1 2015-02-09 00:16:41 <Luke-Jr> bliljerk101: when he's on irc (rare), he'd be in #bitcoin-wiki - and I don't know his sex
  2 2015-02-09 00:21:12 <bliljerk101> Thanks Luke-Jr
  3 2015-02-09 01:55:46 <grim210> hello all.  I found this: https://github.com/bitcoin/bitcoin/issues/2511 and was wondering if anyone knows what bitcoind will be going to in the future as opposed to BDB
  4 2015-02-09 02:03:06 <sipa> grim210: some simple custom append-only key-value store format
  5 2015-02-09 02:05:10 <grim210> so the dev team is attempting to make a custom key-value database specifically for bitcoin core?
  6 2015-02-09 02:05:24 <grim210> and i don't mean that in a deragatory way either..
  7 2015-02-09 02:06:42 <gmaxwell> grim210: "database" is really the wrong word for something that is an append only list of records.
  8 2015-02-09 02:07:15 <sipa> grim210: also, no guarantee when that will ever happen, but that seems the most viable strategy
  9 2015-02-09 02:07:27 <sipa> it's pretty trivial, but there is not so much work going into the wallet implementation
 10 2015-02-09 02:07:37 <gmaxwell> it doesn't need basically any of the features that you'd normally identify for a database, except the file sticking around after unclean shutdowns without corruption (something many 'database' fail pretty miserably at :) )
 11 2015-02-09 02:08:00 <grim210> it's definitely something you don't want so screw up :) i just get frustrated by the use of dbd 4.8 still
 12 2015-02-09 02:08:07 <grim210> bdb *
 13 2015-02-09 02:09:36 <sipa> understandably
 14 2015-02-09 02:56:10 <earlz> Will Debian 7.8.0 work for gitian builds?
 15 2015-02-09 02:56:27 <earlz> Also, is 7.7.0 even available for download anywhere?
 16 2015-02-09 02:56:37 <sipa> indubitably
 17 2015-02-09 02:56:55 <earlz> that... is a word I've never seen before lol
 18 2015-02-09 02:57:37 <fanquake> earlz Yes 7.8 will work
 19 2015-02-09 02:58:26 <sipa> earlz: that's one of the reasons for using it :)
 20 2015-02-09 02:58:29 <earlz> going to attempt to get gitian working again.. hopefully the virtualbox guide on github works better this time than last time
 21 2015-02-09 03:53:53 <SubCreative> <3
 22 2015-02-09 03:55:06 <PRab> earlz: I just followed it a week or 2 ago and it worked for me.
 23 2015-02-09 04:17:08 <earlz> I know last time I tried to follow that guide I ran into tons of problems. I'm hoping things are better now
 24 2015-02-09 06:01:00 <earlz> What format is a gitian key in?
 25 2015-02-09 06:01:08 <earlz> for the SIGNER variable
 26 2015-02-09 06:02:45 <sipa> your name, whatever you like
 27 2015-02-09 06:02:59 <earlz> ah, figured it'd be a public key of some sort
 28 2015-02-09 06:02:59 <sipa> though it should be something your gpg understands to be you
 29 2015-02-09 07:46:09 <michagogo> earlz: well, it's two things
 30 2015-02-09 07:46:21 <michagogo> It's the name of the directory in gitian.sigs
 31 2015-02-09 07:46:34 <michagogo> So it should be something human-recognizable as you
 32 2015-02-09 07:46:48 <michagogo> e.g. your name or nick or whatever
 33 2015-02-09 07:47:50 <michagogo> But it's also the argument passed to GPG for which key to use, so make sure that GPG can figure out from that which key you want to use
 34 2015-02-09 07:50:09 <jonasschnelli> needs GUI and LOW-PRIO tag: https://github.com/bitcoin/bitcoin/pull/5772
 35 2015-02-09 08:16:17 <Krellan> Question about the hardcoded constant of 8 outbound connections: considering making this a command-line option. Good or bad idea?
 36 2015-02-09 08:16:37 <Krellan> Is there something magic about the number 8 that makes it a constant that we want to keep cast in stone, or not?
 37 2015-02-09 08:17:39 <Diablo-D3> heh
 38 2015-02-09 08:17:57 <Diablo-D3> all I want is upstream limiting
 39 2015-02-09 08:18:27 <Krellan> Would be nice
 40 2015-02-09 08:22:41 <jonasschnelli> is there any pre-work (implementation or conceptional work) regarding adding a SPV mode for bitcoind?
 41 2015-02-09 08:25:32 <gmaxwell> Krellan: It should not be configurable.
 42 2015-02-09 08:25:48 <gmaxwell> Krellan: increasing the number just abuses the network. One can decrease it by just adjusting max connections.
 43 2015-02-09 08:26:22 <gmaxwell> Users are often mistaken about advantages they'd gain by increasing it; so offering something to adjust it would only be an attractive nussance.
 44 2015-02-09 08:26:35 <gmaxwell> (8 is already very high as well)
 45 2015-02-09 08:26:51 <phantomcircuit> <Diablo-D3> all I want is upstream limiting
 46 2015-02-09 08:26:53 <phantomcircuit> que?
 47 2015-02-09 08:51:54 <Diablo-D3> phantomcircuit: bittorrent style, I tell bitcoin how much bandwidth to use
 48 2015-02-09 09:05:34 <jitenm> hi folks!
 49 2015-02-09 09:08:57 <phantomcircuit> Diablo-D3, protocol doesn't support it
 50 2015-02-09 09:09:02 <phantomcircuit> youd have to disconnect clients
 51 2015-02-09 09:09:05 <phantomcircuit> nasty
 52 2015-02-09 09:09:50 <wumpus> phantomcircuit: now with parallel download that's no longer true, you could just throttle as well
 53 2015-02-09 09:13:13 <phantomcircuit> wumpus, plenty of older clients
 54 2015-02-09 09:14:10 <wumpus> sure, but at some point they don't matter anymore, and initially the number of nodes using throttling would be very low too
 55 2015-02-09 09:14:24 <Krellan> gmaxwell: Thanks. I want to actually lower it, not increase. I wanted to try only 4 connections (2 in, 2 out).
 56 2015-02-09 09:15:08 <wumpus> (if ,say, throttling would be introduced in 0.11)
 57 2015-02-09 09:15:28 <Krellan> As it is now, when I lower maxconnections below 8, my node becomes 100% outbound, no inbound.
 58 2015-02-09 09:16:53 <wumpus> jonasschnelli: lol, how did we get there storing the window position/geometry for the about dialog inthe first place
 59 2015-02-09 09:21:01 <jonasschnelli> wumpus, Yes. Somehow funny. The "HelpMessageDialog" should be renamed to HelpAndAboutDialog ...
 60 2015-02-09 09:21:03 <wumpus> Krellan: there is a pull that allows changing # of outbound connections separately
 61 2015-02-09 09:21:13 <Diablo-D3> [04:09:14] <phantomcircuit> Diablo-D3, protocol doesn't support it
 62 2015-02-09 09:21:15 <wumpus> Krellan: (obviously, up to 8)
 63 2015-02-09 09:21:20 <Diablo-D3> phantomcircuit: I only care about upstream
 64 2015-02-09 09:21:25 <Diablo-D3> phantomcircuit: downstream is fine
 65 2015-02-09 09:21:45 <wumpus> Diablo-D3: right now just use QoS limiting in your router or OS, there is an example script for that in the repository somewhere
 66 2015-02-09 09:22:11 <Diablo-D3> wumpus: that doesnt really work because of the bittorrent problem: too many independent connections
 67 2015-02-09 09:24:28 <wumpus> linux qos doesn't handle that very well?
 68 2015-02-09 09:24:51 <wumpus> thought it allows defining arbitrary buckets, but it's been a while
 69 2015-02-09 09:29:39 <wumpus> also you can of course limit the number of connections
 70 2015-02-09 09:33:06 <fishdaemon> Hello, I am quite new to bitcoins and the blockchain.   Could anyone point me in the right direction to a python library and maybe some examples on scripting on the blockchain and using it as a distributed database
 71 2015-02-09 09:34:45 <wumpus> as for python library look no further than python-bitcoinlib
 72 2015-02-09 09:35:03 <wumpus> using the block chain as a distributed database sounds like an awful idea though
 73 2015-02-09 09:35:19 <fishdaemon> this one? https://pypi.python.org/pypi/python-bitcoinlib
 74 2015-02-09 09:35:46 <wumpus> yes
 75 2015-02-09 09:36:36 <fishdaemon> thanks
 76 2015-02-09 09:36:43 <fishdaemon> I'll start there
 77 2015-02-09 09:36:43 <jonasschnelli> wumpus, do you know if there was any work on adding a SPV mode for bitcoind? Conceptional or implementation?
 78 2015-02-09 09:36:49 <phantomcircuit> Diablo-D3, older clients will stall if you throttle upload
 79 2015-02-09 09:37:00 <phantomcircuit> that will change as more clients go to 0.10.x
 80 2015-02-09 09:37:07 <Diablo-D3> phantomcircuit: okay let them stall then
 81 2015-02-09 09:37:17 <Diablo-D3> I currently cant run bitcoin because it destroys my connection
 82 2015-02-09 09:38:44 <Luke-Jr> fishdaemon: the blockchain is not a database of any sort at all
 83 2015-02-09 09:42:31 <fishdaemon> Is it not possible to put arbitary information on it?
 84 2015-02-09 09:43:53 <wumpus> jonasschnelli: not that I know of
 85 2015-02-09 09:44:07 <jonasschnelli> okay. Thanks.
 86 2015-02-09 09:45:41 <wumpus> fishdaemon: have you looked into other distributed databases and distributed database research? using the block chain for this because blockchains are 'hot' right now makes zero sense
 87 2015-02-09 09:45:43 <Luke-Jr> fishdaemon: yes and no
 88 2015-02-09 09:47:15 <Luke-Jr> it'd be like encoding your data in DTMF, then "storing" it by making a phone call and sending those DTMF
 89 2015-02-09 09:48:05 <wumpus> hehe
 90 2015-02-09 09:48:07 <jonasschnelli> lol @ DTMF analogy
 91 2015-02-09 09:48:07 <Luke-Jr> sure, *maybe* the NSA will save your data somewhere. but good luck trying to put it to any use
 92 2015-02-09 09:48:48 <fishdaemon> What i would like to do is to associate a short message , like a tweet with a wallet
 93 2015-02-09 09:49:10 <Luke-Jr> fishdaemon: check out the paymetn protocol
 94 2015-02-09 09:49:13 <Luke-Jr> BIP70 IIRC
 95 2015-02-09 09:49:16 <Luke-Jr> it has a field for memos
 96 2015-02-09 09:49:47 <fishdaemon> thx Luke-jr
 97 2015-02-09 10:47:53 <wumpus> so in 0.9, IsSuperMajority simply got the denominator as an argument instead of side-loading it from Params(), that was easier to read
 98 2015-02-09 11:15:46 <fishdaemon> Would it be considered bad practice to make a null transaction if that transaction would most likely result in many real transactions ?
 99 2015-02-09 11:27:23 <wumpus> a null transaction?
100 2015-02-09 11:43:09 <L_Cranston_Shado> so… I accidentally synced from the bitcoin/bitcoin:master fork which added a bunch of ridiculous commits to my PR and I can’t seem to undo it, despite trying revert
101 2015-02-09 11:46:39 <wumpus> if you want to get rid of everything and make your branch equal to master, use git reset --hard origin/master (this gets rid of any changes, but as it is just one line re-doing that is easier than fighting git)
102 2015-02-09 11:50:39 <L_Cranston_Shado> thanks
103 2015-02-09 11:50:57 <L_Cranston_Shado> wumpus, can I steal your idea to create a bumper sticker that says “I thought the git and the git won”?
104 2015-02-09 11:51:08 <L_Cranston_Shado> *I fought the git and the gin won
105 2015-02-09 11:51:11 <L_Cranston_Shado> *git
106 2015-02-09 11:51:15 <L_Cranston_Shado> you know what i meant
107 2015-02-09 11:51:20 <wumpus> hehe, sure :)
108 2015-02-09 11:52:26 <L_Cranston_Shado> had to force the push, but hopefully it worked
109 2015-02-09 11:52:47 <L_Cranston_Shado> it closed the PR
110 2015-02-09 11:53:10 <wumpus> strange; normally it will just show the new commits when you force push
111 2015-02-09 11:53:46 <wumpus> e.g. no need to open a new PR, but I've heard of it going wrong with other people too, I've never had issues force-pushing
112 2015-02-09 11:54:39 <L_Cranston_Shado> I’m going to implement the proposed changes from the PR and push, hopefully that will work
113 2015-02-09 11:59:56 <L_Cranston_Shado> ugh, didn’t reopen and won’t let me reopen it
114 2015-02-09 12:00:10 <L_Cranston_Shado> one of the people with permissions for the repository has to I think
115 2015-02-09 12:00:12 <wumpus> just open a new one
116 2015-02-09 12:00:28 <wumpus> if github closes the issue, no one but github admins can reopen it, not even repo admins
117 2015-02-09 12:00:37 <L_Cranston_Shado> damn
118 2015-02-09 12:00:53 <L_Cranston_Shado> already bugged the github admins enough today
119 2015-02-09 12:01:00 <L_Cranston_Shado> I’ll just do a new one I guess
120 2015-02-09 12:13:24 <L_Cranston_Shado> FYI: https://github.com/bitcoin/bitcoin/pull/5777
121 2015-02-09 12:18:57 <fanquake> wumpus guess we’re having a rc5 in a bit?
122 2015-02-09 12:19:31 <fanquake> eh sorry, none of those merge have been to the 0.10 branch.
123 2015-02-09 12:21:12 <L_Cranston_Shado> can I suggest not doing a merge on any branch with a PR open? Trust me, bad idea :)
124 2015-02-09 12:21:47 <L_Cranston_Shado> assuming you have PR’s open, if you don’t disregard
125 2015-02-09 12:22:18 <L_Cranston_Shado> fanquake, and if you think I’ve gone mad at this point, you missed the issues above that I was having
126 2015-02-09 12:24:32 <wumpus> fanquake: no, I don't plan on doing a rc5 unless new problems turn up
127 2015-02-09 12:25:16 <wumpus> I really hope rc4 can be final and we can stop slipping on the release schedule
128 2015-02-09 12:25:33 <L_Cranston_Shado> wumpus, is there still a demand for gitian builders?
129 2015-02-09 12:25:42 <fanquake> Yep. rc4 looks pretty good.
130 2015-02-09 12:25:57 <fanquake> L_Cranston_Shado Always.
131 2015-02-09 12:26:02 <wumpus> L_Cranston_Shado: more gitian builders are always welcome
132 2015-02-09 12:26:20 <L_Cranston_Shado> I was going to explore it on my next day off since I figure it will take at least several hours to figure out and do properly
133 2015-02-09 12:26:31 <fanquake> Think we’ve got 8 or 9 at the moment. Which is actually more than usual :)
134 2015-02-09 12:27:40 <L_Cranston_Shado> not sure how it will work trying to build on my laptop but it probably shouldn’t be too bad
135 2015-02-09 12:28:26 <L_Cranston_Shado> at some point I’m going to get a Rasberry Pi so I can set it up as an ADS-B receiver (for aviation traffic), which I may be able to double bill to also use as a build box
136 2015-02-09 12:29:29 <wumpus> building bitcoind on a rpi is an exercise in masochism, you need lots of memory and CPU to compile it
137 2015-02-09 12:30:06 <L_Cranston_Shado> very well, laptop it is then
138 2015-02-09 12:35:35 <wumpus> L_Cranston_Shado: oh, you force-pushed *before* re-commiting your changes, oops, yes that will irrevocably close the pull
139 2015-02-09 12:35:45 <jonasschnelli> L_Cranston_Shado, regarding #5777 ... Yes. Github sometimes mix up things. So no worries.
140 2015-02-09 12:36:11 <jonasschnelli> L_Cranston_Shado, i think in that case you could still consider using git reflog
141 2015-02-09 12:36:30 <wumpus> reflog doesn't help there
142 2015-02-09 12:36:56 <L_Cranston_Shado> at this point I’m happy to let it die, this is almost a new PR anyway since it actually has a coherent description and the code you submitted… works
143 2015-02-09 12:38:55 <wumpus> right, there's just nothing you can do to resurrect a pull after github thinks it was merged (or removed), besides opening a new PR which is easy enough :)
144 2015-02-09 12:40:11 <L_Cranston_Shado> It’s one of many things that github shouldn’t let you do but does. On the technical side, Git and by extension Github, has to be one of the most mistake unfriendly systems possible
145 2015-02-09 12:40:54 <L_Cranston_Shado> They know their audience though, it’s designed for hardcore coders
146 2015-02-09 12:41:19 <jonasschnelli> L_Cranston_Shado, so you created a new github profile? L-Cranston-Shadow: Joined on Feb 7, 2015
147 2015-02-09 12:42:21 <L_Cranston_Shado> No, I just joined the other day
148 2015-02-09 12:43:10 <wumpus> well it's quite ok once you get used to it, I don't know if you used any of the other repo hosting systems but github is one of the more user friendly ones
149 2015-02-09 12:43:55 <wumpus> (e.g. there's gitorious, bitbucket, and a few others too, as well as some software you can host on your own server like trac)
150 2015-02-09 12:45:13 <L_Cranston_Shado> Maybe it’s just the few bad experiences I’ve had in the last two days, this was minor compared to the debacle I went through before, thankfully solved by a very helpful GitHub admin. I’ld rather not describe it on a logged channel though.
151 2015-02-09 12:47:50 <L_Cranston_Shado> to its credit though, interacting with other users is incredibly easy and friendly which is great
152 2015-02-09 12:48:43 <L_Cranston_Shado> Jonasschnelli, I’m The Shadow though, so I may be new but I know what evil lurks in the hearts of men
153 2015-02-09 12:50:03 <jonasschnelli> ;)
154 2015-02-09 12:50:48 <L_Cranston_Shado> So few people get the reference nowadays, it’s sad
155 2015-02-09 12:51:49 <L_Cranston_Shado> wumpus, jonasschnelli, here’s a question, assuming that Satoshi is still active in the project, do you guys ever try to figure out which contributor on GitHub to the project is actually Satoshi?
156 2015-02-09 12:52:49 <jonasschnelli> L_Cranston_Shado, The Shadow is from the 30ties, don't expect that this lives forever.
157 2015-02-09 12:53:35 <L_Cranston_Shado> jonasschnelli, unfortunately the only Shadow project in recent history was a mediocre (to be generous) movie starring Alec Baldwin in the 90's
158 2015-02-09 12:53:57 <L_Cranston_Shado> http://www.imdb.com/title/tt0111143/?ref_=fn_al_tt_1
159 2015-02-09 12:59:39 <L_Cranston_Shado> so is there no way to get up to speed with master on a fork without screwing up PRs?
160 2015-02-09 12:59:56 <L_Cranston_Shado> branches rather, nto forks
161 2015-02-09 13:00:00 <L_Cranston_Shado> *not
162 2015-02-09 13:00:20 <jonasschnelli> git rebase
163 2015-02-09 13:00:31 <jonasschnelli> but if there are conflict you need to resolve these manually
164 2015-02-09 13:00:47 <jonasschnelli> some conflicts needs attentions from humans
165 2015-02-09 13:01:55 <wumpus> I already mentioned that on your original issue, git pull --rebase
166 2015-02-09 13:03:17 <L_Cranston_Shado> there will be a conflict since one of the modified files is utilitydialog.cpp
167 2015-02-09 13:04:01 <jonasschnelli> Could be, just try to solve it directly in the source.... should be easy
168 2015-02-09 13:18:14 <Tykling> how can I get the full command line bitcoin-core was started with? is there something like GetBoolArg which returns the lot? thanks!
169 2015-02-09 13:20:14 <Tykling> I need it inside src/qt/guiutil.cpp if that matters
170 2015-02-09 13:20:57 <jonasschnelli> Tykling, i guess you could use "mapArgs"
171 2015-02-09 13:21:14 <jonasschnelli> std::map<std::string, std::string> mapArgs;
172 2015-02-09 13:21:27 <wumpus> no, the original command line is not stored (although it could be, if that's really useful)
173 2015-02-09 13:21:41 <wumpus> mapArgs also includes input from bitcoin.conf
174 2015-02-09 13:21:55 <wumpus> as well as options overridden by the GUI
175 2015-02-09 13:22:13 <jonasschnelli> Tykling, what would you like to do in guiutil.cpp?
176 2015-02-09 13:23:36 <Tykling> jonasschnelli: I need the command line because I am working on a bug where the autostart shortcut created when you check "Start Bitcoin on system login" doesn't start a testnet instance even if it is checked in a testnet bitcoin-core
177 2015-02-09 13:24:05 <jonasschnelli> Tykling, ah, That one.
178 2015-02-09 13:24:15 <Tykling> like, if you run a mainnet and a testnet at the same time and check  "Start Bitcoin on system login" in both, two mainnet bitcoin-core starts up
179 2015-02-09 13:24:27 <Tykling> and you get an error and no testnet instance
180 2015-02-09 13:24:48 <Tykling> I figured the best way to solve it was to create the shortcut using the same arguments that was used to start the program
181 2015-02-09 13:25:23 <Tykling> I've created this issue https://github.com/bitcoin/bitcoin/issues/5778 and a simple patch that solves it for the -testnet case, but I think it would be more correct to use the full commandline..
182 2015-02-09 13:25:27 <jonasschnelli> Tykling, sure... hmm... i don't have the understanding of how windows does autostart at login...
183 2015-02-09 13:27:12 <Tykling> jonasschnelli: the program creates a .lnk shortcut file in which the commandline arguments are set, https://github.com/bitcoin/bitcoin/blob/master/src/qt/guiutil.cpp#L600
184 2015-02-09 13:27:43 <Tykling> in line 600 in guiutil.cpp it sets the argument to "-min" nomatter what network it is on
185 2015-02-09 13:29:08 <wumpus> I'd recommend putting 'persistent' settings in bitcoin.conf
186 2015-02-09 13:29:15 <wumpus> and not working around this in this way
187 2015-02-09 13:29:42 <Tykling> wumpus: regardless it is wrong to create a shortcut that starts a mainnet instance from a testnet instance
188 2015-02-09 13:29:49 <wumpus> remembering the original command line seems smart, but is also a good way to generate confusion, as it is unexpected persistence
189 2015-02-09 13:30:21 <Tykling> this simple patch fixes my problem but I think it should be extended to include -regtest and other network types too https://github.com/tykling/bitcoin/commit/b61cd38948f66b9882f0ecb77c8a38f27123fe8e
190 2015-02-09 13:30:23 <wumpus> Tykling: yes, for mainnet/testnet it makes sense
191 2015-02-09 13:30:41 <Tykling> which makes me think that just using the original command line was easier, and here I am :)
192 2015-02-09 13:30:50 <wumpus> as they change the instance type at a root level; but not for other command line options
193 2015-02-09 13:31:00 <wumpus> no, I'd really recommand against that
194 2015-02-09 13:31:28 <Tykling> ok
195 2015-02-09 13:31:35 <Tykling> well, then perhaps my current patch is fine
196 2015-02-09 13:31:38 <wumpus> single out the network type, and pass a command line argument based on that
197 2015-02-09 13:32:16 <wumpus> apart from that , command line options are meant for temporary setting changes, not for anything that should be persisted
198 2015-02-09 13:33:06 <Tykling> ok
199 2015-02-09 13:33:27 <Tykling> but there is more than just testnet vs. mainnet, you have regtest as well, and others?
200 2015-02-09 13:33:41 <wumpus> to make persistent configuration changes put them in bitcoin.conf, or in case of the GUI in the options dialog
201 2015-02-09 13:33:59 <Tykling> sure
202 2015-02-09 13:34:04 <wumpus> no, not others.
203 2015-02-09 13:36:31 <Tykling> since you need to set testnet before the program starts we can't do it in the options dialog
204 2015-02-09 13:36:43 <wumpus> yes
205 2015-02-09 13:36:59 <wumpus> it indeed makes sense to add an argument based on the network type
206 2015-02-09 13:37:14 <wumpus> so that the automatically launched instance will have the network type that you configured it in
207 2015-02-09 13:37:17 <Tykling> even if I put testnet=1 in the config file it will still create the wrong shortcut and thus start the wrong bitcoin instance
208 2015-02-09 13:37:22 <Tykling> yeah, ok we agree then
209 2015-02-09 13:37:34 <L_Cranston_Shado> it’s not accepting my push after I do a rebase, can I safely do a pull after a rebase and then the push?
210 2015-02-09 13:37:46 <jonasschnelli> L_Cranston_Shado, did you use --force?
211 2015-02-09 13:37:48 <wumpus> yes, I agree on that part, just not on doing it for other arguments
212 2015-02-09 13:38:08 <L_Cranston_Shado> jonasschnelli, I’m pretty sure that’s what caused me problems last time which makes me somewhat reluctant
213 2015-02-09 13:38:22 <Tykling> That means that my patch will have to check what network we are running in, instead of checking if -testnet was passed, because it needs to work even if testnet=1 was specified in the config file and not on the command line
214 2015-02-09 13:38:30 <wumpus> putting -testnet or -regtest in bitcoin.conf does work, but it is a bit strange
215 2015-02-09 13:38:32 <wumpus> Tykling: yep
216 2015-02-09 13:38:46 <xabbix> Block sync on v0.9.x was done via 1 peer alone?
217 2015-02-09 13:38:55 <xabbix> s/was/is
218 2015-02-09 13:39:32 <L_Cranston_Shado> jonasschnelli, yep https://github.com/bitcoin/bitcoin/pull/5777
219 2015-02-09 13:40:28 <jonasschnelli> L_Cranston_Shado, something you must do different than others...
220 2015-02-09 13:41:00 <L_Cranston_Shado> I just did git rebase master, and then the force push
221 2015-02-09 13:41:03 <wumpus> Tykling: qt has some per-network information in NetworkStyle, e.g. you could add the argument to launch with a certain network to that
222 2015-02-09 13:41:42 <jonasschnelli> L_Cranston_Shado, but don't close again, create a new local branch out of master, cherry pick your commit, push --force
223 2015-02-09 13:41:49 <jonasschnelli> this will work
224 2015-02-09 13:43:45 <wumpus> Tykling: if you want it to work in the case of regtest=X and testnet=X in the bitcoin.conf you have to path *both* though, to avoid one of them being overridden (or both set, in which case it will crash), so -testnet=0 -regtest=0 to force mainnet :p
225 2015-02-09 13:43:58 <wumpus> that's pretty annoying and why  https://github.com/bitcoin/bitcoin/pull/5229 exists to unify into one argument
226 2015-02-09 13:44:16 <wumpus> so in that case your patch would just have to specify -network=<curnetworkname>
227 2015-02-09 13:45:05 <wumpus> (well, as long as that takes precedence over testnet=X and regtest=X of course... it could make the determination even more complex...)
228 2015-02-09 13:45:34 <Tykling> yeah
229 2015-02-09 13:45:36 <Tykling> ugh
230 2015-02-09 13:46:25 <Tykling> this is like when you yank on a piece of string in a sweater and before you know it you are naked
231 2015-02-09 13:47:33 <wumpus> yes. if this were to be designed from scratch, -network=<network> would obviously make the most sense, but there's a lot of historical legacy, and back in the day testnet was introduced no one was thinking that there would be a third chain at some point
232 2015-02-09 13:48:00 <Tykling> yeah
233 2015-02-09 13:48:04 <Tykling> classic :)
234 2015-02-09 13:49:23 <Tykling> ok to sum up, what is the best way of determining which network we are running on while inside guiutil.cpp in your opinion?
235 2015-02-09 13:49:38 <wumpus> in any case - it would be acceptable for this to just not work for regtest
236 2015-02-09 13:50:03 <wumpus> it's meant for regression testing, not for running persistently in the background
237 2015-02-09 13:50:16 <L_Cranston_Shado> jonasschnelli, didn’t work
238 2015-02-09 13:50:29 <jonasschnelli> L_Cranston_Shado, what happened?
239 2015-02-09 13:50:38 <wumpus> so just handle the testnet and mainnet case, and ignore testnet (ie, do nothing if on regtest, or unittest, or anything weirder)
240 2015-02-09 13:50:48 <wumpus> s/ignore testnet/ignore regtest/
241 2015-02-09 13:51:22 <L_Cranston_Shado> I manually put in the changes to the two files I needed to change, I used the client to do the commit so I didn’t have to do the reason in terminal, which is a pain in the butt, and then in terminal I did the force push
242 2015-02-09 13:51:29 <L_Cranston_Shado> jonasschnelli
243 2015-02-09 13:52:16 <wumpus> after #5229 is merged it could be adapted to just pass -network=<current network id>, making sure that -network always takes precedence over loose -regtest and -testnet options
244 2015-02-09 13:52:51 <wumpus> (something #5229 doesn't do at this point, so maybe comment on this usecase)
245 2015-02-09 13:53:38 <wumpus> it does make doing what you want to do lots easier
246 2015-02-09 13:53:43 <Tykling> yes
247 2015-02-09 13:55:01 <jonasschnelli> L_Cranston_Shado, i would do the following: don't use a Github-GUI-Client for now. Assume "origin" is https://github.com/bitcoin/bitcoin, do a "git reset --hard origin/master", then "git branch new_local_branch", then "git checkout new_local_branch". Then do you changes, then "git add <your 2 files>", then "git commit -m '<commitmsg>'", then "git push your-remote new_local_branch:removing-about-dialog-window-saving"
248 2015-02-09 13:55:24 <jonasschnelli> maybe you have to check that you have a remote pointing to "github.com/bitcoin/bitcoin"
249 2015-02-09 13:56:00 <jonasschnelli> if your origin points to your bitcoin fork, make sure adding another remote "git add remote add bitcoin https://github.com/bitcoin/bitcoin".
250 2015-02-09 13:56:16 <jonasschnelli> Then do "git fetch bitcoin" "git reset --hard bitcoin/master"
251 2015-02-09 14:03:21 <wumpus> Tykling: thinking of it, I think right now the obvious solution will just work. e.g. strprintf("-testnet=%i -regtest=%i", GetBoolArg("-testnet",0), GetBoolArg("-regtest",0))
252 2015-02-09 14:03:45 <wumpus> Tykling: GetBoolArg takes whatever value holds, no matter whether it comes from the command line or bitcoin.conf
253 2015-02-09 14:03:52 <Tykling> oh!
254 2015-02-09 14:04:12 <Tykling> clever
255 2015-02-09 14:04:13 <wumpus> so that will just reproduce exactly the current network
256 2015-02-09 14:04:18 <Tykling> yes
257 2015-02-09 14:04:29 <Tykling> I'll modify my patch and do a pull request later tonight
258 2015-02-09 14:04:51 <Tykling> thank you very much for your help :)
259 2015-02-09 14:07:31 <L_Cranston_Shado> jonasschnelli, that last step doesn’t work, it treats the first part as a hostname
260 2015-02-09 14:07:35 <L_Cranston_Shado> good try though
261 2015-02-09 14:09:28 <L_Cranston_Shado> jonas, you have full perms on my repository so feel free to fiddle if you want, although don’t feel any pressure to. I’ll leave a note on the request noting what happened and maybe try to figure it again out when I get back on
262 2015-02-09 14:11:04 <jonasschnelli> L_Cranston_Shado, try "git push origin new_local_branch:removing-about-dialog-window-saving"
263 2015-02-09 14:12:20 <L_Cranston_Shado> it rejected it as being behind the remote version (which is utterly ridiculous considering I just updated my master from the remote) so I did a force
264 2015-02-09 14:12:31 <jonasschnelli> L_Cranston_Shado, try "git push origin new_local_branch:removing-about-dialog-window-saving --force"
265 2015-02-09 14:13:24 <jonasschnelli> L_Cranston_Shado, what does "git remote show" says?
266 2015-02-09 14:13:35 <L_Cranston_Shado> bitcoin and origin
267 2015-02-09 14:14:43 <L_Cranston_Shado> maybe use —mirror and add it?
268 2015-02-09 14:15:09 <L_Cranston_Shado> I can never be sure whether it’s pulling from my master fork or from the /bitcoin/bitcoin master fork
269 2015-02-09 14:15:39 <L_Cranston_Shado> incidentally, that last attempt worked in that it pushed it, but it still has 86 commits and 28 changed files
270 2015-02-09 14:15:51 <jonasschnelli> do the same steps again but make sure to do "git fetch bitcoin", "git reset --hard bitcoin/master" before doing the steps
271 2015-02-09 14:16:36 <L_Cranston_Shado> is there any way to delete forks btw?
272 2015-02-09 14:16:37 <jonasschnelli> but your last push is about 1h ago... so your "git push origin new_local_branch:removing-about-dialog-window-saving --force" didn't worked?
273 2015-02-09 14:16:59 <jonasschnelli> L_Cranston_Shado, you mean deleting branches on your remote?
274 2015-02-09 14:17:08 <jonasschnelli> git push origin --delete <branchName>
275 2015-02-09 14:17:08 <L_Cranston_Shado> I meant on the server side
276 2015-02-09 14:17:24 <jonasschnelli> but in your case there is no need to.
277 2015-02-09 14:17:29 <jonasschnelli> just over-push it with --force
278 2015-02-09 14:17:48 <jonasschnelli> but first you need a proper branch, created from a recent bitcoin/master
279 2015-02-09 14:18:19 <L_Cranston_Shado> git fetch bitcoin
280 2015-02-09 14:18:22 <L_Cranston_Shado> err whoops
281 2015-02-09 14:20:30 <Tykling> wumpus: does this look right?
282 2015-02-09 14:20:32 <Tykling> TCHAR pszArgs[5] = TEXT("-min");
283 2015-02-09 14:20:34 <Tykling> becomes
284 2015-02-09 14:20:35 <L_Cranston_Shado> jonasshcnelli, just got a notice that I’ve diverged from the server version which is a good sign
285 2015-02-09 14:20:40 <Tykling> TCHAR pszArgs[5] = strprintf("-min -testnet=%i -regtest=%i", GetBoolArg("-testnet",0), GetBoolArg("-regtest",0))
286 2015-02-09 14:21:12 <Tykling> or do I need to wrap it in TEXT(...); ?
287 2015-02-09 14:25:17 <wumpus> that looks scary
288 2015-02-09 14:25:51 <wumpus> I'd first assign the result of strprintf to std::string args
289 2015-02-09 14:26:01 <Tykling> ok
290 2015-02-09 14:26:26 <wumpus> then do something like TCHAR *pszArgs = args.c_str();
291 2015-02-09 14:26:44 <wumpus> (I have no idea what the TEXT macro does or whether you need it)
292 2015-02-09 14:26:52 <L_Cranston_Shado> jonasschnelli, thank you very much, it worked
293 2015-02-09 14:27:25 <jonasschnelli> L_Cranston_Shado, no rename commit to a msg that makes sense
294 2015-02-09 14:27:38 <jonasschnelli> s/commit/your commit
295 2015-02-09 14:27:58 <jonasschnelli> git commit --amend -m "New commit message"
296 2015-02-09 14:29:28 <L_Cranston_Shado> what was that last update, 10.01?
297 2015-02-09 14:30:37 <L_Cranston_Shado> nm, doesn’t matter
298 2015-02-09 14:34:11 <L_Cranston_Shado> fuck, I just did a push because the amend wasn’t going through and it reverted my fix
299 2015-02-09 14:35:05 <L_Cranston_Shado> something really screwy is going on
300 2015-02-09 14:35:07 <paveljanik> L_Cranston_Shado, have you read Pro Git book?
301 2015-02-09 14:35:20 <paveljanik> it is a wonderful introduction into Git.
302 2015-02-09 14:35:45 <L_Cranston_Shado> I’ll look for it at some point
303 2015-02-09 14:35:51 <paveljanik> esp. chapter Remote branching.
304 2015-02-09 14:36:28 <L_Cranston_Shado> paveljanik, jonasschnelli and I have been trying this for awhile now, this last update to the master tree keeps creating phantom diffs where there are none
305 2015-02-09 14:37:10 <L_Cranston_Shado> we just got it fixed and then trying to amend the last change brought it back, despite the fact that I was pulling the latest one from the master
306 2015-02-09 14:38:31 <L_Cranston_Shado> I’me exausted, I have to get to bed, I’ll leave this window open when I disconnect so I can easily scroll back and try again when I wake up or tonight and in the meantime I’ll leave a note on the PR
307 2015-02-09 14:39:15 <Chenser>  /quote user opident opident opident opident
308 2015-02-09 14:45:07 <Tykling> wumpus: I don't have a build environment for windows here so I can't really test this but I think I have it now
309 2015-02-09 14:46:57 <jonasschnelli> Tykling, i can give it a testshot on my windows vm if you give my your branch url
310 2015-02-09 14:48:26 <Tykling> jonasschnelli: cool! hang on
311 2015-02-09 14:49:53 <Tykling> jonasschnelli: https://github.com/tykling/bitcoin/tree/fix-testnet-autostart I am not a c++ guy so there may be stupid mistakes, apologies in advance :)
312 2015-02-09 14:50:54 <Tykling> I found out that the linux desktop shortcut thing has the same problem so I did the same fix there
313 2015-02-09 14:51:29 <jonasschnelli> Tykling, sorry, i did not follow the discussion. How should i test it?
314 2015-02-09 14:51:48 <jonasschnelli> Just built a bitcoin-qt.exe,...
315 2015-02-09 14:52:10 <Tykling> jonasschnelli: see it it builds and if it does, try starting with -testnet and check the "start on system startup" in options, and see if the generated shortcut has -testnet=1 in it
316 2015-02-09 14:52:35 <jonasschnelli> Tykling, okay. Let me try
317 2015-02-09 14:52:48 <Tykling> the shortcut in the start menu Startup folder I mean
318 2015-02-09 14:55:01 <Tykling> before the patch it would generate a shortcut with just "-min" and no other arguments, regardless of whether it was running in testnet mode or not, this should fix it so the generated shortcut starts the network the program was running under when the checkbox was checked :)
319 2015-02-09 14:57:02 <jonasschnelli> Tykling, build error: qt/guiutil.cpp:603:41: error: invalid conversion from ‘const char*’ to ‘TCHAR* {aka char*}’ [-fpermissive]
320 2015-02-09 14:57:34 <Tykling> figures.. ok I'll have to setup a build env myself and get it sorted, what do you use?
321 2015-02-09 14:58:49 <Tykling> is there a guide somewhere on how to setup a build env on windows
322 2015-02-09 15:01:07 <jonasschnelli> Tykling, don't build on windows.
323 2015-02-09 15:01:16 <jonasschnelli> Tykling, cross build window builds on linux
324 2015-02-09 15:01:54 <jonasschnelli> Or create a VirtualBox ubuntu VM on window to create windows builds... sounds stupid but will be much less pain
325 2015-02-09 15:02:49 <Tykling> mkay
326 2015-02-09 15:16:51 <wumpus> Tykling: maybe that's what TEXT(...) macro does, cast const char* to char*
327 2015-02-09 15:29:07 <aliakbar> im new here..who can help me with building bitcoin-0.1.0 on linux?
328 2015-02-09 15:30:00 <jonasschnelli> aliakbar, did you had a look here: https://github.com/bitcoin/bitcoin/blob/master/doc/build-unix.md
329 2015-02-09 15:30:41 <aliakbar> @jonasschnelli: ya but i think it
330 2015-02-09 15:30:52 <aliakbar> is referring to the current code
331 2015-02-09 15:30:57 <aliakbar> isnt it?
332 2015-02-09 15:31:24 <jonasschnelli> aliakbar, so you don't want to build 0.10?
333 2015-02-09 15:31:53 <aliakbar> no..i want to build the very first release of bitcoin
334 2015-02-09 15:32:22 <jonasschnelli> aliakbar, hu.. Okay.... why you want to do this?
335 2015-02-09 15:33:11 <aliakbar> i know that the first release is only for Windows platform..but is there any instruction on how to build the first code on unix platforms?
336 2015-02-09 15:33:25 <jonasschnelli> aliakbar, maybe look at https://github.com/bitcoin/bitcoin/tree/v0.1.5
337 2015-02-09 15:33:31 <aliakbar> jonasschnelli, its for my bachelor
338 2015-02-09 15:33:37 <aliakbar> bachelors thesis
339 2015-02-09 15:33:41 <znetGeZ> aliakbar: salam maleikum
340 2015-02-09 15:33:59 <znetGeZ> aliakbar: b. Sci ?
341 2015-02-09 15:34:04 <jonasschnelli> aliakbar, i think 0.1 was not meant to run on compile/linux
342 2015-02-09 15:34:10 <jonasschnelli> *run or compile on linux
343 2015-02-09 15:34:12 <aliakbar> im trying to get familiar with the code and its parameters as in the difficulty target etc.
344 2015-02-09 15:34:41 <aliakbar> znetGeZ, aleikum salam, yes B.Sci.
345 2015-02-09 15:34:57 <znetGeZ> Informatics?
346 2015-02-09 15:34:59 <aliakbar> jonasschnelli, yes I know
347 2015-02-09 15:35:08 <aliakbar> no mech. eng.
348 2015-02-09 15:35:19 <znetGeZ> aliakbar: ah ok
349 2015-02-09 16:23:34 <earlz> does gitian build things in parallel? (ie, -j2 or some such)
350 2015-02-09 16:24:13 <earlz> and/or what can I do to speed it up
351 2015-02-09 16:34:15 <sipa> earlz: yes, exactly that
352 2015-02-09 16:34:28 <sipa> also, read the documentation
353 2015-02-09 16:41:36 <earlz> Well, I'm kinda timid since small things can break determinism in teh build
354 2015-02-09 16:41:59 <earlz> do you mean the actual gitian documentation? Cause I've read basically everything in the bitcoin github over gitian
355 2015-02-09 16:42:46 <gmaxwell> well if it breaks it, change back and try again.
356 2015-02-09 16:43:55 <sipa> earlz: i mean gbuild --help
357 2015-02-09 16:44:23 <sipa> and parallellism should not break the determinisim; i always build with -j 5
358 2015-02-09 16:46:24 <earlz> I'll try that then thanks
359 2015-02-09 16:47:05 <earlz> Also, if I have a password protected key, is there a way to only provide my key password once?
360 2015-02-09 16:53:32 <wumpus> walletpassphrase with a really long timeout?
361 2015-02-09 16:56:21 <earlz> no, not for my wallet, for gpg/gsign.. maybe I should read teh gpg documentation bleh
362 2015-02-09 16:58:31 <wumpus> look into gnome-keyring, gpg-agent (but off topic here)
363 2015-02-09 17:49:20 <davout> ohey, does anyone here have some spare testnet coins? mvcPejPSDKtH3FGESjQcpCCCvGTS68qV9R
364 2015-02-09 18:07:20 <cfields> wumpus: over the weekend i managed to get freebsd (and part of openbsd) cross-build working from linux. Any interest in hooking those up to c-i ?
365 2015-02-09 18:08:17 <wumpus> cfields: hehe why not, is there a limit to the number of test environments?
366 2015-02-09 18:08:42 <wumpus> can it also *run* fbsd/obsd executables and not just build them?
367 2015-02-09 18:08:48 <cfields> wumpus: no, it just adds to the total time it takes
368 2015-02-09 18:09:09 <cfields> only build them afaik. no clue what the wrapper situation is there
369 2015-02-09 18:09:14 <wumpus> (I remember at one time linux was able to)
370 2015-02-09 18:09:25 <wumpus> ok, well I'd say pick the one that's more popular
371 2015-02-09 18:10:03 <cfields> but it pointed out a few code incompatibilities, so it's worth doing _something_ with imo. Though it may not be justifiable to run for each build, maybe just add it to depends and run manually every now and then
372 2015-02-09 18:10:50 <cfields> based on googling for help with the issues, seems freebsd is the most viable target.
373 2015-02-09 18:11:02 <wumpus> otoh freebsd is most similar to osx
374 2015-02-09 18:11:11 <wumpus> (but maybe not the toolchain, just the kernel)
375 2015-02-09 18:11:12 <ajweiss> isn't openbsd more "bitcoinish"?
376 2015-02-09 18:11:24 <ajweiss> security focused, blah blah
377 2015-02-09 18:11:25 <wumpus> yes I guess so
378 2015-02-09 18:11:29 <cfields> well as another data point...
379 2015-02-09 18:11:43 <cfields> this started when nerdcore volunteered to help get depends working for openbsd
380 2015-02-09 18:11:58 <cfields> while he was testing stuff, i started on a freebsd cross build, since the toolchain is easier from linux
381 2015-02-09 18:12:00 <wumpus> ok
382 2015-02-09 18:12:13 <cfields> freebsd ended up being dead-simple, openbsd took some work
383 2015-02-09 18:12:15 <wumpus> well what is actual interest in is a good data point
384 2015-02-09 18:12:53 <cfields> so, yes, your argument may be valid. the fact that it was very little work may actually point to it being _less_ desirable to build via c-i
385 2015-02-09 18:14:08 <cfields> either way they weren't a ton of work. I'll clean up the changes for depends and the minor changes to bitcoin and PR them. From there, we can discuss if it's worth trying to automate at all. Sound reasonable?
386 2015-02-09 18:16:43 <wumpus> yes, sounds reasonable
387 2015-02-09 18:17:12 <cfields> roger, thanks
388 2015-02-09 18:17:13 <wumpus> at this point openbsd sounds preferable to me when we want to add an addiational build environment to c-i, espeically if we can manage to execute the tests somehow
389 2015-02-09 18:18:03 <cfields> i would agree with that. main hurdle (as far as i've seen so far) is that binutils doesn't have openbsd support baked in, like it does for freebsd
390 2015-02-09 18:18:09 <cfields> i could be wrong on that, though
391 2015-02-09 18:57:39 <petertodd> Luke-Jr: re replace-by-fee patch set, I'm going to do one up now that v0.10 is about to be released; had a few other people ask me about it as well
392 2015-02-09 18:59:30 <petertodd> Luke-Jr: would you be interested in a version applied against the Eligus patches? seems that a lot of miners are interested in that as many are now running your fork
393 2015-02-09 19:11:32 <jonasschnelli> does rpc call listtransactions has a sort order?
394 2015-02-09 19:16:31 <harding> jonasschnelli: I thought it was most-recently-added-to-wallet first, but I'm not 100% sure about that.
395 2015-02-09 19:16:56 <jonasschnelli> harding, okay. thanks. Let me test this.
396 2015-02-09 19:23:32 <jonasschnelli> for the records: sort order is -> wtx->nOrderPos
397 2015-02-09 19:41:15 <sipa> ;;blocks
398 2015-02-09 19:41:16 <gribble> 342747
399 2015-02-09 20:23:16 <rubensayshi> hmm, "receive version message: /BitCoinJ:0.12SNAPSHOT/Satoshi:0.2.0/: version 70001, blocks=342690, us=127.0.0.1:8333, them=127.0.0.1:8333, peer=204.232.165.160:45167" keeps repeating in my log and http'ing to that IP gives me bc.info (I think phising) site ... anyone else seeing similar things?
400 2015-02-09 20:25:42 <dhill> i am
401 2015-02-09 20:27:14 <rubensayshi> it's unusual for a node to send version more than once right?
402 2015-02-09 20:29:01 <dhill> a proper node won't.
403 2015-02-09 20:31:33 <jonasschnelli> / Each connection can only send one version message
404 2015-02-09 20:31:47 <jonasschnelli> main.cpp line 3463+
405 2015-02-09 20:31:55 <jonasschnelli> pfrom->PushMessage("reject", strCommand, REJECT_DUPLICATE, string("Duplicate version message"));
406 2015-02-09 20:34:59 <ajweiss> i think you actually get dos points if you do
407 2015-02-09 20:35:05 <ajweiss> oh duh hahaha
408 2015-02-09 20:36:37 <ajweiss> you get one whole dos point
409 2015-02-09 20:42:18 <ajweiss> that's also the message you see when a node connects
410 2015-02-09 20:43:09 <ajweiss> it could be a bitcoinj/thinclient with a flakey network connection... although i would assume that it would rotate its peer candidates on reconnect
411 2015-02-09 20:44:07 <ajweiss> you know what though
412 2015-02-09 20:44:13 <ajweiss> incidentally, i am seeing the same thing
413 2015-02-09 20:44:15 <ajweiss> interesting.
414 2015-02-09 21:00:33 <hearn> ajweiss: that UA is pretty weird
415 2015-02-09 21:01:19 <hearn> ajweiss: Satoshi:0.2.0 as the next part? yes, bitcoinj won't normally hammer on the same peer repeatedly, but it's a library. it can be configured to do all sorts of things.
416 2015-02-09 21:03:56 <ajweiss> well there's no ptr record
417 2015-02-09 21:03:59 <ajweiss> it's hosted at rackspace
418 2015-02-09 21:33:34 <ajweiss> here we go again...
419 2015-02-09 21:33:58 <ajweiss> looks like a 2-5s interval
420 2015-02-09 22:31:22 <Luke-Jr> petertodd: you mean ljrP? sure
421 2015-02-09 22:31:28 <Luke-Jr> ljrP*
422 2015-02-09 22:32:37 <kanzure> what's the name for the step between "coin selection" and "signaturehash signing"? when you are retrieving the correct private key. does that have a name?