 2013-02-14 17:34:01 <Tatsuya> oh, thanks Scrat. I see, I was looking at the API calls list
234 2013-02-14 17:46:15 <muhoo> wow, satoshidice is like the farmville of bitcoin
235 2013-02-14 17:50:33 <muhoo> tcatm: actually, systemd is more like a replacement for init
236 2013-02-14 17:51:30 <muhoo> if all ProfMac wants is to keep a database of logs, switching from sysv init to systemd might be a bit.... disruptive
237 2013-02-14 17:52:26 <sipa> muhoo: how do you mean (re: farmville of bitcoin)?
238 2013-02-14 17:53:43 <muhoo> sipa: meaning most of the traffic appears to be that :-)
239 2013-02-14 17:54:03 <helo> causes a massive spike in popularity, followed by exodus and widespread annoyance
240 2013-02-14 17:54:19 <muhoo> i only log into FB every few months or so, and almost all of what i see is farmville or mafia wars or the like
241 2013-02-14 17:59:18 <sipa> muhoo: either i hace a different group of friends, or i happen to have blocked most of that stuff, but i haven't seen any farmville stuff for many months i think
242 2013-02-14 17:59:37 <sipa> or maybe i've just inconsciously learned to ignore it
243 2013-02-14 17:59:39 <muhoo> heh, maybe it's been that long since i logged in.
244 2013-02-14 18:00:10 <muhoo> and, my friends don't do that shit. my facebook "friends"-- most of whom are people i haven't seen or talked to in 20-30 years, do.
245 2013-02-14 18:00:26 <BCB> what is the cmd to check if an bitcoin address exists in my wallet?
246 2013-02-14 18:00:59 <muhoo> BCB: from memory, listaddressesbyaccount?
247 2013-02-14 18:01:12 <muhoo> or getaddressesbyacccount icanneverrememberthosenames
248 2013-02-14 18:01:14 <BCB> muhoo: what if i don't know the account
249 2013-02-14 18:01:24 <muhoo> i iterate through all the accounts
250 2013-02-14 18:01:32 <BCB> muhoo: that's annoying
251 2013-02-14 18:01:41 <muhoo> well, i don't iterate, i map/reduce, but same result :-)
252 2013-02-14 18:02:35 <BCB> mhuoo: how
253 2013-02-14 18:05:11 <BCB> muhoo: how do I iterate on the cmd line
254 2013-02-14 18:05:43 <muhoo> write a bash script. or, use some other language, and access the JSON API that way (which is what i do)
255 2013-02-14 18:06:42 <sipa> BCB: validateaddress
256 2013-02-14 18:06:48 <helo> woah. reading satoshi dice use case in https://gist.github.com/gavinandresen/4120476 indicates massive awesomeness in the payment protocol :D
257 2013-02-14 18:07:44 <BCB> sipa looking for an address that exists in my wallet
258 2013-02-14 18:08:27 <BCB> muhoo: I'm using JRPC to access with which i'm familar. I'm looking to do this down and dirty on the cmd line
259 2013-02-14 18:08:27 <sipa> BCB: i don't think you can list all addresses in a wallet
260 2013-02-14 18:08:35 <BCB> sipa: boo
261 2013-02-14 18:08:44 <sipa> you can check whether a given address is in it though
262 2013-02-14 18:08:51 <BCB> how do I list ALL of my transactions regardles of account
263 2013-02-14 18:08:53 <BCB> ??
264 2013-02-14 18:08:59 <BCB> sipa: how
265 2013-02-14 18:09:08 <helo> listtransactions
266 2013-02-14 18:09:12 <sipa> listtransactions will list all transactions
267 2013-02-14 18:09:34 <sipa> validateaddress will tell you (among other things) whether an address is in the wallet
268 2013-02-14 18:09:39 <helo> ACTION commends the person that came up with that name
269 2013-02-14 18:09:52 <muhoo> BCB: https://en.bitcoin.it/wiki/API_reference_%28JSON-RPC%29
270 2013-02-14 18:09:54 <sipa> satoshi, i suppose
271 2013-02-14 18:10:00 <muhoo> BCB: https://en.bitcoin.it/wiki/Elis-API
272 2013-02-14 18:10:08 <helo> s/person/extraterrestrial/
273 2013-02-14 18:10:11 <muhoo> BCB: https://en.bitcoin.it/wiki/Original_Bitcoin_client/API_Calls_list
274 2013-02-14 18:10:44 <muhoo> thenamingoffunctionsinthebitcoinjsonapiisunfortunate
275 2013-02-14 18:10:49 <BCB> sipa but only 100
276 2013-02-14 18:10:52 <BCB> I have thousands
277 2013-02-14 18:11:00 <sipa> use a higher limit
278 2013-02-14 18:11:24 <BCB> sipa can you post and example. I can't get it o return anything
279 2013-02-14 18:11:42 <BCB> muhoo: ihavetosaythatiagreeiwthyouverymuch
280 2013-02-14 18:11:50 <andytoshi> how hard would it be to patch in a "list all addresses" API call
281 2013-02-14 18:12:06 <andytoshi> maybe we should have a patch for developers who need it
282 2013-02-14 18:12:18 <sipa> andytoshi: very easy
283 2013-02-14 18:12:54 <gmaxwell> andytoshi: do you mean something that would expose the keypool? that would be bad because it would instantly invalidate backups???
284 2013-02-14 18:12:56 <sipa> bcb: ./bitcoind help listtransactions
285 2013-02-14 18:13:10 <andytoshi> gmaxwell: that's right, we wouldn't put it in the main tree
286 2013-02-14 18:13:24 <andytoshi> just give a link for "i'm trying to do something awful with my own code" people
287 2013-02-14 18:13:29 <sipa> gmaxwell: not being able to list used change address is a shortcoming right now though
288 2013-02-14 18:13:34 <muhoo> why would it invalidate backups?
289 2013-02-14 18:13:53 <andytoshi> muhoo: because whenever you do pretty-much anything, a new address is generated
290 2013-02-14 18:13:57 <sipa> muhoo: not being able to know future addresses is essential for backups
291 2013-02-14 18:14:18 <sipa> muhoo: otherwise they may be used without them being marked as such
292 2013-02-14 18:15:06 <gmaxwell> sipa: Agreed on change addresses.
293 2013-02-14 18:18:04 <gavinandresen> andytoshi : listaddressgroupings might give you what you are looking for
294 2013-02-14 18:18:12 <gavinandresen> ??? depending on what you're looking for
295 2013-02-14 18:18:40 <BCB> sipa: listtransactions asks for an account - I want all transactions in all accounts
296 2013-02-14 18:19:08 <gavinandresen> listtransactions '*' 99999
297 2013-02-14 18:20:18 <discrete> ;;ticker
298 2013-02-14 18:20:19 <gribble> BTCUSD ticker | Best bid: 26.46986, Best ask: 26.46999, Bid-ask spread: 0.00013, Last trade: 26.46986, 24 hour volume: 143560.16996625, 24 hour low: 21.72100, 24 hour high: 26.67000, 24 hour vwap: 25.04641
299 2013-02-14 18:20:25 <BCB> gavinandresen: you are a genius
300 2013-02-14 18:20:32 <BCB> I've been looking for that for months!
301 2013-02-14 18:20:35 <Luke-Jr> lol
302 2013-02-14 18:20:47 <gmaxwell> ... but thats how you're supposed to use it.
303 2013-02-14 18:20:50 <sipa> ;;genrate 2000000
304 2013-02-14 18:20:51 <BCB> That's why the pay him the big bucks
305 2013-02-14 18:20:51 <gribble> The expected generation output, at 2000000.0 Mhps, given difficulty of 3275464.58657, is 307.074922131 BTC per day and 12.7947884221 BTC per hour.
306 2013-02-14 18:20:59 <BCB> and he's a nice guy to boot
307 2013-02-14 18:21:06 <gmaxwell> ACTION ponders what docs need to be updated.
308 2013-02-14 18:22:06 <BCB> gavinandresen: it works! so simple
309 2013-02-14 18:22:22 <gavinandresen> Easy When You Know How(???)
310 2013-02-14 18:23:23 <tigger0> yo what up
311 2013-02-14 18:23:24 <gavinandresen> ACTION wonders if Colloquy knows about copyright, too: ?? <-- ohh, shiny!
312 2013-02-14 18:23:25 <tigger0> when my friend runs dumpprivkey <address> it returns
313 2013-02-14 18:23:31 <Happzz> where do i complain about the inflation?
314 2013-02-14 18:23:31 <tigger0> Safe mode: Warning: Displayed transactions may not be correct! You may need to upgrade, or other nodes may need to upgrade. (code -2)
315 2013-02-14 18:23:37 <tigger0> the address is in his client
316 2013-02-14 18:23:41 <tigger0> he's on 0.8 rc1
317 2013-02-14 18:25:37 <sipa> hmm, that's the first case of a safe mode triggered in 0.8 i know of
318 2013-02-14 18:25:41 <gavinandresen> hmm??? dumpprivkey should be allowed in safe mode, I think. He's in safe mode because his idea of the blockchain doesn't match the rest of the network's for some reason
319 2013-02-14 18:26:36 <gavinandresen> tigger0: workaround: re-run bitcoind or bitcoin-qt with the -disablesafemode switch
320 2013-02-14 18:26:37 <tigger0> ok
321 2013-02-14 18:27:34 <gavinandresen> tigger0: ??? but he shouldn't send any coins until he fixes whatever problem is putting him into safe mode
322 2013-02-14 18:27:50 <tigger0> ok
323 2013-02-14 18:28:51 <andytoshi> gavinandresen: if i post a patch to enable dumpprivkey in safe mode, will that be pulled? or is it probably not worth the trouble?
324 2013-02-14 18:29:31 <gavinandresen> andytoshi: it'd probably be pulled, it'd definitely be safe and I can't think of a reason to disallow it.
325 2013-02-14 18:30:01 <gmaxwell> yea, that should be allowed: dumping out private keys might be how you're going to recover from a totally goofed client
326 2013-02-14 18:30:10 <Luke-Jr> might make sense to go over all the other RPCs added in the last year or two as well
327 2013-02-14 18:30:32 <Luke-Jr> might be wise to fix the deadlock on GBT too
328 2013-02-14 18:30:44 <andytoshi> coolio, i'll post an initial pull request to give somewhere to discuss other rpcs
329 2013-02-14 18:30:58 <Luke-Jr> (IIRC, we drop into safe mode if no new block in <N> minutes, which then disables GBT making finding blocks impossible)
330 2013-02-14 18:31:28 <gavinandresen> Luke-Jr: is there an issue on that?
331 2013-02-14 18:32:15 <gmaxwell> 0_o
332 2013-02-14 18:33:21 <Luke-Jr> gavinandresen: not sure; it probably needs discussion, since -disablesafemode might be more appropriate for dealing with it should it ever occur
333 2013-02-14 18:33:35 <tigger0> ok works with -disablesafemode thanks
334 2013-02-14 18:34:39 <Luke-Jr> gavinandresen: I only know about it because I ran into it in testnet-in-a-box - IIRC, the chance of it happening on mainnet is extremely unlikely, and it may very well be better to alert miners to a problem if it seems to
335 2013-02-14 18:34:56 <Luke-Jr> gotta pick up kids from school tho, bbiab
336 2013-02-14 18:35:23 <Guest9274> hello
337 2013-02-14 18:35:52 <Guest9274> Luke-Jr can you direct me to a link where i can find the code to have prices show up in bit coin as well as USD
338 2013-02-14 18:35:56 <gmaxwell> [OT] some folks here may find this interesting: https://password-hashing.net/
339 2013-02-14 18:36:34 <Guest9274> or gmaxwell
340 2013-02-14 18:37:42 <gmaxwell> Guest9274: I'm not sure what you're talking about.
341 2013-02-14 18:37:51 <tigger0> he wants exchange rate in his client lol
342 2013-02-14 18:38:05 <Guest9274> yes
343 2013-02-14 18:38:10 <tigger0> why not just go to mtgox.com
344 2013-02-14 18:38:19 <Guest9274> so the customer can see the price both in BTC and USD
345 2013-02-14 18:38:28 <gmaxwell> Customer? price?
346 2013-02-14 18:38:29 <sipa> 2013-02-14 12:16:52 - Connect 685 transactions: 3043.89ms (4.444ms/tx, 2.715ms/txin)
347 2013-02-14 18:38:32 <sipa> 2013-02-14 12:16:52 - Verify 1121 txins: 3161.93ms (2.821ms/txin)
348 2013-02-14 18:38:34 <sipa> 2013-02-14 12:16:52 - Connect: 14622.80ms
349 2013-02-14 18:38:35 <Guest9274> tigger0 no it's for my store i'm doing
350 2013-02-14 18:38:39 <sipa> :o
351 2013-02-14 18:39:00 <gavinandresen> gmaxwell: I'll be interested in the password hashing stuff in Q3 2014 when they select the finalists???.
352 2013-02-14 18:39:19 <gmaxwell> Guest9274: mtgox has an API??? though using raw quotes may have unexpected results??? current prices may not have much to do with prices available to you later.
353 2013-02-14 18:39:35 <Guest9274> ok
354 2013-02-14 18:39:54 <gmaxwell> (I assume other exchanges have APIs too)
355 2013-02-14 18:41:01 <gavinandresen> ACTION is still tempted to put a 24-hour-volume-weighted-multi-exchange-price into the blockchain once an hour....
356 2013-02-14 18:41:35 <gmaxwell> uh. because that needs to be preserved for all time. Good job.
357 2013-02-14 18:41:41 <gmaxwell> :-/
358 2013-02-14 18:42:00 <helo> lol
359 2013-02-14 18:42:10 <jgarzik> I'm sure you're aware of methods to make transactions provably prunable
360 2013-02-14 18:42:27 <sipa> i have no clue why that exchange data needs to be in the blockchain?
361 2013-02-14 18:42:33 <gmaxwell> jgarzik: Uh. I'm sure you're aware that this doesn't actually prevent having to keep the data forever.
362 2013-02-14 18:42:35 <gavinandresen> I just think it'd be cool.
363 2013-02-14 18:42:47 <gmaxwell> jgarzik: it just gets it out of the working set.
364 2013-02-14 18:42:53 <sipa> 'cool' like having prayers and obituaries there?
365 2013-02-14 18:43:14 <gavinandresen> I'd create a vanity address for it, then once a (hour/day/week/whatever) send exactly $1 worth of BTC to it.
366 2013-02-14 18:43:25 <gavinandresen> ??? and empty it (so is pruned) regularly.
367 2013-02-14 18:43:41 <sipa> how about posting it to twitter instead?
368 2013-02-14 18:44:01 <sipa> or an rss feed
369 2013-02-14 18:44:08 <gavinandresen> Transmitting it over the network as a transaction means less code to write if you want to keep track of the price.
370 2013-02-14 18:44:30 <sipa> i'm sure people are able to write code to fetch a url
371 2013-02-14 18:44:44 <gavinandresen> ok, ok???. I didn't say I was actually going to do it
372 2013-02-14 18:45:05 <gmaxwell> gavinandresen: I'd suggest just making nlocktimed transactions for the infinite future.... except....
373 2013-02-14 18:45:30 <gavinandresen> oooh! we could do futures pricing in the chain???.
374 2013-02-14 18:45:35 <gmaxwell> but really, even locked transactions... flodding gavin's moderately interesting quote of the day doesn't scale unless it were just limited to gavin.
375 2013-02-14 18:47:24 <gavinandresen> "Where can I find historical bitcoin prices?" "Just look at this address in blockexplorer.com or any other blockchain address viewer???." Gotta admit that'd be cool.....
376 2013-02-14 18:47:43 <andytoshi> should listaddressgroupings also be allowed in safe mode?
377 2013-02-14 18:47:47 <andytoshi> it's hard to test without it :P
378 2013-02-14 18:48:02 <discrete> ;;buy 10
379 2013-02-14 18:48:02 <gribble> (buy [--long] <amount> <thing> [at|@] <priceperunit> <otherthing> [<notes>]) -- Logs a buy order for <amount> units of <thing>, at a price of <price> per unit, in units of <otherthing>. Use the optional <notes> field to put in any special notes. <price> may include an arithmetical expression, and {mtgox(ask|bid|last)} to index price to mtgox ask, bid, or last price. May also include (1 more message)
380 2013-02-14 18:48:29 <gmaxwell> gavinandresen: Where can I find historical bitcoin prices? Just look at this address in blockchain.info ... http://blockchain.info/charts/market-price
381 2013-02-14 18:49:16 <gmaxwell> :P
382 2013-02-14 18:50:07 <gavinandresen> gmaxwell: mmm. Actually, I've repeatedly looked for historical prices going all the way back to when bitcoinmarket.com first started trading, and have a hard time finding them.
383 2013-02-14 18:52:20 <gmaxwell> gavinandresen: though for most uses you're going to want more frequent/detailed quotes than weekly volume-weighed-last or what have you.
384 2013-02-14 18:52:48 <gmaxwell> Having some "Bitcoin archivist project" would be a good thing to have... something collecting all this data.
385 2013-02-14 18:52:50 <helo> ACTION likes this volume-weighted idea
386 2013-02-14 18:53:42 <gmaxwell> esp since bitcoin sites are so volatile...
387 2013-02-14 18:53:44 <gavinandresen> Yes. If anybody is listening: I can dig out a spreadsheet I've got that has all of the earliest trades from bitcoinmarket.com
388 2013-02-14 18:53:45 <helo> i only ever see bid/ask/last
389 2013-02-14 18:54:28 <gmaxwell> helo: huh? everything gives a volumed weighed too.
390 2013-02-14 18:54:31 <gmaxwell> ;;ticker
391 2013-02-14 18:54:31 <gribble> BTCUSD ticker | Best bid: 26.80000, Best ask: 26.90941, Bid-ask spread: 0.10941, Last trade: 26.95000, 24 hour volume: 146303.35982553, 24 hour low: 21.72100, 24 hour high: 26.95000, 24 hour vwap: 25.09208
392 2013-02-14 18:54:40 <gmaxwell> Last datum.
393 2013-02-14 18:54:49 <gmaxwell> 25.09208
394 2013-02-14 18:55:00 <gavinandresen> is gribble's ticker multi-exchange, or just Mt. Gox ?
395 2013-02-14 18:55:36 <gavinandresen> (and geepers creepers, almost $27 today?)
396 2013-02-14 18:55:51 <gmaxwell> I believe it's just mtgox. The value of multi-exchange is debatable even when volume weighed though..
397 2013-02-14 18:56:31 <helo> ahh, good to know
398 2013-02-14 18:56:45 <gmaxwell> E.g. I find some included but dead exchange with no users... I now do a bizillion back and forth trades at $10000/btc and distort the prices. Means are sensitive to outliers. :)
399 2013-02-14 18:57:02 <gavinandresen> I just assume that mt. gox will eventually be surpassed by some other exchange, although I don't know nuthin about currency exchange markets and how concentrated the big players tend to be
400 2013-02-14 18:57:04 <sipa> mtgox' volume is really high though
401 2013-02-14 18:57:26 <helo> well under a bizillion though
402 2013-02-14 18:57:34 <gmaxwell> sipa: yea. but if I'm the only party on the exchange I can trade at $10,000 for hours, at whatever speed the api will take.
403 2013-02-14 18:57:47 <gmaxwell> Volume doesn't by itself imply efficiency.
404 2013-02-14 18:58:07 <sipa> i'm more referring to popularity than efficiency
405 2013-02-14 18:58:14 <gmaxwell> oh sure.
406 2013-02-14 18:58:34 <sipa> i remember that a month ago i was surprised to see mtgox back a $1M/day
407 2013-02-14 18:58:39 <sipa> but now it's far above that
408 2013-02-14 18:58:44 <gavinandresen> trimming outliers would definitely be a Good Idea, given that exchange feed regularly have nasty bugs reporting bogus prices
409 2013-02-14 19:01:41 <sipa> ;;bc,blocks
410 2013-02-14 19:01:51 <gmaxwell> sipa: I'd almost suggest explicitly distorting your progress bar to make sure it gets faster at the end. :P
411 2013-02-14 19:02:09 <gribble> 221170
412 2013-02-14 19:03:01 <sipa> gmaxwell: in practice, my largest own (psychological...) problem with it, is that it's so slow in the beginning (even if doing thousands of blocks/s) compared to what i'm used to, that it seems stcuk
413 2013-02-14 19:04:10 <sipa> perhaps i shouldn't be measuring/estimating nTransactions, but nTransactions+5*nBlocks or so
414 2013-02-14 19:04:14 <gmaxwell> sipa: from a UI perspective the QT client could probably benefit from some 'hollywood' features. some progress display that had rapidly moving meaningless numbers.
415 2013-02-14 19:04:17 <sipa> as blocks have some constant overhead
416 2013-02-14 19:04:45 <sipa> gmaxwell: if you know some hollywood programmers...
417 2013-02-14 19:05:04 <gmaxwell> E.g. even something as simple as a 'Best block: 000000....DEADBEEF'
418 2013-02-14 19:05:33 <gmaxwell> then it wouldn't look stuck at least.
419 2013-02-14 19:06:11 <sipa> actually
420 2013-02-14 19:06:40 <sipa> just a field somewhere in the bottom bar saying "Last block#: 197324", updated once a second, would be great
421 2013-02-14 19:07:23 <gmaxwell> "What block are you on?" ... "no no leave the mouse still" "just don't touch it" "die die die" (snippits of me providing tech support on IRC)
422 2013-02-14 19:08:41 <sipa> when i was doing the benchmark, i couldn't believe that the last checkpoint corresponded to only 30% progress
423 2013-02-14 19:08:51 <sipa> but looking at the resulting graph, it's remarkably accurate
424 2013-02-14 19:09:41 <sipa> andytoshi: are you dutch?
425 2013-02-14 19:10:11 <andytoshi> sipa: yes, 1/4
426 2013-02-14 19:10:12 <gavinandresen> gmaxwell: ascii strings from coinbase transactions displayed during IBD??? could be a new revenue source for miners....
427 2013-02-14 19:10:16 <andytoshi> and 100% of my name :P
428 2013-02-14 19:10:44 <sipa> andytoshi: yes, i noticed :)
429 2013-02-14 19:11:17 <Luke-Jr> gavinandresen: lol, do it ;)
430 2013-02-14 19:11:33 <gmaxwell> gavinandresen: there should be a bitcoin screensaver.
431 2013-02-14 19:12:51 <gmaxwell> something that displays transaction dependencies like gitk history (https://people.xiph.org/~greg/ss-gitk.png) and coinbases and varrious stats.
432 2013-02-14 19:12:59 <gavinandresen> bah, too many interesting ideas. I think next week I should turn off IRC and just write code.
433 2013-02-14 19:13:50 <gmaxwell> "Of course I run a full node??? how else can you get the awesome screensaver"
434 2013-02-14 19:14:36 <Luke-Jr> gmaxwell: hmm :>
435 2013-02-14 19:14:52 <andytoshi> yeah, that's probably a much better idea than you intended it..
436 2013-02-14 19:15:19 <sipa> it's depressing that i'm reindexing and at block 200k, and my progress is at 20%...
437 2013-02-14 19:16:02 <gmaxwell> thats honest, and better than the "wtf, it's been at 90% for like two hours!@#?!"
438 2013-02-14 19:16:10 <sipa> yes, indeed
439 2013-02-14 19:16:39 <andytoshi> maybe we should write a message to the blockchain at each block saying what % of the full chain it is
440 2013-02-14 19:16:41 <andytoshi> ACTION ducks
441 2013-02-14 19:17:08 <sipa> andytoshi: LOI
442 2013-02-14 19:18:59 <sipa> damnit
443 2013-02-14 19:19:17 <sipa> typed rm -f ~/.bitcoin/debug.log instead of tail -f ~/.bitcoin/debug.log
444 2013-02-14 19:21:44 <gmaxwell> sipa: I've done that a zillion times. ctrl-r.log<enter> "crap!"
445 2013-02-14 19:23:21 <andytoshi> ACTION discovers ctrl-r
446 2013-02-14 19:24:00 <gmaxwell> andytoshi: now you can make mistakes 40-50 times faster.
447 2013-02-14 19:24:30 <andytoshi> yeah, i'm excited
448 2013-02-14 19:24:32 <gmaxwell> I tried excluding rm from the shell history, but that was a bit annoying.
449 2013-02-14 19:24:54 <andytoshi> if only i could force rm to require i write "-r" before the filenames, like on mac..
450 2013-02-14 19:24:54 <Luke-Jr> sipa: yay SIGHUP? :p
451 2013-02-14 19:25:11 <Luke-Jr> I like PgUp more than Ctrl-R - slightly harder to mess up
452 2013-02-14 19:25:39 <Luke-Jr> andytoshi: I aliased 'rm' to 'echo NO U' so I have to type out /bin/rm <.<
453 2013-02-14 19:25:46 <andytoshi> i often do "up-up enter" as root to toggle my screen brightness
454 2013-02-14 19:25:54 <Luke-Jr> I fear soon I need to make /bin/rm exempt from my history tho
455 2013-02-14 19:25:57 <andytoshi> and half the time that's "init 0"..
456 2013-02-14 19:26:12 <gmaxwell> andytoshi: can't you run xbacklight as a normal user?
457 2013-02-14 19:26:44 <andytoshi> hmm, i'll look that up..it'd be much nicer than my /proc-poking shell script
458 2013-02-14 19:27:01 <gmaxwell> whatever command you need, enable it in sudo and then alias b='sudo backlightcommand' then b<enter> is fewer keys and can't screw you.
459 2013-02-14 19:27:27 <gmaxwell> thats how I do my proc-poking suspend.
460 2013-02-14 19:28:33 <Luke-Jr> my proc-poking suspend is too complicated for sudo :P
461 2013-02-14 19:38:26 <Eliel_> gmaxwell: it'd be nice if you could define a "dangerous command list" for ctrl-r so that when one of those gets recalled, it will ignore enter keypresses until the line's been shown for at least 2 seconds.
462 2013-02-14 19:38:33 <gmaxwell> sipa: what is your memory usage like after your syncs?
463 2013-02-14 19:38:56 <sipa> sync or reindex?
464 2013-02-14 19:39:02 <Luke-Jr> Eliel_: IMO, just prepend the line with a # in history
465 2013-02-14 19:39:54 <Luke-Jr> err, I guess ctrl-R doesn't let you edit the command before running it:/
466 2013-02-14 19:40:09 <Eliel_> Luke-Jr: the idea is to make sure you have time to register what the command does, not to give you extra steps to memorize so you can again make mistakes much faster :P
467 2013-02-14 19:40:17 <sipa> Luke-Jr: sure does
468 2013-02-14 19:40:18 <Luke-Jr> Eliel_: :P
469 2013-02-14 19:40:24 <Luke-Jr> sipa: ?
470 2013-02-14 19:40:36 <MCM-Mike> use histignore to to run epic rm commands over and over again
471 2013-02-14 19:40:49 <sipa> Luke-Jr: it doesn't work for you?
472 2013-02-14 19:40:59 <Luke-Jr> ah, it does if I press an arrow key first
473 2013-02-14 19:41:06 <sipa> ctrl-r, then use backspace/left/right to move/edit
474 2013-02-14 19:41:21 <Luke-Jr> sipa: I'm used to PgUp where I can just keep typing :P
475 2013-02-14 19:41:44 <gmaxwell> sipa: whatever you have? I finally went through Jouke's debug log in detail, and I don't see any reason he was using so much memory. I'm wondering how reproducable the issue is. (I still don't really have anything to run bitcoin on right now)
476 2013-02-14 19:41:44 <sipa> ic
477 2013-02-14 19:42:01 <gmaxwell> Luke-Jr: ctrl-r lets you edit.
478 2013-02-14 19:42:18 <gmaxwell> ctrl-r substring <rightarrow>
479 2013-02-14 19:43:29 <sipa> gmaxwell: on my VPS, which just reindexed, 543MB of RES, 99 connections, dbcache=200, maxreceivebuffer=1200, maxsendbuffer=1600
480 2013-02-14 19:44:02 <gmaxwell> Eliel_: I think even more often than rm I'll ctrl-r reinvoke ./really.long.command -o output_file_I_really_didnt_want_to_overwrite
481 2013-02-14 19:44:25 <sipa> on my desktop where i did the same, 429 RES, 1 connection, -dbcache=200
482 2013-02-14 19:44:49 <Eliel_> gmaxwell: well, unless you want to delay everything by 2 seconds, you can't really shield against that.
483 2013-02-14 19:45:20 <Eliel_> even one second could be enough... maybe
484 2013-02-14 19:46:31 <gmaxwell> Eliel_: which was my point. ... a versioning filesystem would protect against that.
485 2013-02-14 19:47:03 <Luke-Jr> gmaxwell: I'd love to see a versioning filesystem without versions :P
486 2013-02-14 19:47:06 <gmaxwell> I've also got a lot of my really long running commands put their output on stdout.
487 2013-02-14 19:47:40 <Luke-Jr> so eg, an editor's Undo buffer is basically stored in the filesystem, and Save becomes Tag
488 2013-02-14 19:48:24 <Eliel_> Luke-Jr: you mean integrate an editor with git?
489 2013-02-14 19:48:31 <Luke-Jr> Eliel_: git doesn't behave this way.
490 2013-02-14 19:48:43 <Luke-Jr> git only commits when you tell it to
491 2013-02-14 19:49:04 <Luke-Jr> but sure, a git filesystem is a step forward
492 2013-02-14 19:50:23 <Eliel_> I mean, the git editor could do an automatic commit when you save the file. Although perhaps do it in a way that allows you to combine the commits at the end into a bigger one.
493 2013-02-14 19:51:47 <kytv> Hi. Would the channel ops/others be opposed to this channel being relayed to the I2P network as #bitcoin is?
494 2013-02-14 19:52:53 <gmaxwell> Luke-Jr: when doing r&d work, I often run something like while true; do $EDITOR program.c ; gcc program && ./program && git commit -a ; done
495 2013-02-14 19:54:55 <ProfMac> I try to only save code that gives a clean compile, and "correct" output. I refine the program to do more and more agressive levels of correct work.
496 2013-02-14 19:56:48 <helo> kytv: they probably are opposed to it, since they wouldn't be able to selectively ban/kick i2p users
497 2013-02-14 19:57:28 <sipa> meh, i don't mind - if there's really problems we can always kick the relay bot
498 2013-02-14 19:58:11 <kytv> fwiw annoying users can be blocked from being relayed if it comes to that.
499 2013-02-14 19:58:30 <kytv> (assuming the relay was allowed, that is)
500 2013-02-14 19:58:41 <gmaxwell> kicking the bot works fine. if people who are ops here are given ops on I2P they might be convinced to log in and punt problems instead of the realy.
501 2013-02-14 20:00:45 <kytv> and if there are problems in #bitcoin, I can be highlighted in case i'm AFK (killyourtv or kytv) and I'll take care of the problem users. Those that caused the relay to be k/b'd in the past were active when I was AFK (as luck would have it)
502 2013-02-14 20:03:51 <Luke-Jr> Eliel_: I mean, it should "commit" even before saving
503 2013-02-14 20:04:42 <i2pRelay> <KillYourTV@i2p> alright, let's see how this goes (I don't want 'us' to add to the 'noise')
504 2013-02-14 20:06:57 <BlueMatt> sipa: ahh, ok just wondered
505 2013-02-14 20:08:14 <sipa> BlueMatt: thanks for reminding me, though :)
506 2013-02-14 20:16:31 <ProfMac> I can't start bitcoin-qt. It says I can't bind to port 8333. netstat -W | grep 8333 doesn't return anything. I'm totally paranoid about being hacked, I wonder why I can't use that port.
507 2013-02-14 20:17:11 <BlueMatt> does netstat -W return listening ports?
508 2013-02-14 20:17:23 <BlueMatt> it doesnt appear to on my system...
509 2013-02-14 20:17:35 <BlueMatt> you're looking for netstat -l
510 2013-02-14 20:17:45 <BlueMatt> probably -ln
511 2013-02-14 20:18:26 <i2pRelay> <K1773R@i2p> sudo netstat -n --listening --inet -p | grep 8333
512 2013-02-14 20:18:38 <i2pRelay> <K1773R@i2p> shows u which process/pid is listening on port 8333
513 2013-02-14 20:19:37 <discrete> ;;ticker
514 2013-02-14 20:19:37 <gribble> BTCUSD ticker | Best bid: 27.13099, Best ask: 27.15000, Bid-ask spread: 0.01901, Last trade: 27.12824, 24 hour volume: 159840.67127667, 24 hour low: 21.72100, 24 hour high: 27.59690, 24 hour vwap: 25.30244
515 2013-02-14 20:20:16 <BlueMatt> getting volatile much?
516 2013-02-14 20:21:20 <ProfMac> sudo netstat -n --listening --inet -p | grep 8333 returns nothing. As does --inet6 The last change I made was to comment out the onlynet=IPv6 line in bitcoind.conf.
517 2013-02-14 20:21:57 <discrete> ;;getrate 130
518 2013-02-14 20:21:57 <gribble> Error: "getrate" is not a valid command.
519 2013-02-14 20:22:09 <i2pRelay> <K1773R@i2p> whats in debug.log?
520 2013-02-14 20:22:17 <discrete> ;;genrate 130
521 2013-02-14 20:22:18 <gribble> The expected generation output, at 130.0 Mhps, given difficulty of 3275464.58657, is 0.0199598699385 BTC per day and 0.000831661247437 BTC per hour.
522 2013-02-14 20:22:46 <discrete> ;;genrate
523 2013-02-14 20:22:46 <gribble> (genrate <hashrate> [<difficulty>]) -- Calculate expected bitcoin generation rate using <hashrate> Mhps, at current difficulty. If optional <difficulty> argument is provided, expected generation time is for supplied difficulty.
524 2013-02-14 20:23:00 <discrete> ;;genrate 60000
525 2013-02-14 20:23:01 <gribble> The expected generation output, at 60000.0 Mhps, given difficulty of 3275464.58657, is 9.21224766392 BTC per day and 0.383843652663 BTC per hour.
526 2013-02-14 20:24:12 <Scrat> discrete: you're spamming. you can pm the bot
527 2013-02-14 20:24:51 <Scrat> in other news asicminer is now @ 1.3 TH/s. oh lawdy
528 2013-02-14 20:29:46 <Scrat> 2.3*
529 2013-02-14 20:30:20 <broomkorn> damn
530 2013-02-14 20:38:59 <sipa> gavinandresen: how did you manage to pick a checkpoint with so many 11's in its total transaction since genesis count? (11011160)
531 2013-02-14 20:40:29 <gmaxwell> lol!
532 2013-02-14 20:40:30 <gavinandresen> ooh, is that the last one? generated at 11:11 on 1/11 ?
533 2013-02-14 20:40:45 <gavinandresen> (sorry, 11/1 for europeans)
534 2013-02-14 20:41:00 <sipa> yes
535 2013-02-14 20:41:05 <sipa> number 216116 :)
536 2013-02-14 20:41:49 <gavinandresen> that is a lot of elevenses. Must be an omen, price is headed to $111.11 ???..
537 2013-02-14 20:42:00 <sipa> ... or $0.11
538 2013-02-14 20:42:08 <gavinandresen> nah. Maybe $11.11
539 2013-02-14 20:42:47 <andytoshi> haha
540 2013-02-14 20:44:12 <gavinandresen> picking a favorite number is a good way of demonstrating confirmation bias; I see and notice 11's everywhere. (e.g. my computer says it is 4:44 right now....)
541 2013-02-14 20:44:54 <sipa> here it's 22:44 !!!
542 2013-02-14 20:45:05 <gavinandresen> what are the chances!
543 2013-02-14 20:45:15 <i2pRelay> <K1773R@i2p> same
544 2013-02-14 20:45:34 <sipa> K1773R: wow!
545 2013-02-14 20:46:16 <gavinandresen> of course, if my favorite number was 1773 I'm sure I'd see it a lot less often
546 2013-02-14 20:46:48 <i2pRelay> <K1773R@i2p> so ur favorite number is 11?
547 2013-02-14 20:46:59 <gavinandresen> yes, my favorite number is eleven
548 2013-02-14 20:49:45 <i2pRelay> <K1773R@i2p> so this address would be a dream for you: http://blockchain.info/address/1111111111111111111114oLvT2
549 2013-02-14 20:51:47 <gavinandresen> mmm??? that'll never happen. I wonder if there is a combination of "11" and "eleven" and "ELEVEN'" that has a checksum of "1111" in base58 ???.
550 2013-02-14 20:52:57 <gavinandresen> wait: lower-case-L isn't allowed???, would have to be e1even
551 2013-02-14 20:52:59 <i2pRelay> <K1773R@i2p> i dont think so
552 2013-02-14 20:53:44 <echius> what kind of sum are you talking about?
553 2013-02-14 20:54:04 <gavinandresen> the 4-byte checksum at the end that makes http://blockchain.info/address/1111111111111111111114oLvT2 ugly
554 2013-02-14 20:59:11 <i2pRelay> <K1773R@i2p> an address that represents part of PI would be funny
555 2013-02-14 21:00:46 <Luke-Jr> I own 3P14159f73E4gFr7JterCCQh9QjiTjiZrG
556 2013-02-14 21:03:14 <sipa> nice :)
557 2013-02-14 21:03:20 <sipa> not very accurate though :p
558 2013-02-14 21:03:29 <i2pRelay> <K1773R@i2p> 3P1? which altcoin is this
559 2013-02-14 21:03:36 <kjj> p2sh
560 2013-02-14 21:03:39 <sipa> bitcoin p2sh address
561 2013-02-14 21:04:11 <Luke-Jr> K1773R: it's a standard Bitcoin address ;)
562 2013-02-14 21:04:39 <i2pRelay> <K1773R@i2p> never heard about p2sh
563 2013-02-14 21:05:21 <kjj> normal addresses are hashes of pubkeys. p2sh addresses are hashes of scripts. most commonly used in multisig
564 2013-02-14 21:05:21 <sipa> see BIP13 for the idea, BIP16 for the implementation
565 2013-02-14 21:05:27 <gmaxwell> Luke-Jr: not 3P243F6A8885A ?
566 2013-02-14 21:05:33 <kjj> ha!
567 2013-02-14 21:05:58 <sipa> haha :)
568 2013-02-14 21:06:07 <sipa> or how about Pi in base58?
569 2013-02-14 21:07:48 <Luke-Jr> gmaxwell: I should vanitygen that too
570 2013-02-14 21:07:51 <i2pRelay> <K1773R@i2p> does anyone have a base58 calc, standalone C/py/bash? interested to feed it into vanitygen :P
571 2013-02-14 21:08:15 <Luke-Jr> K1773R: python-base58 ?
572 2013-02-14 21:08:34 <sipa> pi in base58: 4.9DKRhUz
573 2013-02-14 21:09:38 <sipa> well, correctly rounded it's 4.9DKRhV1
574 2013-02-14 21:09:46 <i2pRelay> <K1773R@i2p> luke i guess u mean this one: http://gitorious.org/bitcoin/python-base58/blobs/master/base58.py right?
575 2013-02-14 21:09:52 <Luke-Jr> right
576 2013-02-14 21:09:55 <i2pRelay> <K1773R@i2p> the dot is a problem :S
577 2013-02-14 21:10:03 <Luke-Jr> I did P for Point
578 2013-02-14 21:10:07 <sipa> then do 1/Pi or so
579 2013-02-14 21:10:09 <sipa> or Pi/4
580 2013-02-14 21:10:11 <i2pRelay> <K1773R@i2p> ty
581 2013-02-14 21:10:31 <gmaxwell> 1/?? or ??/4 lose the 3 upfront.
582 2013-02-14 21:10:47 <sipa> aka ??-3
583 2013-02-14 21:11:00 <Luke-Jr> sipa: what if you set the 20 bytes to pi in binary (exponent implied) and encode it as an address? :P
584 2013-02-14 21:11:36 <Scrat> ??
585 2013-02-14 21:13:04 <echius> gavinandresen: ok, arent't those bitcoin addresses the result of multiple SHA256 hashings? or is this a custom-script-based address?
586 2013-02-14 21:13:05 <i2pRelay> <K1773R@i2p> lol
587 2013-02-14 21:13:05 <i2pRelay> <K1773R@i2p> $ ./oclvanitygen 1Pi4P9DKRhV1
588 2013-02-14 21:13:06 <i2pRelay> <K1773R@i2p> Difficulty: 9883693997182075238
589 2013-02-14 21:13:09 <i2pRelay> <K1773R@i2p> [29.05 Mkey/s][total 335544320][Prob 0.0%][50% in 7477.3y]
590 2013-02-14 21:13:48 <sipa> Luke-Jr: C90FDAA22168C234C4C6628B80DC1CD129024E08 = base58(uint160(pi/4*2^10))
591 2013-02-14 21:13:51 <sipa> eh
592 2013-02-14 21:13:58 <sipa> Luke-Jr: 1KL7p7i8nsXqjWHsCKLT8gFb8AaSQRod8k
593 2013-02-14 21:14:08 <sipa> 2^160
594 2013-02-14 21:14:09 <echius> gavinandresen: can people actually generate addresses that customized? i thought just making vanity addresses with just a few bytes chosen by you was quite the task
595 2013-02-14 21:14:37 <sipa> echius: it's much easier if you don't need a matching private key
596 2013-02-14 21:14:49 <sipa> echius: which is what was done for 1111111111111111111114oLvT2
597 2013-02-14 21:15:06 <kjj> hmm. once I turn my GPUs from mining towards vanitygenning, I might just have to patch importprivkey to import into the keypool
598 2013-02-14 21:15:07 <i2pRelay> <K1773R@i2p> same goes for BitcoinEaterDontSend
599 2013-02-14 21:15:13 <echius> ah, i get it. i guess those bitcoins are gone then
600 2013-02-14 21:15:15 <gavinandresen> yes, if you just want an address that you can never spend it is easy. the 1111???vT2 address corresponds to "keypair whos public key hashes to all zeroes"
601 2013-02-14 21:15:42 <Scrat> only need to bruteforce ~2 billion hashes on average
602 2013-02-14 21:15:52 <sipa> Scrat: for what?
603 2013-02-14 21:16:04 <andytoshi> If you take the word "pi" and sha256 it 10 million times, then encode that, you get 16kAaYc3v6DUwbsRdRc2D27u4nL1mDGENt
604 2013-02-14 21:16:10 <Scrat> sipa is gonna kill me
605 2013-02-14 21:16:16 <Scrat> the address checksum
606 2013-02-14 21:16:19 <sipa> /kill Scrat
607 2013-02-14 21:16:26 <sipa> /kill -9 Scrat
608 2013-02-14 21:16:29 <andytoshi> oops, i shoulda wrecked that a bit to prevent anyone trying to spend to it..
609 2013-02-14 21:16:44 <i2pRelay> <K1773R@i2p> why kill? kill -STOP :P
610 2013-02-14 21:16:56 <i2pRelay> <K1773R@i2p> and wake him up with -CONT 10 years later :D
611 2013-02-14 21:16:57 <gavinandresen> phew, andytoshi almost guessed my brainwallet ("pi" sha256 hashed 11 million times)
612 2013-02-14 21:16:59 <gavinandresen> (kidding)
613 2013-02-14 21:17:06 <andytoshi> haha
614 2013-02-14 21:17:13 <sipa> gavinandresen: it's 11111111 times obviously
615 2013-02-14 21:17:19 <gavinandresen> d'oh!
616 2013-02-14 21:19:31 <i2pRelay> <K1773R@i2p> wasnt it 11, hashed 11 millions times?
617 2013-02-14 21:22:17 <sipa> sssssh
618 2013-02-14 21:23:01 <sipa> (super-savvy-sexy-secure-shell)
619 2013-02-14 21:24:24 <Luke-Jr> hmm
620 2013-02-14 21:24:41 <Luke-Jr> Is there a term for the SHA256 of pi? :P
621 2013-02-14 21:26:04 <Eliel_> ... that doesn't sound like something that's possible to calculate... At least unless you're talking about sha256 of an approximate value of pi.
622 2013-02-14 21:26:35 <andytoshi> haha
623 2013-02-14 21:26:55 <Eliel_> ... unless there's some exotic number system that can actually express pi without needing to approximate.
624 2013-02-14 21:27:10 <andytoshi> base pi would do it
625 2013-02-14 21:27:10 <sipa> Eliel_: sure, it's "10" in base Pi
626 2013-02-14 21:27:25 <i2pRelay> <K1773R@i2p> thats easy, its "Pi"
627 2013-02-14 21:28:32 <Eliel_> ... how does a number system work when the base is not an integer...
628 2013-02-14 21:28:51 <sipa> Eliel_: in exactly the same way
629 2013-02-14 21:29:19 <sipa> abc.de in base Pi is a*Pi^2 + b*Pi + c + d*Pi^-1 + e*Pi^-2
630 2013-02-14 21:29:27 <andytoshi> Eliel_: as long as you use a positive real number, it's fairly clear what's going on
631 2013-02-14 21:29:28 <sipa> it's not unambiguous
632 2013-02-14 21:29:49 <sipa> but base 10 isn't always unambiguous either :p
633 2013-02-14 21:30:02 <andytoshi> if you use an integer, and only digits less than that integer, it's almost unambiguous
634 2013-02-14 21:30:22 <sipa> unambiguous almost everywhere, yes
635 2013-02-14 21:30:36 <sipa> but not everywhere
636 2013-02-14 21:30:39 <andytoshi> oh, yeah, that's right
637 2013-02-14 21:30:50 <BlueMatt> gmaxwell: you dont happen to have that script lying around which tests for test_bitcoin, etc coverage?
638 2013-02-14 21:30:58 <Eliel_> ACTION wonders if there's unambigous number system that can express pi exactly.
639 2013-02-14 21:31:12 <BlueMatt> gmaxwell: I figure Ive waited long enough to make sure that pull-tester complains for every pull req that decreases total coverage :)
640 2013-02-14 21:31:25 <sipa> Eliel_: there is no unambiguous number system
641 2013-02-14 21:31:54 <Eliel_> sipa: well ok, a number system that's not more ambiguous than base-10 then :)
642 2013-02-14 21:32:21 <Eliel_> although, I'm not sure in what ways base-10 is ambiguous.
643 2013-02-14 21:32:26 <sipa> Eliel_: 0.99999... == 1
644 2013-02-14 21:32:36 <andytoshi> that didn't occur to me
645 2013-02-14 21:32:41 <andytoshi> "a general proof can be given that almost all real numbers are normal"..."this proof is not constructive and only very few specific numbers have been shown to be normal"
646 2013-02-14 21:32:41 <andytoshi> this is maybe of interest: https://en.wikipedia.org/wiki/Normal_number
647 2013-02-14 21:32:43 <andytoshi> here "normal" means that all strings occur with roughly the expected probability in every base
648 2013-02-14 21:33:19 <jaakkos> is the reference client doing a crazy amount of fsync or what's going on...
649 2013-02-14 21:33:29 <gmaxwell> BlueMatt: install lcov. compile with --coverage run lcov -c -d `pwd` -b `pwd` -t test_bitcoin -o test_bitcoin.info; genhtml -s test_bitcoin.info -o coverage.report/
650 2013-02-14 21:33:35 <sipa> jaakkos: yes
651 2013-02-14 21:33:36 <BlueMatt> gmaxwell: thanks
652 2013-02-14 21:33:50 <gmaxwell> BlueMatt: it reports the coverage on stdout in text and generates useful html.
653 2013-02-14 21:34:00 <jaakkos> that's not very nice on ZFS ;)
654 2013-02-14 21:34:07 <BlueMatt> gmaxwell: thanks, Ill try to get that working next few days
655 2013-02-14 21:34:12 <jaakkos> because it forces a txg to be pushed to the disk...
656 2013-02-14 21:34:21 <sipa> jaakkos: is 0.8 as bad?
657 2013-02-14 21:34:38 <jaakkos> so downloading blockchain 700 kB/s causes 250MB/s disk writes
658 2013-02-14 21:35:18 <BlueMatt> gmaxwell: how does one merge the coverage reports again?
659 2013-02-14 21:35:23 <Eliel_> sipa: ok, then that gives a concise definition for what I'd mean with unambiguous. a number system where two numbers, both expressable without infinity sequenes are equal only if they are identical.
660 2013-02-14 21:35:33 <jaakkos> sipa: haven't tried - this is 0.7.2.0 from debian sid
661 2013-02-14 21:35:53 <gmaxwell> BlueMatt: lcov -c report report report -o newreport
662 2013-02-14 21:36:02 <BlueMatt> gmaxwell: thanks
663 2013-02-14 21:36:44 <jaakkos> sipa: i'm temporarily using 'eatmydata' to eliminate fsync while downloading the chain
664 2013-02-14 21:37:14 <gmaxwell> BlueMatt: you may want to also do lcov -c -i -o baseline.info before running any thing, so that totally uncovered stuff shows up.
665 2013-02-14 21:39:59 <jaakkos> i'm not sure if crazy amount of fsync is a good idea... considering that many cheapo physical storage ignores barriers anyway
666 2013-02-14 21:40:09 <jaakkos> but then again i should be trying 0.8 i suppose
667 2013-02-14 21:40:36 <BlueMatt> gmaxwell: alrighty, thanks
668 2013-02-14 21:41:22 <gmaxwell> jaakkos: take that up with oracle, it's not our code thats super fsync happy.
669 2013-02-14 21:44:42 <Luke-Jr> Eliel_: base pi works
670 2013-02-14 21:46:12 <sipa> 4 can be expressed in base pi in a (countably) infinite number of ways
671 2013-02-14 21:46:30 <sipa> every number except 0, actually
672 2013-02-14 21:48:04 <jaakkos> gmaxwell: this could be a zfsonlinux bug... looking to it.
673 2013-02-14 21:48:34 <jaakkos> sync=disabled or eatmydata didn't help, surprisingly.
674 2013-02-14 21:49:24 <Scrat> jaakkos: zfsonlinux? you're brave
675 2013-02-14 21:49:45 <jaakkos> Scrat: it's actually quite good. i would trust it over btrfs any time (though that's not much...)
676 2013-02-14 21:50:04 <Luke-Jr> sipa: yes, but pi is a simple 10
677 2013-02-14 21:50:08 <Scrat> indeed :p Ts'o taking his sweet time
678 2013-02-14 21:55:11 <sipa> Luke-Jr: or 3.01102111002...
679 2013-02-14 21:55:23 <Luke-Jr> sipa: hm?
680 2013-02-14 21:55:41 <jaakkos> oh the client uses berkeley db
681 2013-02-14 21:55:48 <Luke-Jr> jaakkos: not for long!
682 2013-02-14 21:55:48 <sipa> jaakkos: up to 0.7.x
683 2013-02-14 21:55:56 <jaakkos> zfs has a known regression with that
684 2013-02-14 21:55:56 <sipa> well, 0.8 still uses bdb for the wallet
685 2013-02-14 21:55:57 <Luke-Jr> sipa: well, 0.8.x uses it for wallet still :<
686 2013-02-14 21:56:31 <sipa> Luke-Jr: check it; Pi in base-Pi lies between 3.01102111002 and 3.01102111003
687 2013-02-14 21:56:42 <sipa> in addition to being 10
688 2013-02-14 21:56:47 <Luke-Jr> ???
689 2013-02-14 21:57:01 <sipa> trust me :p
690 2013-02-14 21:57:02 <jaakkos> sipa, Luke-Jr: i see
691 2013-02-14 21:57:50 <sipa> ;;calc 3+1/pi^2+1/pi^3+2/pi^5+1/pi^6+1/pi^7+1/pi^8+2/pi^11
692 2013-02-14 21:57:51 <gribble> Error: Something in there wasn't a valid number.
693 2013-02-14 21:58:12 <Luke-Jr> sipa: it's python
694 2013-02-14 21:59:25 <sipa> ;;calc pi=3.14159265358979323; 3+1/pi^2+1/pi^3+2/pi^5+1/pi^6+1/pi^7+1/pi^8+2/pi^11
695 2013-02-14 21:59:26 <gribble> Error: invalid syntax (<string>, line 1)
696 2013-02-14 21:59:50 <gribble> Error: invalid syntax (<string>, line 1)
697 2013-02-14 21:59:50 <sipa> ;;calc pi=3.14159265358979323; 3+1/pi**2
698 2013-02-14 22:03:35 <andytoshi> sipa: cool, it works in octave
699 2013-02-14 22:03:53 <andytoshi> i'm kinda surprised you were able to do that without using digits greater than 3
700 2013-02-14 22:05:09 <sipa> andytoshi: that's the point of course
701 2013-02-14 22:05:40 <sipa> andytoshi: the digit range is 0-3 (which means 4 possibilities), but you only need 3.1415... possibilities to cover every number in base pi
702 2013-02-14 22:06:01 <sipa> not that the fractional number of possibities is possible of course
703 2013-02-14 22:06:22 <sipa> but it results in every non-integer base in having an infinite number of ways to express every non-zero number
704 2013-02-14 22:09:10 <andytoshi> oh :P i forgot that 0,1,2,3 was four numbers..
705 2013-02-14 22:09:53 <BlueMatt> gmaxwell: hmm...why am I missing branch coverage? I seem to remember that being there?
706 2013-02-14 22:10:40 <gmaxwell> BlueMatt: your lcov is old.
707 2013-02-14 22:11:00 <BlueMatt> gmaxwell: nope, Im on 1.10, which appears to be newest
708 2013-02-14 22:11:14 <BlueMatt> (though I havent tested on jenkins, which may have different results)
709 2013-02-14 22:11:14 <gmaxwell> not that branch coverage is super useful on our codebase. :( all the C++ magic creates ooodles of untestable heap allocation failure checks.
710 2013-02-14 22:11:22 <BlueMatt> yes, but still....
711 2013-02-14 22:12:12 <gmaxwell> BlueMatt: IIRC it's on by default since 1.09 or something. I'm running 1.11 (some CVS version) because I hit a crashbug in something.
712 2013-02-14 22:12:28 <BlueMatt> gmaxwell: Im on 1.10 though....
713 2013-02-14 22:13:13 <BlueMatt> well Ill see whats up with jenkins, but I suppose its no big deal without
714 2013-02-14 22:56:31 <jaakkos> ACTION created ext4 on zvol, fixed the bdb issue
715 2013-02-14 22:59:43 <sipa> Luke-Jr: smaller representation of Pi in base Pi: 3.0033332333111332313333122033213033233312132103331323221213121120332232113111122133221313032233331113
716 2013-02-14 23:00:06 <Luke-Jr> ACTION hides from sipa
717 2013-02-14 23:00:38 <Luke-Jr> sipa: please tell me you didn't spend the last 2 hours on that
718 2013-02-14 23:00:44 <sipa> I didn't
719 2013-02-14 23:00:59 <sipa> now, my status bar at this point in irssi (and it's not intentional):
720 2013-02-14 23:01:07 <sipa> [01:00] [sipa(+i)] [17:freenode/#bitcoin-dev(+Ccnt)] [Act: 3,14,15,39]
721 2013-02-14 23:01:20 <sipa> i have no channel numbered 92, unfortunately