1 2015-01-10 02:51:32 <dfletcher> configure: error: PROTOC not found <-- what's protoc ?
2 2015-01-10 02:51:41 <sipa> protobuf compiler
3 2015-01-10 02:56:24 <dfletcher> ahhh thanks sipa --with-protoc-bindir and we're in business :>
4 2015-01-10 04:38:42 <gmaxwell> http://sourceforge.net/p/bitcoin/mailman/message/33221963/ Re: earlier compiler errors brought up in here today; sorry I couldn't respond helpfully then but I was still working on making sure that we didn't have a greater security vulnerablity here.
5 2015-01-10 04:59:50 <Arnavion> As I just discovered, it doesn't even build
6 2015-01-10 05:00:04 <Arnavion> It has a return(cflags); that should be return(CFLAGS);
7 2015-01-10 05:00:09 <Arnavion> Quality software
8 2015-01-10 05:00:32 <gmaxwell> Arnavion: 1.0.1? yea. 1.0.0 builds.
9 2015-01-10 05:00:45 <Arnavion> Yeah, 1.0.1k
10 2015-01-10 05:06:29 <cfields> Arnavion: i saw that change in the changelog
11 2015-01-10 05:06:51 <cfields> did their fix actually break something?
12 2015-01-10 05:07:05 <Arnavion> What, they did it on purpose?
13 2015-01-10 05:07:09 <Arnavion> I thought it was a typo
14 2015-01-10 05:07:21 <Arnavion> cflags isn't defined, but CFLAGS is
15 2015-01-10 05:07:34 <Arnavion> and the rest of the file follows the convention #ifdef FOO return(FOO)
16 2015-01-10 05:08:23 <cfields> Arnavion: https://github.com/openssl/openssl/commit/5c5e7e1a7eb114cf136e1ae4b6a413bc48ba41eb
17 2015-01-10 05:08:45 <Arnavion> Well it's not in 1.0.1k
18 2015-01-10 05:09:30 <cfields> ah, yep
19 2015-01-10 05:09:52 <cfields> return(CFLAGS) just sounds terrifying, btw. I'll save looking at that for another day.
20 2015-01-10 05:10:09 <Arnavion> For now, https://github.com/hexchat/gtk-win32/blob/master/openssl/openssl-fix-compilation.patch is sufficient
21 2015-01-10 05:10:43 <cfields> mm, we didn't need it for win32
22 2015-01-10 05:10:55 <cfields> mingw, though. i suppose you're building with msvc or something?
23 2015-01-10 05:10:59 <Arnavion> Yes
24 2015-01-10 05:11:11 <cfields> ok
25 2015-01-10 05:12:00 <cfields> nnite
26 2015-01-10 09:32:00 <fanquake> afternoon
27 2015-01-10 09:37:33 <wumpus> morning
28 2015-01-10 09:42:40 <fanquake> wumpus I marked a bunch of issues for closure a few days ago, any chance you saw them?
29 2015-01-10 09:48:05 <wumpus> fanquake: if they are still open, probably not
30 2015-01-10 09:49:47 <wumpus> can you please give the #'s again?
31 2015-01-10 09:51:38 <fanquake> wumpus in the bips repo #120 can be merged. #116 closed (doesnât actually fix the typo).
32 2015-01-10 09:53:52 <wumpus> done
33 2015-01-10 09:55:50 <fanquake> wumpus In the bitcoin repo, #3003 can be closed. #2413 seems to be a duplicate. #4625 You can decide about.
34 2015-01-10 10:01:36 <wumpus> done, good catch
35 2015-01-10 15:43:20 <maaku> well it seems we may see a minimal bip 62 soft fork in the near future
36 2015-01-10 15:43:42 <maaku> are there any other non-controversial soft-forks that should be considered, I wonder?
37 2015-01-10 15:44:01 <maaku> e.g. checklocktimeverify
38 2015-01-10 16:09:18 <gmaxwell> Absolutely not.
39 2015-01-10 16:09:57 <gmaxwell> And no, it will not be BIP62, as that is clearly not mature.
40 2015-01-10 16:17:28 <maaku> gmaxwell: i'm not talking about bundling them together, rather opening voting on more than one at the same time
41 2015-01-10 16:17:43 <maaku> since we're getting the attention of miners regarding the vote anyway
42 2015-01-10 16:17:52 <maaku> (one ballot, multiple propositions)
43 2015-01-10 16:18:14 <maaku> and yes, hence "minimal bip 62". i know we're talking about just the restrictions which are already non-standard
44 2015-01-10 19:23:39 <PRab> I'm trying to setup gitian and am following the instructions at https://github.com/bitcoin/bitcoin/blob/master/doc/gitian-building.md. When I ran "bin/make-base-vm --lxc --arch amd64 --suite precise" it gave an error "E: Couldn't download packages: bsdutils busybox-initramfs eject file". Any ideas?
45 2015-01-10 19:23:53 <PRab> I had a solid internet connection the entire time.
46 2015-01-10 19:34:09 <Luke-Jr> PRab: did you skip the apt-cacher-ng?
47 2015-01-10 19:35:40 <PRab> Luke-Jr: I don't think so. Let me try to install it again.
48 2015-01-10 19:36:10 <PRab> apt-cacher-ng is already the newest version.
49 2015-01-10 19:37:45 <warren> PRab: is apt-cacher-ng running?
50 2015-01-10 19:38:46 <petertodd> maaku: I'm strongly against including CLTV as part of a minimal BIP62 soft-fork
51 2015-01-10 19:39:48 <petertodd> maaku: also, if you want to see CLTV go live sooner, write some demo applications using it and publish your work
52 2015-01-10 19:40:20 <petertodd> maaku: I have a repo myself with a CLTV micropayment channel demo: https://github.com/petertodd/checklocktimeverify-demos
53 2015-01-10 19:40:28 <PRab> warren: I think so, "ps aux | grep apt-cache-ng" returned a line.
54 2015-01-10 19:41:36 <PRab> I'll try running it agaoin.
55 2015-01-10 19:45:08 <PRab> Looks like it might be working. I'm at "Calling hook: configure_os"
56 2015-01-10 20:05:54 <PRab> It wasn't kidding that it might take a while. ~20 minutes in and still chugging.
57 2015-01-10 21:36:09 <michagogo> PRab: yeah, I think debootstrap just takes a long time for some reason
58 2015-01-10 21:36:24 <PRab> It finished now.
59 2015-01-10 21:36:49 <PRab> running gbuild is still giving an error, but I figured I would tackle that later.
60 2015-01-10 21:37:04 <michagogo> PRab: what error?
61 2015-01-10 21:37:09 <michagogo> (And what host?)
62 2015-01-10 21:37:12 <PRab> ./bin/gbuild:21:in `system!': failed to run on-target setarch x86_64 bash -x < var/build-script > var/build.log 2>&1 (RuntimeError)
63 2015-01-10 21:37:13 <iwilcox> I'm a bit confused by "--suite precise" in a Debian VM.
64 2015-01-10 21:37:35 <michagogo> PRab: pastebin var/build.log?
65 2015-01-10 21:37:38 <michagogo> iwilcox: Hm?
66 2015-01-10 21:37:59 <michagogo> You're telling make-base-vm to create a 12.04 image
67 2015-01-10 21:38:02 <iwilcox> michagogo: Just a failing command PRab pasted earlier. I'm probably missing something.
68 2015-01-10 21:38:06 <PRab> iwilcox: The way I read that is that bootstrap builds an Ubuntu VM. I assume you get a vm inside of a VM.
69 2015-01-10 21:38:16 <michagogo> (As opposed to the default, which is 10.04 iirc)
70 2015-01-10 21:38:35 <michagogo> PRab: not *exactly* a VM, but yes
71 2015-01-10 21:39:08 <michagogo> If you do it the way that that guide describes, you're using LXC for gitian
72 2015-01-10 21:39:25 <michagogo> Which is a container, not quite a VM
73 2015-01-10 21:40:17 <michagogo> PRab: that error isn't very descriptive. Can you pastebin the var/build.log file?
74 2015-01-10 21:40:24 <PRab> michagogo: I'm trying to get the build.log.
75 2015-01-10 21:40:34 <PRab> I'm not super familiar with scp
76 2015-01-10 21:40:50 <michagogo> PRab: why not just use pastebinit?
77 2015-01-10 21:41:00 <PRab> I've never heard of it.
78 2015-01-10 21:41:09 <michagogo> It's a pretty handy script
79 2015-01-10 21:41:22 <michagogo> It can pastebin either a file or stdin
80 2015-01-10 21:41:51 <michagogo> Just `sudo apt-get install pastebinit`
81 2015-01-10 21:42:06 <michagogo> But it's not very hard: `scp user@host:/path/to/file .` should do it, I think
82 2015-01-10 21:42:09 <PRab> http://paste.debian.net/140086/
83 2015-01-10 21:42:37 <michagogo> PRab: Ah, I see the issue
84 2015-01-10 21:42:49 <michagogo> The container doesn't have Internet access
85 2015-01-10 21:42:59 <michagogo> I think this was a known issue in some cases
86 2015-01-10 21:43:10 <PRab> Ok
87 2015-01-10 21:43:18 <michagogo> PRab: if you look in release-process, I think there's a command there to prefetch the file
88 2015-01-10 21:43:21 <michagogo> Files*
89 2015-01-10 21:43:40 <PRab> Ah, I thought that didn't apply to me because I was going to keep my machine online.
90 2015-01-10 21:43:43 <michagogo> Something like `make -c depends download`
91 2015-01-10 21:43:56 <michagogo> But with more arguments or something
92 2015-01-10 21:44:16 <michagogo> PRab: I'm not 100% sure, I think maybe it should work, but doesn't for some reason
93 2015-01-10 21:44:24 <michagogo> Btw, what's your real host?
94 2015-01-10 21:44:31 <PRab> Windows
95 2015-01-10 21:44:35 <michagogo> Ah
96 2015-01-10 21:44:40 <michagogo> Wait, scp?
97 2015-01-10 21:44:54 <michagogo> Cygwin or something?
98 2015-01-10 21:44:56 <PRab> scp from inside of putty.
99 2015-01-10 21:45:08 <michagogo> Inside putty? Hm?
100 2015-01-10 21:45:31 <PRab> I was just guessing that it could push a file back, but hadn't actually done it before.
101 2015-01-10 21:45:34 <michagogo> Wait, what were you scping from where to where?
102 2015-01-10 21:45:54 <michagogo> ACTION is confused
103 2015-01-10 21:46:23 <PRab> I had ssh'd into the debian VM and wanted to run scp to push the file back out onto the host. I don't know if thats actually possible.
104 2015-01-10 21:47:08 <michagogo> PRab: you can copy a file to a remote host with scp, yes, but I don't think that works with windows
105 2015-01-10 21:47:28 <michagogo> Unless you have sshd running on the host
106 2015-01-10 21:47:48 <michagogo> You probably want to use WinSCP to get files out
107 2015-01-10 21:47:58 <PRab> Ah, I was hoping it would be smart enough to just use the already established ssh (via putty) link.
108 2015-01-10 21:48:21 <michagogo> PRab: putty doesn't do files, I don't think
109 2015-01-10 21:48:56 <michagogo> Maybe that would work sshing linux to linux, but I don't actually know
110 2015-01-10 21:49:14 <PRab> Ok, I'm a bit of a linux noob.
111 2015-01-10 21:49:22 <PRab> I know enough to be dangerous.
112 2015-01-10 21:49:26 <uiop> michagogo: putty has an scp-equivalent iirc
113 2015-01-10 21:49:37 <uiop> (command line though)
114 2015-01-10 21:49:41 <uiop> (iirc)
115 2015-01-10 21:50:13 <PRab> I like the little pastebinit script, that worked great.
116 2015-01-10 21:51:11 <PRab> do I run "make -C ../bitcoin/depends download SOURCES_PATH=`pwd`/cache/common" from inside ./gitian-builder?
117 2015-01-10 21:53:14 <PRab> I just went for it. It looks like its working.
118 2015-01-10 21:56:13 <elichai2> petertodd, here?
119 2015-01-10 21:57:10 <elichai2> that's interesting, in petertodd's `replace-by-fee-tools` he sends the dust coins to 1C7zdTfnkzmr13HfA2vNm5SJYRK6nEKyq8
120 2015-01-10 21:57:27 <elichai2> and this address had total of 21.8 BTC!!
121 2015-01-10 21:57:45 <elichai2> and it looks like most of it was dust money
122 2015-01-10 21:58:21 <phantomcircuit> he's getting rich of course
123 2015-01-10 21:59:11 <phantomcircuit> elichai2, (the private key for that address is well known)
124 2015-01-10 21:59:20 <elichai2> phantomcircuit, really?
125 2015-01-10 21:59:24 <elichai2> where?
126 2015-01-10 21:59:44 <PRab> elichai2: Um, https://blockexplorer.com/tx/f4aa1b6ad9aad17e436c1efd278bd0d52198b3bc2dbeacc953c169afb3b84786#o0 appears to be most of that.
127 2015-01-10 22:00:10 <phantomcircuit> elichai2, battery horse staple
128 2015-01-10 22:00:13 <phantomcircuit> compressed key
129 2015-01-10 22:00:18 <elichai2> ohh, ok :)
130 2015-01-10 22:00:36 <elichai2> so i guess there are a lot of bots that stealing all the dust money :)
131 2015-01-10 22:04:06 <PRab> Smack me if this is too offtopic, but I wish that there was a more standardized mirrored download protocol. It looks like the make command that I'm running picked a slow mirror (:http://anychimirror101.mirrors.tds.net) to download bdb from.
132 2015-01-10 22:05:52 <PRab> Something like download://filename=NAME&hash=HASH. I guess what I am basically looking for is to replace mirrors with torrent magnet links backed by DHT.
133 2015-01-10 22:11:19 <elichai2> PRab, maybe you can help me understand this tx? https://blockexplorer.com/tx/78e143fa57e8e75f66e92a06d0ea84712fccf276747e7f16aad4777057c1d2b6
134 2015-01-10 22:11:33 <Luke-Jr> PRab: yes, that sounds fairly off-topic
135 2015-01-10 22:11:33 <phantomcircuit> PRab, would be nice if bittorrent became the standard
136 2015-01-10 22:11:49 <Luke-Jr> does wget even support magnet?
137 2015-01-10 22:12:40 <phantomcircuit> Luke-Jr, no
138 2015-01-10 22:13:54 <gmaxwell> torrent DHT is basically unworkably unreliable with small swarms in any case; as we found with the bootstrap torrent. needs a tracker (or N) to not spin uselessly for hours.
139 2015-01-10 22:15:18 <Luke-Jr> it's a shame the whole torrent system is built with each torrent essentially isolated
140 2015-01-10 22:16:18 <Luke-Jr> someone was recently talking about extending it with Bitcoin, but I think it could be remade much nicer than merely extending, with bitcoinâ¦
141 2015-01-10 22:17:39 <Luke-Jr> (it would be interesting to see how such an alternative played out, since it would make copyright infringement use harder - all of a sudden that would be a criminal act most everywhere since it's commercial)
142 2015-01-10 22:20:53 <uiop> Luke-Jr: if it made copyright infringement difficult, no one would use it
143 2015-01-10 22:21:25 <Luke-Jr> uiop: not everyone infringes on copyright
144 2015-01-10 22:22:16 <PRab> More on topic, that make command appears to have solved my issue. It is now chugging away building.
145 2015-01-10 22:22:35 <uiop> Luke-Jr: sure, but probably like over 80% of torent traffic is
146 2015-01-10 22:22:49 <PRab> Does the bitcoin build benefit from multiple cores?
147 2015-01-10 22:23:03 <Luke-Jr> PRab: if you use -j
148 2015-01-10 22:23:10 <uiop> Luke-Jr: hence, your new awesome protocol would be used by like 7 guys that're downloading linux distrib isos :)
149 2015-01-10 22:23:15 <Luke-Jr> ⦠note you should have a number after -j or you will crash your system
150 2015-01-10 22:23:35 <Luke-Jr> uiop: dunno..
151 2015-01-10 22:23:56 <PRab> ... and people wonder why I'm not a huge fan of linux.
152 2015-01-10 22:24:23 <PRab> Oh well, after this build I'll add another core or 2 to the VM and try it again with -j #.
153 2015-01-10 22:26:15 <Luke-Jr> PRab: as if Windows is better ;P
154 2015-01-10 22:26:48 <PRab> I'm just glad that I finally got off my butt and built bitcoin from scratch.
155 2015-01-10 22:26:53 <sipa> good!
156 2015-01-10 22:26:58 <PRab> I should have done this a year or 2 ago.
157 2015-01-10 22:27:18 <uiop> PRab: if having multiple cores had single most useful application, it'd be make -jN
158 2015-01-10 22:27:34 <PRab> oh, no space?
159 2015-01-10 22:28:15 <PRab> and can I pass -jN to gbuild?
160 2015-01-10 22:28:22 <sipa> yes
161 2015-01-10 22:28:57 <uiop> most command line progs deal with no space before the argument to single-dash-args
162 2015-01-10 22:29:48 <uiop> (risking pedantry since you said you're not that familiar with unix)
163 2015-01-10 22:30:09 <michagogo> ACTION finds that confusing, when combined with the fact that single-dash switches can be combined
164 2015-01-10 22:30:52 <uiop> michagogo: the sane thing to do there is have the arg apply to last char
165 2015-01-10 22:30:57 <PRab> I'm familiar with using the command line in windows, so most of that knowledge transfers over. Some of the conventions are just a bit different (especially when it comes to arguments)
166 2015-01-10 22:32:10 <PRab> Just googled it. looks like --jobs is the long version so I'll probably use that. I don't mind typing out more if it makes it easier to read/understand.
167 2015-01-10 22:32:26 <michagogo> PRab: Erm, about that
168 2015-01-10 22:32:33 <michagogo> Maybe that works for make
169 2015-01-10 22:32:43 <michagogo> I don't think that works for gbuild
170 2015-01-10 22:32:53 <PRab> good to know...
171 2015-01-10 22:32:59 <uiop> michagogo: oh yeah, in case arg is non-numeric need a space (unless you were super leet and disambiguated in the case of arg beginning with alpha char that isn't a valid single-char switch ;)
172 2015-01-10 22:33:00 <gribble> devrandom/gitian-builder · GitHub: <https://github.com/devrandom/gitian-builder>; gitian-builder/gbuild at master · devrandom/gitian-builder · GitHub: <https://github.com/devrandom/gitian-builder/blob/master/bin/gbuild>; bitcoin/gitian-building.md at master · bitcoin/bitcoin · GitHub: <https://github.com/bitcoin/bitcoin/blob/master/doc/gitian-building.md>
173 2015-01-10 22:33:00 <michagogo> ;;Google gitian-builder gbuild github
174 2015-01-10 22:34:30 <michagogo> PRab: looks like either -j or --num-make
175 2015-01-10 22:35:21 <michagogo> (For gbuild)
176 2015-01-10 22:35:26 <PRab> Thanks!
177 2015-01-10 22:38:09 <uiop> PRab: i'd use -j since it's most portable
178 2015-01-10 22:38:32 <uiop> apparently
179 2015-01-10 22:38:54 <PRab> I guess I'll do that for this, but how do you avoid making alphabet soup with all these commands.
180 2015-01-10 22:39:50 <PRab> I can never remember what "ps aux" stands for, but I know I need to run it to check what processes are running on linux.
181 2015-01-10 22:40:12 <sipa> same
182 2015-01-10 22:40:35 <sipa> i also don't remember what the netstat -ltnp flags stand for, but they list what applications are listening on which tcp ports
183 2015-01-10 22:40:59 <uiop> PRab: oh yeah, ps args are the irregular verbs of unix arg language
184 2015-01-10 22:41:30 <uiop> (iirc ps uses bsd (historical?) argument conventions)
185 2015-01-10 22:42:04 <PRab> sipa: I should try to memorize that one, I have rebooted a linux box way too many times because I couldn't figure out what was holding a port.
186 2015-01-10 22:42:42 <uiop> oh, i randomly saw this amazing netstat arg combo somewhere random
187 2015-01-10 22:42:45 <uiop> netstat -tuna
188 2015-01-10 22:42:48 <uiop> SO LEET
189 2015-01-10 22:45:12 <PRab> Is it normal for a bitcoin build to take 20+ minutes?
190 2015-01-10 22:46:19 <gmaxwell> first time on a system that isn't super fast? sure. if you have ccache it's faster on later builds.
191 2015-01-10 22:47:00 <PRab> how do I check if I have ccache on?
192 2015-01-10 22:47:06 <uiop> PRab: you'd love compiling gcc
193 2015-01-10 22:47:32 <PRab> I think I'm going to run back to my C#/Java world.
194 2015-01-10 22:48:23 <gmaxwell> PRab: type ccache<enter> to see if you have it, the configure script will use it if you do.
195 2015-01-10 22:49:01 <elichai2> hey, what is the default DNS that the core asks for peers?
196 2015-01-10 22:49:37 <PRab> gmaxwell: Darn, looks like I don't.
197 2015-01-10 22:50:00 <uiop> PRab: you can always install it
198 2015-01-10 22:50:06 <uiop> sudo yum install -y ccache
199 2015-01-10 22:50:14 <gmaxwell> elichai2: if it doesn't know any, it queries
200 2015-01-10 22:50:16 <gmaxwell> vSeeds.push_back(CDNSSeedData("bitcoin.sipa.be", "seed.bitcoin.sipa.be"));
201 2015-01-10 22:50:19 <gmaxwell> vSeeds.push_back(CDNSSeedData("bluematt.me", "dnsseed.bluematt.me"));
202 2015-01-10 22:50:20 <PRab> uiop: just did (sudo apt-get install ccache)
203 2015-01-10 22:50:22 <gmaxwell> vSeeds.push_back(CDNSSeedData("dashjr.org", "dnsseed.bitcoin.dashjr.org"));
204 2015-01-10 22:50:25 <gmaxwell> vSeeds.push_back(CDNSSeedData("bitcoinstats.com", "seed.bitcoinstats.com"));
205 2015-01-10 22:50:28 <gmaxwell> vSeeds.push_back(CDNSSeedData("xf2.org", "bitseed.xf2.org"));
206 2015-01-10 22:50:29 <uiop> i haven't used debian for so long the apt command escapes me
207 2015-01-10 22:50:42 <uiop> PRab: nice
208 2015-01-10 22:50:58 <gmaxwell> elichai2: why?
209 2015-01-10 22:51:20 <elichai2> gmaxwell, i need some IP's of the bigs Bitcoin Nodes
210 2015-01-10 22:51:33 <elichai2> to pull some data and compare
211 2015-01-10 22:51:44 <Luke-Jr> there are no big nodes. except luck and abuse
212 2015-01-10 22:52:18 <elichai2> it doesn't really matter, i want to check for double spends in mempools
213 2015-01-10 22:52:29 <gmaxwell> elichai2: there is nothing big, those dns results return random nodes. Often the peers returned are not even responding immediately.
214 2015-01-10 22:52:36 <gmaxwell> okay.
215 2015-01-10 22:52:43 <elichai2> (so i'll want big nodes, and maybe some pool nodes)
216 2015-01-10 22:53:02 <gmaxwell> elichai2: there are (except for some misguided misconfiguration) no "big" nodes.
217 2015-01-10 22:53:14 <elichai2> hmmm
218 2015-01-10 22:53:25 <elichai2> and can i "talk" with the pool nodes?
219 2015-01-10 22:53:31 <elichai2> *with some pool nodes
220 2015-01-10 22:53:40 <gmaxwell> among all nodes that accept connections they should be more or less equal. (there are some misguided folks who crank their max connection counts up, they usually don't end up with many more inbound than anyone else though)
221 2015-01-10 22:54:11 <PRab> elichai2: You would have to contact the pool owner to find which IPs they are using.
222 2015-01-10 22:54:14 <gmaxwell> elichai2: the pools' actual nodes themselves aren't reachable (prevents DOS attacks) though pools may run other publically accessible nodes that you can reach.
223 2015-01-10 22:54:31 <PRab> I believe most of them hide that info to prevent DDOS attacks.
224 2015-01-10 22:54:39 <elichai2> gmaxwell, btw, you know sipa dns allows directory listening?
225 2015-01-10 22:54:44 <Luke-Jr> PRab: most don't have public access to their pool node
226 2015-01-10 22:54:52 <elichai2> gmaxwell, that's sounds good
227 2015-01-10 22:54:58 <Luke-Jr> elichai2: ⦠that doesn't make sense.
228 2015-01-10 22:55:18 <gmaxwell> elichai2: what exactly do you mean by 'directory listening' ?
229 2015-01-10 22:56:09 <gmaxwell> elichai2: do you just mean you loaded the url and saw something? if so you're not seeing sipa's host: you're seeing a website on a random bitcoin node.
230 2015-01-10 22:56:18 <elichai2> gmaxwell, i forgot the exact expression, but you can browse all the files on the server
231 2015-01-10 22:56:30 <elichai2> https://seed.bitcoin.sipa.be/
232 2015-01-10 22:56:35 <Luke-Jr> it's not a server, it's a DNS name
233 2015-01-10 22:56:36 <elichai2> it's not sipa's?
234 2015-01-10 22:56:46 <phantomcircuit> nope
235 2015-01-10 22:56:52 <phantomcircuit> that's just some random bitcoin nodes ip
236 2015-01-10 22:56:57 <elichai2> Luke-Jr, yeah, but this DNS point to a server
237 2015-01-10 22:57:00 <elichai2> lol
238 2015-01-10 22:57:08 <Luke-Jr> elichai2: no, it points to a random set of bitcoin nodes
239 2015-01-10 22:57:20 <elichai2> ok, but how something like this can happen?
240 2015-01-10 22:57:31 <elichai2> that through this i can browse this guy's files
241 2015-01-10 22:57:32 <PRab> elichai2: Every time you resolve that DNS name you will get a new ipaddress.
242 2015-01-10 22:57:44 <michagogo> PRab: installing ccache on the debian vm is irrelevant here
243 2015-01-10 22:57:49 <Luke-Jr> elichai2: his node happens to have a webserver too
244 2015-01-10 22:58:17 <PRab> michagogo: Oh, it has to be in the Ubuntu container?
245 2015-01-10 22:58:25 <michagogo> The first gbuild for each platform will take a long time, yeah, but that's because it builds the deps
246 2015-01-10 22:58:43 <sipa> PRab: gitian does not use ccache
247 2015-01-10 22:58:47 <michagogo> PRab: no, ignore the ccache advice (for gbuilding)
248 2015-01-10 22:58:55 <sipa> intentionally
249 2015-01-10 22:59:17 <michagogo> I assume gmaxwell didn't realize you were talking about a gbuild
250 2015-01-10 22:59:49 <PRab> oh, I think I understand. So if I'm going to build bitcoin regularly, I should use make directly instead of gitian?
251 2015-01-10 23:00:07 <sipa> yes
252 2015-01-10 23:00:38 <PRab> Got it, I thought I was supposed to always use gitian.
253 2015-01-10 23:00:49 <Luke-Jr> only for deterministic output
254 2015-01-10 23:00:58 <Luke-Jr> or if you have broken system OpenSSL..
255 2015-01-10 23:01:03 <sipa> gitian is for release builds, where reproducability is important
256 2015-01-10 23:03:02 <PRab> I knew gitian was reproducible, but I didn't realize that gitian is slower/harder than regular builds.
257 2015-01-10 23:03:21 <PRab> I figured I would just use 1 build system all the time.
258 2015-01-10 23:03:25 <michagogo> PRab: well, it depends
259 2015-01-10 23:03:41 <michagogo> What are you using the builds for?
260 2015-01-10 23:04:26 <Luke-Jr> PRab: gitian does cache dependencies now, so it's only slow if you are doing development
261 2015-01-10 23:04:32 <michagogo> If it's just personal stuff you can use the depends system on your system
262 2015-01-10 23:04:40 <PRab> Honestly, I'm mainly building to prove that I can. I'll probably install it for a day or 2 then go back to using the official binary.
263 2015-01-10 23:04:46 <sipa> PRab: you can also use the depends system, which allows to build pretty much like for releases, only without the determinism
264 2015-01-10 23:05:02 <michagogo> PRab: well, gbuilding gives you the official binary exactly
265 2015-01-10 23:05:08 <michagogo> Unless something's wrong
266 2015-01-10 23:05:14 <sipa> michagogo: when applies to release tags
267 2015-01-10 23:05:20 <michagogo> sipa: right
268 2015-01-10 23:05:32 <PRab> If I get really motivated I might try to actually do some development.
269 2015-01-10 23:05:52 <michagogo> (Minus authenticode/gatekeeper, ofc)
270 2015-01-10 23:06:14 <PRab> (I'm a C#/java developer, so C/C++ is not my strongest language)
271 2015-01-10 23:06:17 <michagogo> PRab: if you have gitian set up, your sigs would be appreciated
272 2015-01-10 23:06:25 <PRab> That I can do.
273 2015-01-10 23:06:26 <Luke-Jr> PRab: even if you don't do development, ⦠what michagogo said ^
274 2015-01-10 23:06:36 <michagogo> (Follow release-process.md)
275 2015-01-10 23:06:51 <PRab> I'm about an hour into the build, so I might as well sign it.
276 2015-01-10 23:07:21 <michagogo> While it runs, you can go fork and clone gitian.sigs
277 2015-01-10 23:07:36 <michagogo> Also, future gbuild's will take much less time
278 2015-01-10 23:07:38 <PRab> I'm building 0.10.0rc1.
279 2015-01-10 23:07:53 <michagogo> The vast majority of build time is Qt, if I'm not mistaken
280 2015-01-10 23:08:29 <michagogo> PRab: right -- so e.g. v0.10.0rc2 should only take a few minutes
281 2015-01-10 23:08:45 <michagogo> Since the dependencies (other than openssl) will be cached
282 2015-01-10 23:08:56 <PRab> Good to know. I thought because of no ccache it would take the full time again.
283 2015-01-10 23:09:04 <michagogo> PRab: nope
284 2015-01-10 23:09:25 <Luke-Jr> PRab: ccache is file-level caching - gitian still caches entire builds though
285 2015-01-10 23:09:40 <michagogo> Gitian, combined with our depends system, caches the dependency builds
286 2015-01-10 23:10:04 <michagogo> So you'll still be rebuilding Bitcoin Core from scratch
287 2015-01-10 23:10:06 <sipa> PRab: gitian will always build bitcoin completely from scratch, intentionally - but it will cache dependencies (openssl, qt, ...)
288 2015-01-10 23:10:17 <PRab> Oh, I should learn more about gitian. I thought it just did stuff like set file timestamps and seeded random number generators.
289 2015-01-10 23:10:20 <michagogo> But you'll already have Qt built and cacher
290 2015-01-10 23:10:25 <michagogo> cached
291 2015-01-10 23:10:34 <PRab> Got it.
292 2015-01-10 23:10:34 <uiop> make is dreamy with good make files
293 2015-01-10 23:10:46 <michagogo> PRab: well, basically there's a new gitian feature
294 2015-01-10 23:11:05 <michagogo> It lets descriptors save files to a cache folder
295 2015-01-10 23:11:17 <michagogo> Gitian then copies that cache out and saves it
296 2015-01-10 23:11:36 <michagogo> And when you next build, it restores that cache back into the clean VM
297 2015-01-10 23:12:38 <michagogo> So when your build process is written well, and knows how to take advantage of that persistent directory correctly, it can cache things like dependencies
298 2015-01-10 23:13:00 <PRab> So if I'm signing a release, I should wait for them to commit it, right? Reading https://github.com/bitcoin/bitcoin/blob/master/doc/release-process.md implies that I should make my own tag.
299 2015-01-10 23:13:14 <michagogo> PRab: no, don't make your own tag
300 2015-01-10 23:13:22 <michagogo> Start from the building part
301 2015-01-10 23:13:44 <michagogo> (Also, follow v0.10.0rc1's r-p.md)
302 2015-01-10 23:14:32 <PRab> Also, for "export VERSION=(new version, e.g. 0.8.0)", am I supposed to include the "v"or not?
303 2015-01-10 23:14:39 <sipa> no
304 2015-01-10 23:14:56 <michagogo> Specifically not
305 2015-01-10 23:15:11 <michagogo> Because some places need it without the v
306 2015-01-10 23:15:21 <michagogo> And where the v is required, it's added
307 2015-01-10 23:15:35 <PRab> Ok, I included it in the export, but then fixed up the "./bin/gbuild --commit bitcoin=v${VERSION} ..." command to "./bin/gbuild --commit bitcoin=${VERSION} ..."
308 2015-01-10 23:15:38 <elichai2> what message type should i send in order to get the mempool?
309 2015-01-10 23:15:42 <PRab> Next time, I leave it out.
310 2015-01-10 23:15:47 <sipa> elichai2: rpc or p2p?
311 2015-01-10 23:15:54 <elichai2> sipa, p2p
312 2015-01-10 23:16:00 <michagogo> PRab: no
313 2015-01-10 23:16:05 <sipa> see bip 35
314 2015-01-10 23:16:06 <elichai2> "The mempool message sends a request to a node asking for information about transactions it has verified but which have not yet confirmed. The response to receiving this message is an inv message containing the transaction hashes for all the transactions in the node's mempool."
315 2015-01-10 23:16:08 <michagogo> gbuild does need the v
316 2015-01-10 23:16:19 <michagogo> Oh, I see what you mean
317 2015-01-10 23:16:31 <elichai2> sipa, it's called mempool?
318 2015-01-10 23:16:34 <michagogo> Gbuild needs the v because it's referring to the tag
319 2015-01-10 23:16:37 <sipa> elichai2: that's what it says
320 2015-01-10 23:16:49 <michagogo> But then e.g. gsign needs to not have the v
321 2015-01-10 23:17:09 <elichai2> sipa, yeah buy, `sends a request to a node asking for information about transactions` what command do that?
322 2015-01-10 23:17:19 <Luke-Jr> "mempool"
323 2015-01-10 23:17:23 <elichai2> :\
324 2015-01-10 23:17:27 <sipa> elichai2: dude, please read BIP 35
325 2015-01-10 23:17:30 <sipa> it's well specified
326 2015-01-10 23:17:31 <elichai2> bitcore doesn't have `mempool`
327 2015-01-10 23:17:31 <elichai2> https://github.com/bitpay/bitcore/blob/master/lib/transport/messages.js
328 2015-01-10 23:17:39 <elichai2> sipa, yeah, i'm reading right now
329 2015-01-10 23:17:41 <Luke-Jr> elichai2: then open an issue in bitcore
330 2015-01-10 23:17:41 <PRab> Yeah, I don't think I did too much damage and I know how to undo it if necessary.
331 2015-01-10 23:17:45 <michagogo> elichai2: so complain to thrm
332 2015-01-10 23:17:47 <michagogo> them
333 2015-01-10 23:18:04 <elichai2> Luke-Jr, the wiki just isn't clear enough :)
334 2015-01-10 23:18:11 <Luke-Jr> for some reason, bitcore devs hang out in #bitcore instead of here :/
335 2015-01-10 23:18:14 <sipa> elichai2: wiki?
336 2015-01-10 23:18:26 <elichai2> it sounds like mempool is only an answer message, and not a request
337 2015-01-10 23:18:31 <elichai2> sipa, en.bitcoin.it/wiki/
338 2015-01-10 23:18:42 <sipa> i wouldn't rely on that; it's very outdated
339 2015-01-10 23:19:07 <elichai2> sipa, yeah...
340 2015-01-10 23:19:10 <elichai2> but thanks anyway
341 2015-01-10 23:19:13 <Luke-Jr> sipa: eh? only niches
342 2015-01-10 23:19:16 <sipa> ok
343 2015-01-10 23:20:40 <PRab> If I'm doing gitian sigs do I have to do osx, or can I just do Windows and Linux?
344 2015-01-10 23:20:57 <sipa> do whatever you like :)
345 2015-01-10 23:21:01 <Luke-Jr> PRab: you don't *have* to do any
346 2015-01-10 23:21:49 <PRab> Ok, I'll probably skip osx then because I would have to register for the Apple SDK, and I have no interest in doing osx development.
347 2015-01-10 23:23:45 <Luke-Jr> totally understandable
348 2015-01-10 23:25:35 <elichai2> good night
349 2015-01-10 23:29:37 <midnightmagic> huh? you have to register to get xcode?
350 2015-01-10 23:29:47 <sipa> yes
351 2015-01-10 23:30:23 <midnightmagic> i don't recall registering. huh.
352 2015-01-10 23:39:45 <PRab> woohoo, build (for windows) finished!
353 2015-01-10 23:40:35 <midnightmagic> sweet!
354 2015-01-10 23:41:08 <PRab> and they match https://gist.github.com/laanwj/737988993f2431ba65f2.
355 2015-01-10 23:46:50 <michagogo> midnightmagic: yeah, pretty sure you need to be registered
356 2015-01-10 23:47:09 <michagogo> ACTION whispers: or get a copy from someone else
357 2015-01-10 23:47:49 <michagogo> It's not complicated, it's not like it's signing up to either of the $99/year subscriptions, but it is registration