1 2013-02-11 00:23:41 <B0g4r7> Ugh, Ubuntu Software Center offers to install bitcoind 0.3.24.
  2 2013-02-11 00:28:18 <rdponticelli> lol
  3 2013-02-11 00:28:28 <rdponticelli> That's not bitcoin anymore
  4 2013-02-11 00:28:30 <gmaxwell> B0g4r7: please ask them to remove it, it is insecure and will get people robbed.
  5 2013-02-11 00:34:29 <andytoshi> yikes
  6 2013-02-11 00:36:35 <B0g4r7> I wonder how hard it'd be to get 0.7.2 into the repo.
  7 2013-02-11 00:37:37 <B0g4r7> and keep it up to date
  8 2013-02-11 00:37:52 <Luke-Jr> B0g4r7: it might be easier to get 0.6.4 in
  9 2013-02-11 00:48:33 <petertodd> Does anyone have data on orphan rates?
 10 2013-02-11 00:48:59 <petertodd> Looking at blockchain.info the orphan rate appears to be about 1-2%
 11 2013-02-11 00:49:20 <petertodd> Just wondering if a more concerted effort has been made to determine this.
 12 2013-02-11 01:38:05 <eckey> Does anyone here rune Bitcoin-Qt on MacBook OSX 10.8.2?
 13 2013-02-11 02:11:33 <sgornick> gmaxwell: Ping
 14 2013-02-11 02:12:31 <sgornick> re: > I asked for permissions to unlock it but haven't heard back??? will prod.  <--- Bitcoin wiki BIP article.
 15 2013-02-11 02:12:46 <sgornick> That article isn't locked, is that resolved for you?
 16 2013-02-11 02:13:09 <sgornick> Wiki does use that BitcoinPayment anti-spam now, but other than that you should be able to edit that article.
 17 2013-02-11 02:24:38 <B0g4r7> I run Bitcoin-Qt on 10.6 and 10.7.  Not on 10.8 tho.
 18 2013-02-11 08:31:33 <Raccoon> What paste sites (eg, pastebin) do you guys use on IRC? I'm compiling a list of ones people actively use. Please /msg me your reply; cross-posted.
 19 2013-02-11 12:05:57 <Belkaar> is it ok to hard- or symlink the new blk????.dat and rev????.dat between two instances of bitcoind 0.8 (provided it is not the highest number) ?
 20 2013-02-11 12:07:36 <sipa> Belkaar: should be, yes
 21 2013-02-11 12:07:56 <Belkaar> so once they are "full" they are not written to again?
 22 2013-02-11 12:08:15 <sipa> indeed
 23 2013-02-11 12:09:12 <Belkaar> But as I understand it they contain unpruned blocks, so in future updates they could be altered again by the pruning feature?
 24 2013-02-11 12:10:58 <Luke-Jr> Belkaar: you can hardlink the last one too. just new blocks won't be shared
 25 2013-02-11 12:11:38 <Belkaar> Luke-Jr: wont both bitcoinds try to write to it at the same time when a new block arrives?
 26 2013-02-11 12:11:48 <Luke-Jr> Belkaar: no, only old ones will
 27 2013-02-11 12:11:53 <Luke-Jr> Belkaar: new ones cap out blocks at 256 MB
 28 2013-02-11 12:12:28 <Belkaar> Luke-Jr
 29 2013-02-11 12:12:57 <Belkaar> Luek-Jr: sorry I don't get it, lets say the currently not full one is blk0012.dat
 30 2013-02-11 12:13:26 <Luke-Jr> Belkaar: as long as it's over 256 MB, the new one will start blk00012.dat
 31 2013-02-11 12:13:37 <Luke-Jr> Belkaar: but it's not intended to scale beyond blk0003.dat
 32 2013-02-11 12:14:25 <Belkaar> Luke-Jr: I'm talking about a newly synced block database from 0.8 with its currently about 40 block files
 33 2013-02-11 12:15:10 <Luke-Jr> Belkaar: oh
 34 2013-02-11 12:15:41 <Belkaar> Luke-Jr: So I have blk0000.dat-blk0042.dat with 128MB and blk0043.dat with 16MB . So my question is: can I symlink 0000-00042 only or all of them
 35 2013-02-11 12:15:50 <Belkaar> Luke-Jr: *hard-link
 36 2013-02-11 12:15:59 <Luke-Jr> why would you make any links?
 37 2013-02-11 12:16:16 <Luke-Jr> also, 0.8+ uses blk00000.dat etc, not blk0001.dat etc
 38 2013-02-11 12:16:47 <Belkaar> Luke-Jr: So I don't need 3GB of disk space containing the same data for each running node
 39 2013-02-11 12:17:39 <Luke-Jr> Belkaar: nobody can predict how 0.9 will be compatible with 0.8
 40 2013-02-11 12:24:24 <sipa> Belkaar: if pruning is ever added (it may well be), it will just be deleting blk* and rev* files
 41 2013-02-11 12:26:19 <Belkaar> Ok. Thank you for the clarification
 42 2013-02-11 12:26:39 <sipa> (that's the primary reason for smaller block files, so you have better granularity for pruning)
 43 2013-02-11 12:45:13 <Happzz> is the testnet also limited to 21mil, etc'?
 44 2013-02-11 12:46:47 <kjj> the problem in that question is the "the".  it would be better to say that "each testnet is also limited"
 45 2013-02-11 12:47:21 <Happzz> are there more than 1 testnet?
 46 2013-02-11 12:47:24 <kjj> but it would be trivial to change that in your own testnet, should you ever wish to do so
 47 2013-02-11 12:47:55 <kjj> yes, there is a testnet-in-a-box that anyone can use to set up their own private test network
 48 2013-02-11 12:48:37 <kjj> there is also some agreement to make a distributed test network if you want to connect to it
 49 2013-02-11 12:49:32 <kjj> if you are using the testnet mode of the stock client, it follows mostly the same rules as the real network
 50 2013-02-11 12:49:51 <Happzz> okay
 51 2013-02-11 13:11:03 <Belkaar> kjj: On that note: I'm running a Testnet node 24/7 at the moment. Is that still usefull?
 52 2013-02-11 13:19:38 <kjj> yeah, I think I've seen the guys talking about wanting more long lived nodes on the distributed testnet
 53 2013-02-11 13:20:09 <kjj> I've been meaning to set one up, but my poor decade old colo box just can't take it
 54 2013-02-11 13:22:15 <jgarzik> Belkaar: yes, it is use.  Consider CPU mining on testnet, with a single core or so, if you can afford it.
 55 2013-02-11 13:24:59 <Belkaar> jgarzik: It's running on a raspberry pi with cpu minig on. Does 177kH/s. So it will ensure a block every 4 hours or so
 56 2013-02-11 15:59:12 <jgarzik> 27262 jgarzik   39  19 1791m 227m  16m S  8.6  5.8  38:59.35 bitcoind
 57 2013-02-11 15:59:24 <jgarzik> not bad at all!  Quite nice memory usage, for a full node + mining.
 58 2013-02-11 16:03:18 <Guest85802> 227mb?
 59 2013-02-11 16:10:04 <jgarzik> correct
 60 2013-02-11 16:11:58 <EskimoBob> Hi, how do I check out the latest stable version from github? Or is v0.8.0rc1 OK for regular use?
 61 2013-02-11 16:13:00 <sipa> rc = release candidate
 62 2013-02-11 16:13:07 <sipa> so not intended for production use yet
 63 2013-02-11 16:13:25 <sipa> v0.7.2 is the latest release
 64 2013-02-11 16:13:33 <EskimoBob> OK, let me rephrase my question. How to I get 0.7.2 from github
 65 2013-02-11 16:13:57 <EskimoBob> make that version current or what eve it is in git speak :)
 66 2013-02-11 16:15:09 <Scrat> https://github.com/bitcoin/bitcoin/archive/0.7.2.zip
 67 2013-02-11 16:15:32 <EskimoBob> there is no way to do it with git?
 68 2013-02-11 16:15:41 <Scrat> if you want to clone it git://git.github.com/bitcoin/bitcoin/tree/0.7.2
 69 2013-02-11 16:15:50 <rdponticelli> EskimoBob: Maybe what you're looking for is: git checkout v0.7.2
 70 2013-02-11 16:16:06 <rdponticelli> If you already are on a cloned repository
 71 2013-02-11 16:16:59 <EskimoBob> rdponticelli: yes. That's the word
 72 2013-02-11 16:17:12 <rdponticelli> ;)
 73 2013-02-11 16:17:19 <EskimoBob> git tag -l
 74 2013-02-11 16:17:47 <EskimoBob> $ git checkout v0.7.2
 75 2013-02-11 16:17:51 <EskimoBob> is that correct?
 76 2013-02-11 16:18:00 <rdponticelli> Should be
 77 2013-02-11 16:18:00 <sipa> EskimoBob: yes
 78 2013-02-11 16:18:13 <EskimoBob> thank you for your help
 79 2013-02-11 16:20:35 <EskimoBob> "for production use" do you mean mining? Is it OK to use it for moving millions? ;)
 80 2013-02-11 16:21:38 <EskimoBob> fuck: A fatal error occurred. Bitcoin can no longer continue safely and will quit.
 81 2013-02-11 16:21:40 <EskimoBob> EXCEPTION: St13runtime_error
 82 2013-02-11 16:21:42 <EskimoBob> CLevelDB(): error opening database environment IO error: lock /home/mart/.bitcoin/blocks/index/LOCK: Resource temporarily unavailable
 83 2013-02-11 16:21:44 <EskimoBob> bitcoin in Runaway exception
 84 2013-02-11 16:22:48 <EskimoBob> that was stupid, I guess back to downloading that fkn 4 Gb monster files
 85 2013-02-11 16:23:18 <EskimoBob> and it not my PC :(
 86 2013-02-11 16:24:09 <sipa> EskimoBob: sounds like it's already running
 87 2013-02-11 16:55:35 <EskimoBob> it sure did some damage. 14K bloks needs to be downloaded
 88 2013-02-11 16:59:08 <ne0futur> EskimoBob: diskspace ?
 89 2013-02-11 17:01:56 <EskimoBob> ne0futur: no, I started 0.8.0rc1 qt while 0.7.something was running with -daemon
 90 2013-02-11 17:03:07 <EskimoBob> so now its sucking down blocks from last 100 days
 91 2013-02-11 17:04:24 <ne0futur> arf
 92 2013-02-11 17:05:18 <EskimoBob> and at this rate, it will take about 237 min...
 93 2013-02-11 17:05:19 <discrete> who hear is mining on bitcoin.cz
 94 2013-02-11 17:05:35 <EskimoBob> or 4 hours
 95 2013-02-11 17:05:58 <discrete> I am not able to retrieve work from RPC server
 96 2013-02-11 17:06:11 <discrete> http://api.bitcoin.cz:8332
 97 2013-02-11 17:06:24 <discrete> could it be that I am new to the pool
 98 2013-02-11 17:06:44 <discrete> I am using rpcminer
 99 2013-02-11 17:07:09 <discrete> can someone help me out?
100 2013-02-11 17:13:03 <Eliel_> discrete: you might have more luck asking on the pool's own irc-channel.
101 2013-02-11 17:16:27 <ProfMac> ;ident ProfMac
102 2013-02-11 17:23:31 <ProfMac> ;;ident profmac
103 2013-02-11 17:23:32 <gribble> Nick 'profmac', with hostmask 'ProfMac!42ea7b1c@gateway/web/freenode/ip.66.234.123.28', is identified as user ProfMac, with GPG key id 4B238EBE02617AB2, key fingerprint 6AEA1BC998E702854BD3BD0C4B238EBE02617AB2, and bitcoin address 1fdAs2wtgAUbNmaLMeFrABQUpBSjhLhV8
104 2013-02-11 17:24:18 <ProfMac> does anyone have a utility that will dump peers.dat to a human readable file?
105 2013-02-11 17:56:09 <jgarzik> ProfMac: http://bitcoin.sipa.be/seeds.txt
106 2013-02-11 17:56:49 <casascius> wondering if I can direct some bitcointalk mod attention to the following thread: https://bitcointalk.org/index.php?topic=92969.660;topicseen
107 2013-02-11 17:56:54 <gmaxwell> jgarzik: well, thats not peers.dat...
108 2013-02-11 18:01:19 <casascius> specifically, the bitcoin gift certificate someone has created featuring what looks like a 6-year-old girl.  image likely problematic if it spreads
109 2013-02-11 18:02:55 <Luke-Jr> casascius: wrong place to report it
110 2013-02-11 18:05:17 <sipa> ProfMac: i once had a patch for bitcoind to dump peers.dat in a human-readable format
111 2013-02-11 18:05:22 <sipa> not sure if i still have it
112 2013-02-11 18:06:13 <jgarzik> casascius: not much forum modding going on in here.  We all think the forum is a ghetto ;p
113 2013-02-11 18:06:19 <jgarzik> casascius: try reporting it
114 2013-02-11 18:06:35 <casascius> reported a while back
115 2013-02-11 18:08:34 <casascius> it may well be a ghetto, having that image not circulate on the internet is in the project's best interest, anyone able to pull it ought to.  there, I've done my part =)
116 2013-02-11 18:09:19 <ProfMac> Thanks jgarzik.  I have also teleported to Ubuntu land and used grep 'version message' < debug.log to get a similar thing.  I will probably wade into the data structure of peers.dat and do my own dump utility if one does not exist.  Sigh, I haven't even compiled the reference, so this is not a 1 hour task.
117 2013-02-11 18:09:28 <Luke-Jr> casascius: theymos is big on not moderating
118 2013-02-11 18:09:44 <jgarzik> ProfMac: it's not an easy task, as you must grok the serialized data format
119 2013-02-11 18:09:46 <ProfMac> sipa, I'd love to know more about your patch.
120 2013-02-11 18:09:57 <jgarzik> ProfMac: peers.dat is serialized CAddrMan object
121 2013-02-11 18:11:07 <webb> so if i buy a 1500$ Bitcoin miner GPU i can make 200$ per month out of it?
122 2013-02-11 18:11:14 <ProfMac> I don't know cpp.  But the Satoshi client seems to be really well done, I almost understand it.  Only two more levels to go before I grok.
123 2013-02-11 18:11:50 <Luke-Jr> webb: doubtful
124 2013-02-11 18:12:57 <discrete> - Im finally running multiple workers
125 2013-02-11 18:13:13 <discrete> is 240 mhash/s decent?
126 2013-02-11 18:13:31 <discrete> will I be able to generate 2-3 bitcoins a month?
127 2013-02-11 18:15:17 <Luke-Jr> 
ACTION redirects webb and discrete over to #bitcoin-mining

128 2013-02-11 18:15:46 <sipa> ;;genrate 240
129 2013-02-11 18:15:47 <gribble> The expected generation output, at 240.0 Mhps, given difficulty of 3275464.58657, is 0.0368489906557 BTC per day and 0.00153537461065 BTC per hour.
130 2013-02-11 18:17:42 <discrete> thanks!
131 2013-02-11 18:18:17 <discrete> so that puts me at 1 btc per month
132 2013-02-11 18:18:18 <discrete> lol
133 2013-02-11 18:18:38 <discrete> :'(
134 2013-02-11 18:19:06 <sipa> discrete: also, if ASIC usage starts soon, global hash rate may rise very significantly
135 2013-02-11 18:19:12 <sipa> making GPU mining even less profitable
136 2013-02-11 18:44:16 <velige> hi jgarzik, how are you?
137 2013-02-11 18:45:19 <velige> and luke-jr too. how is our luke-jr?
138 2013-02-11 18:45:45 <velige> my two favorite people
139 2013-02-11 18:47:57 <velige> aren't you two feeling well today? isn't everybody here ok with them? don't we all love them?
140 2013-02-11 18:49:17 <velige> i am a big luke-jr and jgarzik fan
141 2013-02-11 18:49:28 <velige> i even got an autograph
142 2013-02-11 18:50:10 <jgarzik> velige: on topic, please
143 2013-02-11 18:53:08 <discrete> ;;genrate 3026000
144 2013-02-11 18:53:09 <gribble> The expected generation output, at 3026000.0 Mhps, given difficulty of 3275464.58657, is 464.604357184 BTC per day and 19.3585148827 BTC per hour.
145 2013-02-11 19:11:05 <discrete> I have access to free electricity
146 2013-02-11 19:11:25 <sipa> do you also have free hardware?
147 2013-02-11 19:11:36 <discrete> nope wanna invest in ASICs
148 2013-02-11 19:11:38 <discrete> lol
149 2013-02-11 19:11:57 <discrete> or is it to late
150 2013-02-11 19:12:14 <discrete> is the marketplace taken over by the bigpools
151 2013-02-11 19:12:25 <discrete> and the leaders within
152 2013-02-11 19:12:29 <gmaxwell> 0_o
153 2013-02-11 19:12:43 <gmaxwell> This conversation would probably be better had in #bitcoin.
154 2013-02-11 19:14:24 <gmaxwell> 
ACTION invites discrete to #bitcoin

155 2013-02-11 20:57:16 <ProfMac> ident
156 2013-02-11 20:57:27 <gribble> You are identified as user ProfMac, with GPG key id 4B238EBE02617AB2, key fingerprint 6AEA1BC998E702854BD3BD0C4B238EBE02617AB2, and bitcoin address 1fdAs2wtgAUbNmaLMeFrABQUpBSjhLhV8
157 2013-02-11 20:57:27 <ProfMac> ;; ident
158 2013-02-11 21:13:11 <MC1984> 100 blocks to go and bitcoin is not really pegging my CPU
159 2013-02-11 21:13:14 <MC1984> is that normal
160 2013-02-11 21:14:27 <MC1984> thrashing HDD like a cat o nine tails though
161 2013-02-11 21:14:34 <sipa> 0.8?
162 2013-02-11 21:14:42 <MC1984> c1
163 2013-02-11 21:14:45 <MC1984> rc1
164 2013-02-11 21:16:22 <johnywin> -dev
165 2013-02-11 21:17:38 <johnywin> #bitcoin-dev
166 2013-02-11 21:17:44 <TD> 
ACTION brings up a new 0.8 node on a bitvps

167 2013-02-11 21:18:01 <TD> let's hope it doesn't crater this server like it did for my last one :(
168 2013-02-11 21:18:23 <MC1984> also i kp forgetting to ask, does filesystem fragmentation have a great effect on IBD
169 2013-02-11 21:19:32 <johnywin> #bitcoin-dev
170 2013-02-11 21:20:17 <MC1984> umm yes you ae here bro
171 2013-02-11 21:21:04 <MC1984> MFW ive got 4gb free space left on this drive and im stuffing it full of others peoples gambling habits
172 2013-02-11 21:25:06 <TD> sipa: what's up with the renamed directories?
173 2013-02-11 21:27:26 <TD> sipa: btw one thing i thought is that nodes shouldn't listen during ibd
174 2013-02-11 21:27:37 <MC1984> coins was pretty ambiguous
175 2013-02-11 21:27:46 <MC1984> like "wait, MY coins?"
176 2013-02-11 21:40:50 <gmaxwell> TD: they don't announce themselves during the IBD.
177 2013-02-11 21:40:58 <TD> hm
178 2013-02-11 21:41:08 <gmaxwell> Not exactly the same as not listening, but for a node which wasn't previously announced it's equal.
179 2013-02-11 21:41:13 <TD> i sometimes see nodes that are a long way beyind appear in dns seeds
180 2013-02-11 21:41:16 <TD> perhaps they were reset
181 2013-02-11 21:41:30 <gmaxwell> TD: There are nodes that are simply stuck.
182 2013-02-11 21:41:33 <gmaxwell> They'll never recover.
183 2013-02-11 21:41:42 <TD> i saw one that was 0.8
184 2013-02-11 21:41:52 <gmaxwell> Sure??? nodes get stuck due to disk corruption too.
185 2013-02-11 21:42:01 <gmaxwell> well, hm. that one might have just be reset as you say.
186 2013-02-11 21:46:03 <sipa> TD: there was an incompatible-with-unreleased-code change that was made anyway (drop support for non-checksummed undo files), so that was an opportunity to change the confusing names
187 2013-02-11 21:46:09 <TD> ok
188 2013-02-11 21:47:07 <MC1984> Connect 692 transactions: 10435.01ms (15.079ms/tx, 9.145ms/txin)
189 2013-02-11 21:48:27 <sipa> MC1984: is that single-threaded?
190 2013-02-11 21:48:44 <MC1984> yes
191 2013-02-11 21:49:06 <sipa> do you have similar ms/txin numbers during reindex?
192 2013-02-11 21:49:31 <MC1984> i dont know yet
193 2013-02-11 21:49:44 <MC1984> its after the checkpoint
194 2013-02-11 21:49:52 <sipa> during reindex, after the last checkpoint, i mean
195 2013-02-11 21:50:54 <MC1984> ill reindex overnight once the sync is finished
196 2013-02-11 21:51:02 <sipa> ok
197 2013-02-11 21:51:15 <MC1984> are those numbers bad?
198 2013-02-11 21:51:22 <sipa> if that 9ms/txin is (mostly) due to signature checking, it's really slow
199 2013-02-11 21:51:45 <MC1984> wellbitcoin is hammering the cpu now
200 2013-02-11 21:52:01 <sipa> i see around 0.6ms/txin here
201 2013-02-11 21:52:14 <TD> does -par get set automatically, or do i have to set it myself?
202 2013-02-11 21:52:29 <sipa> TD: by default it's equal to the number of detected cores
203 2013-02-11 21:52:33 <TD> ok, cool
204 2013-02-11 21:52:52 <MC1984> what would cause it to be that slow
205 2013-02-11 21:52:58 <sipa> MC1984: what CPU?
206 2013-02-11 21:52:59 <MC1984> this cpu is a Dothan 1.6ghz
207 2013-02-11 21:53:30 <sipa> MC1984: can you find your cpu here: http://www.cpubenchmark.net/cpu_list.php ?
208 2013-02-11 21:54:39 <MC1984> that site just crashed my browser lol
209 2013-02-11 21:54:55 <sipa> i assume it's a pentium m
210 2013-02-11 21:55:23 <MC1984> penitum m dothan 1.6
211 2013-02-11 21:56:24 <sipa> not too surprising, if i see the benchmarks on that site
212 2013-02-11 21:56:55 <MC1984> what cpu have you got|
213 2013-02-11 21:56:57 <Scrat> ancient in-order cpu
214 2013-02-11 21:57:06 <Scrat> their IPC is incredibly low
215 2013-02-11 21:57:17 <MC1984> its not in order
216 2013-02-11 21:57:17 <sipa> naively interpreting the per-core performance numbers there, i expect your cpu to be around 7 times slower
217 2013-02-11 21:58:06 <MC1984> is bitcoin optimised for a certain amount of cache
218 2013-02-11 21:58:15 <MC1984> this has got 2mb whih is still respectable
219 2013-02-11 21:58:21 <sipa> not really
220 2013-02-11 21:58:28 <Scrat> MC1984: you're right. but it's scary how much juice per mhz you get today compared to 8 years ago
221 2013-02-11 21:58:31 <sipa> (optimized for a certain amount)
222 2013-02-11 21:58:56 <MC1984> what i the verification engine fit in the cache?
223 2013-02-11 21:59:03 <MC1984> mite b cool
224 2013-02-11 22:00:27 <MC1984> i remember people going bananas at the first 512kb cache chips back in the day, because a seti@home work unit fit completely in it
225 2013-02-11 22:00:29 <MC1984> heh
226 2013-02-11 22:01:51 <MC1984> well this is a pain in the ass, but not so much if it was being done low priority in the background
227 2013-02-11 22:01:55 <MC1984> hint hint
228 2013-02-11 22:02:24 <sipa> see #2199
229 2013-02-11 22:03:04 <MC1984> Verify 2608 txins: 26257.76ms (10.068ms/txin) 26 seconds lawd
230 2013-02-11 22:11:03 <serp> hello fellers
231 2013-02-11 22:12:19 <amiller> sup serp
232 2013-02-11 22:12:20 <serp> I installed bitcoin-qt 0.8 fresh on a win7 machine.. on day 3 of syncing.. i was questioning whether this is still expected given the updates with bloom filtering and db improvements
233 2013-02-11 22:13:07 <sipa> bloom filtering won't help you - it's for light clients fetching transactions from you
234 2013-02-11 22:13:18 <sipa> and 3 days is certainly unexpectedly slow
235 2013-02-11 22:13:40 <MC1984> i thought i had it bad
236 2013-02-11 22:13:42 <serp> that was   normal for prior versions for me
237 2013-02-11 22:13:48 <serp> at the same hardware
238 2013-02-11 22:14:20 <serp> at least going from scratch
239 2013-02-11 22:15:47 <sipa> what hardware/disk/network connection?
240 2013-02-11 22:17:28 <serp> intel core 2 duo, 4gb ram, probably a 7200 rpm disk... network is pretty fast cable (at least for US)
241 2013-02-11 22:18:30 <sipa> and is it making progress?
242 2013-02-11 22:18:35 <sipa> how far are you?
243 2013-02-11 22:18:38 <serp> yes it's getting close now
244 2013-02-11 22:18:48 <serp> i can sit and watch it and it'll maek progress
245 2013-02-11 22:18:51 <sipa> how many left?
246 2013-02-11 22:18:59 <serp> 5034 blocks
247 2013-02-11 22:19:33 <sipa> ;;calc 216116-[bc,blocks]
248 2013-02-11 22:19:34 <gribble> -4617
249 2013-02-11 22:20:08 <sipa> have you tried exiting and restarting?
250 2013-02-11 22:20:22 <serp> i have not done that.. just started it and let it go
251 2013-02-11 22:20:27 <Scrat> serp: something wrong with your system. my laptop has the same specs and it took around 7 hrs
252 2013-02-11 22:20:54 <sipa> you may be unlucky and got a slow peer to download from
253 2013-02-11 22:21:08 <sipa> the sync mechanism is pretty stupid and sometimes gets confused
254 2013-02-11 22:21:26 <sipa> stopping and restarting (unfortunately) really helps sometimes
255 2013-02-11 22:21:33 <serp> k i will try
256 2013-02-11 22:22:02 <sipa> Scrat: i would even consider that ridiculously slow
257 2013-02-11 22:22:19 <serp> uhh after restart it's now down to 2700 left
258 2013-02-11 22:22:38 <sipa> it's only a guess
259 2013-02-11 22:22:49 <sipa> it doesn't know the real number of blocks until it has them
260 2013-02-11 22:23:08 <serp> right
261 2013-02-11 22:23:14 <serp> but it halfed when i restarted
262 2013-02-11 22:23:18 <serp> well almost
263 2013-02-11 22:23:30 <serp> oh now it's back up .. disregard :)
264 2013-02-11 22:24:58 <serp> what is a reasonable expectation for a full sync now on such a machine
265 2013-02-11 22:25:06 <serp> assuming a decent peer is chosen
266 2013-02-11 22:25:09 <Scrat> sipa: in a VM :)
267 2013-02-11 22:25:41 <sipa> serp: in perfect conditions, half an hour or so
268 2013-02-11 22:26:04 <sipa> though in practice i haven't seen any reports for less than an hour
269 2013-02-11 22:29:05 <serp> what are the light clients... is this/will this be built into bitcoin-qt or is it to support like electrum?  or perhaps something different?
270 2013-02-11 22:29:52 <sipa> electrum is technically not a real client, as it doesn't connect to the network (it connects to a special server)
271 2013-02-11 22:30:11 <sipa> though it certainly classifies as light
272 2013-02-11 22:30:33 <sipa> multibit and bitcoin wallet for android are light clients, and they will probably soon support bloom filtering
273 2013-02-11 22:30:49 <sipa> bitcoin-qt implements a full node, which is the opposite of a light client
274 2013-02-11 22:30:50 <Scrat> will an SPV client gain anything by storing the UXTO database?
275 2013-02-11 22:31:04 <sipa> an SPV client doesn't verify transactions
276 2013-02-11 22:31:12 <sipa> no point for it to have an UTXO database
277 2013-02-11 22:31:19 <Scrat> ye
278 2013-02-11 22:32:10 <gmaxwell> By definition a SPV client does "simplified payment validation" it never checks if txn are valid??? it only checks that they're in a chain??? hopefully a valid chain (but they wouldn't know??? except by hoping that there is a unique longest chain and that its valid)
279 2013-02-11 22:33:15 <sipa> 
ACTION benchmarks a sync from scratch over the internet

280 2013-02-11 23:07:02 <Scrat> bleh, wanted to do a fast sync on a 2 x octacore xeon and it just stood there with 8 connections using <1% cpu
281 2013-02-11 23:07:10 <Scrat> addnode doesn't do anything
282 2013-02-11 23:07:54 <sipa> if you want a fast sync, use -connect
283 2013-02-11 23:08:06 <sipa> (i.e., a single fast peer)
284 2013-02-11 23:08:35 <sipa> and addnode won't help you get more than 8 outgoing connections
285 2013-02-11 23:09:53 <Scrat> must've got 8 dialups
286 2013-02-11 23:09:58 <gmaxwell> Scrat: connections has ~nothing to do with sync speed.  You can -connect a known fast node to pull fast, or loadblock the data from a blockchain file.
287 2013-02-11 23:10:16 <gmaxwell> s/8/1/
288 2013-02-11 23:11:39 <midnightmagic> there is a bootstrap thing jgarzik is seeding. https://bitcointalk.org/index.php?topic=117982.0
289 2013-02-11 23:11:49 <midnightmagic> so you could bootstrap it with a torrent download.
290 2013-02-11 23:11:53 <midnightmagic> :-D
291 2013-02-11 23:12:08 <Scrat> I know, was just testing unaided sync
292 2013-02-11 23:12:11 <gmaxwell> though good luck with the torrent, people frequently complain that it takes forever to download.
293 2013-02-11 23:12:18 <Scrat> jgarzik needs to upload latest one
294 2013-02-11 23:13:39 <midnightmagic> 0.8.x is fast enough that an eeepc with 2GB ram (701 model no less) can still be a full node. :)
295 2013-02-11 23:14:29 <Scrat> well average tps is <1
296 2013-02-11 23:14:33 <lianj> midnightmagic: that model has only 4gigs or diskspace :P but yea
297 2013-02-11 23:16:58 <midnightmagic> lianj: Mine has 2G ram, and I was using an external drive.
298 2013-02-11 23:18:24 <phantomcircuit> gmaxwell, the problem is the default is dht only
299 2013-02-11 23:18:35 <phantomcircuit> which is basically useless when there's only 8 peers
300 2013-02-11 23:19:03 <MC1984> you know you can pass trackers in magent uri
301 2013-02-11 23:19:40 <gmaxwell> phantomcircuit: but DHT's fix everything!
302 2013-02-11 23:19:45 <phantomcircuit> yes im aware but the magnet link in the first post doesn't have any
303 2013-02-11 23:19:52 <phantomcircuit> so most of the peers are only on dht
304 2013-02-11 23:20:00 <phantomcircuit> and the link that does lists like 8 trackers
305 2013-02-11 23:20:02 <MC1984> dht works
306 2013-02-11 23:20:05 <MC1984> eventually
307 2013-02-11 23:20:14 <phantomcircuit> most clients will skip a tracker if it returns 0 peers
308 2013-02-11 23:20:25 <gmaxwell> It was intentionally a trackerless torrent.
309 2013-02-11 23:20:37 <phantomcircuit> so with only a handful of peers and a handful of trackers you can basically end up in a huge ridiculous loop
310 2013-02-11 23:20:52 <phantomcircuit> gmaxwell, im aware
311 2013-02-11 23:21:02 <phantomcircuit> just saying it's not super effective with a small number of peers
312 2013-02-11 23:21:12 <MC1984> it should still sort itself out eventually
313 2013-02-11 23:21:21 <phantomcircuit> eventually can be hours
314 2013-02-11 23:21:29 <MC1984> only if theres connectivity issues will it never get the torrent
315 2013-02-11 23:21:32 <gmaxwell> Or never, depending on which software you're using.
316 2013-02-11 23:21:32 <phantomcircuit> and most people just assume it wont work after a few minutes tops
317 2013-02-11 23:21:44 <MC1984> so fuck those people
318 2013-02-11 23:21:46 <HM> j
319 2013-02-11 23:21:49 <MC1984> they can IBD like chumps
320 2013-02-11 23:22:06 <MC1984> its not supposed to be widespread anyway
321 2013-02-11 23:22:20 <midnightmagic> MC1984: what's not supposed to be widespread?
322 2013-02-11 23:22:30 <gmaxwell> e.g. rtorrent??? which works fine on many popular DHT'ed torret will happily go weeks without getting anything.
323 2013-02-11 23:22:33 <MC1984> the bootstrap
324 2013-02-11 23:22:43 <gmaxwell> MC1984: people were advocating making it _the_ only way to bootstrap bitcoin.
325 2013-02-11 23:23:05 <phantomcircuit> gmaxwell, yeah the only way to make rtorrent work with dht is to seed the bootstrap with magnet links that contain trackers
326 2013-02-11 23:23:06 <MC1984> and thats why development is not a democracy
327 2013-02-11 23:23:30 <MC1984> fwiw uttoernt seems to have made some dht changes reently
328 2013-02-11 23:23:34 <midnightmagic> MC1984: Why isn't it supposed to be widespread?
329 2013-02-11 23:23:39 <MC1984> i got 2000 nodes once
330 2013-02-11 23:23:55 <MC1984> the chagnelog said something about improving metadata time
331 2013-02-11 23:23:58 <gavinandresen> 
ACTION grumbles about boost::filesystem incompatibilities biting him AGAIN....

332 2013-02-11 23:23:58 <gmaxwell> phantomcircuit: rtorrent has dht support, and it can work??? it's just super unreliable with small numbers of nodes it seems.
333 2013-02-11 23:24:12 <phantomcircuit> gmaxwell, the problem is it doesn't have any bootstrap method
334 2013-02-11 23:24:22 <MC1984> midnightmagic cos it breaks the trust model or somethin
335 2013-02-11 23:24:28 <gmaxwell> phantomcircuit: Yes, even if you _manually_ boostrap the DHT it doesn't work right.
336 2013-02-11 23:24:28 <MC1984> also its cheating
337 2013-02-11 23:24:34 <midnightmagic> MC1984: ...
338 2013-02-11 23:24:35 <phantomcircuit> so you pretty much have to go start downloading some popular content to seed the bootstrap
339 2013-02-11 23:24:46 <phantomcircuit> the manual add_peer thing doesn't work
340 2013-02-11 23:24:52 <phantomcircuit> i dont think it ever has
341 2013-02-11 23:25:13 <gmaxwell> May explain why it seemed inconsistent to me: worked on some hosts not on others.
342 2013-02-11 23:25:16 <MC1984> sounds like rtorrent DHT is just fucked up tbh
343 2013-02-11 23:25:27 <MC1984> ive not once had an issue with dht in bittorent
344 2013-02-11 23:25:31 <gmaxwell> MC1984: the bootstrap doesn't impact security in the least.
345 2013-02-11 23:25:42 <Scrat> transmission does DHT like a pro
346 2013-02-11 23:25:49 <gmaxwell> So long as the normal stuff still actually works reliably.
347 2013-02-11 23:25:51 <phantomcircuit> MC1984, it works well once you've got the peer database seeded
348 2013-02-11 23:25:55 <MC1984> well its still cheating
349 2013-02-11 23:25:58 <phantomcircuit> but actually making that happen isn't trivial
350 2013-02-11 23:26:04 <gmaxwell> Scrat: yea, go download the bitcoin torrent and report back how long it took you.
351 2013-02-11 23:26:07 <phantomcircuit> and there's some setting you have to add for it to survive a restart
352 2013-02-11 23:26:14 <phantomcircuit> which im guessing nobody has set
353 2013-02-11 23:26:16 <moore> have you looked a libtorrent http://www.rasterbar.com/products/libtorrent/
354 2013-02-11 23:26:20 <MC1984> doesnt rtorrent use the normal bootstrap nodes?
355 2013-02-11 23:26:30 <phantomcircuit> MC1984, no it doesn't that's my point
356 2013-02-11 23:26:36 <phantomcircuit> it doesn't do http bootstrap for dht
357 2013-02-11 23:26:44 <phantomcircuit> which utorrent and everybody else does
358 2013-02-11 23:26:48 <phantomcircuit> (they all use the utorrent node)
359 2013-02-11 23:26:55 <gmaxwell> MC1984: it's not??? it really has the same security model.  It would be a fine thing to use if bittorrent had actually solved the problems we haven't solved, but for the most part it hasn't.
360 2013-02-11 23:27:32 <MC1984> it still requires someone go and make a new bootstrap once in a while
361 2013-02-11 23:27:39 <Scrat> gmaxwell: just did, and it took 20 seconds
362 2013-02-11 23:28:03 <phantomcircuit> gmaxwell, on vps @ momentovps im getting about 20mbps down
363 2013-02-11 23:28:06 <MC1984> if you guys figure out a way to put bittorrent into satoshi and make it produce a new bootstrap automatically then great
364 2013-02-11 23:28:14 <MC1984> sounds like a pain in hte ass though
365 2013-02-11 23:28:19 <phantomcircuit> either other peers are all home connections or they're throttled
366 2013-02-11 23:28:22 <moore> if you have a wish list for bittorrent features I know people there
367 2013-02-11 23:28:30 <midnightmagic> MC1984: You should code up a DHT participant and submit a patch.
368 2013-02-11 23:28:58 <phantomcircuit> the only reason to use the bittorrent dht is to enable random ip scanning
369 2013-02-11 23:29:01 <MC1984> if i could i would just to troll greg
370 2013-02-11 23:29:09 <phantomcircuit> (yes there are enough peers that you can do that and actually succeed)
371 2013-02-11 23:29:22 <midnightmagic> MC1984: The awesome part would be if it were well-written, clean, and worked well.
372 2013-02-11 23:29:34 <midnightmagic> MC1984: But you have to assume in advance that it won't be accepted. :)
373 2013-02-11 23:29:37 <phantomcircuit> otherwise piggybacking on the bittorrent dht is just an excellent way for some journalist to write a horrible article
374 2013-02-11 23:30:14 <MC1984> phantomcircuit the bittorrent dht is THE dht
375 2013-02-11 23:30:26 <MC1984> except for the weird azureus one
376 2013-02-11 23:30:45 <phantomcircuit> MC1984, my point is that it doesn't add much to the bootstrap for bitcoin
377 2013-02-11 23:31:10 <phantomcircuit> since the bootstrap for the bittorrent dht is simply a fixed dns entry
378 2013-02-11 23:31:32 <MC1984> well
379 2013-02-11 23:31:38 <MC1984> youg otta start somewhere
380 2013-02-11 23:31:48 <phantomcircuit> that's pointless
381 2013-02-11 23:32:01 <phantomcircuit> actually it's worse than pointless
382 2013-02-11 23:32:11 <MC1984> wait what are you talkin about
383 2013-02-11 23:32:15 <phantomcircuit> your linking all the political failure modes of bittorrent with bitcoin and vice versa
384 2013-02-11 23:32:39 <phantomcircuit> "THEYRE JUST PIRATES!"
385 2013-02-11 23:32:45 <phantomcircuit> "THEYRE JUST PRINTING FAKE MONEY!"
386 2013-02-11 23:32:51 <MC1984> oh ffs
387 2013-02-11 23:32:54 <MC1984> its just technology
388 2013-02-11 23:32:54 <Scrat> 20 seconds to grab all metadata obviously
389 2013-02-11 23:33:01 <Scrat> luck?
390 2013-02-11 23:33:05 <phantomcircuit> uh huh
391 2013-02-11 23:33:14 <MC1984> hack journos gonna hack
392 2013-02-11 23:33:20 <phantomcircuit> Scrat, 20 seconds is slow for grabbing metadata
393 2013-02-11 23:33:36 <Scrat> re: bootstrap torrent
394 2013-02-11 23:33:40 <phantomcircuit> but yes it's fast in a lot of clients
395 2013-02-11 23:33:46 <phantomcircuit> but it's not in a lot of others
396 2013-02-11 23:34:16 <MC1984> well if hed make a new one i will seed it
397 2013-02-11 23:34:47 <MC1984> i heard the old bootstrap actually doesnt work anymore in 0.8
398 2013-02-11 23:35:09 <phantomcircuit> gmaxwell, is that the right ip?
399 2013-02-11 23:35:17 <phantomcircuit> MC1984, which one?
400 2013-02-11 23:35:38 <phantomcircuit> dnsseed/fixed fallback nodes/irc bootstrap (which i hope nobody is using)
401 2013-02-11 23:35:50 <MC1984> bootstrap.dat
402 2013-02-11 23:36:06 <gmaxwell> Scrat: it sure as heck didn't take 20 seconds to download a 4gbyte file...
403 2013-02-11 23:36:09 <phantomcircuit> i cant imagine why
404 2013-02-11 23:36:18 <phantomcircuit> bootstrap.dat is just skipping the network stuff
405 2013-02-11 23:36:23 <gmaxwell> MC1984: it works fine.
406 2013-02-11 23:36:29 <phantomcircuit> it follows the same code paths as if you received it from the network
407 2013-02-11 23:36:43 <freewil> is it legal to have a transaction with more than 1 output to the same address?
408 2013-02-11 23:36:45 <MC1984> oh i thought i read it gets stuck or something
409 2013-02-11 23:36:51 <phantomcircuit> gmaxwell, he's just talking about the metadata lookup
410 2013-02-11 23:37:00 <phantomcircuit> freewil, yes
411 2013-02-11 23:37:21 <phantomcircuit> i think you can even generate that with the current client
412 2013-02-11 23:37:39 <MC1984> lolwho you bannin
413 2013-02-11 23:37:43 <gmaxwell> freewil: it's perfectly legal, there is no exposed way to do it in the reference client.
414 2013-02-11 23:37:45 <MC1984> your old friend?
415 2013-02-11 23:37:46 <freewil> phantomcircuit, bitcoind gives me an error
416 2013-02-11 23:37:56 <freewil> thanks gmaxwell and phantomcircuit
417 2013-02-11 23:38:00 <gmaxwell> MC1984: stamit has gone nuts again.
418 2013-02-11 23:38:13 <phantomcircuit> lol that's funny it does give an error
419 2013-02-11 23:38:22 <phantomcircuit> i'll try and do it with rawtransaction
420 2013-02-11 23:38:29 <gmaxwell> phantomcircuit: nope, can't do it that way either.
421 2013-02-11 23:38:34 <MC1984> yeah i got some dickhead who pings me 100 times when i login to another server
422 2013-02-11 23:38:40 <MC1984> IRC weirdos take it to another level
423 2013-02-11 23:38:46 <freewil> really? cant you pretty much do anything with raw?
424 2013-02-11 23:38:46 <phantomcircuit> gmaxwell, really
425 2013-02-11 23:38:48 <phantomcircuit> that's unfortunate
426 2013-02-11 23:39:14 <gmaxwell> phantomcircuit: no, it's not??? discourages stupid behavior
427 2013-02-11 23:39:32 <phantomcircuit> i guess
428 2013-02-11 23:40:10 <phantomcircuit> lol
429 2013-02-11 23:40:22 <gmaxwell> (the precrime detectors are going strong today)
430 2013-02-11 23:41:03 <freewil> not sure what the use case would be
431 2013-02-11 23:41:28 <Scrat> gmaxwell: it's 2.5 gb, and I was talking about transmission's DHT implementation
432 2013-02-11 23:41:49 <MC1984> gmaxwell is that an autoban?
433 2013-02-11 23:41:51 <gmaxwell> Scrat: getting the metadata isn't enough??? you actually have to find some peers.
434 2013-02-11 23:42:00 <MC1984> every non-dictionary 5 character nick?
435 2013-02-11 23:42:13 <Scrat> 14 MB/s down, that good enough?
436 2013-02-11 23:42:41 <gmaxwell> MC1984: no, he's showing up and crapping on other channels. In the past he'd flood here too once banned elsewhere, so I'm just broadcasting.
437 2013-02-11 23:42:50 <gmaxwell> Scrat: Fair enough.
438 2013-02-11 23:43:11 <gmaxwell> Scrat: from now on I appoint you tech support for people who show up on IRC saying it doesn't work. :)
439 2013-02-11 23:45:36 <phantomcircuit> Scrat, you wouldn't happen to be in .nl would you
440 2013-02-11 23:47:30 <Scrat> phantomcircuit: correct
441 2013-02-11 23:48:24 <Scrat> (server is in NL that is)
442 2013-02-11 23:48:41 <phantomcircuit> Scrat, i just added a seeder at ovh on a 200 mbps line
443 2013-02-11 23:48:46 <phantomcircuit> so that was probably just my node
444 2013-02-11 23:49:18 <phantomcircuit> i guess i'll leave it running
445 2013-02-11 23:54:46 <Scrat> phantomcircuit: 23 peers
446 2013-02-11 23:54:56 <Scrat> can't see number of seeds (web interface :p)