1 2014-06-16 00:02:51 <dsnrk> Luke-Jr: "modifying it to function as a totally different pool were easy to do" < an altcoin does this, they have three p2pool shards. might be working looking into how they work.
2 2014-06-16 00:06:00 <dsnrk> Luke-Jr: it does support being sharded quite well. there's options to change the magic for different altcoins and testnet. all you need to do is add a new network and magic, and you're good to go. https://github.com/forrestv/p2pool/blob/f0eeb48c31abd3a59b65ab060e4daa0895772ddd/p2pool/networks.py#L12
3 2014-06-16 00:07:33 <Luke-Jr> dsnrk: sharding it doesn't count
4 2014-06-16 00:07:44 <Luke-Jr> dsnrk: the rules/system is what matters
5 2014-06-16 00:08:00 <Luke-Jr> one should be able to pull out PPLNS and put in CPPSRB without messing with the p2p logic
6 2014-06-16 00:08:26 <dsnrk> Luke-Jr: nothing stopping you from changing the magic so p2pool and p3pool don't talk to each other, then it doesn't matter what rules you have modified.
7 2014-06-16 00:08:42 <Luke-Jr> dsnrk: my point is that modifying the rules is difficult
8 2014-06-16 00:08:49 <Luke-Jr> or really modifying any of the logic
9 2014-06-16 00:08:51 <dsnrk> got it, sorry.
10 2014-06-16 00:10:25 <dsnrk> maybe somebody will fund a modular p3pool in C :)
11 2014-06-16 00:11:48 <buZz> vertcoin has 3 p2pools
12 2014-06-16 00:17:48 <reipr> petertodd: care to talk about this in a pm?
13 2014-06-16 00:19:24 <reipr> I'd like to
14 2014-06-16 00:19:33 <reipr> future proof my implementation
15 2014-06-16 00:36:51 <reipr> actually nvm
16 2014-06-16 00:45:35 <copumpkin> is coin control standard in bitcoin-qt now?
17 2014-06-16 00:45:46 <copumpkin> I seem to have it on but I can't remember if I did anything weird to get it
18 2014-06-16 00:46:45 <phantomcircuit> copumpkin, yes it is
19 2014-06-16 00:46:52 <phantomcircuit> you have to enable expert mode or something
20 2014-06-16 00:47:13 <copumpkin> ah okay
21 2014-06-16 00:55:07 <Hondura> Hello.
22 2014-06-16 00:55:15 <Hondura> I'm looking for some help, im at right place?
23 2014-06-16 01:02:26 <copumpkin> Hondura: depends what kind of help you're after...
24 2014-06-16 01:02:53 <Hondura> Ok.´here it goes.
25 2014-06-16 01:03:19 <Hondura> As a ''coder'' i want leanr more.
26 2014-06-16 01:03:56 <Hondura> And I decided to create my own coin, not to launch but just to make the process behind it and try mine it and make everything that a coin does.
27 2014-06-16 01:04:02 <Hondura> Based on bitcoin..
28 2014-06-16 01:04:17 <Hondura> can i pm you please?
29 2014-06-16 01:06:07 <Luke-Jr> !ops Hondura has already been informed #bitcoin* is for Bitcoin, and expressed that he doesn't even know how to compile code in ##altcoin-dev before coming here
30 2014-06-16 01:06:50 <gwillen> Hondura: sorry, this isn't the right place for that. This channel is only for bitcoin, and only for discussion of development of the bitcoin client software.
31 2014-06-16 01:07:19 <Hondura> gwillen may I pm me and take 2min of your time?
32 2014-06-16 01:07:20 <Hondura> I sware
33 2014-06-16 01:07:53 <Luke-Jr> Hondura: if people wanted to deal with you, they'd be answering you in ##altcoin-dev
34 2014-06-16 01:07:57 <gwillen> Hondura: Sorry, no.
35 2014-06-16 01:08:38 <Hondura> Luke and cant you help me like 10min? of ur time?
36 2014-06-16 01:09:04 <Hondura> I just want learn..
37 2014-06-16 01:09:16 <yano> not everyone in here is a developer for bitcoin
38 2014-06-16 01:09:31 <Hondura> I know.
39 2014-06-16 01:09:33 <Hondura> yano.
40 2014-06-16 01:09:45 <yano> anyone is free to join the channel as it appears
41 2014-06-16 01:10:15 <phantomcircuit> idiot
42 2014-06-16 01:10:24 <gwillen> phantomcircuit: be nice :-(
43 2014-06-16 01:10:39 <phantomcircuit> gwillen, nah
44 2014-06-16 01:12:11 <SomeoneWeird> heh
45 2014-06-16 01:12:18 <jgarzik> - request.push_back(Pair("id", id));
46 2014-06-16 01:12:18 <jgarzik> - request.push_back(Pair("method", strMethod));
47 2014-06-16 01:12:18 <jgarzik> - request.push_back(Pair("params", params));
48 2014-06-16 01:12:19 <jgarzik> + request.pushKV("id", id);
49 2014-06-16 01:12:19 <jgarzik> + request.pushKV("method", strMethod);
50 2014-06-16 01:12:19 <jgarzik> + request.pushKV("params", params);
51 2014-06-16 01:12:25 <jgarzik> die, json-spirit, die
52 2014-06-16 01:12:29 <copumpkin> o.O
53 2014-06-16 01:12:43 <Apocalyptic> die in a fire
54 2014-06-16 01:55:25 <Luke-Jr> anyone know how I can debug why BFGMiner is causing 25% CPU to be wasted on context switching? :| jgarzik?
55 2014-06-16 01:56:17 <jgarzik> Luke-Jr, are you constantly talking to the kernel, making ioctl calls for each USB packet or somesuch?
56 2014-06-16 01:56:39 <Luke-Jr> jgarzik: no USB, just SPI - and I think I've got it minimised down to as little as possible
57 2014-06-16 01:56:47 <Diablo-D3> Luke-Jr: spi?
58 2014-06-16 01:58:12 <jgarzik> Luke-Jr, Forgive if this is known... context switch == jumping back and forth between your program and the kernel. You will context switch if you are waiting on I/O, talking to a kernel device driver of any sort (USB, SPI, anything).
59 2014-06-16 01:58:40 <Luke-Jr> jgarzik: right. their miner seems to do fine with 10% ctx switching :x
60 2014-06-16 01:58:42 <jgarzik> It seems likely that is heading towards a pattern of "I'm running into problems because I did not want to write a kernel driver to solve this problem"
61 2014-06-16 01:59:08 <jgarzik> You want to batch work. Be interrupt driven, not polling.
62 2014-06-16 01:59:29 <jgarzik> Do not get more than 1 interrupt per work returned from hardware, ideally.
63 2014-06-16 01:59:41 <jgarzik> (depends on hardware interface, obviously)
64 2014-06-16 02:00:01 <jgarzik> If you are directly driving serial or USB at the signal or packet level, that is a difficult path to sustain.
65 2014-06-16 02:01:50 <Luke-Jr> Linux doesn't seem to have a "continually do SPI and notify me when there's data", I think :/
66 2014-06-16 02:02:29 <Luke-Jr> jgarzik: any way to confirm which syscalls are causing it the most?
67 2014-06-16 02:03:10 <jgarzik> Luke-Jr, Sure Linux does; many kernel drivers do that.
68 2014-06-16 02:03:25 <Luke-Jr> seems I have 1.91x involunary ctx switch and 2.96x voluntary
69 2014-06-16 02:03:32 <Luke-Jr> jgarzik: for userspace? ;)
70 2014-06-16 02:04:50 <jgarzik> Luke-Jr, how do you talk to SPI? ioctl(...SPI_IOC_MESSAGE(..)...) ?
71 2014-06-16 02:05:04 <jgarzik> Luke-Jr, do you ever call poll(2) ?
72 2014-06-16 02:06:00 <Luke-Jr> SPI_IOC_MESSAGE yes
73 2014-06-16 02:06:18 <Luke-Jr> poll() specifically, no; just select
74 2014-06-16 02:09:30 <jgarzik> Luke-Jr, it looks quite dangerous to have ioctl and direct memory map open and going at the same time
75 2014-06-16 02:09:55 <Luke-Jr> jgarzik: direct memory map for SPI? O.o
76 2014-06-16 02:10:45 <jgarzik> Luke-Jr, you open /dev/mem
77 2014-06-16 02:10:49 <Luke-Jr> for GPIO, yes
78 2014-06-16 02:11:53 <Luke-Jr> but is that going to interfere with unrelated SPI?
79 2014-06-16 02:12:19 <jgarzik> Luke-Jr, doing anything through /dev/mem is usually a red flag
80 2014-06-16 02:12:27 <jgarzik> Luke-Jr, sadly sometimes necessary in the embedded world
81 2014-06-16 02:12:48 <Luke-Jr> jgarzik: my understanding is that it's a best practice for GPIO since the other interface has a ton of overhead
82 2014-06-16 02:13:22 <jgarzik> and thus crap piles upon crap, when one chooses to drive hardware from userspace
83 2014-06-16 02:13:38 <jgarzik> all this is just so damn easy in the kernel
84 2014-06-16 02:14:19 <jgarzik> proper kernel design is a driver, even One Big Multi-Vendor Driver will do, plus a tiny userspace module. kernel presents a unified interface across a single device class of hardware.
85 2014-06-16 02:15:34 <jgarzik> design the kernel<->userland API properly. Choose whether you need a file descriptor a la http://man7.org/linux/man-pages/man2/timerfd_create.2.html as a central point for interfacing with the kernel driver.
86 2014-06-16 02:16:43 <Luke-Jr> but HURD support! </troll> :p
87 2014-06-16 02:17:00 <jgarzik> everything is nice and interrupt driven inside the kernel. or, if the hardware is too stupid to support any sort of async signaling that may be pretended to be an interrupt, the kernel has nice fallback facilities which __protect system resources__ while polling for the data you need.
88 2014-06-16 02:17:41 <jgarzik> all this is an easy library call, inside the kernel. You focus on your hardware-specific details only.
89 2014-06-16 02:18:52 <Luke-Jr> jgarzik: in general, I worry about: 1) not being able to provide something that works in distros today; and 2) Linus's review process (maybe I'm paranoid about it, but I imagine it's super-strict for adding a whole new device class)
90 2014-06-16 02:19:10 <jgarzik> Thou Shalt Not Bitbang From Userspace. Still holds true today. Your little program will be far more stupid than the kernel's core process scheduler, I promise.
91 2014-06-16 02:19:52 <jgarzik> People are happy to add working drivers for in-the-field hardware.
92 2014-06-16 02:22:24 <Luke-Jr> SPI ioctl is bitbanging now? :|
93 2014-06-16 02:33:17 <SomeoneWeird> pushed my gitian sigs \o/
94 2014-06-16 02:39:04 <phantomcircuit> Luke-Jr, what are you trying to drive?
95 2014-06-16 02:39:25 <Luke-Jr> phantomcircuit: A1 chips
96 2014-06-16 02:40:46 <phantomcircuit> Luke-Jr, what are those on?
97 2014-06-16 02:40:56 <Luke-Jr> phantomcircuit: ?
98 2014-06-16 02:41:16 <phantomcircuit> oh that's bitmain.ch
99 2014-06-16 02:41:46 <Luke-Jr> well, I have non-Bitmain A1 chips, but yes same thing
100 2014-06-16 02:44:01 <phantomcircuit> Luke-Jr, im guessing that's the same spi daisy chaining stuff bitfury uses?
101 2014-06-16 02:44:03 <phantomcircuit> just give up
102 2014-06-16 02:44:11 <phantomcircuit> you're going to be using tons of cpu time driving them
103 2014-06-16 02:44:18 <Luke-Jr> phantomcircuit: yes, I am. But it's possible.
104 2014-06-16 02:44:32 <Luke-Jr> but no, *different* SPI chaining stuff :p
105 2014-06-16 03:21:01 <phantomcircuit> Luke-Jr, what sort of power performance do you see on those at the wall?
106 2014-06-16 03:25:50 <Luke-Jr> phantomcircuit: no idea
107 2014-06-16 05:49:53 <bsm117532> Hi folks, I'm trying to get bitcoind to sync with the network. I have bootstrap.dat from April, and it keeps crashing. (using git head)
108 2014-06-16 05:50:29 <wumpus> can you get a traceback on the crash?
109 2014-06-16 05:51:27 <bsm117532> http://pastebin.com/H44pAkkR
110 2014-06-16 05:52:03 <bsm117532> That contains seven crashes (they're not identical), and I've also included the last line(s) of debug.log
111 2014-06-16 05:53:09 <bsm117532> What can I do to help? Debugging such a large data structure is...daunting.
112 2014-06-16 05:53:35 <bsm117532> Is CCoinsViewCache new code? Most crashes are at coins.cpp:134.
113 2014-06-16 05:54:38 <wumpus> it looks like it is most like a synchronization/locking issue
114 2014-06-16 05:55:04 <wumpus> it's not new code but there have been a few changes regarding cs_main locking lately to reduce contention
115 2014-06-16 05:55:18 <wumpus> this happens while it is still syncing from the bootstrap?
116 2014-06-16 05:55:52 <bsm117532> Yes. It's at 2014-06-16 05:49:19 UpdateTip: new best=00000000000000046ecff1fada9be8c94eb6e040463686eb647700d56692de73 height=269451 log2_work=73.845876 tx=26990218 date=2013-11-13 19:40:34 progress=0.343208
117 2014-06-16 05:55:58 <bsm117532> which is still within bootstrap.dat
118 2014-06-16 05:56:50 <bsm117532> I've checked that bootstrap.dat is not corrupt. (but bittorrent does checksumming anyway...)
119 2014-06-16 05:57:26 <wumpus> a corrupt bootstrap.dat cannot result in these problems, the blocks are validated as if they come from the network
120 2014-06-16 05:58:26 <bsm117532> I was previously having problems syncing with the network. It was pointed out that my blockchain somehow got corrupted (unclean shutdown?) I was surprised this could happen...but anyway started over from bootstrap.dat.
121 2014-06-16 05:58:28 <wumpus> seems that in those tracebacks there are various cases of heap corruption
122 2014-06-16 05:58:38 <bsm117532> yes...that would seem to be the acse.
123 2014-06-16 06:00:14 <wumpus> it's certainly possible for the block chain to be corrupted, but usually it's only the index in which case a -reindex will solve it, no need to restart from a bootstrap file
124 2014-06-16 06:00:34 <bsm117532> Hmmm well too late now. ;-)
125 2014-06-16 06:00:45 <bsm117532> It would be cool if bitcoind could detect this case...
126 2014-06-16 06:01:03 <wumpus> if you really want to start over you need to delete all the files in .bitcoin as well (apart from the wallet ofc)
127 2014-06-16 06:01:17 <bsm117532> Yes, I did.
128 2014-06-16 06:01:23 <wumpus> it's not clear what 'this case' is, it could be a bug with master as well, that's a risk you have with running the bleeding edge
129 2014-06-16 06:01:40 <bsm117532> Yeah, well I thought you guys would want to know.
130 2014-06-16 06:01:54 <bsm117532> Trying to get into bitcoin myself, so throwing myself at the bleeding edge.
131 2014-06-16 06:02:05 <wumpus> I'd suggest to try with 0.9.2
132 2014-06-16 06:02:13 <wumpus> git checkout v0.9.2
133 2014-06-16 06:02:51 <dsnrk> for what it's worth I am running master with no glaring issues.
134 2014-06-16 06:03:26 <wumpus> dsnrk: me too, but if you have issues, it helps to compare with the latest stable release
135 2014-06-16 06:03:30 <bsm117532> Have you tried rebuilding the entire blockchain?
136 2014-06-16 06:03:37 <bsm117532> Trying 0.9.2 now...
137 2014-06-16 06:03:55 <dsnrk> yes, I built a fresh testnet node a few hours ago.
138 2014-06-16 06:04:09 <bsm117532> How big is the blockchain on testnet?
139 2014-06-16 06:05:30 <bsm117532> FWIW these crashes are happening at height =~ 200000
140 2014-06-16 06:05:38 <bsm117532> And not at the same place each time.
141 2014-06-16 06:05:39 <dsnrk> 1.3G
142 2014-06-16 06:06:04 <bsm117532> I'm guessing that's a height < 30000
143 2014-06-16 06:06:19 <dsnrk> blocks=140700
144 2014-06-16 06:06:49 <bsm117532> The lowest height I saw this crash was at height=173245
145 2014-06-16 06:07:07 <bsm117532> Whatever it is, it's rare.
146 2014-06-16 06:07:52 <bsm117532> Anyway, I'll report on 0.9.2 in an hour or so.
147 2014-06-16 06:07:53 <wumpus> I've opened an issue for it https://github.com/bitcoin/bitcoin/issues/4345
148 2014-06-16 06:08:05 <wumpus> hopefully it's ok with 0.9.2
149 2014-06-16 06:08:41 <bsm117532> Thanks, I'll post further to that bug.
150 2014-06-16 06:18:56 <wumpus> bsm117532: btw what kind of system is it, esp. how many cores?
151 2014-06-16 06:19:17 <bsm117532> 4 core, Intel Core 2 quad, 8G ram.
152 2014-06-16 06:19:24 <wumpus> ok thanks
153 2014-06-16 06:43:29 <michagogo> Wow, nice to see so many gitian builders
154 2014-06-16 06:43:41 <michagogo> Is everyone matching, other than Gavin?
155 2014-06-16 06:51:13 <wumpus> michagogo: everything matches except gavin OSX
156 2014-06-16 06:51:37 <SomeoneWeird> yup
157 2014-06-16 06:55:00 <michagogo> Has he been around yet?
158 2014-06-16 06:55:10 <michagogo> ;;seen gavinandresen
159 2014-06-16 06:55:11 <gribble> gavinandresen was last seen in #bitcoin-dev 1 day, 5 hours, 54 minutes, and 46 seconds ago: <gavinandresen> If you're able to gitian-build releases⦠now would be a great time to fire up your virtual machine and build the 0.9.2 release.
160 2014-06-16 06:55:15 <gribble> gavinandresen was last seen in #bitcoin-dev 1 day, 5 hours, 54 minutes, and 51 seconds ago: <gavinandresen> If you're able to gitian-build releases⦠now would be a great time to fire up your virtual machine and build the 0.9.2 release.
161 2014-06-16 06:55:15 <SomeoneWeird> ;;seen gavinandresen
162 2014-06-16 06:55:16 <SomeoneWeird> heh
163 2014-06-16 06:55:24 <SomeoneWeird> nop! :p
164 2014-06-16 06:58:59 <michagogo> What was sipa's topic change at 01:27 UTC?
165 2014-06-16 06:59:20 <michagogo> Uh, sorry, make that 00:27 -- keep forgetting we're +3 in the summer
166 2014-06-16 06:59:41 <michagogo> (I wish irccloud did topic history)
167 2014-06-16 06:59:47 <dsnrk> think it just made the topic more forceful
168 2014-06-16 07:51:49 <petertodd> getting a lot of: " ProcessMessages(tx, 1 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length" <- is this some DoS attack I haven't heard about?
169 2014-06-16 07:52:16 <petertodd> (on my replace-by-fee nodes, which would explain why they're being attacked...)
170 2014-06-16 07:54:41 <dsnrk> nothing in any of my logs like that. lots and lots of invalid pubkey errors though.
171 2014-06-16 07:54:56 <petertodd> dsnrk: what version are you running?
172 2014-06-16 07:55:08 <dsnrk> 0.9.1 / 0.9.1 / master
173 2014-06-16 07:55:24 <petertodd> same, so I'm probably getting attacked
174 2014-06-16 07:55:36 <petertodd> (or someone's software is shite...)
175 2014-06-16 07:55:39 <Apocalyptic> probably
176 2014-06-16 07:55:58 <Apocalyptic> though it doesn't seem to be a very disruptive attack
177 2014-06-16 07:56:22 <petertodd> Apocalyptic: one seems to have crashed, the others are ok, so yeah, not very disruptive
178 2014-06-16 07:56:25 <dsnrk> I've been trying to track it down but it's one of those things where as soon as you try to get a pcap of the traffic it stops. I don't know that's causing it, but I have a whole bunch of different pubkey errors in floods.
179 2014-06-16 07:56:52 <petertodd> well, pubkey errors aren't anything surprising; this is because I'm being sent definitely invalid txs
180 2014-06-16 07:57:25 <dsnrk> well same I assume, a normal node wouldn't relay junk so it's either an attack or shitty software.
181 2014-06-16 07:57:29 <petertodd> yup
182 2014-06-16 07:57:53 <dsnrk> oddly I'm seeing them on a node that doesn't listen too.
183 2014-06-16 07:58:10 <petertodd> oh, curious
184 2014-06-16 07:58:20 <petertodd> dsnrk: er, you mean my attack or the invalid pubkeys?
185 2014-06-16 07:58:53 <dsnrk> invalid pubkey spam, I've checked the logs of all three and not seen any exceptions like that.
186 2014-06-16 07:59:27 <petertodd> dsnrk: like i say, the invalid pubkey tests are something that's been added, so seeing invalid ones isn't a surprise
187 2014-06-16 07:59:59 <dsnrk> what would be making them though?
188 2014-06-16 08:00:27 <petertodd> dsnrk: old software
189 2014-06-16 08:00:41 <petertodd> dsnrk: maybe data-in-chain things that don't use pubkey hiding?
190 2014-06-16 08:01:31 <dsnrk> I thought that, but none of my connections on this node are anything but >= 0.8.0. or at least they claim to be that.
191 2014-06-16 08:01:44 <petertodd> dsnrk: what's the exact error message you're seeing again?
192 2014-06-16 08:02:55 <dsnrk> a mix of the non-canonical public key errors. wrong length, neither compressed nor uncompressed.
193 2014-06-16 08:03:22 <petertodd> dsnrk: yeah, i forget when checks for that were added, but anyway, nothing surprising to see that
194 2014-06-16 08:03:34 <petertodd> dsnrk: I think counterparty is still producing those, among others
195 2014-06-16 08:04:06 <dsnrk> petertodd: happier to know a cause, thanks.
196 2014-06-16 08:23:18 <gdm85> I am using 'onlynet=ipv4
197 2014-06-16 08:23:34 <gdm85> in the configuration file, shouldnt that limit listening ports to IPv4?
198 2014-06-16 08:24:34 <wumpus> no, that doesn't affect binding, just what it connects to
199 2014-06-16 08:24:59 <gdm85> ah, clear. any idea how to force binding only on specific network?
200 2014-06-16 08:25:20 <wumpus> -bind
201 2014-06-16 08:25:38 <gdm85> makes sense..thanks
202 2014-06-16 08:26:12 <gdm85> do you know if it's an option that can be specified multiple times? I'd say yes
203 2014-06-16 08:26:21 <wumpus> yes
204 2014-06-16 08:26:54 <gdm85> but will -bind affect the RPC or the P2P? or both? it can't be both if you can specify the port..
205 2014-06-16 08:27:06 <wumpus> it affects P2P
206 2014-06-16 08:27:23 <gdm85> ah! sorry, I was unclear. I am looking at the interface bound for RPC
207 2014-06-16 08:27:26 <wumpus> in master there is rpcbind as well, which works the same but for RPC
208 2014-06-16 08:27:41 <gdm85> so 0.9.1 doesn't have that?
209 2014-06-16 08:27:49 <wumpus> in 0.9.x it's either bind local or bind everything
210 2014-06-16 08:27:52 <gdm85> I see. that's why is not in the wiki
211 2014-06-16 08:35:54 <gdm85> wumpus: I am using listen=0 and server=1 in 0.9.1, and even with bind=10.0.5.3 it still binds RPC port on IPv6's ::
212 2014-06-16 08:35:59 <gdm85> ACTION is puzzled
213 2014-06-16 08:36:12 <wumpus> bind and listen do not affect RPC
214 2014-06-16 08:36:29 <wumpus> they are P2P options
215 2014-06-16 08:36:32 <wumpus> just like onlynet
216 2014-06-16 08:36:48 <gdm85> wumpus: so when you say "bind local or bind everything" you mean either bind to 127.0.0.1 or bind on all networks
217 2014-06-16 08:36:55 <wumpus> you shouldn't confuse the options for those two services, they are completely separate
218 2014-06-16 08:37:31 <wumpus> gdm85: when you specify rpcallowip it binds on all networks, if you don't specify it it only binds locally
219 2014-06-16 08:37:43 <wumpus> gdm85: no combination of other options will influence that
220 2014-06-16 08:39:22 <gdm85> wumpus: I guess I am missing the part where it binds on all networks. I have specified rpcallowip but it's binding RPC only to IPv6
221 2014-06-16 08:39:35 <gdm85> could this be related to the fact I am using a non-root user to run bitcoind?
222 2014-06-16 08:39:42 <wumpus> *only* to IPv6?!
223 2014-06-16 08:39:51 <gdm85> yep
224 2014-06-16 08:39:59 <wumpus> no, you should definitely never run bitcoind as root
225 2014-06-16 08:40:09 <gdm85> the port number is correct, but it's on ::
226 2014-06-16 08:40:20 <gdm85> good to hear :)
227 2014-06-16 08:40:22 <wumpus> are you sure it's not doing a IPv4+IPv6 bind?
228 2014-06-16 08:40:25 <xch__> hey guys, quick question -- what format is the output of the dumpprivkey key in?
229 2014-06-16 08:40:37 <lianj> xch__: wip
230 2014-06-16 08:40:49 <gdm85> wumpus: I am using 'netstat -talnp' to check open ports, I see only one entry for that port on the :: interface
231 2014-06-16 08:40:51 <wumpus> are you running on some strange operating system or non-standard security configuration?
232 2014-06-16 08:40:56 <ubuntuDoubts> HI
233 2014-06-16 08:40:56 <xch__> lianj: is WIP different than WIF?
234 2014-06-16 08:41:05 <wumpus> gdm85: have you *tried* whether the port is open, with nc for example?
235 2014-06-16 08:41:07 <lianj> xch__: eh sorry, wif
236 2014-06-16 08:41:12 <gdm85> wumpus: it's Linux.
237 2014-06-16 08:41:14 <ubuntuDoubts> Im loking for some help to compile bitcoind + qt on ubuntu. Does some1 can help me out ?
238 2014-06-16 08:41:31 <xch__> lianj: that's what I thought -- but for some reason the new Coinpocket app can't read it
239 2014-06-16 08:41:57 <ubuntuDoubts> Im quite new to ubuntu, I was trying to do on windows but many people said me to compile on Ubuntu, cause it will be much easier.
240 2014-06-16 08:42:11 <ubuntuDoubts> Does anybody get me a good guide to follow ? Step by step ?
241 2014-06-16 08:42:22 <xch__> ubuntuDoubts: are you trying to compile Bitcoin Core?
242 2014-06-16 08:43:03 <ubuntuDoubts> xch_ ok, not bitcoin, my own coin ( not to launch , and im not a dev) , im just a programmer who wish to learn more about cryptos..
243 2014-06-16 08:43:10 <ubuntuDoubts> Based on bitcoin.
244 2014-06-16 08:43:11 <gdm85> wumpus: you're right, even if it's not listed it seems open with nc :s
245 2014-06-16 08:43:26 <ubuntuDoubts> I just changed the limit of coins nothing more.
246 2014-06-16 08:43:28 <ubuntuDoubts> Ah and the name ..
247 2014-06-16 08:43:48 <ubuntuDoubts> i passed whole 'bitcoin' terms to 'testcoin'
248 2014-06-16 08:43:53 <gdm85> I guess I'll have to understand how this IPv4+IPv6 binding works..
249 2014-06-16 08:43:58 <Dagger> gdm85/wumpus: if you have net.ipv6.bindv6only left at the default 0, v6 sockets listening on :: can accept v4 connections too
250 2014-06-16 08:44:11 <wumpus> Dagger: yep
251 2014-06-16 08:44:12 <gdm85> Dagger: that's one bit of information I was missing!
252 2014-06-16 08:44:16 <xch__> lianj: is there a difference between private keys that start with 5 and those that start with K?
253 2014-06-16 08:44:24 <xch__> lianj: or are they both wallet import format?
254 2014-06-16 08:48:50 <lianj> xch__: K is a compressed pubkey address
255 2014-06-16 08:48:58 <lianj> 5 is an uncompressed one
256 2014-06-16 08:49:07 <Jouke> xch__: both wif, starting with 5 in uncompressed, K or L is compressed public key
257 2014-06-16 08:50:13 <xch__> lianj: Bitcoin Core outputs both though, from dumpprivkey -- is there a way to specify only uncompressed keys?
258 2014-06-16 08:50:31 <xch__> lianj: for instance, sometimes dumpprivkey returns something like KwF6FhkCCa6rD6NBFdG3kYqns5SKUPeq6ai48Fa6cmrcQu3D7Am3
259 2014-06-16 08:50:38 <xch__> lianj: other times it outputs 5JEM98FTNnWT6Fc3aYx97JTUJ9Mt229DF8Q9LWAZYExnf2Ek6Eh
260 2014-06-16 08:50:41 <Jouke> Than you have an old wallet.
261 2014-06-16 08:51:29 <Jouke> A compressed pubkey creates an other bitcoinaddress than a umcompressed pubkey from the same private key.
262 2014-06-16 08:51:47 <gmaxwell> You should not be using uncompressed keys, generally.
263 2014-06-16 08:52:22 <xch__> gmaxwell: I am running Bitcoin Core 0.9.1 -- just trying to create private keys that start with 5
264 2014-06-16 08:52:39 <lianj> if coinpocket app doesn't suppor compressed pubkey addresses then coinpocket is the problem
265 2014-06-16 08:52:49 <gmaxwell> xch__: It will no longer generate private keys that start with 5, not at least if you're using a current wallet version.
266 2014-06-16 08:53:05 <xch__> gmaxwell : I see, then Coinpocket is the problem
267 2014-06-16 08:53:19 <gmaxwell> Uncompressed keys result in considerably larget transactions (and thus higher tx fees in aggregate)
268 2014-06-16 08:53:47 <gmaxwell> er larger*
269 2014-06-16 08:54:15 <xch__> gmaxwell : I see, makes sense. So the correct format should be KwF6FhkCCa6rD6NBFdG3kYqns5SKUPeq6ai48Fa6cmrcQu3D7Am3 for all WIF clients going forward?
270 2014-06-16 08:56:30 <gdm85> thanks guys, I could troubleshoot the other end of the problem :)
271 2014-06-16 08:56:48 <gmaxwell> xch__: Or L*
272 2014-06-16 08:57:49 <xch__> gmaxwell: Thank you
273 2014-06-16 08:58:22 <lianj> well, parsing 5 should still be supported. the format didn't really change. you unpack the base58 and then it tells you what it is
274 2014-06-16 08:58:31 <ubuntuDoubts_> im back
275 2014-06-16 09:05:52 <ubuntuDoubts_> Does anyone give me a good guide to compile coin on linux ?
276 2014-06-16 09:06:10 <sipa> rtfm
277 2014-06-16 09:06:14 <lianj> https://github.com/bitcoin/bitcoin/blob/master/doc/build-unix.md
278 2014-06-16 09:06:16 <ubuntuDoubts_> Im quite new to this OS, and im just using it cause was a bit complicated to complile on windows.
279 2014-06-16 09:06:38 <ubuntuDoubts_> lianj, that will help me to build not bitcoin but other coin based on bitcoin ?
280 2014-06-16 09:06:45 <buZz> yes
281 2014-06-16 09:06:48 <gdm85> Dagger: IPv6's '::' does not correspond to 0.0.0.0 correct?
282 2014-06-16 09:06:55 <ubuntuDoubts_> Just a dumb quastion, where should i dump my coin code?
283 2014-06-16 09:07:15 <buZz> i use ~/coinname/src
284 2014-06-16 09:07:33 <lianj> /dev/null
285 2014-06-16 09:07:40 <Dagger> gdm85: it's the same idea. listen on :: to accept connections to all local addresses rather than a specific one
286 2014-06-16 09:08:31 <Dagger> (all v6 addresses, but it also has that special behavior where v4 connections can be accepted too)
287 2014-06-16 09:09:13 <gdm85> Dagger: ok, I understand. but having one rpcallowip= line with a non-local address, should bitcoind listen on 0.0.0.0?
288 2014-06-16 09:11:23 <Dagger> gdm85: I'm not very familiar with it... but I don't think rpcallowip is for specifying what IP to bind to
289 2014-06-16 09:11:40 <Dagger> gdm85: as in, it'd be valid to listen for all connections but reject any that come from an IP not listed in rpcallowip
290 2014-06-16 09:12:14 <gdm85> Dagger: second what wumpus said, either binds locally or to all interfaces. for some reason it's binding locally here
291 2014-06-16 09:12:32 <sipa> you need rpcbind for that i guess
292 2014-06-16 09:12:32 <ubuntuDoubts_> lianj may I pm ya ?
293 2014-06-16 09:12:33 <gdm85> Dagger: yes that's what it would do with a working rpcallowip
294 2014-06-16 09:12:49 <gdm85> sipa: I am sure that will be much easier to use
295 2014-06-16 09:13:22 <sipa> rpcbind defines what you bind to, rpcallowip filters incoming connects after they are established
296 2014-06-16 09:14:07 <gdm85> sipa: thus rpcallowip should first bind to all interfaces, then filter. but it's binding to localhost only here..
297 2014-06-16 09:14:27 <warren> "Even worse, they can double-trade coins in a way that no one would ever find out, even if they dropped back below 51%."
298 2014-06-16 09:14:35 <ubuntuDoubts_> guys, first of all i should install all dependecnies
299 2014-06-16 09:14:51 <warren> ACTION facepalm
300 2014-06-16 09:16:13 <gdm85> eh, P2P port is going correctly on 0.0.0.0...
301 2014-06-16 09:21:09 <sipa> gdm85: are you specifying rpcbind?
302 2014-06-16 09:21:23 <gdm85> sipa: it's not master, it's 0.9.1/0.9.2rc2
303 2014-06-16 09:21:52 <sipa> that code has not changed in years
304 2014-06-16 09:21:56 <sipa> i may misremember though
305 2014-06-16 09:22:04 <sipa> but afaik by default it only binds to localhost
306 2014-06-16 09:22:41 <gdm85> sipa: unfortunately the rpcbind option is not there in 0.9.1. but I am doing some other tests on client side
307 2014-06-16 09:25:39 <gdm85> sipa: in the end I managed to have it working
308 2014-06-16 09:29:49 <gdm85> mainly what I learnt is that when looking at netstat ':::18332' does not mean it's listening on localhost only
309 2014-06-16 09:30:15 <gdm85> although listed as 'tcp6', requests are coming in perfectly as if it were 0.0.0.0
310 2014-06-16 09:30:22 <ubuntuDoubts_> Someone help me please?
311 2014-06-16 09:30:30 <ubuntuDoubts_> I'm stucked at compiling..
312 2014-06-16 09:30:40 <ubuntuDoubts_> Im quite new to ubuntu mates..
313 2014-06-16 09:30:57 <ubuntuDoubts_> Anyway seems much easier than compile on windows.
314 2014-06-16 09:32:00 <gdm85> ubuntuDoubts_: people is more likely willing to help if you put a bit more effort in using existing documentation/resources. nothing is inherently 'easy', although that doesn't mean you should give up early
315 2014-06-16 09:32:32 <ubuntuDoubts_> gdm85 i 'll never give up! Im doing this for the past 2 weeks.
316 2014-06-16 09:33:09 <ubuntuDoubts_> gdm85 my problem is, i cant find a good guide.. I passed last week trying compile on windows, i couldn't..
317 2014-06-16 09:33:20 <gdm85> ubuntuDoubts_: lianj gave you a link: https://github.com/bitcoin/bitcoin/blob/master/doc/build-unix.md
318 2014-06-16 09:33:23 <ubuntuDoubts_> LAst night some people here told me to try ubuntu.
319 2014-06-16 09:33:24 <wumpus> people also cannot help if you don't tell what the specific problem is
320 2014-06-16 09:33:49 <wumpus> or even what you are trying to do, do you want to compile for ubuntu itself or do a gitian build for all platforms?
321 2014-06-16 09:34:00 <ubuntuDoubts_> ok, my problem atm is: I dont know where i should start.. I already dump my source code to ubuntu drive.
322 2014-06-16 09:34:22 <ubuntuDoubts_> gitian build for al plataforms seems better.
323 2014-06-16 09:34:50 <ubuntuDoubts_> Anyway, this is only to understand and learn more.. Im 21 php coder.. I jsut want to learn new things.
324 2014-06-16 09:35:00 <ubuntuDoubts_> wumpus my I pm you ?
325 2014-06-16 09:35:17 <wumpus> rather not, unless you have a confidential issue to report
326 2014-06-16 09:35:36 <wumpus> gitian building is described in https://github.com/bitcoin/bitcoin/blob/master/doc/gitian-building.md
327 2014-06-16 09:36:34 <sipa> wumpus: he just doesn't know how to do a bitcoin complation
328 2014-06-16 09:36:46 <sipa> or anything in linux
329 2014-06-16 09:36:57 <sipa> i suspect
330 2014-06-16 09:37:22 <gdm85> ubuntuDoubts_: did you check that official guide?
331 2014-06-16 09:37:40 <gdm85> it might not be extremely detailed or exactly a 'for dummies' guide, but it's a start
332 2014-06-16 09:38:11 <wumpus> in that case you should follow the instructions in doc/build-unix.md; ie, install the dependencies using apt-get, and do ./autogen.sh ./configure <flags>, and then make
333 2014-06-16 09:38:42 <wumpus> I see gdm85 already quoted them
334 2014-06-16 09:39:00 <ubuntuDoubts_> ok, wumpus does i need instal gitian ?
335 2014-06-16 09:39:05 <ubuntuDoubts_> Or i can do that from ubuntu ?
336 2014-06-16 09:39:24 <gdm85> sipa, wumpus, Dagger regarding RPC listening: perhaps I have just been confusing :: with ::1 :s my bad.. now everything is working as expected. (although having IPv4 routed through IPv6 adds to the confusion)
337 2014-06-16 09:39:38 <gdm85> ubuntuDoubts_: I would start without gitian, it's a tad more complex
338 2014-06-16 09:39:44 <ubuntuDoubts_> Ok
339 2014-06-16 09:40:02 <ubuntuDoubts_> But without gitian i willn't able to run the client on windows right?
340 2014-06-16 09:40:15 <ubuntuDoubts_> this client will only wun on linux?
341 2014-06-16 09:40:22 <gdm85> ubuntuDoubts_: yes
342 2014-06-16 09:40:27 <ubuntuDoubts_> Sorry all for my newbie questions!
343 2014-06-16 09:40:49 <ubuntuDoubts_> ahhhh so, i will have anyway to compile it on windows to got it work sucessufu on windows ?
344 2014-06-16 09:41:09 <sipa> no, you can use gitian for that
345 2014-06-16 09:41:16 <wumpus> ipv4 is not "routed through" ipv6, it's just that the case that programs want to bind to both networks is so common that it is the default in linux and can be done in one bind statement...
346 2014-06-16 09:41:17 <ubuntuDoubts_> thansk sipa.
347 2014-06-16 09:41:19 <sipa> which crosscompiles linux to windows
348 2014-06-16 09:41:23 <ubuntuDoubts_> Ok!
349 2014-06-16 09:41:32 <ubuntuDoubts_> U cryptocurrency ubuntunderstood now, much better,
350 2014-06-16 09:41:41 <ubuntuDoubts_> For now i will try just compile it for ubuntu.
351 2014-06-16 09:41:51 <sipa> i would suggest that
352 2014-06-16 09:42:01 <ubuntuDoubts_> that will be the next objective.
353 2014-06-16 09:42:10 <ubuntuDoubts_> thanks for this help people..
354 2014-06-16 09:42:26 <ubuntuDoubts_> Sipa, gdm85 and wumpus.. thansk your patience.
355 2014-06-16 09:42:38 <ubuntuDoubts_> I will let know how this runs.
356 2014-06-16 09:43:12 <gdm85> wumpus: yeah now I know..
357 2014-06-16 09:49:05 <WalletMigrator> Hi, I have an old wallet that (for a variety of reasons, including: I accidentally getnewaddressed a few thousand times, and it's still using uncompressed keys) I want to replace with a fresh wallet. What's the best way to do this? Send everything to the new wallet all at once in one tx? Send everything to the new wallet in a bunch of txes? Export the privkeys that have unspent outputs and move them into the new wallet?
358 2014-06-16 09:49:30 <sipa> is the wallet encrypted?
359 2014-06-16 09:49:40 <WalletMigrator> It is
360 2014-06-16 09:50:06 <WalletMigrator> why?
361 2014-06-16 09:50:17 <sipa> because encrypting flushes the keypool
362 2014-06-16 09:51:03 <WalletMigrator> ah, I see what you;re saying, if it weren't i could encrypt it and fix the uncompressed thing at least
363 2014-06-16 09:51:57 <sipa> you can make new keys be compressed by startwith -upgradewalet
364 2014-06-16 09:52:09 <sipa> by starting with -upgradewallet
365 2014-06-16 09:52:15 <WalletMigrator> upgradewallet? what does that do?
366 2014-06-16 09:52:32 <WalletMigrator> " -upgradewallet Upgrade wallet to latest format on startup"
367 2014-06-16 09:52:37 <WalletMigrator> there are different formats? O_o
368 2014-06-16 09:52:47 <sipa> not really different formats
369 2014-06-16 09:52:54 <sipa> but a few optional features that old clients don't support
370 2014-06-16 09:53:17 <sipa> by using -upgradewallet, you set the version marker to the latest available
371 2014-06-16 09:53:19 <WalletMigrator> like compressed keys? :)
372 2014-06-16 09:53:25 <WalletMigrator> anything else?
373 2014-06-16 09:53:28 <sipa> yes, and that's the only feature really
374 2014-06-16 09:53:36 <sipa> encryption is one too, but you already have that
375 2014-06-16 09:55:13 <WalletMigrator> well, the thing is
376 2014-06-16 09:55:19 <WalletMigrator> the wallet is 5.15 MB
377 2014-06-16 09:55:29 <sipa> so?
378 2014-06-16 09:55:32 <Jouke> so?
379 2014-06-16 09:55:36 <WalletMigrator> it takes a long time to start up
380 2014-06-16 09:55:57 <sipa> ok, then move to a new wallet
381 2014-06-16 09:56:03 <sipa> if you don't mind losing the old keys
382 2014-06-16 09:56:15 <WalletMigrator> right, that's what I wanted to do. so what's the best way to do that?
383 2014-06-16 09:56:22 <WalletMigrator> in terms of safety and privacy
384 2014-06-16 09:56:28 <sipa> make a new wallet, send your money there :)
385 2014-06-16 09:56:44 <Jouke> are you sure it is the wallet's fault?
386 2014-06-16 09:56:44 <sipa> ah, privacy... you'll likely be aggregating all of your coins
387 2014-06-16 09:56:55 <WalletMigrator> should I send each utxo I have in a transaction? or make it separate? or something else?
388 2014-06-16 09:57:01 <Jouke> I have a wallet of over 500M that's still ok to start
389 2014-06-16 09:58:09 <WalletMigrator> hm
390 2014-06-16 09:58:10 <WalletMigrator> 2014-06-16 09:50:00 wallet 663ms
391 2014-06-16 09:58:24 <WalletMigrator> maybe that's not the problem...
392 2014-06-16 09:58:27 <sipa> that sounds acceptable :p
393 2014-06-16 09:58:39 <WalletMigrator> :-/
394 2014-06-16 09:58:39 <WalletMigrator> 2014-06-16 09:50:00 block index 32460ms
395 2014-06-16 09:59:12 <WalletMigrator> 2014-06-16 09:24:13 Verifying last 288 blocks at level 3 2014-06-16 09:38:22 No coin database inconsistencies in last 114 blocks (39253 transactions) 2014-06-16 09:38:23 block index 1119410ms
396 2014-06-16 09:59:12 <WalletMigrator> and last time:
397 2014-06-16 09:59:22 <sipa> what hardware?
398 2014-06-16 09:59:24 <WalletMigrator> why does the block index take a long time
399 2014-06-16 09:59:32 <WalletMigrator> i7-3610QM
400 2014-06-16 09:59:34 <WalletMigrator> 8gb ram
401 2014-06-16 09:59:43 <sipa> what disk/os/fs?
402 2014-06-16 10:00:10 <WalletMigrator> win7, ntfs afaik
403 2014-06-16 10:00:15 <WalletMigrator> hd that came in the laptop
404 2014-06-16 10:00:35 <gmaxwell> perhaps some crazy AV is breaking it.
405 2014-06-16 10:00:37 <WalletMigrator> 750gb, idk what model
406 2014-06-16 10:00:50 <gmaxwell> because, cold cache here: 2014-06-16 08:42:42 block index 16254ms
407 2014-06-16 10:00:54 <WalletMigrator> hm, this time just now it only took 30 secs
408 2014-06-16 10:01:00 <WalletMigrator> last time it toom 1119 :/
409 2014-06-16 10:01:02 <WalletMigrator> took
410 2014-06-16 10:01:08 <sipa> well you can decrease the selfcheck depth with -checkblocks=N (to only verify N blocks, but 0 means all, so use at least 1)
411 2014-06-16 10:01:20 <sipa> but if that's so slow, i think it's indicative of another problem