1 2012-10-12 00:00:01 <jgarzik> sipa: if you restart with the magnet link I just posted, you should get going again
2 2012-10-12 00:00:11 * jgarzik 's node was seeding via IPv6, even
3 2012-10-12 00:00:18 <MC1984> surely there has to be a more recent seedbox software thing?
4 2012-10-12 00:00:42 <kjj_> probably, but until like an hour ago, I didn't really care about DHT
5 2012-10-12 00:01:05 <MC1984> well trackers are on the way out so
6 2012-10-12 00:01:50 <MC1984> so are torrent sites actually, the new hotness is search engines that mine the DHT directly for infohashes
7 2012-10-12 00:02:05 lggr has quit (Ping timeout: 244 seconds)
8 2012-10-12 00:02:56 <sipa> jgarzik: nothing yet
9 2012-10-12 00:03:51 davout has quit (Remote host closed the connection)
10 2012-10-12 00:04:40 <sipa> i hate not finding bugs when i want to sleep :(
11 2012-10-12 00:05:41 lggr has joined
12 2012-10-12 00:06:01 <gmaxwell> sipa: I hate it when I think I've almost got it and stay up a bit longer, and then I still haven't fixed it and the sun is rising.
13 2012-10-12 00:06:04 <gmaxwell> :(
14 2012-10-12 00:07:03 copumpkin has joined
15 2012-10-12 00:08:08 Mobius_ is now known as MobiusL
16 2012-10-12 00:08:10 <sipa> sunrise is ok, as long as you don't have to get up :)
17 2012-10-12 00:08:14 nathaniel has quit (Ping timeout: 255 seconds)
18 2012-10-12 00:08:35 <gmaxwell> haha
19 2012-10-12 00:08:40 nathaniel has joined
20 2012-10-12 00:08:52 <sipa> the problem is that now i can't tell whether there is a bug in ultraprune, or in my additional checks during disconnect
21 2012-10-12 00:11:52 lggr has quit (Ping timeout: 245 seconds)
22 2012-10-12 00:12:50 da2ce779 is now known as da2ce7
23 2012-10-12 00:14:13 optimator_ has quit (Ping timeout: 252 seconds)
24 2012-10-12 00:14:50 lggr has joined
25 2012-10-12 00:16:05 <jgarzik> bootstrap.dat should be seeding as before, <= 10pm EST (2 hours from now)
26 2012-10-12 00:16:23 <jgarzik> there are two nodes on there right now, but DHT is still laggard
27 2012-10-12 00:16:29 <jgarzik> maybe kjj_ can seed a bit, too ;p
28 2012-10-12 00:18:03 <sipa> as soon as i get it downloaded, i'll seed it :)
29 2012-10-12 00:20:16 <jgarzik> a hah, a peer. finally.
30 2012-10-12 00:20:23 lggr has quit (Ping timeout: 245 seconds)
31 2012-10-12 00:21:06 <jgarzik> and a peer on the other node. finally.
32 2012-10-12 00:21:23 graingert has quit (Ping timeout: 246 seconds)
33 2012-10-12 00:22:45 chappy has joined
34 2012-10-12 00:23:05 <Luke-Jr> gmaxwell: meh, I thought it wasn't any faster than network IBD?
35 2012-10-12 00:23:45 lggr has joined
36 2012-10-12 00:23:51 <sipa> currently -loadblock is certainly faster than IBD
37 2012-10-12 00:23:56 <sipa> even from localhost
38 2012-10-12 00:24:29 <midnightmagic> seeds:0 peers:0 ..
39 2012-10-12 00:24:29 <jgarzik> sipa: same here, IFF the input file is on a separate hard drive
40 2012-10-12 00:24:58 <jgarzik> midnightmagic: infohash 0bb0521942f586ed96203c6f4d136324756f8a9a ?
41 2012-10-12 00:25:02 <sipa> jgarzik: all my ultraprune benchmarks are done a laptop with a single non-SSD encrypted hard drive
42 2012-10-12 00:25:17 <sipa> loadblock is much faster than downloading from network :)
43 2012-10-12 00:25:48 <Impaler> new idea for proofing, I call it proof of network
44 2012-10-12 00:25:49 <sipa> just because the connection process isn't interrupted by downloading
45 2012-10-12 00:25:55 <jgarzik> my favorite import tends to be from 127.0.0.1
46 2012-10-12 00:26:53 <Impaler> have a long series of signatory wallets, each one produces a nonrandom hash that determines what subset of wallets are valid for the next signature
47 2012-10-12 00:28:06 <Impaler> say for example the next signatory address must match X characters/bits from the hash, the higher the difficulty the lower the percentage of all wallets that are valid signatories
48 2012-10-12 00:29:50 <Impaler> block bounces around accumulating signatures untill the original signatory meets the validity requirement to sign the end, any one can determine this and it is effectivly random
49 2012-10-12 00:29:59 lggr has quit (Ping timeout: 265 seconds)
50 2012-10-12 00:30:25 <MC1984> u wot m8
51 2012-10-12 00:32:25 <Impaler> thoughts?
52 2012-10-12 00:32:53 lggr has joined
53 2012-10-12 00:33:24 paraipan has joined
54 2012-10-12 00:33:40 BitcoinBaltar has joined
55 2012-10-12 00:33:57 <MC1984> im not qualified
56 2012-10-12 00:36:02 mmoya has quit (Ping timeout: 260 seconds)
57 2012-10-12 00:38:51 att has quit (Ping timeout: 255 seconds)
58 2012-10-12 00:39:22 lggr has quit (Ping timeout: 245 seconds)
59 2012-10-12 00:41:47 mmoya has joined
60 2012-10-12 00:42:17 lggr has joined
61 2012-10-12 00:44:18 ThomasV has joined
62 2012-10-12 00:47:27 bcb has joined
63 2012-10-12 00:47:46 <bcb> is backupwallet on linux essentially a mv function
64 2012-10-12 00:48:21 lggr has quit (Ping timeout: 265 seconds)
65 2012-10-12 00:48:37 ne0futur has quit (Read error: Operation timed out)
66 2012-10-12 00:48:44 ne0futur has joined
67 2012-10-12 00:48:45 ne0futur has quit (Changing host)
68 2012-10-12 00:48:45 ne0futur has joined
69 2012-10-12 00:50:36 chappy has quit (Quit: Page closed)
70 2012-10-12 00:51:43 lggr has joined
71 2012-10-12 00:54:07 da2ce7 has quit (Ping timeout: 276 seconds)
72 2012-10-12 00:58:25 lggr has quit (Ping timeout: 244 seconds)
73 2012-10-12 00:59:06 <gmaxwell> 17:11 < Luke-Jr> gmaxwell: meh, I thought it wasn't any faster than network IBD?
74 2012-10-12 00:59:43 <gmaxwell> torrent + boostrap is probably faster at least with ultraprune for most people just due to our network suckyness.
75 2012-10-12 01:00:16 <gmaxwell> e.g. your chosen peer disconnects.. and now you're doing nothing for a few minutes to a half hour before you start to pull again..
76 2012-10-12 01:00:26 <gmaxwell> then that peer goes away and you're stuck again.
77 2012-10-12 01:00:34 <gmaxwell> Or you pick some peer on 28.8k dialup in africa.
78 2012-10-12 01:00:56 <gmaxwell> though I note, DHT download of this thing is still not going for me
79 2012-10-12 01:01:15 mmoya has quit (Ping timeout: 260 seconds)
80 2012-10-12 01:01:40 <gmaxwell> with BDB backend there is less difference... the bdb part is so slow you're less likely to be bottlenecked by slow peers.
81 2012-10-12 01:02:10 lggr has joined
82 2012-10-12 01:03:02 darsk1ez has quit (Ping timeout: 246 seconds)
83 2012-10-12 01:04:49 one_zero has joined
84 2012-10-12 01:07:53 lggr has quit (Ping timeout: 245 seconds)
85 2012-10-12 01:09:28 eoss has quit (Remote host closed the connection)
86 2012-10-12 01:10:19 ThomasV has quit (Ping timeout: 252 seconds)
87 2012-10-12 01:10:48 <jgarzik> gmaxwell: still on rtorrent?
88 2012-10-12 01:11:16 da2ce7 has joined
89 2012-10-12 01:11:18 lggr has joined
90 2012-10-12 01:11:26 <gmaxwell> Yup. I know I've done trackerless torrents with it in the past, it seems to not be bootstraping onto the dht however.
91 2012-10-12 01:12:14 <jgarzik> gmaxwell: http://libtorrent.rakshasa.no/wiki/RTorrentUsingDHT seems to hint DHT defaults to off
92 2012-10-12 01:12:43 TigrBot has quit (Ping timeout: 260 seconds)
93 2012-10-12 01:13:02 <gmaxwell> yep, turned it on.. and it required a session directory
94 2012-10-12 01:13:18 <gmaxwell> ah, got it.
95 2012-10-12 01:13:27 <gmaxwell> well its now talking to the dht.
96 2012-10-12 01:13:34 <gmaxwell> I had to manually give it a dht seed node.
97 2012-10-12 01:13:44 <gmaxwell> And it didn't work from the commandline.
98 2012-10-12 01:14:17 <MC1984> why on earth does the IBD try to pull every block from one node?
99 2012-10-12 01:14:24 davout has joined
100 2012-10-12 01:14:24 davout has quit (Changing host)
101 2012-10-12 01:14:24 davout has joined
102 2012-10-12 01:14:33 <gmaxwell> I had to rtorrent -odht=on,session=/home/gmaxwell/.rtorrent and then in rtorrent press ctrl-x dht_add_node=router.bittorrent.com:6881 ... then quit and restart.
103 2012-10-12 01:14:44 <MC1984> why not connect to 8 and pull blocks round robin fashion
104 2012-10-12 01:15:15 <jgarzik> MC1984: hysterical raisins
105 2012-10-12 01:15:28 <MC1984> wut
106 2012-10-12 01:15:30 <jgarzik> MC1984: plenty of ideas floating around, sure
107 2012-10-12 01:15:41 bcb has quit (Quit: Page closed)
108 2012-10-12 01:15:59 <gmaxwell> MC1984: it just doesn't mostly because the fetching and validating are run sequentially and validating is inherently serial.
109 2012-10-12 01:16:07 <jgarzik> MC1984: historical reasons. that's the way they ended up, and there were higher priority issues to tackle first.
110 2012-10-12 01:16:17 <MC1984> oh
111 2012-10-12 01:16:21 darsk1ez has joined
112 2012-10-12 01:16:24 <gmaxwell> If they were decoupled so it downloaded with only hash validation in one thread and then followed behind validating that would work fine.
113 2012-10-12 01:16:39 agricocb has joined
114 2012-10-12 01:16:41 <gmaxwell> and would be enormously faster.
115 2012-10-12 01:16:54 <MC1984> stick it on the wishlist boys
116 2012-10-12 01:17:01 <gmaxwell> 17:36 < bcb> is backupwallet on linux essentially a mv function
117 2012-10-12 01:17:03 <gmaxwell> ^ no.
118 2012-10-12 01:17:36 <gmaxwell> it's closes and flushes the database so it can be safely copied without corruption, and copies that out to a new file.
119 2012-10-12 01:18:15 lggr has quit (Ping timeout: 255 seconds)
120 2012-10-12 01:18:36 <gmaxwell> jgarzik: so now I'm happily talking to the dht but it's not finding it yet still.
121 2012-10-12 01:18:57 davout has quit (Ping timeout: 245 seconds)
122 2012-10-12 01:19:24 <jgarzik> gmaxwell: using magnet link or torrent file from http://gtf.org/garzik/bitcoin/bootstrap.dat.torrent ?
123 2012-10-12 01:19:34 <MC1984> gmaxwell the DHT needs a bit of time to spin up
124 2012-10-12 01:19:35 <gmaxwell> magnet link
125 2012-10-12 01:19:43 <MC1984> usually 300 ish nodes
126 2012-10-12 01:20:09 <gmaxwell> well, it did a whole searching 18/25 nodes thing. I'll just leave it be and see if it gets going on its own.
127 2012-10-12 01:20:12 <jgarzik> magnet:?xt=urn:btih:0bb0521942f586ed96203c6f4d136324756f8a9a&dn=bootstrap.dat
128 2012-10-12 01:20:17 <gmaxwell> I'm mostly just fetching this to see if it works
129 2012-10-12 01:20:25 <gmaxwell> jgarzik: yep.
130 2012-10-12 01:20:31 <gmaxwell> my client shows * 0BB0521942F586ED96203C6F4D136324756F8A9A.meta
131 2012-10-12 01:20:39 <gmaxwell> which is what it shows before its resolved the torrent file.
132 2012-10-12 01:20:50 * jgarzik has no clue about the torrent upload process
133 2012-10-12 01:20:56 <MC1984> metalink?
134 2012-10-12 01:21:09 <MC1984> metalinks are different things
135 2012-10-12 01:21:11 <jgarzik> I was going to add the .torrent file to the forum post. It is ~24k and should be small enough.
136 2012-10-12 01:21:13 lggr has joined
137 2012-10-12 01:21:52 <jgarzik> s/torrent upload/torrent file -> magnet magic/
138 2012-10-12 01:22:11 <gmaxwell> MC1984: thats just how rtorrent works for magnet links.
139 2012-10-12 01:22:21 <MC1984> oh
140 2012-10-12 01:22:58 <MC1984> you know this magnet link busines is just a way to id a peer to pul the .torrent from, instead of a website
141 2012-10-12 01:23:01 <gmaxwell> 0BB0521942F586ED96203C6F4D136324756F8A9A is the key in the DHT for the torrent file's entry. And it'll fetch that then display it like a normal torrent file.
142 2012-10-12 01:23:53 <gmaxwell> "distributed systems are hard to make work right, lets just centeralize things!" :P
143 2012-10-12 01:23:58 <MC1984> yeah it displays the hash until its pulled the .torrent, whence it carries on as normal
144 2012-10-12 01:24:16 Impaler_ has joined
145 2012-10-12 01:24:21 <MC1984> lol can i qute you on that gmaxwell
146 2012-10-12 01:24:25 <gmaxwell> besides, if my connectivity to the dht is poor enough that I can't pull the torrent itself, peer discovery isn't going to work great.
147 2012-10-12 01:24:46 <gmaxwell> MC1984: I'm mocking the recourse to torrent files when the dht stuff is failtastic.
148 2012-10-12 01:24:52 <gmaxwell> (and trackers)
149 2012-10-12 01:25:11 <MC1984> not once have i ever had an issue with dht, it just werks
150 2012-10-12 01:25:20 <MC1984> but then i dont run rtorrent and stuff
151 2012-10-12 01:25:39 <gmaxwell> MC1984: a _lot_ of torrrents out there have a whole pile of trackers in them.
152 2012-10-12 01:26:50 <MC1984> does everyone have peer exchange enabled?
153 2012-10-12 01:27:10 yellowhat has quit (Quit: yellowhat)
154 2012-10-12 01:27:23 <MC1984> dht is kind of a half measure without pex i read
155 2012-10-12 01:27:38 <gmaxwell> In any case, I'll stop dorking with it, the purposes of me running is is mostly as an idiot test. If it won't work for me then it's not working super well.
156 2012-10-12 01:27:51 lggr has quit (Ping timeout: 260 seconds)
157 2012-10-12 01:27:51 Impaler has quit (Ping timeout: 260 seconds)
158 2012-10-12 01:30:15 lggr has joined
159 2012-10-12 01:31:12 da2ce7 has quit (Read error: Connection reset by peer)
160 2012-10-12 01:32:03 Impaler_ has quit (Remote host closed the connection)
161 2012-10-12 01:36:59 optimator_ has joined
162 2012-10-12 01:37:10 att has joined
163 2012-10-12 01:37:10 lggr has quit (Ping timeout: 244 seconds)
164 2012-10-12 01:38:19 graingert has joined
165 2012-10-12 01:40:43 lggr has joined
166 2012-10-12 01:46:51 da2ce7 has joined
167 2012-10-12 01:47:03 lggr has quit (Ping timeout: 245 seconds)
168 2012-10-12 01:49:51 optimator_ has quit (Remote host closed the connection)
169 2012-10-12 01:49:55 lggr has joined
170 2012-10-12 01:51:10 maaku has joined
171 2012-10-12 01:52:29 D34TH has quit (Read error: Connection reset by peer)
172 2012-10-12 01:52:38 freakazoid has quit (Ping timeout: 265 seconds)
173 2012-10-12 01:54:17 da2ce7 has quit (Ping timeout: 246 seconds)
174 2012-10-12 01:56:57 lggr has quit (Ping timeout: 255 seconds)
175 2012-10-12 01:57:56 <jgarzik> MC1984: I imagine far more do peer exchange than DHT, actually
176 2012-10-12 02:00:08 lggr has joined
177 2012-10-12 02:07:02 lggr has quit (Ping timeout: 252 seconds)
178 2012-10-12 02:10:15 lggr has joined
179 2012-10-12 02:16:14 <kjj_> very strange that there isn't a DHT proxy
180 2012-10-12 02:16:48 lggr has quit (Ping timeout: 265 seconds)
181 2012-10-12 02:18:32 <jgarzik> a poor spread of DHT takes a little while to heal, as the DHT is coded to prefer longer lived nodes... and there are a few occasions where you would rather they swapped them out of buckets more rapidly
182 2012-10-12 02:19:42 lggr has joined
183 2012-10-12 02:23:54 Joric has joined
184 2012-10-12 02:24:27 <kjj_> no, I mean I'm surprised that there isn't a daemon I can run that talks to DHT, and then presents a pseudo-tracker on a local port
185 2012-10-12 02:26:02 lggr has quit (Ping timeout: 245 seconds)
186 2012-10-12 02:26:36 MC1984 has quit (Quit: Leaving)
187 2012-10-12 02:26:54 <kjj_> oh god, the copyright blockchain thread managed to get even WORSE while I wasn't paying attention
188 2012-10-12 02:27:27 <jgarzik> kjj_: heh, yeah that kind of DHT proxy would be neat
189 2012-10-12 02:28:53 lggr has joined
190 2012-10-12 02:29:13 <Diablo-D3> lol
191 2012-10-12 02:32:21 AlexWaters has quit (Remote host closed the connection)
192 2012-10-12 02:34:07 Zarutian has quit (Quit: Zarutian)
193 2012-10-12 02:35:39 lggr has quit (Ping timeout: 255 seconds)
194 2012-10-12 02:37:06 <asciilifeform> Please comment: "Shitcoin: a Modest Proposal." (http://www.loper-os.org/?p=988)
195 2012-10-12 02:38:47 lggr has joined
196 2012-10-12 02:38:52 Joric has quit ()
197 2012-10-12 02:40:38 <kjj_> jgarzik: you going to post your torrent on the forums?
198 2012-10-12 02:41:08 <jgarzik> kjj_: I wanted to give it several hours to get healthy in the DHT
199 2012-10-12 02:41:17 <jgarzik> kjj_: only just finished uploading to my two seeds 29 minutes ago
200 2012-10-12 02:41:44 * jgarzik was just running azureus/vuze, and poking its DHT
201 2012-10-12 02:41:49 <kjj_> heh. if you'd included a tracker, I'd have had two seeds for you ready as soon as you started
202 2012-10-12 02:42:01 <jgarzik> looks like azureus sees the magnet link and downloads the torrent very quickly
203 2012-10-12 02:42:26 <graingert> jgarzik: where did you put the torrent?
204 2012-10-12 02:43:02 <jgarzik> graingert: magnet:?xt=urn:btih:0bb0521942f586ed96203c6f4d136324756f8a9a&dn=bootstrap.dat or http://gtf.org/garzik/bitcoin/bootstrap.dat.torrent
205 2012-10-12 02:43:12 <jgarzik> with the proviso that it does not include trackers... requires DHT
206 2012-10-12 02:43:27 <graingert> no trackers?
207 2012-10-12 02:43:28 <graingert> y
208 2012-10-12 02:43:29 fiesh has quit (Ping timeout: 246 seconds)
209 2012-10-12 02:44:51 lggr has quit (Ping timeout: 244 seconds)
210 2012-10-12 02:45:22 <graingert> jgarzik: adding trackers, it just works tm
211 2012-10-12 02:45:46 <kjj_> he was worried about it being a DOS target
212 2012-10-12 02:45:56 <graingert> kjj_: in what way?
213 2012-10-12 02:46:24 fiesh has joined
214 2012-10-12 02:46:25 <kjj_> bitcoin attracts a lot of attention from pricks
215 2012-10-12 02:47:04 <graingert> so if you download the torrent then you get DDOS
216 2012-10-12 02:47:17 <graingert> bull
217 2012-10-12 02:47:39 <kjj_> no, if you host the tracker for it, someone might
218 2012-10-12 02:47:39 lggr has joined
219 2012-10-12 02:47:54 <graingert> udp://tracker.openbittorrent.com:80 has enough oomph
220 2012-10-12 02:48:14 <kjj_> heh. I built mine with three public trackers embedded in it
221 2012-10-12 02:48:22 <kjj_> http://www.jerviss.org/bitcointalk.org/bootstrap_193000.dat.torrent
222 2012-10-12 02:48:42 <kjj_> please don't link that in the forums. it'll be bad enough when the logbot finds it
223 2012-10-12 02:51:03 <BlueMatt> yay, a bitcoin seed torrent...and no one has bothered to fix any of the many ibd issues...
224 2012-10-12 02:51:43 <kjj_> ahh, shit. don't use my link
225 2012-10-12 02:51:55 <graingert> BlueMatt a bitcoin seed torrent?
226 2012-10-12 02:52:01 <graingert> BlueMatt what does it seed?
227 2012-10-12 02:52:11 <BlueMatt> block data torrent, I meant
228 2012-10-12 02:52:16 <graingert> IC
229 2012-10-12 02:52:36 <graingert> I always thought it would be handy to piggy back off of trackers to get peers
230 2012-10-12 02:52:46 <kjj_> I just noticed that my two seeds don't understand UDP either. haha
231 2012-10-12 02:52:58 <BlueMatt> ie my usual "lets fix the root of the problem instead of hacking on more crap to work around it" complaint
232 2012-10-12 02:53:00 <graingert> kjj_: UTP
233 2012-10-12 02:53:40 <kjj_> graingert: no, UDP
234 2012-10-12 02:53:59 lggr has quit (Ping timeout: 246 seconds)
235 2012-10-12 02:54:07 <Diablo-D3> utp? wtf?
236 2012-10-12 02:54:08 <gmaxwell> BlueMatt: on the plus side the torrent doesn't appear to work for many (most?) people, without adding centeralization
237 2012-10-12 02:54:08 <graingert> uTP?
238 2012-10-12 02:54:17 <Diablo-D3> MICRO TOLIET PAPER
239 2012-10-12 02:54:22 <Diablo-D3> FOR UNUSUALLY SMALL ASSES
240 2012-10-12 02:54:26 <graingert> utorrent protocol
241 2012-10-12 02:55:06 <BlueMatt> gmaxwell: well, that wont stop people from using it...
242 2012-10-12 02:55:06 <Diablo-D3> graingert: er, which protocol is that?
243 2012-10-12 02:55:29 Eslbaer has joined
244 2012-10-12 02:55:35 <kjj_> well fuck, I guess my job for tomorrow is to get rtorrent
245 2012-10-12 02:55:38 <gmaxwell> BlueMatt: I was being silly. The torrent is harmless other than distracting us from fixing our networking code
246 2012-10-12 02:55:40 <graingert> Micro Transport Protocol or µTP (sometimes also uTP) is an open UDP-based variant of the BitTorrent peer-to-peer file sharing protocol intended to mitigate poor latency and other congestion control issues found in conventional BitTorrent over TCP, while providing reliable, ordered delivery.
247 2012-10-12 02:55:47 <gmaxwell> BlueMatt: it doesn't violate the security at all.
248 2012-10-12 02:55:58 <Diablo-D3> graingert: er
249 2012-10-12 02:56:10 <BlueMatt> gmaxwell: of course it doesnt violate bitcoin security, but it keeps us from having to fix the crap that is ibd logic...
250 2012-10-12 02:56:12 <Diablo-D3> I thought all clients already spoke that including the main torrent client?
251 2012-10-12 02:56:24 <kjj_> graingert: I'm talking about the tracker connection being over UDP instead of http, not the bulk transfer
252 2012-10-12 02:56:25 <gmaxwell> BlueMatt: it doesn't really, we're only seeding up to the highest checkpoint.
253 2012-10-12 02:56:35 * BlueMatt feels like hes become the old curmudgeon even more than Diablo-D3 here...
254 2012-10-12 02:56:42 dvide has quit ()
255 2012-10-12 02:56:45 <kjj_> BlueMatt: I don't think it'll help performance much either way
256 2012-10-12 02:56:46 <gmaxwell> BlueMatt: people will still suffer from IBD but a little less.
257 2012-10-12 02:57:03 <gmaxwell> (and we concurrently improved IBD a little)
258 2012-10-12 02:57:03 <BlueMatt> gmaxwell: it doesnt really what?
259 2012-10-12 02:57:06 <Diablo-D3> bluematt: wat
260 2012-10-12 02:57:14 <gmaxwell> BlueMatt: remove the need to fix ibd.
261 2012-10-12 02:57:21 <Diablo-D3> no one is more curmugeony than me
262 2012-10-12 02:57:26 lggr has joined
263 2012-10-12 02:57:37 <BlueMatt> gmaxwell: no, it doesnt, but it eases the pressure on doing it a bit more...
264 2012-10-12 02:57:38 <graingert> ibd?
265 2012-10-12 02:57:43 <BlueMatt> initial block download
266 2012-10-12 02:57:51 <graingert> ah
267 2012-10-12 02:57:58 da2ce7_d has quit (Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/)
268 2012-10-12 02:58:03 <graingert> yeah just torrent it it's fine LOLOLOLOL!!!11111oneone
269 2012-10-12 02:58:06 <BlueMatt> kjj_: it will help many people because of the crap logic we have now (got a slow peer? sucks for you!!!)
270 2012-10-12 02:58:07 TheEslbear has quit (Ping timeout: 245 seconds)
271 2012-10-12 02:58:08 AlexWaters has joined
272 2012-10-12 02:58:12 <gmaxwell> BlueMatt: A little. besides there are people who are just fixated on doing the download externally no matter how fast it is
273 2012-10-12 02:58:33 <gmaxwell> this gives them a way that doesn't give them a @!#! insecure index, and is still ultraprune compatible.
274 2012-10-12 02:58:40 <BlueMatt> gmaxwell: yep, which is why I always complain when people bring up torrents for ibd...
275 2012-10-12 02:59:08 <gmaxwell> BlueMatt: and magically ibd is not fixed yet.
276 2012-10-12 02:59:17 <gmaxwell> Although it has improved a little.
277 2012-10-12 02:59:33 <BlueMatt> gmaxwell: I dont call shipping a bootstrap.dat improved a little
278 2012-10-12 02:59:40 <BlueMatt> I call it taking a step backwards...
279 2012-10-12 03:00:06 <gmaxwell> Though we're now seeing a fairly high rate of reports of windows users who can't manage to not press the @#$#@ reset button once an hour getting corrupted block indexes and having to try to sync several times in the 12 hours its taking them.
280 2012-10-12 03:00:19 <gmaxwell> BlueMatt: er. Thats _not_ what I'm talking about.
281 2012-10-12 03:00:37 <gmaxwell> Improved a little is that we don't pick braindead nodes as our pull point.
282 2012-10-12 03:00:43 <gmaxwell> (the height check)
283 2012-10-12 03:00:47 <BlueMatt> (sorry doing like 5 things at once, just got back in after cloning to a new drive, trying to test what was lost and copy it back off backups, etc, etc)
284 2012-10-12 03:00:58 <gmaxwell> s'okay.
285 2012-10-12 03:01:22 <gmaxwell> ultraprune also greatly improves IBD and will be pulled any day now.
286 2012-10-12 03:01:28 <BlueMatt> gmaxwell: oh, we actually do height check now? well thats somewhat better, but that doesnt really solve the incredibly slow ibd anyway...
287 2012-10-12 03:01:50 <gmaxwell> well it prevent the IBD for getting stuck over and over again because it's picked a stupid peer that can't give it blocks anyways.
288 2012-10-12 03:02:12 <BlueMatt> true
289 2012-10-12 03:03:01 <gmaxwell> actually fixing this stuff requires a big restructure for reverse header sync, and it doesn't make sense to even start on an implementation until ultraprune is merged
290 2012-10-12 03:03:06 <gmaxwell> because it needs structural changes.
291 2012-10-12 03:03:59 lggr has quit (Ping timeout: 260 seconds)
292 2012-10-12 03:06:42 lggr has joined
293 2012-10-12 03:07:26 graingert has quit (Read error: Connection reset by peer)
294 2012-10-12 03:07:41 da2ce7 has joined
295 2012-10-12 03:11:00 att has quit (Ping timeout: 255 seconds)
296 2012-10-12 03:13:27 lggr has quit (Ping timeout: 255 seconds)
297 2012-10-12 03:15:07 ForceMajeure has quit (Ping timeout: 248 seconds)
298 2012-10-12 03:16:45 lggr has joined
299 2012-10-12 03:22:36 <BlueMatt> true...if only we had some kind of setup through which major refactors of one set of code didn't effect major refactors of other parts...
300 2012-10-12 03:23:21 lggr has quit (Ping timeout: 255 seconds)
301 2012-10-12 03:23:21 <kjj_> ok, I'm out. long drive home
302 2012-10-12 03:23:36 <gmaxwell> er. block connection is as the heart of that kind of change.
303 2012-10-12 03:24:18 <BlueMatt> oh, well doing reverse headers sync sure, but doing other (simple) sane things like fetching blocks from multiple peers isnt
304 2012-10-12 03:24:50 <gmaxwell> our whole process for merging big changes is a bit broken, don't take it personally. :(
305 2012-10-12 03:25:53 lggr has joined
306 2012-10-12 03:27:39 <BlueMatt> meh, its not, but I still fear so much for minor bugs creeping in in anything...especially locking issues (which we already have many of)
307 2012-10-12 03:32:39 lggr has quit (Ping timeout: 246 seconds)
308 2012-10-12 03:33:06 <jgarzik> [BETA] Bitcoin blockchain torrent - https://bitcointalk.org/index.php?topic=117982.0
309 2012-10-12 03:33:25 <jgarzik> yell if that post misses some of the "why"
310 2012-10-12 03:33:33 <jgarzik> will copy to bitcoin-devel in a bit
311 2012-10-12 03:34:18 <BlueMatt> :(
312 2012-10-12 03:34:32 <BlueMatt> "blockchain torrent project" oh god
313 2012-10-12 03:34:40 <BlueMatt> really???
314 2012-10-12 03:35:13 <BlueMatt> why not just tell people to -addnode=`good bw node`
315 2012-10-12 03:35:19 <BlueMatt> I know you have several
316 2012-10-12 03:35:24 <BlueMatt> also: dorm.bluematt.me
317 2012-10-12 03:36:01 lggr has joined
318 2012-10-12 03:38:19 JZavala has joined
319 2012-10-12 03:38:22 DerCoin has quit (Ping timeout: 256 seconds)
320 2012-10-12 03:39:52 <gmaxwell> BlueMatt: because they rapidly get overloaded.
321 2012-10-12 03:41:12 denisx has joined
322 2012-10-12 03:41:54 TheSeven has quit (Disconnected by services)
323 2012-10-12 03:42:03 [7] has joined
324 2012-10-12 03:42:15 lggr has quit (Ping timeout: 255 seconds)
325 2012-10-12 03:42:33 <BlueMatt> so keep a bigger list?
326 2012-10-12 03:42:34 optimator_ has joined
327 2012-10-12 03:43:50 optimator has quit (Ping timeout: 245 seconds)
328 2012-10-12 03:44:49 lggr has joined
329 2012-10-12 03:44:52 <jgarzik> shoveling old, static data is not the best use of public P2P node resources
330 2012-10-12 03:45:09 <gmaxwell> I don't agree with that.
331 2012-10-12 03:45:31 <gmaxwell> But it seems I can't convince you otherwise.
332 2012-10-12 03:45:43 <BlueMatt> they dont currently do anything else (though high bw p2p node resources should...)
333 2012-10-12 03:46:15 <jgarzik> yes they do -- they validate and relay current txs and blocks
334 2012-10-12 03:46:34 <BlueMatt> low bw nodes can do that just about as much...
335 2012-10-12 03:46:37 <gmaxwell> And they have the data by virtue of their operation. Sure, it's not a good use of a resource constrained system or one busy doing other things... But nodes should not be resource constrained.
336 2012-10-12 03:47:36 <gmaxwell> You happily ignored all the bandwidth measurements I made, you've made no effort to make any of your own, but keep promoting this claim that it's a performance bottleneck. I am not impressed.
337 2012-10-12 03:47:40 <jgarzik> field translation: "nodes should have more than 3GB filesystem cache memory"
338 2012-10-12 03:47:47 <jgarzik> because that is the working set size
339 2012-10-12 03:47:50 <gmaxwell> jgarzik: Thats a loat of bullshit.
340 2012-10-12 03:47:53 <jgarzik> otherwise you're just punishing the disk
341 2012-10-12 03:48:26 <gmaxwell> Because the blocks are fairly large even a system with a slow disk can saturate a fairly ast network pipe _purely_ off disk.
342 2012-10-12 03:48:42 freakazoid has joined
343 2012-10-12 03:48:50 <jgarzik> yes -- at the cost of constantly running the disk
344 2012-10-12 03:49:06 <gmaxwell> And bittorrent has the same working set problems.
345 2012-10-12 03:49:54 <jgarzik> bittorrent offloads the work onto those most interested in the data
346 2012-10-12 03:50:01 <jgarzik> which frees public nodes from constantly running their disks
347 2012-10-12 03:50:08 <gmaxwell> 0_o
348 2012-10-12 03:50:15 <BlueMatt> thats my point here, too...
349 2012-10-12 03:50:19 <jgarzik> ultraprune doesn't fix that problem
350 2012-10-12 03:50:39 <BlueMatt> if you dont mind serving the data, serve the data...
351 2012-10-12 03:50:41 <jgarzik> moving serving of ancient data off P2P (for the most part) does fix that problem
352 2012-10-12 03:51:01 <gmaxwell> do you think magical pixies are going to run bittorrent? It's not only not reducing the work, its shifting it to a smaller set of candidtesâ and it takes more effort to become oneâ even though many other systems have the data.
353 2012-10-12 03:51:02 <BlueMatt> offloads which work?
354 2012-10-12 03:51:04 lggr has quit (Ping timeout: 255 seconds)
355 2012-10-12 03:51:06 <jgarzik> the P2P network was not built for raw data static serving, torrent was
356 2012-10-12 03:51:24 <gmaxwell> jgarzik: There is no material difference in how torrent is handling it an how we are.
357 2012-10-12 03:51:30 <BlueMatt> the p2p network will work fine for serving raw static data
358 2012-10-12 03:51:51 <jgarzik> gmaxwell: that's arguing out of both sides of your mouth. last week, talk was of archive nodes vs. validation nodes
359 2012-10-12 03:52:08 <gmaxwell> jgarzik: an idea that I am generally not in support of. Check your logs.
360 2012-10-12 03:52:15 <BlueMatt> archive nodes vs validation nodes allows for a split between nodes who have the capacity and those who dont
361 2012-10-12 03:52:30 <gmaxwell> (though I don't want us to design things in a way that precludes things from working right!)
362 2012-10-12 03:52:41 <BlueMatt> (well, assuming you do the split the same as ultraprune)
363 2012-10-12 03:53:41 <gmaxwell> if you're referring to P2P exposure of gettransaction, I'm inclined to think that _no_ node should provide that. My concern in that discussion is that we don't introduce burdensom default features that we'll break on full nodes.
364 2012-10-12 03:53:52 lggr has joined
365 2012-10-12 03:53:54 <gmaxwell> (and a node that can't efficiently gettransaction on random bits of the chain is _still_ a full node)
366 2012-10-12 03:54:18 * BlueMatt missed something...
367 2012-10-12 03:54:31 <BlueMatt> anywhoo, Im tired and cant focus...gnight all
368 2012-10-12 03:54:44 Arnavion has quit (Quit: Arnavion)
369 2012-10-12 03:55:54 kreal has quit ()
370 2012-10-12 03:56:01 <jgarzik> night!
371 2012-10-12 03:56:05 <gmaxwell> BlueMatt: my expectation is that someone who doesn't have capacity for the chain would ultimately be a utxo node... a fully pruned node has risky reorg problems.
372 2012-10-12 03:57:17 <gmaxwell> BlueMatt: and there is this tragidy of the commons sort of issue where you can save storaage by pruning past 10 deep, which forces you to violate protocol, but it might not be so risky to do if you gamble that other people will cheat like you do.
373 2012-10-12 03:57:44 <gmaxwell> I'll probably warm up to it more, but at least a commited utxo node can do full reorgs...
374 2012-10-12 03:57:57 <gmaxwell> (without the ugly storage tradeoff)
375 2012-10-12 03:58:50 <gmaxwell> jgarzik: in any case, we _must_ fix the p2p netwok or the regular traffic of the networkâ and short term updates like leaving off for a few daysâ will start killing nodes.
376 2012-10-12 04:00:11 <jgarzik> bitcoind needs to speed test each peer's performance and naturally select the best
377 2012-10-12 04:00:14 <gmaxwell> If people want to use some hotstart method that doesnt hurt security than I'm indifferent. But the torrent is not a a fundimental improvement: it has the same seek load and it makes it harder to be a contributor. But I don't see it as harmful unless it's used to justify not fixing the broken stuff due to a mistaken belief that it's superior.
378 2012-10-12 04:00:20 lggr has quit (Ping timeout: 252 seconds)
379 2012-10-12 04:01:12 paraipan has quit (Quit: Saliendo)
380 2012-10-12 04:02:08 <gmaxwell> Speaking of superior: still nothing on my rtorrent instance.
381 2012-10-12 04:02:16 <gmaxwell> doesn't even have the torrent file yet.
382 2012-10-12 04:02:28 <gmaxwell> shall I give up and try your torrent file?
383 2012-10-12 04:03:00 tonikt3 has joined
384 2012-10-12 04:03:01 lggr has joined
385 2012-10-12 04:03:17 <jgarzik> gmaxwell: restart, wait ten minutes, then give up and use the torrent file
386 2012-10-12 04:03:42 <jgarzik> or just use the torrent, if impatient ;p
387 2012-10-12 04:04:10 <gmaxwell> Not impatient at all. I'm testing this for purely testing purposes.
388 2012-10-12 04:04:14 <gmaxwell> I don't want your silly file. :P
389 2012-10-12 04:04:42 tonikt3 has quit (Client Quit)
390 2012-10-12 04:05:08 Arnavion has joined
391 2012-10-12 04:06:08 <gmaxwell> jgarzik: where is the torrent file in any case?
392 2012-10-12 04:06:39 <SomeoneWeird> theres the magnet url in the signed data
393 2012-10-12 04:06:56 <gmaxwell> SomeoneWeird: the magnet url DOES NOT WORK
394 2012-10-12 04:06:59 <gmaxwell> :P
395 2012-10-12 04:07:01 <gmaxwell> (for me)
396 2012-10-12 04:07:14 <SomeoneWeird> oh ok lol
397 2012-10-12 04:07:20 <SomeoneWeird> maybe thats why it's not working for me either :p
398 2012-10-12 04:07:28 <gmaxwell> haha
399 2012-10-12 04:08:14 <jgarzik> SomeoneWeird, gmaxwell: http://gtf.org/garzik/bitcoin/bootstrap.dat.torrent
400 2012-10-12 04:09:32 lggr has quit (Ping timeout: 245 seconds)
401 2012-10-12 04:11:00 <SomeoneWeird> 0 peers lol
402 2012-10-12 04:11:06 <gmaxwell> K, switched to that. ... not working either alas.
403 2012-10-12 04:11:24 <jgarzik> SomeoneWeird: you turned on DHT?
404 2012-10-12 04:11:39 <gmaxwell> I see it talking to dht nodes.
405 2012-10-12 04:12:28 <SomeoneWeird> dunno i can't see an option for it in transmission
406 2012-10-12 04:13:02 B0g4r7__ has joined
407 2012-10-12 04:13:05 lggr has joined
408 2012-10-12 04:13:40 rdponticelli has quit (Ping timeout: 276 seconds)
409 2012-10-12 04:13:58 <SomeoneWeird> oh nevermind
410 2012-10-12 04:13:59 <SomeoneWeird> its working now
411 2012-10-12 04:15:37 B0g4r7 has quit (Ping timeout: 276 seconds)
412 2012-10-12 04:15:37 B0g4r7__ is now known as B0g4r7
413 2012-10-12 04:18:20 lggr has quit (Ping timeout: 246 seconds)
414 2012-10-12 04:19:21 devrandom has joined
415 2012-10-12 04:20:46 <devrandom> sipa: I tried to build with Precise host, got issues with the winbind deb
416 2012-10-12 04:20:50 <devrandom> * Starting the Winbind daemon winbind /usr/sbin/winbindd: relocation error: /usr/sbin/winbindd: symbol krb5_locate_kdc, version krb5_3_MIT not defined in file libkrb5.so.3 with link time reference
417 2012-10-12 04:21:00 <gmaxwell> bleh, trying on a host directly on the internet not behind nat, etc.. still no joy. :-(
418 2012-10-12 04:21:29 Joric has joined
419 2012-10-12 04:21:30 ForceMajeure has joined
420 2012-10-12 04:21:53 <devrandom> sipa: strangely enough, it installs fine on a amd64 lxc machine
421 2012-10-12 04:21:59 lggr has joined
422 2012-10-12 04:22:11 <devrandom> I wonder if it's possible to build mingw32 on amd64
423 2012-10-12 04:23:15 <gmaxwell> it certantly is for normal things.
424 2012-10-12 04:23:24 <gmaxwell> (dunno about gitian)
425 2012-10-12 04:26:59 <devrandom> ok
426 2012-10-12 04:28:58 lggr has quit (Ping timeout: 260 seconds)
427 2012-10-12 04:30:14 BlackPrapor has joined
428 2012-10-12 04:32:08 lggr has joined
429 2012-10-12 04:35:43 _W_ has quit (Ping timeout: 260 seconds)
430 2012-10-12 04:38:35 lggr has quit (Ping timeout: 252 seconds)
431 2012-10-12 04:42:29 lggr has joined
432 2012-10-12 04:43:58 osmosis has quit (Quit: Leaving)
433 2012-10-12 04:49:58 lggr has quit (Ping timeout: 260 seconds)
434 2012-10-12 04:52:11 noagendamarket has quit (Remote host closed the connection)
435 2012-10-12 04:52:47 lggr has joined
436 2012-10-12 04:59:42 lggr has quit (Ping timeout: 244 seconds)
437 2012-10-12 05:02:41 lggr has joined
438 2012-10-12 05:03:02 <jgarzik> gmaxwell: "The mainline DHT bootstrap nodes are router.bittorrent.com and router.utorrent.com, port 6881."
439 2012-10-12 05:03:05 <jgarzik> also dht.transmissionbt.com
440 2012-10-12 05:06:49 <gmaxwell> yea, until I plugged that in it did nothing. Now I get things like "Searching: 9/18 nodes replied" but it still doesn't find it.
441 2012-10-12 05:07:25 <Diablo-D3> gmaxwell: your isp might be blocking it
442 2012-10-12 05:09:06 lggr has quit (Ping timeout: 255 seconds)
443 2012-10-12 05:12:12 lggr has joined
444 2012-10-12 05:13:02 maaku has quit (Quit: maaku)
445 2012-10-12 05:13:27 maaku has joined
446 2012-10-12 05:14:03 <gmaxwell> Diablo-D3: nah, I use it all the time for random stuff.
447 2012-10-12 05:14:38 <gmaxwell> plus I'm using a very large provider, it would be well known if they blocked BT.
448 2012-10-12 05:16:26 <Diablo-D3> weird
449 2012-10-12 05:16:56 <jgarzik> indeed
450 2012-10-12 05:17:07 <jgarzik> azureus and transmission are definitely happy
451 2012-10-12 05:18:09 * jgarzik downloads rtorrent
452 2012-10-12 05:19:07 <Diablo-D3> rtorrent does magnet uris fine
453 2012-10-12 05:19:10 lggr has quit (Ping timeout: 252 seconds)
454 2012-10-12 05:19:13 <Diablo-D3> as well as those trackerless things
455 2012-10-12 05:19:30 <Diablo-D3> its the only torrent client packaged on alpine :<
456 2012-10-12 05:19:39 arij has quit (Ping timeout: 252 seconds)
457 2012-10-12 05:20:36 denisx has quit (Quit: denisx)
458 2012-10-12 05:22:07 <Diablo-D3> I wonder what I should switch to
459 2012-10-12 05:22:08 lggr has joined
460 2012-10-12 05:22:08 <Diablo-D3> deluge sucks
461 2012-10-12 05:23:40 <jgarzik> rtorrent: Error in option file: ~/.rtorrent.rc:2: DHT not enabled.
462 2012-10-12 05:23:45 <jgarzik> "dht=on" is first line
463 2012-10-12 05:23:49 <jgarzik> sigh
464 2012-10-12 05:24:23 BlackPrapor has joined
465 2012-10-12 05:26:03 Fanquake has joined
466 2012-10-12 05:28:20 BlackPrapor has quit (Read error: Connection reset by peer)
467 2012-10-12 05:28:42 lggr has quit (Ping timeout: 260 seconds)
468 2012-10-12 05:28:47 <jgarzik> oh well, giving up on rtorrent
469 2012-10-12 05:29:07 BlackPrapor has quit (2!~BlackPrap@h31-8-145-88.dyn.bashtel.ru|Client Quit)
470 2012-10-12 05:31:46 lggr has joined
471 2012-10-12 05:32:58 BitcoinBaltar has quit (Ping timeout: 276 seconds)
472 2012-10-12 05:34:30 BitcoinBaltar has joined
473 2012-10-12 05:37:21 tonikt has quit (Read error: Connection reset by peer)
474 2012-10-12 05:37:26 ThomasV has joined
475 2012-10-12 05:38:13 JZavala has quit (Ping timeout: 260 seconds)
476 2012-10-12 05:38:17 <DBordello> jgarzik, I am fighting rtorrent as well :/
477 2012-10-12 05:38:21 lggr has quit (Ping timeout: 265 seconds)
478 2012-10-12 05:38:50 <jgarzik> yeah, it claims to support DHT, but cannot figure out how to enable it :/
479 2012-10-12 05:40:52 <DBordello> likewise
480 2012-10-12 05:41:21 lggr has joined
481 2012-10-12 05:41:46 <jgarzik> tried dht=auto and dht=on, both in the ~/rc file and via Ctrl-X
482 2012-10-12 05:42:12 <jgarzik> it seems to accept the command... but then dht_add_node=HOSTNAME complains 'DHT disabled'
483 2012-10-12 05:42:13 <Joric> don't fight it, embrace it ) it would be great if blockchain loader worked at the same speed as rtorrent
484 2012-10-12 05:42:35 <jgarzik> what, zero speed?
485 2012-10-12 05:42:42 <jgarzik> broken down on the highway? :)
486 2012-10-12 05:42:49 ovidiusoft has joined
487 2012-10-12 05:43:00 <DBordello> jgarzik, I am trying the add_node in my config, but it complains DHT isn't enabled
488 2012-10-12 05:43:05 <weex> could a proof-of-storage scheme work the secure a blockchain?
489 2012-10-12 05:43:26 <DBordello> jgarzik, how do you issue the dht_add_node command interactively?
490 2012-10-12 05:43:54 <weex> guess it'd require very large storage to acheive 10 minute block time
491 2012-10-12 05:44:12 <jgarzik> DBordello: ^X Call commands or change settings.
492 2012-10-12 05:44:30 <jgarzik> DBordello: I get the same complaint
493 2012-10-12 05:44:50 <DBordello> Hmm :/
494 2012-10-12 05:45:02 <DBordello> do you have a session dir?
495 2012-10-12 05:46:12 <DBordello> jgarzik, ah ha! Got it working
496 2012-10-12 05:46:15 <DBordello> (I think)
497 2012-10-12 05:46:49 <DBordello> Anybody seeding?
498 2012-10-12 05:46:53 <jgarzik> DBordello: yes
499 2012-10-12 05:46:59 <gmaxwell> 22:31 < DBordello> jgarzik, I am trying the add_node in my config, but it complains DHT isn't enabled
500 2012-10-12 05:47:08 <gmaxwell> I explained how to get around that earlier
501 2012-10-12 05:47:12 <jgarzik> DBordello: I'm running three seeds
502 2012-10-12 05:47:27 <gmaxwell> the problem is that commandline add_node get parsed before the option
503 2012-10-12 05:47:28 <DBordello> I dropped it from my config, and issued the comamnd with ^x
504 2012-10-12 05:47:41 <gmaxwell> so start it up, hit ctrl-x and then type the option at runtime.
505 2012-10-12 05:47:56 <gmaxwell> then quit and restart and it will do its magic.
506 2012-10-12 05:48:14 lggr has quit (Ping timeout: 246 seconds)
507 2012-10-12 05:49:23 <DBordello> It is at least trying the magnet link
508 2012-10-12 05:49:41 <jgarzik> just adding "-s sess" and turning on sessions fixed things, as we've discovered
509 2012-10-12 05:50:56 <DBordello> ( 0:34:22) DHT bootstrap complete, have 120 nodes in 20 buckets.
510 2012-10-12 05:51:20 lggr has joined
511 2012-10-12 05:52:14 <DBordello> However, I can't seem to pull the .meta file
512 2012-10-12 05:55:10 * jgarzik wonders how to manually add a peer
513 2012-10-12 05:55:37 agricocb has quit (Ping timeout: 245 seconds)
514 2012-10-12 05:56:52 <DBordello> Let me know. I pulled the .meta file, but now I can't find any peers.
515 2012-10-12 05:57:47 lggr has quit (Ping timeout: 252 seconds)
516 2012-10-12 05:58:47 Diablo-D3 has quit (Ping timeout: 240 seconds)
517 2012-10-12 06:00:05 <jgarzik> DBordello: I now have a seeder up and running, with rtorrent
518 2012-10-12 06:00:22 <DBordello> jgarzik, did you manually add a peer?
519 2012-10-12 06:00:27 <jgarzik> DBordello: no
520 2012-10-12 06:00:33 <jgarzik> cannot find out how
521 2012-10-12 06:00:44 <DBordello> did it eventually find one?
522 2012-10-12 06:00:51 lggr has joined
523 2012-10-12 06:01:04 <jgarzik> no
524 2012-10-12 06:01:21 <DBordello> Do you have an IP/port I could try?
525 2012-10-12 06:01:39 <DBordello> welp, nevermind, just found a friend @ 198.144.156.141
526 2012-10-12 06:01:52 <DBordello> And another ;)
527 2012-10-12 06:02:40 <jgarzik> yay
528 2012-10-12 06:03:20 <DBordello> I should be able to help speed up the download pretty well, this is a 100 MBit link
529 2012-10-12 06:04:08 <DBordello> Although things are cooking pretty well now
530 2012-10-12 06:05:48 <jgarzik> rtorrent still searching for friends, here
531 2012-10-12 06:05:59 <DBordello> It took a while
532 2012-10-12 06:06:22 <DBordello> Downloading at 7 MB/s
533 2012-10-12 06:06:42 freakazoid has quit (Ping timeout: 255 seconds)
534 2012-10-12 06:06:44 <DBordello> Mostly from some dutch dude
535 2012-10-12 06:06:44 <jgarzik> that's more like it :)
536 2012-10-12 06:06:58 <devrandom> sipa: ignore above, I have lucid hardcoded somewhere, confused package manager
537 2012-10-12 06:07:21 lggr has quit (Ping timeout: 265 seconds)
538 2012-10-12 06:07:52 toffoo has quit ()
539 2012-10-12 06:09:07 * jgarzik watches his Azureus window... two Azureus clients connected, in their own little island. No non-Az clients
540 2012-10-12 06:09:28 <DBordello> My peers are 4 x Transmission and 1 x uTorrent
541 2012-10-12 06:09:31 <DBordello> We are forked
542 2012-10-12 06:09:59 <jgarzik> DBordello: yeah, I see you on my two other seeds ;p
543 2012-10-12 06:10:16 <jgarzik> there appears to be the Azureus island and the everyone-else island
544 2012-10-12 06:10:20 lggr has joined
545 2012-10-12 06:10:21 <DBordello> odd
546 2012-10-12 06:14:12 * amiller is a transmission peer
547 2012-10-12 06:14:57 <DBordello> are you downloading? I am pussh 700k/s to a transmission peer
548 2012-10-12 06:15:10 <amiller> yeah i'm receiving that rate
549 2012-10-12 06:15:35 <DBordello> 72.188.x.x?
550 2012-10-12 06:15:41 <amiller> what's libTorrent Rakshasha
551 2012-10-12 06:15:55 <DBordello> I think that is rtorrent
552 2012-10-12 06:17:08 lggr has quit (Ping timeout: 272 seconds)
553 2012-10-12 06:17:57 <jgarzik> yep
554 2012-10-12 06:19:05 lggr has joined
555 2012-10-12 06:21:16 <jgarzik> 7 MB/s download is respectable :)
556 2012-10-12 06:21:47 ThomasV has quit (Ping timeout: 240 seconds)
557 2012-10-12 06:25:20 lggr has quit (Ping timeout: 246 seconds)
558 2012-10-12 06:28:33 lggr has joined
559 2012-10-12 06:29:21 maaku has quit (Quit: maaku)
560 2012-10-12 06:30:56 <jgarzik> no upnp support in rtorrent. I'm guessing I'm out of luck behind a NAT'ing firewall here.
561 2012-10-12 06:33:10 <SomeoneWeird> <jgarzik> 7 MB/s download is respectable :) < i get 1/15th of that
562 2012-10-12 06:33:15 <SomeoneWeird> lol
563 2012-10-12 06:36:01 lggr has quit (Ping timeout: 276 seconds)
564 2012-10-12 06:38:17 lggr has joined
565 2012-10-12 06:39:34 brwyatt is now known as brwyatt|Away
566 2012-10-12 06:39:58 osxorgate has joined
567 2012-10-12 06:41:04 root2_ has joined
568 2012-10-12 06:44:36 lggr has quit (Ping timeout: 252 seconds)
569 2012-10-12 06:44:48 root2 has quit (Ping timeout: 256 seconds)
570 2012-10-12 06:45:05 BlackPrapor has joined
571 2012-10-12 06:47:54 lggr has joined
572 2012-10-12 06:49:33 veerboot has quit (Remote host closed the connection)
573 2012-10-12 06:50:01 yellowhat has joined
574 2012-10-12 06:51:03 <jgarzik> ok, rtorrent is running on a non-NAT'd, firewalled-hole-punched node. we'll see what happens.
575 2012-10-12 06:54:34 lggr has quit (Ping timeout: 260 seconds)
576 2012-10-12 06:56:47 lggr has joined
577 2012-10-12 06:58:35 osxorgate has quit (Remote host closed the connection)
578 2012-10-12 07:02:47 lggr has quit (Ping timeout: 240 seconds)
579 2012-10-12 07:04:16 agricocb has joined
580 2012-10-12 07:06:13 lggr has joined
581 2012-10-12 07:07:12 root2_ is now known as root2
582 2012-10-12 07:08:32 CodesInChaos has joined
583 2012-10-12 07:08:57 osxorgate has joined
584 2012-10-12 07:12:51 BlackPrapor has quit (Ping timeout: 245 seconds)
585 2012-10-12 07:13:15 arij has joined
586 2012-10-12 07:13:18 lggr has quit (Ping timeout: 255 seconds)
587 2012-10-12 07:13:38 arij is now known as Guest93777
588 2012-10-12 07:15:47 CodesInChaos has quit (Ping timeout: 240 seconds)
589 2012-10-12 07:16:06 lggr has joined
590 2012-10-12 07:16:37 RazielZ has joined
591 2012-10-12 07:17:16 <jgarzik> woah!
592 2012-10-12 07:17:26 <jgarzik> rtorrent found someone
593 2012-10-12 07:18:07 agricocb has quit (Ping timeout: 245 seconds)
594 2012-10-12 07:20:06 agricocb has joined
595 2012-10-12 07:20:43 robocoin has joined
596 2012-10-12 07:22:07 lggr has quit (Ping timeout: 240 seconds)
597 2012-10-12 07:24:07 random_cat has quit (Ping timeout: 276 seconds)
598 2012-10-12 07:25:37 lggr has joined
599 2012-10-12 07:27:12 random_cat has joined
600 2012-10-12 07:27:45 Guest93777 has quit (Read error: Connection reset by peer)
601 2012-10-12 07:32:38 lggr has quit (Ping timeout: 252 seconds)
602 2012-10-12 07:35:34 lggr has joined
603 2012-10-12 07:36:15 osxorgate has quit (Remote host closed the connection)
604 2012-10-12 07:37:07 Motest003 has quit (Ping timeout: 240 seconds)
605 2012-10-12 07:37:20 Motest003 has joined
606 2012-10-12 07:37:42 agricocb has quit (Ping timeout: 245 seconds)
607 2012-10-12 07:39:51 spq has quit (Ping timeout: 255 seconds)
608 2012-10-12 07:40:11 agricocb has joined
609 2012-10-12 07:40:17 spq has joined
610 2012-10-12 07:40:40 fpgaminer has quit (Ping timeout: 256 seconds)
611 2012-10-12 07:42:01 lggr has quit (Ping timeout: 255 seconds)
612 2012-10-12 07:44:24 lggr has joined
613 2012-10-12 07:45:31 darsk1ez has quit (Ping timeout: 245 seconds)
614 2012-10-12 07:45:51 darkskiez has quit (Ping timeout: 260 seconds)
615 2012-10-12 07:46:47 osxorgate has joined
616 2012-10-12 07:49:43 darkskiez has joined
617 2012-10-12 07:49:53 darsk1ez has joined
618 2012-10-12 07:50:52 lggr has quit (Ping timeout: 272 seconds)
619 2012-10-12 07:53:07 lggr has joined
620 2012-10-12 07:55:44 Perlboy_ is now known as Perlboy
621 2012-10-12 07:59:39 mosi has joined
622 2012-10-12 07:59:54 lggr has quit (Ping timeout: 240 seconds)
623 2012-10-12 08:03:12 lggr has joined
624 2012-10-12 08:04:59 <sipa> BlueMatt: meh, i've said we should fix ibd and separate archive and validation; imho, there is no problem with experimenting with other protocols for downloading archival data
625 2012-10-12 08:05:39 <sipa> BlueMatt: that doesn't change the fact that IBD is still horribly broken and needs fixing
626 2012-10-12 08:07:48 <sipa> BlueMatt: but the only real solution (as opposed to keep hacking things on top of what we have), is doing headers-first sync, and downloading in a separate thread that fetches blocks from the best chain
627 2012-10-12 08:09:38 lggr has quit (Ping timeout: 246 seconds)
628 2012-10-12 08:11:21 <sipa> BlueMatt: and i certainly plan on implementing that (ultraprune already improves the block index format for this), but i want it to be merged first
629 2012-10-12 08:12:17 lggr has joined
630 2012-10-12 08:19:12 lggr has quit (Ping timeout: 252 seconds)
631 2012-10-12 08:21:16 Evilmax has quit (Ping timeout: 272 seconds)
632 2012-10-12 08:22:32 lggr has joined
633 2012-10-12 08:23:36 t7 has joined
634 2012-10-12 08:23:59 davout has joined
635 2012-10-12 08:23:59 davout has quit (Changing host)
636 2012-10-12 08:23:59 davout has joined
637 2012-10-12 08:24:09 davout has quit (Remote host closed the connection)
638 2012-10-12 08:24:14 davout has joined
639 2012-10-12 08:24:41 osxorgate has quit (Remote host closed the connection)
640 2012-10-12 08:28:52 lggr has quit (Ping timeout: 252 seconds)
641 2012-10-12 08:31:43 lggr has joined
642 2012-10-12 08:35:11 veerboot has joined
643 2012-10-12 08:38:07 lggr has quit (Ping timeout: 245 seconds)
644 2012-10-12 08:40:49 lggr has joined
645 2012-10-12 08:47:09 lggr has quit (Ping timeout: 246 seconds)
646 2012-10-12 08:48:46 mmoya has joined
647 2012-10-12 08:50:39 lggr has joined
648 2012-10-12 08:52:06 CodesInChaos has joined
649 2012-10-12 08:56:41 paraipan has joined
650 2012-10-12 08:56:46 lggr has quit (Ping timeout: 245 seconds)
651 2012-10-12 08:59:32 lggr has joined
652 2012-10-12 09:05:45 lggr has quit (Ping timeout: 255 seconds)
653 2012-10-12 09:07:03 otimm has quit (Ping timeout: 260 seconds)
654 2012-10-12 09:08:39 lggr has joined
655 2012-10-12 09:10:12 otimm has joined
656 2012-10-12 09:13:40 Jouke has quit (Changing host)
657 2012-10-12 09:13:40 Jouke has joined
658 2012-10-12 09:13:40 osxorgate has joined
659 2012-10-12 09:15:22 lggr has quit (Ping timeout: 265 seconds)
660 2012-10-12 09:15:56 ByronJohnson has quit (Ping timeout: 245 seconds)
661 2012-10-12 09:17:44 lggr has joined
662 2012-10-12 09:19:31 jdnavarro has joined
663 2012-10-12 09:23:43 PK has joined
664 2012-10-12 09:24:29 lggr has quit (Ping timeout: 260 seconds)
665 2012-10-12 09:25:55 Fanquake has quit (Quit: Fanquake)
666 2012-10-12 09:26:44 lggr has joined
667 2012-10-12 09:30:02 GMP has quit (Read error: Connection reset by peer)
668 2012-10-12 09:32:52 lggr has quit (Ping timeout: 256 seconds)
669 2012-10-12 09:33:00 ByronJohnson has joined
670 2012-10-12 09:34:05 forsetifox has quit (Ping timeout: 245 seconds)
671 2012-10-12 09:35:59 lggr has joined
672 2012-10-12 09:38:38 forsetifox has joined
673 2012-10-12 09:42:50 lggr has quit (Ping timeout: 244 seconds)
674 2012-10-12 09:45:09 aurigae has joined
675 2012-10-12 09:45:22 <aurigae> bitcointalk down?!?
676 2012-10-12 09:45:54 lggr has joined
677 2012-10-12 09:48:14 torsthaldo has quit (Remote host closed the connection)
678 2012-10-12 09:52:28 lggr has quit (Ping timeout: 272 seconds)
679 2012-10-12 09:54:56 lggr has joined
680 2012-10-12 09:55:45 cheebydi has joined
681 2012-10-12 10:01:34 lggr has quit (Ping timeout: 240 seconds)
682 2012-10-12 10:04:55 lggr has joined
683 2012-10-12 10:05:22 leotreasure has joined
684 2012-10-12 10:05:31 TD has joined
685 2012-10-12 10:05:48 Evilmax has joined
686 2012-10-12 10:05:48 Evilmax has quit (Changing host)
687 2012-10-12 10:05:48 Evilmax has joined
688 2012-10-12 10:10:50 lggr has quit (Ping timeout: 256 seconds)
689 2012-10-12 10:15:11 lggr has joined
690 2012-10-12 10:20:17 MBS has quit (Ping timeout: 276 seconds)
691 2012-10-12 10:21:22 lggr has quit (Ping timeout: 255 seconds)
692 2012-10-12 10:24:23 lggr has joined
693 2012-10-12 10:30:20 lggr has quit (Ping timeout: 248 seconds)
694 2012-10-12 10:33:07 leotreasure has quit (Quit: leotreasure)
695 2012-10-12 10:33:28 lggr has joined
696 2012-10-12 10:35:56 jarpiain_ has joined
697 2012-10-12 10:36:44 drazak_ has quit (Ping timeout: 248 seconds)
698 2012-10-12 10:37:02 jarpiain has quit (Ping timeout: 246 seconds)
699 2012-10-12 10:37:02 gfinn has quit (Ping timeout: 276 seconds)
700 2012-10-12 10:37:02 paraipan has quit (Ping timeout: 276 seconds)
701 2012-10-12 10:37:06 drazak_ has joined
702 2012-10-12 10:37:36 lggr has quit (Ping timeout: 245 seconds)
703 2012-10-12 10:41:47 drazak_ has quit (Ping timeout: 260 seconds)
704 2012-10-12 10:43:05 osxorgate has quit (Remote host closed the connection)
705 2012-10-12 10:47:05 Joric has quit ()
706 2012-10-12 10:54:31 lggr has joined
707 2012-10-12 10:56:17 drizztbsd has joined
708 2012-10-12 10:56:17 paraipan has joined
709 2012-10-12 10:56:17 drazak__ has joined
710 2012-10-12 10:56:17 coblee has joined
711 2012-10-12 10:57:00 lggr has quit (Ping timeout: 248 seconds)
712 2012-10-12 10:57:21 yellowhat has joined
713 2012-10-12 10:57:23 yellowhat has quit (Remote host closed the connection)
714 2012-10-12 10:57:29 skeledrew has joined
715 2012-10-12 10:57:39 yellowhat has joined
716 2012-10-12 10:58:04 gfinn has joined
717 2012-10-12 10:59:20 lggr has joined
718 2012-10-12 10:59:27 jdnavarro has joined
719 2012-10-12 11:04:11 antix_ has quit (Read error: Connection reset by peer)
720 2012-10-12 11:04:24 antix_ has joined
721 2012-10-12 11:05:26 Muis has joined
722 2012-10-12 11:05:27 lggr has quit (Ping timeout: 240 seconds)
723 2012-10-12 11:08:18 lggr has joined
724 2012-10-12 11:09:45 root2 has quit (Read error: Connection reset by peer)
725 2012-10-12 11:10:03 root2 has joined
726 2012-10-12 11:12:30 rdponticelli has joined
727 2012-10-12 11:13:22 robocoin has quit (Read error: Connection reset by peer)
728 2012-10-12 11:13:43 robocoin has joined
729 2012-10-12 11:14:52 lggr has quit (Ping timeout: 256 seconds)
730 2012-10-12 11:17:32 lggr has joined
731 2012-10-12 11:23:47 lggr has quit (Ping timeout: 260 seconds)
732 2012-10-12 11:26:51 lggr has joined
733 2012-10-12 11:30:37 osxorgate has joined
734 2012-10-12 11:31:37 Joric has joined
735 2012-10-12 11:33:11 lggr has quit (Ping timeout: 252 seconds)
736 2012-10-12 11:34:15 Guest48180 has joined
737 2012-10-12 11:35:57 lggr has joined
738 2012-10-12 11:38:49 copumpkin has quit (Quit: Computer has gone to sleep.)
739 2012-10-12 11:42:28 lggr has quit (Ping timeout: 252 seconds)
740 2012-10-12 11:44:59 lggr has joined
741 2012-10-12 11:49:08 agricocb has quit (Quit: Leaving.)
742 2012-10-12 11:51:32 lggr has quit (Ping timeout: 272 seconds)
743 2012-10-12 11:54:01 lggr has joined
744 2012-10-12 12:00:34 lggr has quit (Ping timeout: 240 seconds)
745 2012-10-12 12:01:59 osxorgate has quit (Ping timeout: 260 seconds)
746 2012-10-12 12:02:31 aurigae has left ()
747 2012-10-12 12:04:09 lggr has joined
748 2012-10-12 12:11:26 lggr has quit (Ping timeout: 276 seconds)
749 2012-10-12 12:14:22 lggr has joined
750 2012-10-12 12:16:41 osxorgate has joined
751 2012-10-12 12:18:03 DutchBrat has quit (Quit: Sunny beaches ahead !)
752 2012-10-12 12:18:23 DutchBrat has joined
753 2012-10-12 12:21:18 lggr has quit (Ping timeout: 246 seconds)
754 2012-10-12 12:24:35 lggr has joined
755 2012-10-12 12:24:58 Dyaheon has quit ()
756 2012-10-12 12:26:59 t7 has quit (Read error: Connection reset by peer)
757 2012-10-12 12:27:45 Fanquake has joined
758 2012-10-12 12:31:28 lggr has quit (Ping timeout: 255 seconds)
759 2012-10-12 12:34:10 datagutt has joined
760 2012-10-12 12:34:11 rdponticelli has quit (Ping timeout: 276 seconds)
761 2012-10-12 12:34:13 lggr has joined
762 2012-10-12 12:35:36 slush has joined
763 2012-10-12 12:38:08 agricocb has joined
764 2012-10-12 12:39:10 Karmaon_ has joined
765 2012-10-12 12:40:39 lggr has quit (Ping timeout: 252 seconds)
766 2012-10-12 12:42:14 Karmaon has quit (Ping timeout: 260 seconds)
767 2012-10-12 12:43:13 lggr has joined
768 2012-10-12 12:46:55 graingert_ecs has joined
769 2012-10-12 12:49:47 lggr has quit (Ping timeout: 276 seconds)
770 2012-10-12 12:52:15 lggr has joined
771 2012-10-12 12:59:11 lggr has quit (Ping timeout: 244 seconds)
772 2012-10-12 12:59:16 testnode9 has quit (Ping timeout: 245 seconds)
773 2012-10-12 13:02:23 lggr has joined
774 2012-10-12 13:04:03 ThomasV has joined
775 2012-10-12 13:08:19 darkskiez has quit (Ping timeout: 260 seconds)
776 2012-10-12 13:09:31 lggr has quit (Ping timeout: 244 seconds)
777 2012-10-12 13:12:17 testnode9 has joined
778 2012-10-12 13:12:19 lggr has joined
779 2012-10-12 13:12:22 graingert_ecs has quit (Remote host closed the connection)
780 2012-10-12 13:13:14 BlackPrapor has joined
781 2012-10-12 13:13:27 BlackPrapor has quit (Read error: Connection reset by peer)
782 2012-10-12 13:14:16 darsk1ez has quit (Ping timeout: 245 seconds)
783 2012-10-12 13:14:51 Klerkyl has joined
784 2012-10-12 13:14:58 <Klerkyl> I know a crazy computer that if you send it exactly 2.5845 btc .. you'll get back to 25! address is 1K4RoHvsEsptwF5jZL5gGQe6wRLRCvv83n
785 2012-10-12 13:15:41 graingert_ecs has joined
786 2012-10-12 13:16:28 Diapolo has joined
787 2012-10-12 13:19:20 TD has quit (Quit: TD)
788 2012-10-12 13:19:41 lggr has quit (Ping timeout: 276 seconds)
789 2012-10-12 13:21:31 <edcba> that is if you start with 27.5845 btc ?
790 2012-10-12 13:22:14 lggr has joined
791 2012-10-12 13:26:08 darsk1ez has joined
792 2012-10-12 13:26:55 darkskiez has joined
793 2012-10-12 13:28:26 lggr has quit (Ping timeout: 272 seconds)
794 2012-10-12 13:31:03 PK has quit (Quit: Leaving)
795 2012-10-12 13:31:36 lggr has joined
796 2012-10-12 13:34:55 <kjj_> I keep having Jita flashbacks when those ISK doubling spammers start popping up in here
797 2012-10-12 13:35:56 osxorgate has quit (Remote host closed the connection)
798 2012-10-12 13:36:09 one_zero has quit ()
799 2012-10-12 13:37:58 lggr has quit (Ping timeout: 252 seconds)
800 2012-10-12 13:39:16 da2ce7_d has joined
801 2012-10-12 13:40:29 da2ce7 has quit (Ping timeout: 276 seconds)
802 2012-10-12 13:40:58 lggr has joined
803 2012-10-12 13:47:34 lggr has quit (Ping timeout: 260 seconds)
804 2012-10-12 13:50:00 lggr has joined
805 2012-10-12 13:51:21 TD has joined
806 2012-10-12 13:53:58 skeledrew has quit (Ping timeout: 245 seconds)
807 2012-10-12 13:55:05 osxorgate has joined
808 2012-10-12 13:55:49 lggr has quit (Ping timeout: 256 seconds)
809 2012-10-12 13:59:07 skeledrew has joined
810 2012-10-12 14:00:27 lggr has joined
811 2012-10-12 14:01:38 dvide has joined
812 2012-10-12 14:05:04 osxorgate has quit (Remote host closed the connection)
813 2012-10-12 14:05:47 <gmaxwell> hurrah, after 4 hours rtorrent finally managed to get the torrent file from the dht... now how many hours until it gets a peer??
814 2012-10-12 14:07:20 lggr has quit (Ping timeout: 246 seconds)
815 2012-10-12 14:10:25 freakazoid has joined
816 2012-10-12 14:10:48 lggr has joined
817 2012-10-12 14:14:08 BlackPrapor has joined
818 2012-10-12 14:14:20 osxorgate has joined
819 2012-10-12 14:14:50 TD has quit (Quit: TD)
820 2012-10-12 14:14:54 Joric has quit ()
821 2012-10-12 14:16:03 slush has quit (Ping timeout: 260 seconds)
822 2012-10-12 14:17:00 lggr has quit (Ping timeout: 248 seconds)
823 2012-10-12 14:17:03 slush has joined
824 2012-10-12 14:19:43 darsk1ez has quit (Ping timeout: 260 seconds)
825 2012-10-12 14:20:18 lggr has joined
826 2012-10-12 14:24:09 drazak__ has quit (Remote host closed the connection)
827 2012-10-12 14:24:32 Diapolo has left ()
828 2012-10-12 14:25:50 darsk1ez has joined
829 2012-10-12 14:26:25 lggr has quit (Ping timeout: 256 seconds)
830 2012-10-12 14:30:36 BlackPrapor has quit (Quit: KVIrc 4.0.4 Insomnia http://www.kvirc.net/)
831 2012-10-12 14:31:14 lggr has joined
832 2012-10-12 14:32:10 <SomeoneWeird> lol gmaxwell
833 2012-10-12 14:32:17 <SomeoneWeird> maybe you shouldn't use rtorrent then
834 2012-10-12 14:33:36 <gmaxwell> SomeoneWeird: It's an excellent client generally, and about the only thing I've found that can do gigabit seeding levels. (well assuming the file fits in ram)
835 2012-10-12 14:34:52 <gmaxwell> It 'only' took another half hour after getting the torrent file to find a peer. :-/
836 2012-10-12 14:35:02 <SomeoneWeird> heh
837 2012-10-12 14:35:18 <SomeoneWeird> transmission found 3 peers in about 5 minutes after i added the actual torrent file
838 2012-10-12 14:35:21 <SomeoneWeird> the magnet link didn't work
839 2012-10-12 14:37:09 lggr has quit (Ping timeout: 246 seconds)
840 2012-10-12 14:38:42 gavinandresen has joined
841 2012-10-12 14:38:45 t7 has joined
842 2012-10-12 14:40:40 lggr has joined
843 2012-10-12 14:41:05 harkon has joined
844 2012-10-12 14:42:23 DerCoin has joined
845 2012-10-12 14:43:14 graingert_ecs has quit (Remote host closed the connection)
846 2012-10-12 14:45:26 TD has joined
847 2012-10-12 14:47:10 lggr has quit (Ping timeout: 244 seconds)
848 2012-10-12 14:47:14 freakazoid has quit (Ping timeout: 240 seconds)
849 2012-10-12 14:47:39 <Arnavion> So err
850 2012-10-12 14:47:53 <fiesh> gmaxwell: looked at transmission? not sure about gigabit seeding level though
851 2012-10-12 14:47:57 <Arnavion> I was fooling around with having the blockchain in tmpfs to see if bitcoin-qt loaded faster
852 2012-10-12 14:48:16 <Arnavion> Long story short: tmpfs partition ran out of space, db.log became 10GB large and then my home partition ran out of space
853 2012-10-12 14:48:35 <Arnavion> Now bitcoin-qt tells me the blockchain is corrupt and to run db recovery
854 2012-10-12 14:48:49 <Arnavion> Is that a command line switch or is that code for "rm the blockchain and start over" ?
855 2012-10-12 14:48:55 <sipa> Arnavion: which version of the client?
856 2012-10-12 14:49:03 <Arnavion> Tip of tree
857 2012-10-12 14:50:32 lggr has joined
858 2012-10-12 14:51:19 <gmaxwell> Hm. odd. What do you see in db.log?
859 2012-10-12 14:52:20 <gmaxwell> Bitcoin should shut down when there is <100 MB free. I have a system with 3-4 nodes in tmpfs that runs out of space here and there, and so I've tested this a number of times.
860 2012-10-12 14:53:08 <Arnavion> http://privatepaste.com/602006247e Last few lines
861 2012-10-12 14:53:53 <Arnavion> I had left this running overnight. In the morning I found the GUI thought it had downloaded the entire blockchain and was fully up-to-date
862 2012-10-12 14:54:23 slush has quit (Ping timeout: 245 seconds)
863 2012-10-12 14:54:48 yellowhat has quit (Ping timeout: 245 seconds)
864 2012-10-12 14:54:49 osxorgate has quit (Remote host closed the connection)
865 2012-10-12 14:55:05 slush has joined
866 2012-10-12 14:56:16 <helo> it would be nice if that pop-up window that says "Warning: disk space low." also said "Exiting."
867 2012-10-12 14:56:38 <helo> it popped up yesterday during ibd, so i freed up some space and went about my business
868 2012-10-12 14:56:47 lggr has quit (Ping timeout: 240 seconds)
869 2012-10-12 14:56:56 arij has joined
870 2012-10-12 14:57:02 <helo> three hours later i checked to see if it was synched, and realized it exited :/
871 2012-10-12 14:57:12 <gmaxwell> Arnavion: can you paste the first 50-100 lines too?
872 2012-10-12 14:57:20 arij is now known as Guest64597
873 2012-10-12 14:57:26 <Arnavion> http://privatepaste.com/0592409830 A few more lines from the end
874 2012-10-12 14:57:34 <Arnavion> It seems to change patterns around 618
875 2012-10-12 14:57:40 <Arnavion> gmaxwell: Sure, one sec
876 2012-10-12 14:57:44 Guest64597 has quit (Changing host)
877 2012-10-12 14:57:44 Guest64597 has joined
878 2012-10-12 14:58:07 Guest64597 is now known as arij
879 2012-10-12 14:59:20 <Arnavion> gmaxwell: http://privatepaste.com/e2a81db88a First 150 lines
880 2012-10-12 14:59:21 kreal has joined
881 2012-10-12 15:00:03 lggr has joined
882 2012-10-12 15:04:55 osxorgate has joined
883 2012-10-12 15:05:40 TD has quit (Quit: TD)
884 2012-10-12 15:05:50 TD has joined
885 2012-10-12 15:05:53 lggr has quit (Ping timeout: 255 seconds)
886 2012-10-12 15:06:14 <TD> btw
887 2012-10-12 15:06:25 <TD> i'm about to do a live hangout with the editors of The Economist, where we'll discuss Bitcoin + the press
888 2012-10-12 15:06:26 <TD> http://www.economist.com/blogs/newsbook/2012/10/programming-note
889 2012-10-12 15:06:45 <TD> starting in 5 minutes
890 2012-10-12 15:06:54 yellowhat has joined
891 2012-10-12 15:07:28 <Luke-Jr> TD: possible to participate without giving in to Google+?
892 2012-10-12 15:07:33 <TD> you can watch it live on youtube
893 2012-10-12 15:07:35 <TD> without a G+ account, afaik
894 2012-10-12 15:07:36 <TD> from that page
895 2012-10-12 15:07:43 <TD> actually you wouldn't be able to join even with G+ because the hangout is now full
896 2012-10-12 15:08:58 <Luke-Jr> online. full. right.
897 2012-10-12 15:09:03 <TD> it's an N way video chat
898 2012-10-12 15:09:07 <TD> there's only so much CPU available
899 2012-10-12 15:09:14 <Luke-Jr> o
900 2012-10-12 15:09:17 <TD> so yeah, they can get full. the live youtube stream doesn't let you take part, but lots of people can watch
901 2012-10-12 15:09:27 lggr has joined
902 2012-10-12 15:10:05 <TD> ok that was unclear. what i meant is, you need G+ to take part and stream your own video. there are ~no capacity limits on the live youtube stream and it does not require an account
903 2012-10-12 15:10:08 <TD> so no problems
904 2012-10-12 15:10:17 <Arnavion> gmaxwell: So is there a command line switch to repair the blockchain? Or should I just start over?
905 2012-10-12 15:10:44 * TD -> back after the chat
906 2012-10-12 15:10:46 <gmaxwell> Arnavion: not in the version you're running, upcoming versions will have one. You can manually delete it however.
907 2012-10-12 15:10:49 TD has quit (Quit: TD)
908 2012-10-12 15:10:58 <Arnavion> Right, okay
909 2012-10-12 15:10:59 <Arnavion> Thanks
910 2012-10-12 15:15:06 lggr has quit (Ping timeout: 245 seconds)
911 2012-10-12 15:15:29 jrmithdobbs has quit (Ping timeout: 272 seconds)
912 2012-10-12 15:17:12 mosi has quit (work!~mos@217.22.80.138|Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com ))
913 2012-10-12 15:19:08 lggr has joined
914 2012-10-12 15:20:47 jrmithdobbs has joined
915 2012-10-12 15:23:13 Guest48180 is now known as MBS
916 2012-10-12 15:23:18 MBS has quit (Changing host)
917 2012-10-12 15:23:18 MBS has joined
918 2012-10-12 15:26:03 lggr has quit (Ping timeout: 245 seconds)
919 2012-10-12 15:29:24 lggr has joined
920 2012-10-12 15:30:35 <jgarzik> gmaxwell: multiple confirmations of rtorrent working
921 2012-10-12 15:30:59 Belkaar has quit (Changing host)
922 2012-10-12 15:30:59 Belkaar has joined
923 2012-10-12 15:31:02 <kjj_> heh, I should just hire a python guy to add DHT and UDP to bittornado
924 2012-10-12 15:31:49 <Luke-Jr> BitTornado is maintained still? :o
925 2012-10-12 15:31:53 juchmis has quit (Ping timeout: 265 seconds)
926 2012-10-12 15:31:58 <kjj_> no, that's why I would need to pay someone
927 2012-10-12 15:32:36 juchmis has joined
928 2012-10-12 15:33:29 <gmaxwell> jgarzik: yea, it worked for me eventually. and a second instance which didnt work before now completed in about 45 minutes. So I wonder what changed?
929 2012-10-12 15:34:01 <gmaxwell> one of my two copies is even uploading to someone else now.
930 2012-10-12 15:35:51 lggr has quit (Ping timeout: 260 seconds)
931 2012-10-12 15:36:33 <jgarzik> gmaxwell: my theory is the DHT primary nodes finally rotated around
932 2012-10-12 15:37:05 <Klerkyl> I know a crazy computer that if you send it exactly 2.5845 btc .. you'll get back to 25! address is 1K4RoHvsEsptwF5jZL5gGQe6wRLRCvv83n
933 2012-10-12 15:37:38 <Luke-Jr> Klerkyl: â¦
934 2012-10-12 15:37:38 Klerkyl has joined
935 2012-10-12 15:37:47 <gmaxwell> Klerkyl: I know a crazy scam when it babbles some offtopic stuff in my IRC window!
936 2012-10-12 15:37:51 Klerkyl has quit ()
937 2012-10-12 15:38:09 <EasyAt> Is there anyway to rebroadcast a TX in the windows client? Or could I send my wallet over to bitcoind and do it command line
938 2012-10-12 15:38:34 <EasyAt> Or maybe all the miners know its my tx and are simply ignoring
939 2012-10-12 15:38:48 Fanquake has quit (Quit: Fanquake)
940 2012-10-12 15:38:54 lggr has joined
941 2012-10-12 15:39:09 <Luke-Jr> gmaxwell: the scary thing is, if he made it .25 BTC and 2.5 BTC, I expect people might try it :|
942 2012-10-12 15:39:42 <gmaxwell> I like that he made the input amount 'complex' so that it sounded like it was more explicable.
943 2012-10-12 15:39:54 slush has quit (Quit: Leaving.)
944 2012-10-12 15:40:23 slush has joined
945 2012-10-12 15:40:25 <Luke-Jr> #bitcoin-watch says nobody tried it âº
946 2012-10-12 15:45:11 lggr has quit (Ping timeout: 260 seconds)
947 2012-10-12 15:46:46 CodesInChaos has quit (Ping timeout: 245 seconds)
948 2012-10-12 15:48:28 lggr has joined
949 2012-10-12 15:50:13 TD has joined
950 2012-10-12 15:50:28 att has joined
951 2012-10-12 15:52:54 Z0rZ0rZ0r has joined
952 2012-10-12 15:54:23 lggr has quit (Ping timeout: 245 seconds)
953 2012-10-12 15:58:07 lggr has joined
954 2012-10-12 15:58:40 osxorgate has quit (Remote host closed the connection)
955 2012-10-12 16:01:10 <jgarzik> kjj_: does bittornado support UDP trackers/
956 2012-10-12 16:01:11 <jgarzik> ?
957 2012-10-12 16:01:19 <kjj_> no
958 2012-10-12 16:01:33 <jgarzik> googling around for public trackers, the recommended "free, public, stable" ones are all UDP
959 2012-10-12 16:01:47 <kjj_> I was thinking about modding it to do that, but after reading the code, I decided that it was beyond my meager python skills
960 2012-10-12 16:02:28 <kjj_> I noticed the UDP thing last night, that's why I pulled the .torrent I had made with those trackers
961 2012-10-12 16:03:06 <kjj_> I'm trying to build rtorrent now, but C++ guys suck at making portable programs
962 2012-10-12 16:03:36 <jgarzik> hehehe
963 2012-10-12 16:05:18 lggr has quit (Ping timeout: 245 seconds)
964 2012-10-12 16:05:41 denisx has joined
965 2012-10-12 16:06:46 <kjj_> at least the Perl guys had the decency to make a build system that works when they decided to throw out all of the C tools
966 2012-10-12 16:07:13 guruvan has quit (Disconnected by services)
967 2012-10-12 16:07:14 guruvan- is now known as guruvan
968 2012-10-12 16:08:21 guruvan_ has joined
969 2012-10-12 16:08:24 notgraingert has joined
970 2012-10-12 16:08:27 lggr has joined
971 2012-10-12 16:09:00 notgraingert has quit (Client Quit)
972 2012-10-12 16:09:18 notgraingert has joined
973 2012-10-12 16:11:23 graingert_ has joined
974 2012-10-12 16:11:23 notgraingert has quit (Read error: Connection reset by peer)
975 2012-10-12 16:11:51 graingert_ is now known as Guest99005
976 2012-10-12 16:11:57 Guest99005 is now known as graingert
977 2012-10-12 16:12:26 graingert has quit (Changing host)
978 2012-10-12 16:12:26 graingert has joined
979 2012-10-12 16:13:10 B0g4r7_ has joined
980 2012-10-12 16:14:39 sgornick has quit (Read error: Connection reset by peer)
981 2012-10-12 16:15:17 <kjj_> so... got rtorrent working. are you supposed to run it in screen or something?
982 2012-10-12 16:15:18 lggr has quit (Ping timeout: 246 seconds)
983 2012-10-12 16:15:22 graingert has quit (Read error: Connection reset by peer)
984 2012-10-12 16:15:50 B0g4r7 has quit (Ping timeout: 276 seconds)
985 2012-10-12 16:15:50 B0g4r7_ is now known as B0g4r7
986 2012-10-12 16:17:59 pusle has joined
987 2012-10-12 16:18:48 lggr has joined
988 2012-10-12 16:20:12 agricocb has quit (Quit: Leaving.)
989 2012-10-12 16:22:18 <kjj_> hey, question. is there a reason that the bootstrap happens AFTER -loadblock ?
990 2012-10-12 16:24:48 lggr has quit (Ping timeout: 245 seconds)
991 2012-10-12 16:27:24 aurigae has joined
992 2012-10-12 16:27:28 harkon has quit (Quit: Konversation terminated!)
993 2012-10-12 16:28:11 B0g4r7 has quit (Ping timeout: 276 seconds)
994 2012-10-12 16:28:27 lggr has joined
995 2012-10-12 16:29:09 BlackPrapor has joined
996 2012-10-12 16:29:26 harkon has joined
997 2012-10-12 16:30:04 JZavala has joined
998 2012-10-12 16:34:24 lggr has quit (Ping timeout: 246 seconds)
999 2012-10-12 16:36:53 harkon has quit (Ping timeout: 245 seconds)
1000 2012-10-12 16:38:10 att has quit (Ping timeout: 255 seconds)
1001 2012-10-12 16:38:13 lggr has joined
1002 2012-10-12 16:39:31 kreal has quit ()
1003 2012-10-12 16:40:51 tonikt has joined
1004 2012-10-12 16:43:08 toffoo has joined
1005 2012-10-12 16:44:13 rdponticelli has joined
1006 2012-10-12 16:44:28 lggr has quit (Ping timeout: 255 seconds)
1007 2012-10-12 16:47:43 lggr has joined
1008 2012-10-12 16:52:51 att has joined
1009 2012-10-12 16:53:39 lggr has quit (Ping timeout: 260 seconds)
1010 2012-10-12 16:54:50 freakazoid has joined
1011 2012-10-12 16:55:06 Diablo-D3 has joined
1012 2012-10-12 16:57:21 lggr has joined
1013 2012-10-12 17:04:35 lggr has quit (Ping timeout: 276 seconds)
1014 2012-10-12 17:04:46 denisx has quit (Quit: denisx)
1015 2012-10-12 17:06:46 lggr has joined
1016 2012-10-12 17:06:46 maaku has joined
1017 2012-10-12 17:07:19 davout has quit (Remote host closed the connection)
1018 2012-10-12 17:07:20 PhantomSpark has quit (Quit: Not all that glitter is gold not all who wander are lost.)
1019 2012-10-12 17:07:44 davout has joined
1020 2012-10-12 17:07:44 davout has quit (Changing host)
1021 2012-10-12 17:07:44 davout has joined
1022 2012-10-12 17:12:19 davout has quit (Ping timeout: 260 seconds)
1023 2012-10-12 17:12:22 drizztbsd has quit (Remote host closed the connection)
1024 2012-10-12 17:13:17 lggr has quit (Ping timeout: 272 seconds)
1025 2012-10-12 17:16:15 lggr has joined
1026 2012-10-12 17:16:31 <jgarzik> <kjj_> so... got rtorrent working. are you supposed to run it in screen or something? <<-- it is an ncurses, full screen client, for whatever that's worth
1027 2012-10-12 17:17:22 <jgarzik> <kjj_> hey, question. is there a reason that the bootstrap happens AFTER -loadblock ? <<-- just following a general "automatic after manual" pattern
1028 2012-10-12 17:17:27 <jgarzik> no hugely important reason
1029 2012-10-12 17:17:49 <kjj_> that one should be the other way around
1030 2012-10-12 17:18:58 <jgarzik> kjj_: <shrug> this way you can be sure you are importing a known good chain before any automatic machinery appears, thereby avoiding any forks
1031 2012-10-12 17:19:10 <jgarzik> but admittedly it is a weak argument either way
1032 2012-10-12 17:19:37 <kjj_> I'd say that the non-official-but-signed torrent should be loaded first, followed by whatever local crap people want to have
1033 2012-10-12 17:22:31 drazak_ has joined
1034 2012-10-12 17:22:34 lggr has quit (Ping timeout: 265 seconds)
1035 2012-10-12 17:24:00 _W_ has joined
1036 2012-10-12 17:25:31 <jgarzik> who is to say that a file named bootstrap.dat is the non-official-but-signed one? :)
1037 2012-10-12 17:25:35 TD has quit (Quit: TD)
1038 2012-10-12 17:25:58 lggr has joined
1039 2012-10-12 17:26:13 <kjj_> no one, but by making the torrents and posting on the forums, we are suggesting that we'd sorta like it to be
1040 2012-10-12 17:27:04 <BlueMatt> sipa: ack on the how-to-fix-ibd and separate archive+validation (at least in a ultraprune-is-validation, archival is just also serving blocks manner) but my problem with experimenting with other protocols for downloading archive data is, unless there is some other new method that Ive missed, using bittorrent/http/etc doesnt add anything and suddenly make the problem go away, the way I see it, the only reason bittorrent is faster for ibd/dow
1041 2012-10-12 17:27:04 <BlueMatt> nloading archive data is because its not as not-at-all optimized as bitcoin's ibd logic is, whereas if bitcoin's were optimized even only a reasonably amount, it should be able to beat torrent-bootstrap.dat+import
1042 2012-10-12 17:27:35 <BlueMatt> (in terms of time-to-sync)
1043 2012-10-12 17:28:12 nym has quit (Remote host closed the connection)
1044 2012-10-12 17:28:13 mndrix has quit (Remote host closed the connection)
1045 2012-10-12 17:29:39 TD has joined
1046 2012-10-12 17:30:33 <jgarzik> BlueMatt: time-to-sync is not the only metric. it is useful to reduce the cost/pain of running a full public node, by enabling some shifting of traffic away from them
1047 2012-10-12 17:31:29 <BlueMatt> if you want to run a public node to serve as a validator/relayer, just disable the NODE_NETWORK flag and enable the NODE_VALIDATOR or whatever
1048 2012-10-12 17:31:52 <BlueMatt> if you would otherwise run a bittorrent seed for bootstrap.dat, enable NODE_NETWORK and seed every block
1049 2012-10-12 17:32:17 <BlueMatt> which, imho, decreases the cost/pain of running seed nodes for both camps
1050 2012-10-12 17:32:26 jdnavarro has quit (Read error: Connection reset by peer)
1051 2012-10-12 17:32:27 lggr has quit (Ping timeout: 246 seconds)
1052 2012-10-12 17:33:07 JZavala has quit (Ping timeout: 240 seconds)
1053 2012-10-12 17:35:12 jdnavarro has joined
1054 2012-10-12 17:35:37 lggr has joined
1055 2012-10-12 17:38:30 <jgarzik> BlueMatt: that's highly technical handwaving about the future, though, ultimately ;p The truth is we have people right now abandoning full nodes, and the network is strongest when the maximum number of people are running full nodes.
1056 2012-10-12 17:38:40 denisx has joined
1057 2012-10-12 17:39:05 <jgarzik> ultraprune, better peer selection, and shifting traffic away are all things that reduce full node cost
1058 2012-10-12 17:39:07 <BlueMatt> its semi-technical handwaving about a future we can achieve reasonably quickly
1059 2012-10-12 17:39:57 <BlueMatt> ultimately, those who are moving off of full nodes should see ultraprune, realize all those costs go away, and come back
1060 2012-10-12 17:40:15 <TD> go away â¦. to some extent
1061 2012-10-12 17:40:17 * Luke-Jr grumbles about having to implement "serialize block height" yet again
1062 2012-10-12 17:40:17 <BlueMatt> in terms of bootstrapping at that point, sure, it may be expensive, but it will be expensive on the few public nodes
1063 2012-10-12 17:40:24 <TD> the future of full nodes is people who explicitly decide to help the network and donate servers
1064 2012-10-12 17:40:52 <gavinandresen> TD: nice job on the economist hangout!
1065 2012-10-12 17:40:55 da2ce7 has joined
1066 2012-10-12 17:41:55 <TD> thanks. in the end they didn't actually let me ask my question!
1067 2012-10-12 17:42:06 <TD> but interestingly, before the live part, tom (the editor) said he watched my talk
1068 2012-10-12 17:42:09 denisx has quit (Client Quit)
1069 2012-10-12 17:42:16 <TD> and he was very surprised and interested that bitcoin can do so many things using undocumented features
1070 2012-10-12 17:42:24 <TD> he was asking if they are part of the spec, etc. seemed pretty clued in.
1071 2012-10-12 17:42:36 <TD> i am semi-hopeful that in future the Economist might cover something contracts related.
1072 2012-10-12 17:42:47 lggr has quit (Ping timeout: 260 seconds)
1073 2012-10-12 17:42:56 <gavinandresen> very cool.
1074 2012-10-12 17:43:12 da2ce7_d has quit (Ping timeout: 260 seconds)
1075 2012-10-12 17:43:21 <gavinandresen> lots of very influential people read the Economist....
1076 2012-10-12 17:43:29 <jgarzik> like me! </grin>
1077 2012-10-12 17:43:36 <BlueMatt> TD: sure costs dont disappear, but the costs jgarzik is referring to (high disk access, can be high bw, etc) do
1078 2012-10-12 17:44:04 <BlueMatt> TD: and, yes, the future is those who want to help, which is why I say make it easy and keep it all in one simple place instead of running multiple servers...
1079 2012-10-12 17:44:05 <TD> haha
1080 2012-10-12 17:44:08 <TD> jgarzik will pass the laws we need
1081 2012-10-12 17:44:12 <jgarzik> BlueMatt: those costs never disappear for those serving the block chain data
1082 2012-10-12 17:44:59 <BlueMatt> jgarzik: not full serving, but validation/relaying recent blocks
1083 2012-10-12 17:44:59 <gmaxwell> jgarzik: they dont 'disappear' for ultraprune either.
1084 2012-10-12 17:45:11 <gmaxwell> they're just made less insane.
1085 2012-10-12 17:45:54 lggr has joined
1086 2012-10-12 17:45:55 jdnavarro has quit (Read error: Connection reset by peer)
1087 2012-10-12 17:46:08 <gmaxwell> What people are running is _not_ multibit from what I can tell, I believe electrum and blockchain.info webwallet are far more popular than that. (yes, sure people doing custom stuff are on bitcoinj more often; but from a client perspective? nah. They're moving onto thin clients)
1088 2012-10-12 17:46:34 <jgarzik> Ultraprune does make the block download faster. Ultraprune does not at all address the stress point of P2P nodes serving old block chain data.
1089 2012-10-12 17:46:57 <gmaxwell> I dub that jeff's imaginary stress point.
1090 2012-10-12 17:46:59 <jgarzik> For those on the P2P network but not on torrent, the cost/pain of running a public node is reduced.
1091 2012-10-12 17:47:04 <TD> yes. we need to fix that :) i don't know how popular electrum is. does he provide any usage figures?
1092 2012-10-12 17:47:53 <jgarzik> gmaxwell: You handwave away the "full node requires constant disk access" point, as if running a disk at 100%, 24/7 were irrelevant
1093 2012-10-12 17:47:59 <gmaxwell> jgarzik: please go do the math for seek counts for pulling vs bandwidth considering our current batching levels.
1094 2012-10-12 17:48:40 jdnavarro has joined
1095 2012-10-12 17:48:46 <BlueMatt> whether its a big pain or not, splitting nodes into full-serving and pruning/validating solves the issue either way...
1096 2012-10-12 17:48:47 <gmaxwell> jgarzik: no, I'm handwaving away that nodes need to be doing hundred+ megabits before you get to "running a disk at 100%, 24/7" and if nodes were doing that much traffic people would be made about that long before the disk!
1097 2012-10-12 17:48:49 <jgarzik> running the disk 24/7 reduces hardware lifetime
1098 2012-10-12 17:49:44 <gmaxwell> and you're handwaving away the fact that putting the same transfer burden on fewer nodes, with fewer motivations to run them _increases_ the hardware load.
1099 2012-10-12 17:50:08 <gavinandresen> gmaxwell jgarzik : ... but we all agree on the next high-priority thing to improve the situation, right? (pull ultraprune)
1100 2012-10-12 17:50:18 <jgarzik> hyperbole: you don't need to be doing "hundred+ megabits" of traffic, you only need to have (a) a public, listens on the Internet node and (b) pagecache memory smaller than the working set size -- quite realistic given current VPS sizes
1101 2012-10-12 17:50:29 <gmaxwell> jgarzik: No. You fail.
1102 2012-10-12 17:50:53 <gmaxwell> jgarzik: I already worked through the math on this for you, and that didn't work because you ignored itâ though I don't know why.
1103 2012-10-12 17:52:00 lggr has quit (Ping timeout: 255 seconds)
1104 2012-10-12 17:52:05 <gmaxwell> gavinandresen: seperate issue, reallyâ though of course, ultraprune needs to go next.
1105 2012-10-12 17:52:06 <gavinandresen> There is some really low-hanging fruit we could pick to increase performance, like making it easy to put the blockchain, wallet, and debug.log on separate physical disks.
1106 2012-10-12 17:52:08 <jgarzik> because I run public nodes and see the request rates and disk waits
1107 2012-10-12 17:52:27 <jgarzik> real world > imaginary
1108 2012-10-12 17:52:39 <gmaxwell> jgarzik: yes, and thats from block validation, and crazy getheaders load, not normal chain pulling.
1109 2012-10-12 17:53:23 <jgarzik> it is getheaders + getdata, not block validation
1110 2012-10-12 17:53:29 <jgarzik> primarily getdata
1111 2012-10-12 17:53:36 <gmaxwell> GO MEASURE IT. I have. Jesus christ.
1112 2012-10-12 17:54:19 <jgarzik> already have. had to increase the RAM in the public nodes before disk waits went away.
1113 2012-10-12 17:54:57 <gmaxwell> Or just do the mathâ producing big diskload from chain pulling simply doesn't make sense, we're reading a megabyte+ at at a time.
1114 2012-10-12 17:55:21 lggr has joined
1115 2012-10-12 17:55:23 <jgarzik> gmaxwell: times 50 clients, all of whom want vastly different parts of a big file
1116 2012-10-12 17:55:33 <gavinandresen> jgarzik: are you getting an extra heavy dose of IBD traffic because of dns seeding or something?
1117 2012-10-12 17:55:45 <jgarzik> gavinandresen: yes
1118 2012-10-12 17:55:57 <jgarzik> gavinandresen: my public nodes tend to see these issues before others, because of DNS seeding
1119 2012-10-12 17:55:59 <gavinandresen> ok. Maybe THAT's the problem that needs solving, then
1120 2012-10-12 17:56:08 <gmaxwell> jgarzik: sure, but it doesnt matter that the requests are random because you're still reading large chunks at once for each.
1121 2012-10-12 17:56:19 <TD> guys, take a chill pill
1122 2012-10-12 17:56:28 <jgarzik> gmaxwell: you are still hitting the disk, not pagecache
1123 2012-10-12 17:56:55 <TD> bitcoinj clients disproportionately hit nodes in the DNS seeds
1124 2012-10-12 17:57:03 <TD> jeff gave me hassle about it before. it's on the giant pile of must-fix issues
1125 2012-10-12 17:57:16 <jgarzik> gavinandresen: better peer selection is definitely needed to spread the load
1126 2012-10-12 17:57:17 <gmaxwell> jgarzik: yes and it doesn't matter because with 1mbyte random reads off disk you can't saturated it without hundreds of megabits of traffic.
1127 2012-10-12 17:57:52 <gmaxwell> Getheaders from SPV nodes is another matter however, because it doesn't send much data, so it doesn't result in many megabits. But torrents don't help that.
1128 2012-10-12 17:58:00 <jgarzik> gmaxwell: it is not an issue of disk saturation
1129 2012-10-12 17:58:08 <jgarzik> gmaxwell: it is simply hitting the disk at all -> reduced disk lifetime
1130 2012-10-12 17:58:20 <gavinandresen> jgarzik: let me know if throwing some high-memory high-bandwidth hardware at the problem might help, I could ask the Foundation for some budget for something like that
1131 2012-10-12 17:58:38 <TD> just one node that can serve the chain entirely off RAM would help
1132 2012-10-12 17:58:50 <jgarzik> gmaxwell: the cost of a full public node, is, therefore, constantly running the disk | ~6GB of cache memory
1133 2012-10-12 17:59:02 <gmaxwell> jgarzik: then you can't run bitcoin because transaction circulation will _always_ make you hit the disk several times per second unless you have enough memory to have the whole working set in ram.
1134 2012-10-12 17:59:22 <gmaxwell> (and that has nothing to do with IBD, it's just validation)
1135 2012-10-12 17:59:28 <TD> ultra prune makes the latter quite feasible even for a long time yet to come
1136 2012-10-12 17:59:41 <jgarzik> gmaxwell: You are getting close
1137 2012-10-12 18:00:23 <jgarzik> gmaxwell: Transaction validation is spread out over time, and likely to be more-recent-than-not.
1138 2012-10-12 18:00:58 <jgarzik> gavinandresen: Foundation should definitely run a few full nodes, with enough RAM (I'd recommend 8GB today) to cache everything
1139 2012-10-12 18:01:16 <jgarzik> gavinandresen: Stick those in the fallnode node list on the wiki
1140 2012-10-12 18:01:25 <jgarzik> gavinandresen: and run a DNS seed (separate from our other seeds)
1141 2012-10-12 18:01:33 <gmaxwell> jgarzik: to giving up on working with you? Look: I measured this. I gave you figured. I ran it theoretically. I don't disagree with what you saw in practice on actual nodes, and TD makes a fine point about ultraprune. But you are dead wrong about torrents doing anything to fix this, and probably wrong that there are not risks created by it. And I would put a substantial 'bet' on any contest but I'd feel bad for taking your money.
1142 2012-10-12 18:02:40 lggr has quit (Ping timeout: 255 seconds)
1143 2012-10-12 18:03:00 harkon has joined
1144 2012-10-12 18:03:52 <jgarzik> Torrents quite clearly shift some costs away from public P2P network nodes. How much cost? That is open to debate and practice. In times of version release, the swarm might even help absorb bursts of new users on the network. The set of torrent seeds is disjoint but intersecting with the P2P network, and torrent seeds are necessarily self-selected volunteers absorbing those costs.
1145 2012-10-12 18:04:05 <jgarzik> If there are very few interested parties, absolutely, it will be of no use.
1146 2012-10-12 18:04:30 <jgarzik> But there are seeders and seeding mechanisms who are not necessarily interested in running full nodes
1147 2012-10-12 18:04:52 <jgarzik> This gives them an easy, known-quantity way to help
1148 2012-10-12 18:05:36 lggr has joined
1149 2012-10-12 18:06:19 <gmaxwell> Indeed, but does this counterbalance the increase likelyhood of failure from resource seperation instead of pooling? I don't know. It doesn't worry me much nowâ I only started protesting again when you again started talking about this as though it were a fix for the broken ibdness.
1150 2012-10-12 18:07:33 <jgarzik> gmaxwell: I do not recall ever claiming that
1151 2012-10-12 18:08:06 <jgarzik> bitcoind should proceed with all possible speed towards better peer selection and ultraprune
1152 2012-10-12 18:08:18 tonikt2 has quit (Remote host closed the connection)
1153 2012-10-12 18:08:27 <gmaxwell> (You're basically taking load which was previously spread across 10,000+ nodes and putting it on a far smaller number of probably better prepared ones but which require additional effort to setup. This sounds quite risky.)
1154 2012-10-12 18:08:56 <jgarzik> from the standpoint of P2P technology, bitcoin is new while torrent is a known quantity and widespread
1155 2012-10-12 18:08:57 <gmaxwell> jgarzik: er, well perhaps I misunderstood. Bluematt complained about IBD and I thought you responded with the torrent.
1156 2012-10-12 18:09:53 tonikt2 has joined
1157 2012-10-12 18:10:02 <gmaxwell> jgarzik: pretty lossy for a known quantity and widespread. :( But regardless, by definition anyone who wants the bitcoin data is a bitcoin user, not so a torrent user... as I mentioned when this came up before: torrent is blocked on my office network; bitcoin is not.
1158 2012-10-12 18:10:54 <jgarzik> gmaxwell: no, never said torrent was a replacement or fix for IBD. I emphasize this in the forum torrent thread as well.
1159 2012-10-12 18:11:05 TD has quit (Quit: TD)
1160 2012-10-12 18:11:08 <jgarzik> it is self-evident that IBD must be fixed for recent blocks
1161 2012-10-12 18:11:16 <BlueMatt> the way I see it: whether jgarzik is right in terms of load or not, we can accomplish the same thing (which is probably a good thing) that the bootstrap seed is trying to accomplish by splitting nodes into pruned and seeding, some people have enough bw to seed, others may not, and we can get the same speed (easily) that bittorrent provides within bitcoin's framework
1162 2012-10-12 18:11:21 <jgarzik> and torrent does not address this at all
1163 2012-10-12 18:11:40 jrmithdobbs has quit (Ping timeout: 260 seconds)
1164 2012-10-12 18:12:21 lggr has quit (Ping timeout: 265 seconds)
1165 2012-10-12 18:12:48 jrmithdobbs has joined
1166 2012-10-12 18:13:00 <jgarzik> BlueMatt: What I'm doing with torrent is cost shifting off the P2P network. Reduce the cost of running a full P2P node, and more people will run full nodes / less people will abandon full nodes.
1167 2012-10-12 18:13:15 <BlueMatt> and, if we keep it within bitcoin's p2p, we have more control of encouraging people to upgrade to full seed nodes, wheras in a torrent, we dont
1168 2012-10-12 18:13:39 <BlueMatt> jgarzik: ack, but I argue it provides a worse experience for initial users anyway
1169 2012-10-12 18:13:52 MC1984 has joined
1170 2012-10-12 18:13:55 <BlueMatt> if the instructions on getting started are: torrent this, run bitcoin with -loadblock, run, its pretty ugly
1171 2012-10-12 18:14:12 <jgarzik> BlueMatt: bootstrap.dat is automatically imported
1172 2012-10-12 18:14:23 <BlueMatt> ok, fine, but you still have to torrent first
1173 2012-10-12 18:14:29 <BlueMatt> also, when did that get merged?
1174 2012-10-12 18:14:37 <jgarzik> BlueMatt: > 0.7
1175 2012-10-12 18:14:45 <BlueMatt> 0.7.1?
1176 2012-10-12 18:14:49 <jgarzik> yes
1177 2012-10-12 18:14:54 <BlueMatt> pull link?
1178 2012-10-12 18:15:17 lggr has joined
1179 2012-10-12 18:15:29 <jgarzik> BlueMatt: #1684 i think
1180 2012-10-12 18:15:48 spq has quit (Read error: Connection reset by peer)
1181 2012-10-12 18:15:50 <gmaxwell> TD[gone]: at current tx out increasing growth rates it looks like it will be under a year before it doesn't fit in 4g ram. hopefully we can dampen this by making the wallet autosweep all the 1e-8 outputs created by dice.
1182 2012-10-12 18:16:16 <BlueMatt> jgarzik: ah, ok
1183 2012-10-12 18:16:18 <gmaxwell> :( (http://bitcoin.sipa.be/pruning-size.png )
1184 2012-10-12 18:16:32 <BlueMatt> in any case, you still are making people torrent initially just to get started, which is very, very broken
1185 2012-10-12 18:17:03 <gmaxwell> BlueMatt: might be better to just have a "jumbo install" of bitcoin thats available only via torrent, which has everything prepackaged.
1186 2012-10-12 18:17:10 <gmaxwell> and installed via a common installer.
1187 2012-10-12 18:17:15 <BlueMatt> gmaxwell: no
1188 2012-10-12 18:17:18 <gmaxwell> then at least it's just one step.
1189 2012-10-12 18:17:20 <BlueMatt> none of that is better
1190 2012-10-12 18:17:27 <gmaxwell> better than two steps.
1191 2012-10-12 18:17:38 umop3plsdn has quit ()
1192 2012-10-12 18:17:43 <gmaxwell> E.g. regular download, vs jumbo download (may reduce install times!).
1193 2012-10-12 18:17:54 <BlueMatt> better is solving the problem
1194 2012-10-12 18:18:12 <BlueMatt> the issue I have with all of this discussion is: we all know (roughly) the problem, but we arent addressing it, we are working around it
1195 2012-10-12 18:18:22 <BlueMatt> and addressing it isnt /that/ hard
1196 2012-10-12 18:18:37 <gmaxwell> BlueMatt: some people really want to preload for varrious reasons.. for example, they want to run bitcoin at home where bandwidth is expensive but can download it at school/office where it's free.
1197 2012-10-12 18:19:10 <gmaxwell> or they want to run over tor... and it's not too kind of us to transfer many gigs of data over tor.
1198 2012-10-12 18:19:18 <BlueMatt> gmaxwell: and I dont see anything wrong with those users tormenting the bootstrap.dat somewhere and using it...if you have a weird use-case, then the setup may be 2-3 steps, fine
1199 2012-10-12 18:19:50 <BlueMatt> but for many users who dont have bw limits, and want to just download and get started with no crazy stuff going on...
1200 2012-10-12 18:19:54 <jgarzik> BlueMatt: I think you (and previously, gmaxwell) are misunderstanding this torrent project as Solving The Problem
1201 2012-10-12 18:20:04 <jgarzik> when it's really just cost shifting and providing additional options
1202 2012-10-12 18:20:08 <gmaxwell> perhaps. Okay. I do agree with you, but I don't think bootstrap is bad on its face. It's just bad as a general solution.
1203 2012-10-12 18:20:20 <jgarzik> a heterogenous environment that spreads blockchain data far as wide is a good thing, IMO
1204 2012-10-12 18:20:30 <gmaxwell> Options are probably okay (but not certantly!).
1205 2012-10-12 18:20:44 <BlueMatt> providing additional options Im all for, but I think its cost shifting out of the p2p net where we should be working on cost shifting between parts of the p2p network
1206 2012-10-12 18:21:15 <BlueMatt> I dont have an issue with providing a torrent as-is, sure go for it
1207 2012-10-12 18:21:32 lggr has quit (Ping timeout: 245 seconds)
1208 2012-10-12 18:21:35 <BlueMatt> but providing a torrent in a situation where people will see it as "this is what we should tell all new users to do to get started" is not ok
1209 2012-10-12 18:21:58 <gmaxwell> (moreover, a future rendition of bootstrap could read a compressed format that knocks half off the size, but doesnt make general sense for parallized network fetches)
1210 2012-10-12 18:22:28 Joric has joined
1211 2012-10-12 18:23:00 mndrix has joined
1212 2012-10-12 18:23:03 <jgarzik> I don't see how "cost shifting between parts of the p2p network" would reduce the costs of running a full node -- the type of nodes we need the most
1213 2012-10-12 18:23:09 <gmaxwell> BlueMatt: I do sense a little irony hereâ weren't you the first person to offer package blockchain downloads? They're very popular. At times I've had a hard time convincing p2pool users not to use them.
1214 2012-10-12 18:23:13 <jgarzik> Better peer selection will reduce their costs
1215 2012-10-12 18:23:15 Raccoon` has joined
1216 2012-10-12 18:23:17 <jgarzik> a lot
1217 2012-10-12 18:23:43 <jgarzik> Torrent will reduce their costs, maybe too small to be outside noise range, maybe not
1218 2012-10-12 18:23:51 <gmaxwell> jgarzik: better IBD logic couplied with peer selection could even allow cache specialization, in ways that bittorrent does not.
1219 2012-10-12 18:24:03 <BlueMatt> gmaxwell: thats fair; lets just say I saw the light :)
1220 2012-10-12 18:24:39 <BlueMatt> jgarzik: it means you can run a full node (that only verifies, not actually serves the full chain) without having the costs you are talking about, which is what we need, serving the chain can be a select few just fine
1221 2012-10-12 18:24:43 <BlueMatt> (as long as we do headers sync)
1222 2012-10-12 18:24:47 lggr has joined
1223 2012-10-12 18:24:52 <gmaxwell> e.g. if you're doing reverse header sync and can pull any blocks you hash(peer IP + block number) to create preference lists for each peer for blocks you're going to fetch.
1224 2012-10-12 18:24:58 <jgarzik> I don't want serving the chain to be a select few
1225 2012-10-12 18:25:01 <jgarzik> that's #FAIL
1226 2012-10-12 18:25:13 <BlueMatt> thats what the torrent does...
1227 2012-10-12 18:25:21 <BlueMatt> (I mean serving the ibd chain, not recent blocks)
1228 2012-10-12 18:25:42 Raccoon has quit (Ping timeout: 245 seconds)
1229 2012-10-12 18:25:43 Raccoon` is now known as Raccoon
1230 2012-10-12 18:26:15 <helo> so torrenting the chain means participants are sending unverified data that is supposedly a relatively recent blockchain?
1231 2012-10-12 18:26:18 <BlueMatt> gmaxwell: re: Ive run it before...mostly the change is, its now more obvious how a nice network split could be easily accomplished instead of it being more "down the road idealism"
1232 2012-10-12 18:26:27 <BlueMatt> helo: not really...
1233 2012-10-12 18:26:28 <jgarzik> helo: no, not at all
1234 2012-10-12 18:26:31 <gmaxwell> helo: it's verfied on import unlike the old snapshots.
1235 2012-10-12 18:27:32 <gmaxwell> BlueMatt: network 'split' is scarry and bad.
1236 2012-10-12 18:28:07 <BlueMatt> ehh, you know what I mean, and its not that bad
1237 2012-10-12 18:28:11 <helo> ok good :)
1238 2012-10-12 18:28:35 spq has joined
1239 2012-10-12 18:29:29 <gmaxwell> BlueMatt: having different kind of nodes opens up some tricky peer selection problems with many implications, it just needs to be handled carefully to not increase partition/sybil vulnerability.
1240 2012-10-12 18:29:48 <MC1984> our terms of service have changed, please accpt before proceeding. If you do not accept, you cant even look at let alone access any of your shit
1241 2012-10-12 18:29:54 <MC1984> the great cloud computing revolution
1242 2012-10-12 18:30:01 <gmaxwell> For example, if a new node only talks to 'archive' nodes.. and only 1% of the total nodes are archive, then we have a tiny fraction of the sybil resistance for new nodes.
1243 2012-10-12 18:30:16 <BlueMatt> gmaxwell: then all nodes serve full set of headers
1244 2012-10-12 18:30:43 agent8423 has joined
1245 2012-10-12 18:30:45 <BlueMatt> its not /that/ hard to get it right, esp if you do header syn
1246 2012-10-12 18:30:45 <BlueMatt> c
1247 2012-10-12 18:31:11 lggr has quit (Ping timeout: 252 seconds)
1248 2012-10-12 18:31:19 <jgarzik> "bitcoind signrawtransaction fails, but bitcoind decoderawtransaction succeeds?" - https://bitcointalk.org/index.php?topic=117992.0
1249 2012-10-12 18:31:22 <agent8423> will the torrent be include in the next release?
1250 2012-10-12 18:31:27 <agent8423> included
1251 2012-10-12 18:31:32 nym has joined
1252 2012-10-12 18:31:51 <BlueMatt> agent8423: NO!
1253 2012-10-12 18:32:12 <agent8423> that sucks
1254 2012-10-12 18:32:17 <BlueMatt> no it doesnt
1255 2012-10-12 18:32:22 <kjj_> jgarzik: hopefully signrawtransaction is unhappy about the extra crap he's sticking into his scripts
1256 2012-10-12 18:32:28 <gmaxwell> agent8423: It's not something thats "included" in any case, it's something you download.
1257 2012-10-12 18:32:37 <MC1984> guys stop talking about archive nodes srsly
1258 2012-10-12 18:32:45 <jgarzik> agent8423: Version 0.7.1 will automatically import bootstrap.dat, the file downloaded from the torrent.
1259 2012-10-12 18:32:49 <BlueMatt> agent8423's comments make exactly clear why we shouldnt have such a torrent that is "official"ish
1260 2012-10-12 18:32:55 <jgarzik> agent8423: so what BlueMatt is saying is not strictly true
1261 2012-10-12 18:33:15 <agent8423> ah so will bitcoinqt torrent the blockchain next release?
1262 2012-10-12 18:33:21 <jgarzik> agent8423: no
1263 2012-10-12 18:33:23 <BlueMatt> MC1984: do you have an argument or do you just like to complain?
1264 2012-10-12 18:33:54 <MC1984> one of either
1265 2012-10-12 18:34:17 <BlueMatt> when jgarzik releases a torrent, its seen as official by many and then it ends up on the "how to get started with bitcoin-qt" threads, which just looks terrible
1266 2012-10-12 18:34:27 lggr has joined
1267 2012-10-12 18:35:12 <jgarzik> BlueMatt: which is just fine -- that is people self-selecting to offload costs from the P2P network
1268 2012-10-12 18:35:14 <agent8423> when the US government needs Bitcoin to comply with antimoney laundering laws will you guys comply in your development?
1269 2012-10-12 18:35:21 <jgarzik> their choice, and it helps full nodes
1270 2012-10-12 18:35:25 <BlueMatt> jgarzik: uhhh...no its not
1271 2012-10-12 18:35:38 <kjj_> ugh.
1272 2012-10-12 18:35:46 <kjj_> agent8423: can't you leave that shit on the forums where it belongs?
1273 2012-10-12 18:35:50 <jgarzik> agent8423: offtopic for #bitcoin-dev
1274 2012-10-12 18:36:17 <agent8423> but youre developers for Bitcoin? who am i supposed to ask?
1275 2012-10-12 18:36:31 <gmaxwell> agent8423: uh. You're confused. There isn't anything like that we could do; nor are there any laws someone should rationally expect to be applicable there. Go ask coal mining equiment makers about their conformance with 'antimoney laundering laws'. :)
1276 2012-10-12 18:36:34 ThomasV_ has joined
1277 2012-10-12 18:36:37 <BlueMatt> jgarzik: its fine if an individual has a need for it and does it, and if its mentioned as an option, but it will no doubt be mentioned as heres how ALL users should get started, which is horrible ux and makes it all seem as poor as it is
1278 2012-10-12 18:36:48 <Luke-Jr> I would appreciate if anyone can provide code reviews of new coinbase-creation code, especially the base58 parsing: https://bitcointalk.org/?topic=78192.msg1267851#msg1267851
1279 2012-10-12 18:37:01 <kjj_> agent8423: it has been asked and answered a million times on the forums. the government can ask us to change math, but we still can't do it
1280 2012-10-12 18:37:02 <Luke-Jr> (the "code" link has the full diff)
1281 2012-10-12 18:37:19 <MC1984> BlueMatt is right
1282 2012-10-12 18:37:20 denisx has joined
1283 2012-10-12 18:37:33 <agent8423> kjj_: you can release software that complies. you may have to fork the chain but you will have to.
1284 2012-10-12 18:37:34 <Luke-Jr> agent8423: Bitcoin is just cash. AML laws are for financial instutition
1285 2012-10-12 18:37:38 <agent8423> we live in a nation of laws
1286 2012-10-12 18:37:41 <MC1984> wary of hacky this becoming de facto standard procedure
1287 2012-10-12 18:38:01 <jgarzik> Luke-Jr: bitcoins are digital messages, not cash :)
1288 2012-10-12 18:38:09 <kjj_> agent8423: math is math. congress can't change math.
1289 2012-10-12 18:38:21 <Luke-Jr> jgarzik: I'm simplifying :p
1290 2012-10-12 18:38:22 <agent8423> schumer doesn't think so
1291 2012-10-12 18:38:24 <gmaxwell> agent8423: in any case, your question isn't concrete enough to form a meaningful answer.
1292 2012-10-12 18:38:29 <BlueMatt> next person to mention aml gets kicked
1293 2012-10-12 18:38:59 <MC1984> he opped up, he means it
1294 2012-10-12 18:39:13 agent8423 has quit (Quit: Page closed)
1295 2012-10-12 18:39:25 <kjj_> BlueMatt: do you object to the use of bootstrap/torrent, or do you just think that the effort would be better spent on improving the IDB?
1296 2012-10-12 18:39:42 <BlueMatt> kjj_: the second
1297 2012-10-12 18:39:48 PhantomSpark has joined
1298 2012-10-12 18:39:53 <BlueMatt> kjj_: there are legitimate uses, sure, but there are also many not-so-legit uses
1299 2012-10-12 18:40:08 lggr has quit (Ping timeout: 240 seconds)
1300 2012-10-12 18:40:13 <gmaxwell> BlueMatt: I suspect the constant flood of trolls on here and on the forum is wearing us out, 'cause its making us rude to each other and random innocent people with silly questionsâ¦
1301 2012-10-12 18:40:13 <BlueMatt> (to be fair, its not /that/ much effort to keep a torrent seeding, but, still...you get the point)
1302 2012-10-12 18:40:32 <BlueMatt> gmaxwell: probably
1303 2012-10-12 18:40:40 <kjj_> BlueMatt: yeah, I just wanted to clarify the nature of the issue.
1304 2012-10-12 18:40:41 tonikt has quit (Quit: Leaving)
1305 2012-10-12 18:42:13 <kjj_> and well, like it or not, the torrent is here to stay. I'd say that if we pulled it down, someone would just put up another, complete with the index, and then people would get no verification at all
1306 2012-10-12 18:42:26 TD has joined
1307 2012-10-12 18:42:39 <gmaxwell> kjj_: you responded to me like that, and it makes me think you're totally confused.
1308 2012-10-12 18:42:49 <gmaxwell> there have _long_ been torrents of the blockchain data.
1309 2012-10-12 18:42:55 <gmaxwell> as in over a year ago.
1310 2012-10-12 18:43:07 <kjj_> hmm, ok. I guess I never looked.
1311 2012-10-12 18:43:15 <gmaxwell> That isn't what we're discussing. And your "like it or not" response is silly in that light.
1312 2012-10-12 18:43:15 <BlueMatt> kjj_: I know there will be torrents, because it helps, my opposition is to there being one which becomes semi or seen as official
1313 2012-10-12 18:43:44 agricocb has joined
1314 2012-10-12 18:44:06 <kjj_> ok, fair enough
1315 2012-10-12 18:44:08 lggr has joined
1316 2012-10-12 18:44:19 agricocb has quit (Client Quit)
1317 2012-10-12 18:44:23 <MC1984> isnt that why the chain comes in 2gb chunks, to move them around by mens such as torrents?
1318 2012-10-12 18:44:39 <BlueMatt> no, its so that you dont choke on fs'es like fat
1319 2012-10-12 18:44:49 <kjj_> MC1984: no, 2gb is the lowest common denominator of filesystems
1320 2012-10-12 18:44:49 <MC1984> oh right
1321 2012-10-12 18:44:56 <gmaxwell> BlueMatt objects to it as a mechnism for IBD because it promotes resource paritioning and discourages solving underlying problems which _must_ be solved. I think I feel more strongly that its not useful, less strongly that it's bad: its more useful for things like tor users.
1322 2012-10-12 18:45:07 <gmaxwell> The torrent has a >2GB file in any case: bootstrap must be a single file.
1323 2012-10-12 18:46:37 <MC1984> its neato but not desireable in lieu of real improvements
1324 2012-10-12 18:46:43 <MC1984> its not the end of bitcoin though
1325 2012-10-12 18:47:01 <jgarzik> it's not in lieu of improvements
1326 2012-10-12 18:47:07 <BlueMatt> I dont think anyone called it the end of bitcoin...
1327 2012-10-12 18:47:28 <MC1984> yeah i know just sayin
1328 2012-10-12 18:47:41 <gmaxwell> I don't even think it's neato. ::shrugs:: It's a copout that increases the install complexity if widely used, makes being a full contributor to the network harder, and dillutes the message of "if you want to contribute to the network run a full node, and potentially mine"
1329 2012-10-12 18:47:54 <Luke-Jr> I think arguing over it is more of a waste than anything else. *shrug*
1330 2012-10-12 18:48:04 <gmaxwell> here here.
1331 2012-10-12 18:48:20 <gmaxwell> I apologize. luke is right on that point. I'll say no more for now. :(
1332 2012-10-12 18:50:17 lggr has quit (Ping timeout: 245 seconds)
1333 2012-10-12 18:51:45 davout has joined
1334 2012-10-12 18:52:30 davout has quit (Remote host closed the connection)
1335 2012-10-12 18:53:34 davout has joined
1336 2012-10-12 18:53:42 lggr has joined
1337 2012-10-12 18:55:23 molecular has quit (Ping timeout: 252 seconds)
1338 2012-10-12 18:55:32 molecular has joined
1339 2012-10-12 18:55:50 mndrix has quit (Remote host closed the connection)
1340 2012-10-12 18:56:36 <MC1984> hey is multisig on general release yet
1341 2012-10-12 18:56:47 <BlueMatt> if you use rawtxn yea
1342 2012-10-12 18:56:55 <gmaxwell> 0_o
1343 2012-10-12 18:57:03 <MC1984> sounds complicated
1344 2012-10-12 18:58:08 D34TH has joined
1345 2012-10-12 18:58:28 sixx has joined
1346 2012-10-12 19:00:02 <gavinandresen> it is complicated. Making it easy is hard.
1347 2012-10-12 19:00:25 <MC1984> no doubt
1348 2012-10-12 19:00:34 lggr has quit (Ping timeout: 240 seconds)
1349 2012-10-12 19:00:57 <MC1984> i just hope it helps stem the tide of fraud once people actually start using it
1350 2012-10-12 19:01:27 <MC1984> then again i think this waterfall of fraud is useful for teaching people not to be so stupid and greedy in a way
1351 2012-10-12 19:01:30 nym has quit (Remote host closed the connection)
1352 2012-10-12 19:01:33 <Luke-Jr> hmm
1353 2012-10-12 19:01:40 <Luke-Jr> I suppose the anti-fraud case might not be too hard to implement
1354 2012-10-12 19:02:11 <Luke-Jr> but meh @ QtGui
1355 2012-10-12 19:02:46 <MC1984> someone else does the gui then eh :P
1356 2012-10-12 19:02:54 <MC1984> division of labour gentlemen
1357 2012-10-12 19:03:27 lggr has joined
1358 2012-10-12 19:03:37 jdnavarro has quit (Ping timeout: 245 seconds)
1359 2012-10-12 19:03:41 <kjj_> currently, only (m-of-n) is supported in p2sh, right?
1360 2012-10-12 19:04:21 <BlueMatt> s/is supported/is standard/
1361 2012-10-12 19:04:27 <kjj_> er yeah, is standard
1362 2012-10-12 19:04:40 pppingme has joined
1363 2012-10-12 19:04:43 davout has quit (Remote host closed the connection)
1364 2012-10-12 19:04:47 <kjj_> but that means that no one will relay anything else, doesn't it?
1365 2012-10-12 19:05:01 <Luke-Jr> MC1984: the GUI is basically all that is needed at this point
1366 2012-10-12 19:05:21 BCB has joined
1367 2012-10-12 19:05:32 <jgarzik> gavinandresen: Actually, as I found while writing the "Atomic coin swapping" example in https://bitcointalk.org/index.php?topic=112007.msg1212356#msg1212356 ... at this to me, bitcoin stuff is sufficiently complicated that there is a dearth of examples -- however complex or difficult to reproduce
1368 2012-10-12 19:05:50 <jgarzik> I haven't seen a step-by-step cookbook example of a multisig transaction, even for advanced programmers
1369 2012-10-12 19:06:01 <jgarzik> the "making it easy" step is far away ;p
1370 2012-10-12 19:06:06 <gmaxwell> jgarzik: now you have: https://people.xiph.org/~greg/escrowexample.txt
1371 2012-10-12 19:06:14 <Luke-Jr> ^ see, I knew it existed! :P
1372 2012-10-12 19:06:34 <jgarzik> neat :)
1373 2012-10-12 19:06:53 <jgarzik> maybe I need to start a "bitcoin cookbook" thread
1374 2012-10-12 19:06:55 <jgarzik> or wiki page
1375 2012-10-12 19:07:06 <gavinandresen> bitcoin cookbook is a great idea.
1376 2012-10-12 19:07:19 <gmaxwell> I also have https://people.xiph.org/~greg/signdemo.txt I realy intended to clean these up and put them on the forum.. but I've apparently been wasting time having dumb pointles arguments here instead, sorry. :(
1377 2012-10-12 19:07:47 nym has joined
1378 2012-10-12 19:08:02 <gmaxwell> I also have a half written one for doing joint input transactions (e.g. alice and malory atomically pay bob)
1379 2012-10-12 19:08:32 <kjj_> I would consider making "A or (B AND C)" standard sooner rather than later. That is one that will spur development of web services, protocols, etc.
1380 2012-10-12 19:09:28 <Joric> vanity pool gets really popular https://vanitypool.appspot.com/solvedWork
1381 2012-10-12 19:09:45 lggr has quit (Ping timeout: 256 seconds)
1382 2012-10-12 19:09:52 <Luke-Jr> Joric: the future for GPU mining?
1383 2012-10-12 19:10:06 nym has quit (Remote host closed the connection)
1384 2012-10-12 19:10:07 pusle has quit ()
1385 2012-10-12 19:10:12 <Joric> you bet
1386 2012-10-12 19:10:18 <kjj_> yeah, that's probably what I'll be doing with mine about 20 difficulty from now
1387 2012-10-12 19:10:47 <kjj_> that site really needs to have a difficulty calculator, and a column showing reward vs. diff
1388 2012-10-12 19:10:48 nym has joined
1389 2012-10-12 19:11:18 bcb_ has joined
1390 2012-10-12 19:11:38 harkon has quit (Quit: Konversation terminated!)
1391 2012-10-12 19:11:58 <Joric> i see a few 8 char addresses (like 1GMaxweLLbo8mdXvnnC19Wt2wigiYUKgEB)
1392 2012-10-12 19:12:12 <Joric> 1ABCDEFGH for instance
1393 2012-10-12 19:12:26 <gmaxwell> Joric: I generated that on my own.
1394 2012-10-12 19:12:43 <Joric> yeah i wonder why they would need a pool for that
1395 2012-10-12 19:12:55 lggr has joined
1396 2012-10-12 19:12:57 <Luke-Jr> â¦
1397 2012-10-12 19:12:58 darsk1ez has quit (Ping timeout: 246 seconds)
1398 2012-10-12 19:13:03 <gmaxwell> huh? thats listed in the pool? someone is trying to clone me?
1399 2012-10-12 19:13:07 <Joric> whats the estimated difficulty for 9 chars
1400 2012-10-12 19:13:29 <Joric> gmaxwell, nope, i meant there are mostly 7 chars
1401 2012-10-12 19:13:53 BCB has quit (Ping timeout: 245 seconds)
1402 2012-10-12 19:14:13 <gmaxwell> Joric: well it takes a 58^8 operations for 8 characters on average.
1403 2012-10-12 19:14:14 <MC1984> the hell is vanity pool
1404 2012-10-12 19:14:25 <Luke-Jr> MC1984: mining for vanity addresses
1405 2012-10-12 19:14:29 <Luke-Jr> MC1984: like 3P14159f73E4gFr7JterCCQh9QjiTjiZrG
1406 2012-10-12 19:14:32 <gmaxwell> Securely.
1407 2012-10-12 19:14:36 <gmaxwell> er securely.
1408 2012-10-12 19:14:37 <Luke-Jr> except I'm not sure that pool supports P2SH yet
1409 2012-10-12 19:14:48 <MC1984> oh wow
1410 2012-10-12 19:14:51 <gmaxwell> Luke-Jr: nope, nor compressed public keys.
1411 2012-10-12 19:15:07 <gmaxwell> which kinda sucks, compressed public keys significantly reduce transaction sizes
1412 2012-10-12 19:15:07 <Joric> Luke-Jr, why p2sh? it's all done at vanitygen level
1413 2012-10-12 19:15:23 <Luke-Jr> Joric: why not?
1414 2012-10-12 19:15:27 darsk1ez has joined
1415 2012-10-12 19:15:28 <Luke-Jr> Joric: everything should be P2SH ;)
1416 2012-10-12 19:15:47 nym has quit (Remote host closed the connection)
1417 2012-10-12 19:16:16 <Joric> huh someone ordered 1Casascius for 3.2 BTC
1418 2012-10-12 19:16:31 mndrix has joined
1419 2012-10-12 19:16:32 mndrix has quit (Changing host)
1420 2012-10-12 19:16:32 mndrix has joined
1421 2012-10-12 19:16:32 mndrix has quit (Changing host)
1422 2012-10-12 19:16:32 mndrix has joined
1423 2012-10-12 19:16:33 root2 has quit (Read error: Connection reset by peer)
1424 2012-10-12 19:16:44 <MC1984> so peple are getting paid to mine bitcoin addresses instead of coins
1425 2012-10-12 19:16:50 <MC1984> lol ww
1426 2012-10-12 19:16:50 root2 has joined
1427 2012-10-12 19:17:10 <Joric> i also like 1Muhammad (donation address? :)
1428 2012-10-12 19:17:22 rcorreia has joined
1429 2012-10-12 19:17:34 <MC1984> is vanitypool more profitable than actual mining or what
1430 2012-10-12 19:17:50 <kjj_> almost certainly not
1431 2012-10-12 19:17:56 <Joric> MC1984, quite possible, depends of the bounty
1432 2012-10-12 19:18:41 mndrix has quit (Remote host closed the connection)
1433 2012-10-12 19:19:28 <Luke-Jr> MC1984: for CPUs probably
1434 2012-10-12 19:19:34 lggr has quit (Ping timeout: 240 seconds)
1435 2012-10-12 19:19:46 <Luke-Jr> also, note that I think they're mining for *all* of these at once (not sure)
1436 2012-10-12 19:19:58 <kjj_> yeah, they are
1437 2012-10-12 19:20:17 <Joric> it's not that popular yet if you're a big miner you could collect all those bounties in a few days
1438 2012-10-12 19:20:21 <kjj_> the CLI client is getting changed to check both compressed and uncompressed keys, doubling the efficiency of the search
1439 2012-10-12 19:21:05 FSAnode43 has joined
1440 2012-10-12 19:21:10 <FSAnode43> https://bitcointalk.org/index.php?topic=118112.0
1441 2012-10-12 19:21:14 FSAnode43 has quit (Client Quit)
1442 2012-10-12 19:21:37 <kjj_> go figure, Atlas
1443 2012-10-12 19:22:59 lggr has joined
1444 2012-10-12 19:23:32 kreal has joined
1445 2012-10-12 19:23:33 root2 has quit (Read error: Connection reset by peer)
1446 2012-10-12 19:23:54 root2 has joined
1447 2012-10-12 19:24:48 <MC1984> wtf is wrong with that atlas guy
1448 2012-10-12 19:25:03 <Luke-Jr> MC1984: he's on a mission to kill Bitcoin apparently
1449 2012-10-12 19:25:06 <kjj_> looks to me like he wants to be un-un-banned
1450 2012-10-12 19:25:26 <MC1984> you can bust dev balls without sounding like a wingnut
1451 2012-10-12 19:25:47 <Luke-Jr> MC1984: he wants Bitcoin to be "Silk Road currency" or something
1452 2012-10-12 19:27:13 harkon has joined
1453 2012-10-12 19:29:08 lggr has quit (Ping timeout: 260 seconds)
1454 2012-10-12 19:30:53 TD has quit (Quit: TD)
1455 2012-10-12 19:31:15 xIsalty has quit (Remote host closed the connection)
1456 2012-10-12 19:32:45 lggr has joined
1457 2012-10-12 19:33:06 <jgarzik> "Bitcoin cookbook" - https://bitcointalk.org/index.php?topic=118126.0
1458 2012-10-12 19:33:19 <jgarzik> Just 3 recipes at the moment... links requested
1459 2012-10-12 19:33:38 xIsalty has joined
1460 2012-10-12 19:33:47 <jgarzik> People are welcome to copy the recipe list to the wiki, but I think given the associated discussion, it is more useful to have on the forum
1461 2012-10-12 19:34:05 <jgarzik> wiki talk pages are even more of a mess than forum threads
1462 2012-10-12 19:34:20 <kjj_> hmm. I hope that grows into a recipe subforum, where each pattern can have a dedicated discussion thread
1463 2012-10-12 19:34:30 <Luke-Jr> jgarzik: why not just make a category on the wiki?
1464 2012-10-12 19:34:48 <jgarzik> kjj_: that's what the post requests people do
1465 2012-10-12 19:35:01 <jgarzik> kjj_: "If you would like to contribute a recipe, please start your own thread, and then post a link in this thread."
1466 2012-10-12 19:35:29 <kjj_> yup, and hopefully enough of them pop up that we'll need to make a subboard
1467 2012-10-12 19:35:31 <phantomcircuit> jgarzik, at least there is hope that they can be organized
1468 2012-10-12 19:35:33 <jgarzik> Luke-Jr: that's fine. but the wiki as the root of content hosting is lame
1469 2012-10-12 19:35:46 <phantomcircuit> i put in a lot of time on the trade page putting things into tables so you can at least try to compare them
1470 2012-10-12 19:35:50 <jgarzik> Luke-Jr: because each of these will inevitably invite questions and discussion
1471 2012-10-12 19:35:59 <phantomcircuit> then people show up and delete their competition
1472 2012-10-12 19:35:59 <phantomcircuit> :|
1473 2012-10-12 19:37:43 Motest031 has joined
1474 2012-10-12 19:37:49 Motest003 has quit (Ping timeout: 246 seconds)
1475 2012-10-12 19:39:08 lggr has quit (Ping timeout: 252 seconds)
1476 2012-10-12 19:39:31 mndrix has joined
1477 2012-10-12 19:39:31 mndrix has quit (Changing host)
1478 2012-10-12 19:39:31 mndrix has joined
1479 2012-10-12 19:39:31 mndrix has quit (Changing host)
1480 2012-10-12 19:39:31 mndrix has joined
1481 2012-10-12 19:42:22 lggr has joined
1482 2012-10-12 19:48:17 lggr has quit (Ping timeout: 256 seconds)
1483 2012-10-12 19:48:41 sgstair has quit (Read error: Connection reset by peer)
1484 2012-10-12 19:49:10 sgstair has joined
1485 2012-10-12 19:50:25 agricocb has joined
1486 2012-10-12 19:53:11 lggr has joined
1487 2012-10-12 19:55:35 nym has joined
1488 2012-10-12 19:59:14 lggr has quit (Ping timeout: 244 seconds)
1489 2012-10-12 20:02:52 lggr has joined
1490 2012-10-12 20:04:55 Diapolo has joined
1491 2012-10-12 20:05:52 <jgarzik> gavinandresen: 0.7.1rc1 reports seem to indicate OK on windows. one mac crash report: https://bitcointalk.org/index.php?topic=117874.msg1266893#msg1266893
1492 2012-10-12 20:06:41 <gavinandresen> mac crash is a WONTFIX: I don't have time to figure out how to continue to support OSX 10.5
1493 2012-10-12 20:07:50 <Diapolo> Hi guys, I just replied to the person with the Mac issue, didn't know it would be a WONTFIX.
1494 2012-10-12 20:08:37 lggr has quit (Ping timeout: 245 seconds)
1495 2012-10-12 20:08:50 <gavinandresen> I just responded in the thread.
1496 2012-10-12 20:08:56 <Diapolo> I would love to get the current translations in before the final is shipped, wumpus shoult take a look at that pull and merge it then IMO.
1497 2012-10-12 20:10:48 <Diapolo> Gavin, should we check for unsupported OS versions and dissallow startup ... just crashing seems not good. Dunno if that is possible.
1498 2012-10-12 20:10:57 <gavinandresen> "patches welcome"
1499 2012-10-12 20:11:22 <gavinandresen> The crash is a missing DLL, so I don't think checking at startup is possible
1500 2012-10-12 20:11:54 <Diapolo> well then that won't work
1501 2012-10-12 20:12:29 lggr has joined
1502 2012-10-12 20:12:53 <gavinandresen> Might be possible to compile it in a way that the OS tells you it can't run... if we spin a rc2 then I'll give that a try.
1503 2012-10-12 20:17:22 m00p has joined
1504 2012-10-12 20:18:27 lggr has quit (Ping timeout: 245 seconds)
1505 2012-10-12 20:21:02 <asa1024> so no more mac clients?
1506 2012-10-12 20:21:08 <asa1024> or just for 10.5?
1507 2012-10-12 20:22:23 <MC1984> would there be any way to code something in tht makes it easier to grab network metrics
1508 2012-10-12 20:22:56 <MC1984> like ndes online, status of thier chain etc
1509 2012-10-12 20:23:16 lggr has joined
1510 2012-10-12 20:23:26 cjtest has joined
1511 2012-10-12 20:23:38 <MC1984> ive seen pages that try but it seems a bit hit and miss
1512 2012-10-12 20:24:54 <jgarzik> asa1024: just <= 10.5
1513 2012-10-12 20:25:25 <asa1024> ok ty sir
1514 2012-10-12 20:27:59 <MC1984> i wonder if there would be a way to do it without some sort of phoning home to a central server panopticon thing which would surely cause a shitstorm on the forums
1515 2012-10-12 20:29:37 lggr has quit (Ping timeout: 272 seconds)
1516 2012-10-12 20:29:53 <jgarzik> all releases will henceforth be digitally signed by Atlas
1517 2012-10-12 20:30:00 <jgarzik> ;p
1518 2012-10-12 20:30:21 <MC1984> hehehe
1519 2012-10-12 20:30:43 BlackPrapor has quit (Quit: KVIrc 4.0.4 Insomnia http://www.kvirc.net/)
1520 2012-10-12 20:31:20 harkon has quit (Quit: Konversation terminated!)
1521 2012-10-12 20:31:54 <MC1984> maybe each node generates a random guid of some sort, and broadcasts emphemeral statistics about its state through the network
1522 2012-10-12 20:32:03 <MC1984> each node relays this to its nodes
1523 2012-10-12 20:32:49 <MC1984> any node could then sit around for like an hour collecting this data to get a decent rough idea of the state of the network
1524 2012-10-12 20:33:12 <MC1984> would increase chatter a lot though
1525 2012-10-12 20:33:26 <MC1984> then again bittorrent is chatty as fuck and no one minds about that
1526 2012-10-12 20:33:57 lggr has joined
1527 2012-10-12 20:34:27 [7] has quit (Ping timeout: 246 seconds)
1528 2012-10-12 20:34:51 cjtest has quit (Quit: Page closed)
1529 2012-10-12 20:35:56 <MC1984> the guids can cycle every few hours too, and probe nodes can take account of this by only collating data for some previous n hour collection window
1530 2012-10-12 20:36:11 <MC1984> for privacy purposes
1531 2012-10-12 20:36:22 <MC1984> never gonna be exact but should be useful?
1532 2012-10-12 20:36:50 jdnavarro has joined
1533 2012-10-12 20:38:05 <MC1984> like useful for the devs to see what needs fixing and improving more urgently in a technical sense
1534 2012-10-12 20:39:12 <Joric> speaking about brainwallets, is there an easy way to confirm send using a passphrase? multisig?
1535 2012-10-12 20:39:52 lggr has quit (Ping timeout: 245 seconds)
1536 2012-10-12 20:40:27 TheSeven has joined
1537 2012-10-12 20:40:30 <Joric> i mean, the way nobody could change the destination
1538 2012-10-12 20:41:21 jdnavarro has quit (Read error: Connection reset by peer)
1539 2012-10-12 20:43:41 lggr has joined
1540 2012-10-12 20:44:04 jdnavarro has joined
1541 2012-10-12 20:46:04 <Diapolo> Which makefile is used to build bitcoind? .linux-mingw and bitcoind.exe .mingw? I tend to forget this always as I normaly don't use em
1542 2012-10-12 20:46:23 <Diablo-D3> no
1543 2012-10-12 20:46:30 <Diablo-D3> .mingw is for native on windows mingw
1544 2012-10-12 20:46:44 <Diablo-D3> .unix is for building on linux, bsd, whatever
1545 2012-10-12 20:46:51 <Diablo-D3> I think you're asking which target
1546 2012-10-12 20:46:53 TheSeven has quit (Ping timeout: 248 seconds)
1547 2012-10-12 20:46:57 <Diablo-D3> you want make -f makefile.whatever bitcoind
1548 2012-10-12 20:47:16 <kjj_> if you specify -connect=X, and that one node feeds you garbage until misbehaving triggers, will it still disconnect?
1549 2012-10-12 20:47:48 <Diapolo> Diablo-D3: I wanted to know which one is used for building with Gitian and if there are several used for different target OSes.
1550 2012-10-12 20:47:50 TheSeven has joined
1551 2012-10-12 20:48:02 <Diablo-D3> Diapolo: its OS targeted.
1552 2012-10-12 20:48:07 <Diablo-D3> so..
1553 2012-10-12 20:48:09 <Diablo-D3> huh.
1554 2012-10-12 20:48:13 <Diablo-D3> Diapolo: we're doomed.
1555 2012-10-12 20:48:29 <Diapolo> Are we?
1556 2012-10-12 20:48:35 <Diablo-D3> yes, very.
1557 2012-10-12 20:48:42 <Diablo-D3> gmaxwell, sipa: why have you doomed us :<
1558 2012-10-12 20:49:11 da2ce703 has joined
1559 2012-10-12 20:49:52 t7 has quit (Remote host closed the connection)
1560 2012-10-12 20:50:04 lggr has quit (Ping timeout: 246 seconds)
1561 2012-10-12 20:51:51 da2ce703 has quit (Client Quit)
1562 2012-10-12 20:52:35 da2ce708 has joined
1563 2012-10-12 20:53:05 <jgarzik> kjj_: it should
1564 2012-10-12 20:53:18 lggr has joined
1565 2012-10-12 20:53:20 da2ce708 has quit (Client Quit)
1566 2012-10-12 20:53:20 t7 has joined
1567 2012-10-12 20:53:40 da2ce720 has joined
1568 2012-10-12 20:54:39 <Diapolo> makefile.linux-mingw is for bitcoind.exe (gitian-win32.yml)
1569 2012-10-12 20:54:40 <Diapolo> makefile.unix is for bitcoind (gitian.yml)
1570 2012-10-12 20:54:40 <Diapolo> Is this correct?
1571 2012-10-12 20:54:41 ovidiusoft has quit (Ping timeout: 276 seconds)
1572 2012-10-12 20:55:01 <Diablo-D3> that'd be awfully weird
1573 2012-10-12 20:55:13 jdnavarro has quit (Remote host closed the connection)
1574 2012-10-12 20:55:29 <Diablo-D3> Diapolo: I mean, I understand why linux-mingw, the build host is linux running a cross compiler
1575 2012-10-12 20:55:50 <Diablo-D3> oh!
1576 2012-10-12 20:55:53 <Diablo-D3> they're both bitcoind
1577 2012-10-12 20:56:00 <Diablo-D3> Diapolo: thats probably fine the way they ahve it
1578 2012-10-12 20:56:13 <Diapolo> just for different OS binaries
1579 2012-10-12 20:57:03 t7 has quit (Remote host closed the connection)
1580 2012-10-12 20:57:30 osmosis has joined
1581 2012-10-12 20:58:52 Joric has quit ()
1582 2012-10-12 20:59:53 lggr has quit (Ping timeout: 276 seconds)
1583 2012-10-12 21:00:57 t7 has joined
1584 2012-10-12 21:03:05 lggr has joined
1585 2012-10-12 21:03:10 jeremias has quit (Remote host closed the connection)
1586 2012-10-12 21:04:12 ThomasV_ has quit (Quit: Quitte)
1587 2012-10-12 21:09:29 bakh has joined
1588 2012-10-12 21:09:35 lggr has quit (Ping timeout: 252 seconds)
1589 2012-10-12 21:11:41 da2ce720 has quit (Ping timeout: 255 seconds)
1590 2012-10-12 21:12:50 lggr has joined
1591 2012-10-12 21:15:13 t7 has quit (Read error: Connection reset by peer)
1592 2012-10-12 21:17:18 xIsalty has quit (Remote host closed the connection)
1593 2012-10-12 21:18:44 BitcoinBaltar has quit (Ping timeout: 276 seconds)
1594 2012-10-12 21:18:58 lggr has quit (Ping timeout: 256 seconds)
1595 2012-10-12 21:20:10 BTCTrader_ has joined
1596 2012-10-12 21:20:59 BTCTrader has quit (Read error: Connection reset by peer)
1597 2012-10-12 21:20:59 BTCTrader_ is now known as BTCTrader
1598 2012-10-12 21:23:00 BitcoinBaltar has joined
1599 2012-10-12 21:23:22 lggr has joined
1600 2012-10-12 21:23:35 att has quit (Ping timeout: 255 seconds)
1601 2012-10-12 21:27:30 t7` has joined
1602 2012-10-12 21:28:42 Joric has joined
1603 2012-10-12 21:30:02 lggr has quit (Ping timeout: 265 seconds)
1604 2012-10-12 21:30:20 t7` is now known as t7
1605 2012-10-12 21:31:23 phma has quit (Quit: Konversation terminated!)
1606 2012-10-12 21:33:05 lggr has joined
1607 2012-10-12 21:34:14 <sipa> BlueMatt: the way i see it: infrastructure for serving bulk amounts of data already exists
1608 2012-10-12 21:34:59 <sipa> BlueMatt: there is technically not a problem why that infrastructure can't be centralized even without giving up zero-trust (it can just be http from some download site)
1609 2012-10-12 21:35:28 <sipa> it's just a fact that the internet is not peer-to-peer (anymore), but to a large extent servers-clients
1610 2012-10-12 21:36:07 <sipa> and at least for now, many of the nodes we want on the p2p network, will be consumers, and not high-bandwidth sites
1611 2012-10-12 21:38:45 <gmaxwell> Torrent is a fairly poor client - server distribution mechanism.
1612 2012-10-12 21:38:51 <sipa> so what i'm trying to get at is that there is no need why the archival of blocks must be exactly inside the p2p network
1613 2012-10-12 21:39:52 lggr has quit (Ping timeout: 245 seconds)
1614 2012-10-12 21:40:07 <sipa> bittorrent is existing infrastructure that works now for us, but maybe in the future it'll just be http services
1615 2012-10-12 21:40:51 <sipa> also, at least for myself: i work on what i think is needed (and what i like...), and the IBD mechanism is something i want to fix
1616 2012-10-12 21:41:09 <gmaxwell> sipa: if they do not then you can kill new bitcoin installs by killing the that source without killing the bitcoin network; or you can kill the bitcoin network. Two points of failure instead of one.
1617 2012-10-12 21:41:18 <gmaxwell> ::shrugs::
1618 2012-10-12 21:41:52 <sipa> sure, i hope we don't become dependent on what particular distribution mechanism
1619 2012-10-12 21:42:49 <sipa> gmaxwell: what do think about "the future of block distribution", actually?
1620 2012-10-12 21:43:08 lggr has joined
1621 2012-10-12 21:49:23 <gmaxwell> I think I've said before that what I'd like to have is full node software that auto adapts its mode based on available resources unless the user overrides.
1622 2012-10-12 21:49:25 lggr has quit (Ping timeout: 272 seconds)
1623 2012-10-12 21:49:53 <helo> so initial launch starts with a quick benchmark to assess resources?
1624 2012-10-12 21:50:43 <sipa> helo: 1) start header-only, 2) download blocks to verify parts of history, 3) keep them is storage allows
1625 2012-10-12 21:50:51 <sipa> s/is/if/
1626 2012-10-12 21:50:54 <sipa> gmaxwell: right?
1627 2012-10-12 21:51:55 <sipa> hmm, 2% of reachable nodes (weighted by uptime) is on IPv6
1628 2012-10-12 21:52:07 <sipa> and 0.3% is on Tor
1629 2012-10-12 21:52:08 <gmaxwell> basically or in more detail: start header onlyâ and based on resources decide to either build your own utxo or get a committed utxo.. then based on resources keep blocks or don't.
1630 2012-10-12 21:52:37 lggr has joined
1631 2012-10-12 21:52:38 <gmaxwell> the whole notion is that _by default_ users should provide as much as we can get away with getting them to provide.
1632 2012-10-12 21:53:12 <sipa> as long as that it is obvious how to decrease resource usage when it bothers
1633 2012-10-12 21:53:40 <sipa> but as long as that is satisfied, that sounds pretty much optimal
1634 2012-10-12 21:54:20 <gmaxwell> I think its important: 'selfish' users will do things that hurt the network if lots of people do it, we need to counterbalance by default-alturism by people who don't mind it.
1635 2012-10-12 21:54:52 <gmaxwell> and sure, if there is some different protocol for bulk fetches or whatever, thats okay, but it should be integrated so its seemless.
1636 2012-10-12 21:56:50 <gmaxwell> 'cause "You want me to go through _more_ effort to give away more of my resources" is a bad sales pitch; some will do it for sure... but I don't think it offsets all the "I'll go through effort so I can run bitcoin on my rasberry-pi VPS". :P
1637 2012-10-12 21:57:57 <sipa> haha
1638 2012-10-12 21:58:58 lggr has quit (Ping timeout: 245 seconds)
1639 2012-10-12 22:02:00 lggr has joined
1640 2012-10-12 22:02:23 <jgarzik> torrent files include an optional list of HTTP seeds
1641 2012-10-12 22:05:18 <jgarzik> Apparently Azureus has its own DHT by default, though it does have a "mlDHT" plugin that works
1642 2012-10-12 22:07:56 <gmaxwell> I vaguely recall that http seeds don't work for some reason (e.g. they require something weird and unlikely from the http server)
1643 2012-10-12 22:08:25 lggr has quit (Ping timeout: 272 seconds)
1644 2012-10-12 22:08:42 MobiusL has quit (Quit: Ex-Chat)
1645 2012-10-12 22:08:45 <jgarzik> maybe http 1.1 byte ranges
1646 2012-10-12 22:08:53 <jgarzik> like those provided by Amazon S3
1647 2012-10-12 22:09:24 <MC1984> i thought vuze gave up with its own little dht thing and joined the main one by default
1648 2012-10-12 22:10:31 <jgarzik> MC1984: seems to do its own thing by default, here
1649 2012-10-12 22:10:40 <gmaxwell> nah nah, my application for them was the initial distributioon of this: http://xiph.org/video/vid1.shtml and I was unable to use the actual files on our own webserver and our http serving mirrors as http seeds.
1650 2012-10-12 22:10:43 <jgarzik> version 4.6.0
1651 2012-10-12 22:11:30 lggr has joined
1652 2012-10-12 22:11:53 <MC1984> i agree with greg re: the auto resource allocation stuff
1653 2012-10-12 22:12:00 <jgarzik> bleh, http seed requires a helper: http://www.bittorrent.org/beps/bep_0017.html
1654 2012-10-12 22:12:09 <MC1984> dont underestimat ethe power of default
1655 2012-10-12 22:12:24 <gmaxwell> (and thats also why we took the torrents down after the initial release: once we were the only seed again bittorrent was pegging the cpu), http serving is much more efficient when you're the only seed.
1656 2012-10-12 22:12:45 datagutt has quit (Quit: Computer has gone to sleep.)
1657 2012-10-12 22:12:51 <gmaxwell> jgarzik: yea, okay, my memory isn't totally useless.
1658 2012-10-12 22:13:16 <MC1984> i thought web seeds worked automatically with any recent httpd
1659 2012-10-12 22:13:41 <MC1984> one that supports the download resume by bytecount or something?
1660 2012-10-12 22:14:03 <MC1984> archive.org just starting seeding about a million torrents of thier stuff via webseed
1661 2012-10-12 22:14:07 <gmaxwell> Maybe it is now, but wasn't two years ago.. obviously we always supported range requests, you can't seek otherwise.
1662 2012-10-12 22:15:37 MobiusL has joined
1663 2012-10-12 22:16:29 optimato_ has joined
1664 2012-10-12 22:17:20 lggr has quit (Ping timeout: 256 seconds)
1665 2012-10-12 22:19:27 xisalty has joined
1666 2012-10-12 22:21:55 lggr has joined
1667 2012-10-12 22:22:40 agricocb has quit (Remote host closed the connection)
1668 2012-10-12 22:25:39 Diapolo has left ()
1669 2012-10-12 22:26:07 bakh has quit (Quit: Ex-Chat)
1670 2012-10-12 22:27:52 lggr has quit (Ping timeout: 260 seconds)
1671 2012-10-12 22:29:51 <jgarzik> interesting
1672 2012-10-12 22:29:55 <jgarzik> watches his node get stuck in real time
1673 2012-10-12 22:30:00 <jgarzik> 10/12/12 22:16:59 socket recv flood control disconnect (5010080 bytes)
1674 2012-10-12 22:30:00 <jgarzik> 10/12/12 22:16:59 disconnecting node 83.254.81.30:8333
1675 2012-10-12 22:30:08 <jgarzik> and then the IBD stops dead in its tracks
1676 2012-10-12 22:30:17 <sipa> it will continue when you get a new block
1677 2012-10-12 22:30:41 <jgarzik> yep
1678 2012-10-12 22:30:46 <sipa> that's not what i call stuck
1679 2012-10-12 22:31:00 <sipa> (though it is mostly inconvenient, and should be fixed)
1680 2012-10-12 22:31:12 <jgarzik> stuck swallowing data that it requested
1681 2012-10-12 22:31:21 <sipa> s/mostly/highly/
1682 2012-10-12 22:31:29 lggr has joined
1683 2012-10-12 22:33:51 <sipa> a summary of my day: spent the morning trying to visit placed where i may have lost my wallet yesterday, spent the afternood trying to visit places i visited this morning where i may have left my bagpack
1684 2012-10-12 22:34:08 <sipa> status: both are still lost
1685 2012-10-12 22:36:04 <wumpus> darn
1686 2012-10-12 22:36:07 optimato_ has quit (Remote host closed the connection)
1687 2012-10-12 22:37:33 lggr has quit (Ping timeout: 272 seconds)
1688 2012-10-12 22:40:50 lggr has joined
1689 2012-10-12 22:45:27 <gmaxwell> jgarzik: you're not faithfully simulating the user there.
1690 2012-10-12 22:45:50 <gmaxwell> jgarzik: when that happens you're supposted to hit the power switch on your computer, and corrupt the blkindex.
1691 2012-10-12 22:46:15 <gmaxwell> jgarzik: then you're supposted to boot up and find it corrupted and delete everything including your wallet based on a misread of some forum advice.
1692 2012-10-12 22:47:02 <gmaxwell> jgarzik: and then you're supposted to spend a half hour downloading the blockchain from eu1.bitcoincharts.. and put it in without a blk index and restart.. when you see 200k blocks to go then you're supposted to complain on IRC.
1693 2012-10-12 22:47:10 lggr has quit (Ping timeout: 260 seconds)
1694 2012-10-12 22:47:15 <sipa> wooow
1695 2012-10-12 22:47:20 <sipa> i never realized this
1696 2012-10-12 22:47:41 <sipa> when disconnecting inputs, it must happen in reverse order even inside a transaction
1697 2012-10-12 22:47:52 <sipa> and not just the transactions within a block
1698 2012-10-12 22:47:58 <gmaxwell> ... Why?
1699 2012-10-12 22:48:06 <gmaxwell> This is not obvious to me.
1700 2012-10-12 22:48:23 <gmaxwell> Then people on IRC will help you extract your private keys so you can import them into blockchain.info...
1701 2012-10-12 22:48:51 <sipa> it's ultraprune-specific: the last input being spent results in undo data which contains the metadata for the transaction itself (height, coinbase, version)
1702 2012-10-12 22:50:07 <sipa> well, technically not a problem that the restore happens in forward order, but with the stricter sanity checks i introduced, it failed on the first disconnect of a block with a transaction which spends the two last remaining outputs of a former tx
1703 2012-10-12 22:50:10 lggr has joined
1704 2012-10-12 22:51:38 <gmaxwell> makes sense.
1705 2012-10-12 22:52:22 <sipa> by the way, i tried to calculate how large the UTXO would be using "naive" serializations: over 50% larger
1706 2012-10-12 22:52:35 <sipa> i expected it to be more, though
1707 2012-10-12 22:52:37 <MC1984> gmaxwell does contempt for your end user a part of being good at coding :P
1708 2012-10-12 22:53:21 <helo> kind of like "the best programmers are lazy"
1709 2012-10-12 22:53:39 <gmaxwell> MC1984: I actually don't have contempt for end users.
1710 2012-10-12 22:53:47 <gmaxwell> The world is a hard place and people have other things to worry about.
1711 2012-10-12 22:54:03 <gmaxwell> I am an end user to, and I do insanely stupid things when viewed from an all-knowning position.
1712 2012-10-12 22:54:07 <gmaxwell> s/to/too/
1713 2012-10-12 22:54:20 <MC1984> hence :P
1714 2012-10-12 22:54:56 <gmaxwell> Worseâ people have taught the end users bad practices. Such as "if a program gets stuck you should reboot" or "passwords should be 8 characters and have at least one digit from each class"
1715 2012-10-12 22:55:18 <sipa> computer science is a struggle between programmers, trying to create better idiot-proof programs, and the universe, trying to create better idiots... so far the universe is winning
1716 2012-10-12 22:56:18 <gmaxwell> and that isn't the end user's fault. What does help is basically assuming people will do every dumb thing possibleâ and I try not to pass judgement if its because they are dumb or just because with enough users all possible mistakes will happen.
1717 2012-10-12 22:56:31 lggr has quit (Ping timeout: 245 seconds)
1718 2012-10-12 22:56:55 <sipa> sure, even very smart people have - in retrospect - said very stupid things
1719 2012-10-12 22:57:29 <MC1984> thats why you need mrons to beta test releases
1720 2012-10-12 22:57:37 <MC1984> thats my queue
1721 2012-10-12 22:57:45 <gmaxwell> I'm also telling the story to add a bit of color to the "well, it'll continue on the next block so it's not stuck" which is true, but no comfort to the, say, 1:100 users that waste another couple hours downloading again, or the 1:10000 that delete a wallet with coin in it.
1722 2012-10-12 22:57:58 TD has joined
1723 2012-10-12 22:58:03 brwyatt is now known as Away!~brwyatt@brwyatt.net|brwyatt
1724 2012-10-12 22:58:28 <MC1984> how much user handholding can you do though
1725 2012-10-12 22:58:35 <sipa> gmaxwell: yes, i agree; i was certainly not implying that we shouldn't fix the problem for that reason, or even use that as answer to a user who complains about the bad user experience
1726 2012-10-12 22:58:48 <MC1984> apple level of handholding? ie basically giving up sovereignity of your machine
1727 2012-10-12 22:59:01 <sipa> just that we've seen a class of "stuck" that is weorse than what jga saw
1728 2012-10-12 22:59:42 lggr has joined
1729 2012-10-12 22:59:52 <gmaxwell> sipa: ::nods:: I find that I almost never disagree with your positions in any case.
1730 2012-10-12 23:00:08 xisalty has quit (Ping timeout: 276 seconds)
1731 2012-10-12 23:00:19 <gmaxwell> So, I had a somewhat productive conversation with ThomasV today wrt: Electrum security. In it he made the comment:
1732 2012-10-12 23:00:22 <gmaxwell> 12:13 < ThomasV_> anyway, I think for the moment electrum is safer than bitcoin-qt, because bitcoin-qt makes it so easy to lose your coins
1733 2012-10-12 23:00:47 <gmaxwell> so perhaps thats a bit of an external view that we ought to be putting a higher priority on determinstic wallets than we have been.
1734 2012-10-12 23:01:04 <sipa> ACK
1735 2012-10-12 23:01:16 <gmaxwell> I think that concern is hard to weigh, users usually blame themselves for dataloss; so I expect its underreported.
1736 2012-10-12 23:01:33 <gmaxwell> whereas they blame us for slow blockchain downloads so thats overreported.
1737 2012-10-12 23:03:03 <sipa> given all complaints about that, i've been surprised by the lack of comments on my progress reports on ultraprune, somewhat :)
1738 2012-10-12 23:03:42 <BlueMatt> sipa: when it comes to refactors, dont expect responses to updates from this group ;)
1739 2012-10-12 23:04:13 <BlueMatt> gmaxwell: best way to find out: ask the users!
1740 2012-10-12 23:04:17 <sipa> BlueMatt: i mean forum posts containing things like "i've done benchmarks that imported the entire chain in under half an hour"
1741 2012-10-12 23:04:23 <BlueMatt> best way to ask the users: nfc...
1742 2012-10-12 23:05:14 <gmaxwell> sipa: did your report include binary links?
1743 2012-10-12 23:05:24 <BlueMatt> sipa: meh, still...in my experience the forums dont respond as much to code as they do to hand waving (but, then, my experience with the forums ended about 18 months ago...)
1744 2012-10-12 23:05:25 <sipa> ha, no
1745 2012-10-12 23:06:39 lggr has quit (Ping timeout: 260 seconds)
1746 2012-10-12 23:06:46 <sipa> hmm
1747 2012-10-12 23:07:27 <gmaxwell> binaries would help. people will run those.
1748 2012-10-12 23:08:17 <sipa> BlueMatt: maybe they just prefer complaining about bad stuff, over rejoicing about improvement :)
1749 2012-10-12 23:08:27 <BlueMatt> sipa: yea, pretty much
1750 2012-10-12 23:08:29 <gmaxwell> I'm running it on one system right now, it's so far working fine, including a out of space event. I wish there was something I could do to check to see if it was still _correct_ but it hassn't gotten stuck yet.
1751 2012-10-12 23:08:31 <BlueMatt> hey, I knew pull-tester would be useful for something...free bins
1752 2012-10-12 23:08:38 optimato_ has joined
1753 2012-10-12 23:08:42 <gmaxwell> there is not much to say other that it's enormously faster.
1754 2012-10-12 23:09:13 <MC1984> users losing wallets is still probably the #1 cause of coin loss
1755 2012-10-12 23:09:13 <BlueMatt> cs_main still exists -> still gonna move (somewhat) slow...
1756 2012-10-12 23:09:20 <MC1984> rather than theft
1757 2012-10-12 23:09:29 <BlueMatt> MC1984: I think everyone agrees there
1758 2012-10-12 23:09:37 <sipa> BlueMatt: sure, planned to fix :p
1759 2012-10-12 23:09:38 lggr has joined
1760 2012-10-12 23:09:48 <gmaxwell> MC1984: I think thats likely. Not sure how much file loss vs crazy file stunts vs crypto key loss balance out.
1761 2012-10-12 23:10:08 <gmaxwell> but all loss kinds combined are surely the biggest cause.
1762 2012-10-12 23:10:16 <BlueMatt> sipa: careful...bitcoin's locking structure is so convoluted and backwards...
1763 2012-10-12 23:10:17 <MC1984> hurrrrrrrrr
1764 2012-10-12 23:10:35 <sipa> BlueMatt: i think i have a pretty good idea now, after rewriting the validation engine :)
1765 2012-10-12 23:10:38 <BlueMatt> would take a major rebase to remove the intricate locking deps...
1766 2012-10-12 23:11:11 <BlueMatt> sipa: there is a lot of crap where wallet things are locked because of crap way up in the network stack...where it doesnt seem like it should be calling wallet stuff...
1767 2012-10-12 23:11:19 <BlueMatt> sipa: though Im sure you know that...you wrote CWallet
1768 2012-10-12 23:11:23 <sipa> yeah :)
1769 2012-10-12 23:11:36 <BlueMatt> anywho...the only way to fix the locking mess would be a neat callback system imho
1770 2012-10-12 23:11:36 <sipa> but indeed, the wallet will remain a pain for a while
1771 2012-10-12 23:11:38 <gmaxwell> But people reaason differently about different kinds of coin lose. Theft _scares_ people, they lose sleep over it, they feel powerless... violatted by even the prospeect of it. The fear of it makes people do risky things with their wallets like not make ample backups. It may be that loss is more monetary harm while theft is still more mental harm.
1772 2012-10-12 23:11:41 <sipa> BlueMatt: indeed
1773 2012-10-12 23:11:49 <gmaxwell> s/lose/loss/
1774 2012-10-12 23:11:51 Gladamas has quit (Ping timeout: 276 seconds)
1775 2012-10-12 23:14:46 forsetifox has left ("Leaving")
1776 2012-10-12 23:15:08 <sipa> by the way, speed of ultraprune decreased slightly as i've increasing fsync calls, and made -loadblock run in a separate thread with smaller locks
1777 2012-10-12 23:15:21 <sipa> i think those are worth it, though
1778 2012-10-12 23:15:33 <MC1984> the current wallet paradigm needs to change a lot
1779 2012-10-12 23:15:46 <sipa> MC1984: explain?
1780 2012-10-12 23:15:48 <MC1984> theres personal responsiblity and then there is computers are unreliable pieces of shit
1781 2012-10-12 23:15:49 <gmaxwell> MC1984: hm? what do you mean?
1782 2012-10-12 23:15:50 <BlueMatt> shame no one 'roud here trusts me enough to do any major refactors...oh well, back to bitcoinj
1783 2012-10-12 23:16:05 BlueMatt has left ("Ex-Chat")
1784 2012-10-12 23:16:30 <MC1984> theyre technically clunky for reasons you guys seem to be much more familiar with than i
1785 2012-10-12 23:16:36 lggr has quit (Ping timeout: 244 seconds)
1786 2012-10-12 23:16:37 <gmaxwell> ^ thats a real bummer. BlueMatt: you coward. :P That really isn't the case why your major refactor wasn't merged..
1787 2012-10-12 23:16:38 <MC1984> and to easy to lose and stuff
1788 2012-10-12 23:17:33 optimato_ has quit (Ping timeout: 252 seconds)
1789 2012-10-12 23:18:30 graingert has joined
1790 2012-10-12 23:19:35 lggr has joined
1791 2012-10-12 23:20:54 robocoin has quit (Quit: (!)/ã½(´ã¼ï½)ã\(!))
1792 2012-10-12 23:21:58 mps has quit (Ping timeout: 246 seconds)
1793 2012-10-12 23:25:44 BlueMatt has joined
1794 2012-10-12 23:25:54 lggr has quit (Ping timeout: 256 seconds)
1795 2012-10-12 23:28:36 prahanormal has joined
1796 2012-10-12 23:29:50 lggr has joined
1797 2012-10-12 23:33:22 mps has joined
1798 2012-10-12 23:35:57 lggr has quit (Ping timeout: 248 seconds)
1799 2012-10-12 23:36:06 gjs278 has quit (Read error: Operation timed out)
1800 2012-10-12 23:38:48 lggr has joined
1801 2012-10-12 23:38:54 gjs278 has joined
1802 2012-10-12 23:41:56 freewil has quit (Quit: Leaving)
1803 2012-10-12 23:45:11 lggr has quit (Ping timeout: 252 seconds)
1804 2012-10-12 23:45:30 maaku has quit (Quit: maaku)
1805 2012-10-12 23:46:23 maaku has joined
1806 2012-10-12 23:46:37 maaku has quit (Client Quit)
1807 2012-10-12 23:47:50 lggr has joined
1808 2012-10-12 23:50:10 RazielZ has quit (Ping timeout: 246 seconds)
1809 2012-10-12 23:52:08 Karmaon_ is now known as Karmaon
1810 2012-10-12 23:54:43 lggr has quit (Ping timeout: 260 seconds)
1811 2012-10-12 23:57:48 lggr has joined