1 2015-02-08 00:36:47 <popiuiti> Hi
2 2015-02-08 00:37:05 <popiuiti> I sent a transaction while Bitcoin Core was still synching with the network
3 2015-02-08 00:37:21 <popiuiti> now that transaction is unbroadcasted, stuck... and the balance is wrong. How can I fix it?
4 2015-02-08 00:39:50 <sturles> It will be broadcast again. To push it, you could do a bitcoind getrawtransaction txid and then push it e.g. here: https://blockchain.info/pushtx
5 2015-02-08 02:15:35 <jtimon> some people may hate me for this: #5768, hopefully others will like it...
6 2015-02-08 02:22:26 <phantomcircuit> jtimon, files changed 86
7 2015-02-08 02:22:28 <phantomcircuit> why
8 2015-02-08 02:22:33 <phantomcircuit> y u do dis
9 2015-02-08 02:25:48 <jtimon> it needs to be done, it says WIP, I don't intent it to be merged as a whole
10 2015-02-08 02:27:58 <jtimon> we can start with #5681 #5696 #5747 or #5709, for example
11 2015-02-08 02:28:32 <jtimon> I just figured ot that some of those PRs may not make that much sense individually with some more context
12 2015-02-08 02:28:44 <jtimon> so there's the context
13 2015-02-08 02:30:40 <jtimon> some of the commits (ike "LOCAL: notes on include main.h") will never make it into master
14 2015-02-08 02:34:29 <phantomcircuit> jtimon, fair enough
15 2015-02-08 05:36:30 <Luke-Jr> re GBT issue: clients do send "maxversion":2, so we could detect that and give version 2 as long as it's valid..
16 2015-02-08 05:36:42 <Luke-Jr> wumpus: ^ worth implementing for 0.10?
17 2015-02-08 05:36:56 <Luke-Jr> might make more sense to just tell miners they must update
18 2015-02-08 05:37:01 <Luke-Jr> at the same time as Core
19 2015-02-08 05:42:50 <Luke-Jr> http://codepad.org/bX4MdOLr look reasonable for block v3 support?
20 2015-02-08 07:46:35 <fanquake> ;;blocks
21 2015-02-08 07:46:36 <gribble> 342525
22 2015-02-08 09:14:45 <wumpus> Luke-Jr: I really don't want to do another rc for 0.10
23 2015-02-08 09:15:12 <Luke-Jr> wumpus: ok, I don't think it's a big deal if we just tell solo miners "you must upgrade Core and BFGMiner at the same time"
24 2015-02-08 09:15:32 <Luke-Jr> would be nice to throw it in release notes, do those need a rc?
25 2015-02-08 09:16:44 <wumpus> lol no, release notes can be changed any time
26 2015-02-08 09:16:57 <Luke-Jr> ok, I'll go for that then
27 2015-02-08 09:17:33 <wumpus> that's why I was initially histitant to maintain the preliminary release notes in git, it's really something written in parallel. Then again, there's advantages too to doing it this way (e.g. easy to find)
28 2015-02-08 09:25:20 <Luke-Jr> wumpus: https://github.com/bitcoin/bitcoin/pull/5771/files
29 2015-02-08 09:27:03 <Luke-Jr> hm
30 2015-02-08 09:27:14 <Luke-Jr> I could add version numbers with v3 support, if that's a good idea?
31 2015-02-08 09:44:28 <wumpus> how do you mean that?
32 2015-02-08 09:44:43 <Luke-Jr> I updated the PR to reference libblkmaker 0.4.2 and 0.5.1 that support v3
33 2015-02-08 09:44:54 <wumpus> ok
34 2015-02-08 09:44:58 <Luke-Jr> will add*
35 2015-02-08 09:45:06 <wumpus> looks good to me
36 2015-02-08 14:05:39 <waxwing> trying to build 0.10rc4, compile failed at: #include "paymentrequest.pb.h" <- no such file
37 2015-02-08 14:05:43 <waxwing> ring any bells?
38 2015-02-08 14:08:37 <fanquake> waxwing Do you have protobuf installed?
39 2015-02-08 14:08:49 <waxwing> prob not
40 2015-02-08 14:08:52 <waxwing> how to check?
41 2015-02-08 14:09:02 <fanquake> Which os?
42 2015-02-08 14:09:09 <waxwing> ubuntu 14.10
43 2015-02-08 14:10:08 <fanquake> apt-get install libprotobuf-dev protobuf-compiler
44 2015-02-08 14:10:55 <waxwing> k, got it, thanks
45 2015-02-08 14:11:20 <waxwing> didn't fix error tho :(
46 2015-02-08 14:11:43 <fanquake> re run autogen and configure
47 2015-02-08 14:11:49 <waxwing> oh right thanks
48 2015-02-08 20:39:29 <L_Cranston_Shado> anyone know what else, if anything, uses HelpMessageDialog besides the about window(s)? Iâve done a search but other than UI files, the only usage I can find is in utilitydialog.h with the about window
49 2015-02-08 20:41:16 <L_Cranston_Shado> to give context, Iâm trying a change to remove restorewindowgeometry for the about dialog, but I donât want to accidentally hit anything else thatâs using the same code
50 2015-02-08 21:21:41 <L_Cranston_Shado> jonasschnelli, I believe the debug window calls that
51 2015-02-08 21:51:47 <sqed> the JSON response to `signrawtransaction` includes a `complete` field which the docs ( https://en.bitcoin.it/wiki/Raw_Transactions ) say should be either 1 or 0, but I'm getting true/false
52 2015-02-08 21:52:33 <sqed> as in `{"complete": true}` instead of `{"complete": 1}`. I'm using bitcoin-qt 0.9.3
53 2015-02-08 22:14:43 <justanotheruser> why was op_xor disabled? Surely it doesn't present a DoS vulnerability?
54 2015-02-08 22:15:38 <sipa> ask satoshi :)
55 2015-02-08 22:16:02 <gribble> The operation succeeded.
56 2015-02-08 22:16:02 <justanotheruser> ;;later tell satoshi why was op_xor disabled? Surely it doesn't present a DoS vulnerability?
57 2015-02-08 22:16:38 <sipa> haha
58 2015-02-08 22:19:56 <phiz_> I imported some addresses into bitcoind with `bitcoin-cli importprivkey $PRIV "" false`
59 2015-02-08 22:20:09 <phiz_> And restarted bitcoin with `bitcoind -rescan`
60 2015-02-08 22:20:13 <phiz_> I get the log message: http://privatepaste.com/ad2087e964
61 2015-02-08 22:20:26 <phiz_> However, it's not importing the addresses
62 2015-02-08 22:20:44 <phiz_> I can verify bitcoind has imported the correct addresses (with `bitcoind dumpprive $ADDR`)
63 2015-02-08 22:21:01 <sipa> what does validateaddress $ADDR say?
64 2015-02-08 22:22:11 <phiz_> sipa: http://privatepaste.com/b604c94881
65 2015-02-08 22:22:45 <sipa> which bitcoin version is this?
66 2015-02-08 22:23:31 <phiz_> sipa: http://privatepaste.com/971c51379d
67 2015-02-08 22:23:53 <sipa> so 0.9.4
68 2015-02-08 22:24:27 <sipa> by 'it's not importing the addresss', you mean it's not finding transactions to that address?
69 2015-02-08 22:24:33 <phiz_> sipa: That's correct
70 2015-02-08 22:24:41 <sipa> are those transactions confirmed?
71 2015-02-08 22:24:46 <phiz_> It doesn't have any unspent, but that address does
72 2015-02-08 22:25:01 <phiz_> As can be seen here: https://blockchain.info/unspent?active=1DgbvhKW1VgT4HcBExT71qosUkpNXYJkcN&format=html
73 2015-02-08 22:25:35 <sipa> this sounds like a bug
74 2015-02-08 22:26:33 <phiz_> sipa: In my logs it shows the rescan takes only 15 seconds
75 2015-02-08 22:26:34 <phiz_> Is that normal?
76 2015-02-08 22:26:52 <phiz_> (From: http://privatepaste.com/ad2087e964 )
77 2015-02-08 22:26:57 <sipa> no
78 2015-02-08 22:27:29 <phiz_> I killed bitcoind, and reran `bitcoind -rescan` and it also takes around 15 seconds
79 2015-02-08 22:27:53 <phiz_> It's running on a computer with a rather slow hard-drive too
80 2015-02-08 22:29:06 <phiz_> sipa: If it's of any help, I've previously used this bitcoind node with `bitcoind importwallet $file`
81 2015-02-08 22:29:16 <phiz_> Perhaps it's only rescanning from that time?
82 2015-02-08 22:31:57 <sqed> the docs say that the `complete` field in a response to `signrawtransaction` is either 1 or 0, but I get `true` or `false` in the JSON response
83 2015-02-08 22:32:12 <sqed> under "signrawtransaction" here: https://en.bitcoin.it/wiki/Raw_Transactions
84 2015-02-08 22:32:22 <sipa> sqed: seems the documentation is wrong
85 2015-02-08 22:32:31 <sipa> true or false do seem more natural imho
86 2015-02-08 22:33:39 <sqed> yeah; i was using a library that expected 1 or 0 and had a comment about how it's bizarre that bitcoind returns 1 or 0, but I couldn't find a commit in the bitcoind history where the value of `complete` was not a bool
87 2015-02-08 22:34:02 <sipa> feel free to submit a pull request to fix the docs
88 2015-02-08 22:35:30 <gmaxwell> justanotheruser: I'd take a bet that it actually was vulnerable, in fact.
89 2015-02-08 22:36:10 <phiz_> sipa: If I had to guess, this looks like the problem code: https://github.com/bitcoin/bitcoin/blob/31dedb463b0ce77364e516239bf1b9c7eed5b3b0/src/wallet.cpp#L944-L947
90 2015-02-08 22:36:11 <gmaxwell> (not DOS but a crash bug; IIRC at least some of the operations which widened inputs were incorrect)
91 2015-02-08 22:36:33 <phiz_> That is the function that's called on wallet startup, when you pass in -rescan
92 2015-02-08 22:36:43 <phiz_> So my guess, is that it's skipping the blocks
93 2015-02-08 22:37:02 <justanotheruser> why would xor widen inputs?
94 2015-02-08 22:37:06 <sqed> sipa: pull request to the wiki (en.bitcoin.it)?
95 2015-02-08 22:37:23 <phiz_> sqed: if it's the wiki, you can just edit it
96 2015-02-08 22:37:25 <gmaxwell> justanotheruser: it must widen the smallest input to the widest one.
97 2015-02-08 22:37:44 <gmaxwell> sqed: the text in the wiki is copying the documentation from the software.
98 2015-02-08 22:37:53 <sipa> sqed: i thought you were talking sabout the help signrawtransaction output
99 2015-02-08 22:38:03 <justanotheruser> too bad he didn't just force the inputs to be the same size :/
100 2015-02-08 22:38:10 <gmaxwell> (unless someone went an seperately broke the wiki)
101 2015-02-08 22:38:12 <sipa> phiz_: looks like it, nice catch
102 2015-02-08 22:38:23 <sipa> phiz_: so the question is why doesn't importprivkkey reset the birthfday
103 2015-02-08 22:38:30 <sipa> *birthday
104 2015-02-08 22:38:37 <phiz_> Let me look at the code
105 2015-02-08 22:39:00 <sipa> at leastt it should recalculate the birthday
106 2015-02-08 22:39:13 <phiz_> How would it be able to recalculate the birthday
107 2015-02-08 22:39:25 <phiz_> If i do importprivkey, it should assume the birthday is block 0, right?
108 2015-02-08 22:39:42 <sipa> there is code to do so already[B[B[Byup
109 2015-02-08 22:40:14 <sqed> sipa: the code docs also claim that `complete` is a numeric and not bool field: https://github.com/bitcoin/bitcoin/blob/master/src/rpcrawtransaction.cpp#L533
110 2015-02-08 22:40:45 <sqed> but https://github.com/bitcoin/bitcoin/blob/master/src/rpcrawtransaction.cpp#L710 has `complete` as a bool, unless something is converting it later on
111 2015-02-08 22:41:05 <sipa> sqed: apparently not, as you're seeing a bool in the output
112 2015-02-08 22:41:07 <sipa> go fix it :)
113 2015-02-08 22:41:10 <sipa> (the docs)
114 2015-02-08 22:42:05 <sqed> sipa: cool, i'll make a PR
115 2015-02-08 22:42:42 <sqed> sipa: i just thought this was a weird inconsistency and was wondering if the int -> bool switch (if there was one) affected other methods as well
116 2015-02-08 22:43:16 <sipa> from what you've told me (i haven't checked the code), it just seems the docs are wrong
117 2015-02-08 22:43:24 <sipa> for this specific function
118 2015-02-08 22:44:25 <gmaxwell> It's just a documentation error. (as you can see, it always outputs a boolean)
119 2015-02-08 22:44:31 <gmaxwell> Just fix the documentation.
120 2015-02-08 22:45:45 <sqed> done, made a pr
121 2015-02-08 22:47:28 <phiz_> Sipa: So I think I found the issue
122 2015-02-08 22:47:48 <phiz_> When you import an address, it sets the nTimeFirstKey: https://github.com/bitcoin/bitcoin/blob/87ecfb0f9675ef46d360014a65886e9556d28e66/src/rpcdump.cpp#L132-L133
123 2015-02-08 22:47:57 <phiz_> However, it never persists it
124 2015-02-08 22:48:10 <phiz_> So if you don't use the rescan
125 2015-02-08 22:48:19 <phiz_> When you close bitcoind, it'll just lose the value
126 2015-02-08 22:48:56 <phiz_> And when you start bitcoind, it will use the old value: https://github.com/bitcoin/bitcoin/blob/31dedb463b0ce77364e516239bf1b9c7eed5b3b0/src/wallet.cpp#L944-L947
127 2015-02-08 22:49:04 <phiz_> And just continue from there
128 2015-02-08 22:50:04 <phiz_> So I think on this line: https://github.com/bitcoin/bitcoin/blob/87ecfb0f9675ef46d360014a65886e9556d28e66/src/init.cpp#L1213 ... it should set nTimeFirstKey = 1
129 2015-02-08 23:14:03 <phiz_> Can someone review this: https://github.com/bitcoin/bitcoin/pull/5774
130 2015-02-08 23:14:11 <phiz_> I don't know C++, so it's just a best-attempt
131 2015-02-08 23:17:19 <Luke-Jr> phiz_: new pull requests get emailed to many of us automaticlly, no need to request review explicitly (at least not immediately)
132 2015-02-08 23:17:40 <phiz_> Just thought people could point out the obvious stuff, so i could fix it
133 2015-02-08 23:17:43 <phiz_> but yeah, fair enough
134 2015-02-08 23:26:28 <bliljerk101> What does Taras go by on bitcoin-dev?
135 2015-02-08 23:38:45 <Luke-Jr> bliljerk101: I dont think he does? you mean IRC or ML?
136 2015-02-08 23:41:54 <bliljerk101> I mean here. Wanted to chat with her/him first before posting on his/her talk page
137 2015-02-08 23:42:11 <bliljerk101> idk why Taras sounds like a girl name to me, but will assume guy since you said so
138 2015-02-08 23:53:23 <sipa> bliljerk101: i don't think anyone here has a clue who taras is
139 2015-02-08 23:58:40 <bliljerk101> sipa Luke-Jr knows, but thanks. fyi, taras is a person who does a lot of editing on bitcoin.it