1 2012-06-23 00:28:43 <jgarzik> huh: http://pastebin.com/MA4CkKz2
2 2012-06-23 00:28:56 <jgarzik> version=60001, blocks=0. connects, then disconnects.
3 2012-06-23 01:11:25 <jgarzik> interesting
4 2012-06-23 01:12:13 <luke-jr> jgarzik: what UA?
5 2012-06-23 01:12:23 <luke-jr> are we logging that?
6 2012-06-23 01:44:46 <etotheipi_> question: what is the "preferred" way to sign installer files in Windows?
7 2012-06-23 01:45:05 <etotheipi_> I can sign *.deb files using GPG and dpkg-sig... is there something similar for windows?
8 2012-06-23 01:45:37 <luke-jr> etotheipi_: probably have to pay Microsoft soon
9 2012-06-23 01:52:42 <etotheipi_> so the answer is: sign the md5sum of it and distribute that?
10 2012-06-23 01:52:49 <etotheipi_> how do you guys do it?
11 2012-06-23 01:53:08 <etotheipi_> for Bitcoin-Qt
12 2012-06-23 01:54:28 <luke-jr> gitian
13 2012-06-23 02:03:55 <GTRsdk> hey, what's the status of big endian bitcoin?
14 2012-06-23 02:07:24 <luke-jr> GTRsdk: I have a branch. It doesn't work even on little endian at the moment.
15 2012-06-23 02:07:51 <GTRsdk> luke-jr: so you've tried big endian hardware, or an emulated one?
16 2012-06-23 02:08:03 <luke-jr> I have access to a PPC server
17 2012-06-23 02:08:12 <luke-jr> but making it work on LE again is the next step :p
18 2012-06-23 02:09:01 <GTRsdk> Did you get it working on BE?
19 2012-06-23 02:12:57 <GTRsdk> luke-jr: did you get it to work on big endian?
20 2012-06-23 02:13:26 <luke-jr> no
21 2012-06-23 02:13:44 <luke-jr> like I said, fixing LE is next
22 2012-06-23 02:15:50 <jgarzik> boy, that snuck up on me
23 2012-06-23 02:15:57 <jgarzik> we're getting close to blk0002.dat time
24 2012-06-23 03:50:18 <gmaxwell> 06/23/12 05:49:20 ProcessBlock: ACCEPTED
25 2012-06-23 03:50:19 <gmaxwell> 06/23/12 05:49:27 SetBestChain: new best=0000000000000811fe39 height=176553 work=299668530422871187606
26 2012-06-23 03:50:25 <gmaxwell> validation in valgrind slllow.
27 2012-06-23 04:00:32 <luke-jr> gmaxwell: 7 seconds isn't that slow compared to normal <.<
28 2012-06-23 04:02:43 <gmaxwell> I turned off valgrind and it's doing more like 7/second.
29 2012-06-23 04:03:12 <gmaxwell> 06/23/12 06:02:15 SetBestChain: new best=000000000000044567ed height=181242 work=332059775500399487250
30 2012-06-23 04:03:17 <gmaxwell> 06/23/12 06:02:16 SetBestChain: new best=00000000000000cc9702 height=181249 work=332111884782648155244
31 2012-06-23 04:04:52 <luke-jr> oh, old blocks
32 2012-06-23 04:16:04 <gribble> New news from bitcoinrss: xanatos opened issue 1506 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/issues/1506>
33 2012-06-23 08:12:48 <gribble> New news from bitcoinrss: Diapolo opened issue 1507 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/issues/1507>
34 2012-06-23 08:12:50 <nolybab> it's 4AM in Utah
35 2012-06-23 08:12:58 <nolybab> http://www.youtube.com/watch?v=ORYKKNoRcDc
36 2012-06-23 12:33:56 <gribble> New news from bitcoinrss: xanatos opened pull request 1508 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/pull/1508>
37 2012-06-23 13:26:34 <jgarzik> public backbone up to 3 beefy machines now
38 2012-06-23 13:26:53 <jgarzik> all listed in bitseed.xf2.org, and serving -tons- of early blocks
39 2012-06-23 13:29:54 <Cory> Public backbone?
40 2012-06-23 14:51:08 <dinox> Looking for history difficulty data, is there a table somewhere or do I need to query the blockchain myself?
41 2012-06-23 14:51:57 <dinox> I practically want data points for the graphs at sipa.bitcoin.be
42 2012-06-23 14:53:12 <dinox> Hehe, I'm sure you do :)
43 2012-06-23 14:53:31 <sipa> just the difficulty data?
44 2012-06-23 14:53:56 <dinox> yep
45 2012-06-23 14:54:04 <dinox> and block nbr or time
46 2012-06-23 14:54:58 <sipa> http://sipa.bitcoin.be/hist.txt
47 2012-06-23 14:55:32 <Matt_von_Mises> One thing that is not straight forward is the FindAndDelete (https://github.com/bitcoin/bitcoin/blob/master/src/script.h#L482). When removing OP_SEPARATORs it just removes the op codes but for the signatures does it only match when pushing data for the whole signature or any occurrence of a signature anywhere in push operations?
48 2012-06-23 14:57:08 <dinox> thanks sipa! :)
49 2012-06-23 14:57:22 <dinox> Is that updateded or static?
50 2012-06-23 14:57:27 <sipa> static
51 2012-06-23 14:57:39 <sipa> though i could have it live
52 2012-06-23 14:57:58 <Matt_von_Mises> Actually for a push operation is seems to remove a signature if it is found immediately but also remove the signauture if it appears again straight afterwards?
53 2012-06-23 14:58:57 <Matt_von_Mises> while getop(){ while foundSig{ erase siganture } }
54 2012-06-23 15:01:13 <Matt_von_Mises> "pc = begin() + (pc2 - begin());" That makes no sense to me
55 2012-06-23 15:02:41 <sipa> pc2 is a const_iterator, while pc is an interator
56 2012-06-23 15:02:50 <sipa> i suppose you can't just assign one to the other
57 2012-06-23 15:10:45 <Matt_von_Mises> I explained what I meant here: https://bitcointalk.org/index.php?topic=89481.0
58 2012-06-23 15:11:21 <Matt_von_Mises> I'm not too sure on the C++ stuff so I need to check with others.
59 2012-06-23 15:49:53 <gribble> New news from bitcoinrss: luke-jr opened issue 1509 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/issues/1509>
60 2012-06-23 15:57:24 <jgarzik> I wonder if nodes listed via DNS seeds are disproportionately punished in terms of bandwidth?
61 2012-06-23 15:57:57 <Eliel> wouldn't surprise me
62 2012-06-23 15:58:02 <jgarzik> For new nodes joining the network, getting addresses only from DNS seeds, will the peer selection code be -likely- to select a node listed via DNS seed as the peer for I.B.D.?
63 2012-06-23 15:58:56 <jgarzik> for the three new nodes I just setup, they are all listed in bitseed.xf2.org, and they all spend a lot of their time handing out old blocks (< block 140,000)
64 2012-06-23 16:02:08 <luke-jr> jgarzik: some (bluematt?) might consider that abuse of a DNS seed, FYI
65 2012-06-23 16:04:17 <Eliel> luke-jr: abuse? as in, spying on nodes joining the network?
66 2012-06-23 16:04:37 <luke-jr> as in, listing specific nodes in the DNS results for any purpose
67 2012-06-23 16:06:33 <Eliel> o.O I'd think some nodes need to be listed there for new nodes to be able to join the network :P
68 2012-06-23 16:08:04 <luke-jr> Eliel: keyword: specific
69 2012-06-23 16:08:24 <Eliel> how do you list an unspecific node in DNS? makes no sense.
70 2012-06-23 16:28:59 <jgarzik> there has got to be some solution that is better than forcing DNS seed-provided nodes to constant serve old blocks
71 2012-06-23 16:29:31 <jgarzik> maybe when the blk0001.dat -> blk0002.dat switch occurs, we can start distributing blk0001.dat statically somewhere, if enough volunteers were available.
72 2012-06-23 16:31:13 <jgarzik> 55 and 91 connections respectively, at two of the nodes
73 2012-06-23 16:35:42 <jine> "distributing blk0001.dat statically somewhere" - bitlc.net can setup a node for that, no worries.
74 2012-06-23 16:36:17 <jine> or well, not node perhaps, but a url or similar.
75 2012-06-23 16:37:18 <Zarutian> jgarzik: how long until that?
76 2012-06-23 16:37:41 <jgarzik> as soon as the file size hits 2G on a node
77 2012-06-23 16:37:54 <jgarzik> it's different for each node, depending on how much garbage (orphans) they have in there
78 2012-06-23 16:42:03 <Zarutian> jgarzik: lets say that a node has no orphans in blk0001.dat then how big would it be today? Mine is 1.88 GigaByte
79 2012-06-23 16:42:10 <jgarzik> jine: we might wind up with a bunch of volunteers carrying a blk0001.dat.bz2 with a well-known SHA256 hash... that way anyone may contribute, no matter the trust level
80 2012-06-23 16:42:42 <jgarzik> Zarutian: it still depends on the incoming traffic rate too, which varies daily. basically The Point At Which You Reach 2.00 GB
81 2012-06-23 16:43:35 <Zarutian> due to various reasons my bitcoin node isnt constanly running. (I usually fire it up every few days and let it sync to the network)
82 2012-06-23 16:45:32 <Zarutian> so, my guestimate is that it will hit 2.00 sometime later this year or the first half of next
83 2012-06-23 16:46:30 <Zarutian> (hit two gigabytes that is)
84 2012-06-23 16:48:59 <jgarzik> need to double check whether that is 2 billion or 2*1024*1024*1024 etc., too
85 2012-06-23 16:52:01 <Zarutian> jgarzik: for bytes it is always 1024 bytes for a kilobyte, 1024 kilobytes for a megabyte and 1024 megabytes for a gigabyte (and a byte being 8 bits)
86 2012-06-23 16:52:49 <Zarutian> unless you are a harddisk maker trying to bolster your disks capacity
87 2012-06-23 17:01:17 <xorgate> that's such a lame practice
88 2012-06-23 17:09:27 <etotheipi_> for reference, someone in the Armory thread mentioned their blk0001.dat file was: 2,048,006 kB
89 2012-06-23 17:09:51 <etotheipi_> they had a double-stacked file -- I guess the client didn't recognize the original copy of blk0001.dat and started redownloading the chain on top of it
90 2012-06-23 17:11:43 <etotheipi_> that is 0x7d001800 bytes -- I believe sipa said 0x7d000000 bytes is the limit above which the client decides to start a new file
91 2012-06-23 17:13:00 <etotheipi_> P.S. -- bitcoind/-qt devs: what causes the double-stacked file? I've had it happen to myself a few times. is it a known bug?
92 2012-06-23 17:19:56 <galambo> why didnt it create a blk0002.dat?
93 2012-06-23 17:20:31 <etotheipi_> sorry, I should've been clearer: the person did have a blk0002.dat file
94 2012-06-23 17:21:17 <galambo> oh ok i thought you were disclosing a huge bug
95 2012-06-23 17:21:36 <etotheipi_> I was just saying that the blk0001.dat file was that size: thus it appears to switch after 0x7d000000 bytes is exceeded
96 2012-06-23 17:21:52 <etotheipi_> which I think matches with what sipa said
97 2012-06-23 17:22:15 <galambo> i dont think that should be a problem
98 2012-06-23 17:23:02 <galambo> everyones data files will have the same overrun (its impossible to know whoat the size of the last block will be on disk)
99 2012-06-23 17:23:31 <galambo> just as long as its below the 2gb limit satoshi was trying to avoid
100 2012-06-23 17:23:50 <etotheipi_> with a max blocksize of 1 MB, it won't be a problem
101 2012-06-23 17:28:23 <galambo> oh nevermind i am wrong. i didnt know the data file stored the orphans
102 2012-06-23 17:59:26 <jgarzik> Zarutian: what matters is the bitcoin source code :)
103 2012-06-23 17:59:41 <jgarzik> Zarutian: i.e. whether or not the bitcoin source code uses 1000 or 1024 as a multiple
104 2012-06-23 18:00:44 <jgarzik> etotheipi_: what is a double-stacked file?
105 2012-06-23 18:01:51 <jgarzik> etotheipi_: if you mean blk0002.dat, the bitcoin client divides up the block file into 2GB shardds
106 2012-06-23 18:02:17 <jgarzik> satoshi was avoiding 2gb/4gb file size limits
107 2012-06-23 18:03:11 <Matt_von_Mises> Is this the right way to remove signatures in scripts: https://bitcointalk.org/index.php?topic=89481.msg984802#msg984802