1 2014-07-04 01:09:34 <davec_> I'm really sad that NODE_BLOOM pr was shot down.  We're working on implementing bloom filters now, but it's super annoying to have to implement such an exploitable feature to support later protocol version with no way to opt out
  2 2014-07-04 01:14:26 <Luke-Jr> davec_: I don't think it's required.
  3 2014-07-04 01:14:33 <Luke-Jr> could just ignore it
  4 2014-07-04 01:16:16 <davec_> The reject message is 70002 I believe and doesn't >= 70001 imply bloom filters?
  5 2014-07-04 01:25:28 <phantomcircuit> davec_, you simply do not apply the filter
  6 2014-07-04 01:25:35 <phantomcircuit> you simply provide the full block
  7 2014-07-04 01:25:39 <phantomcircuit> which sucks for the client
  8 2014-07-04 01:25:40 <phantomcircuit> but oh well
  9 2014-07-04 01:25:49 <phantomcircuit> (they can detect this and disconnect)
 10 2014-07-04 01:28:22 <dhill> do they detect by getdata filteredblock and not receiving a response?
 11 2014-07-04 01:30:50 <dhill> or should the server just send the full block
 12 2014-07-04 01:31:22 <davec_> we'll probably provide a flag for disabling it
 13 2014-07-04 01:33:06 <davec_> phontomcircuit: thanks for the suggestion - that is a reasonable approach
 14 2014-07-04 01:33:47 <davec_> phantomcircuit: ^
 15 2014-07-04 01:34:14 <phantomcircuit> yeah
 16 2014-07-04 01:34:21 <phantomcircuit> blargh
 17 2014-07-04 01:34:22 <phantomcircuit> 50mbps
 18 2014-07-04 01:34:26 <phantomcircuit> this should be 150
 19 2014-07-04 01:34:30 <phantomcircuit> god damn it comcast
 20 2014-07-04 01:52:10 <Luke-Jr> davec_: nothing is implied, nodes can't be assumed to be trustworthy at all
 21 2014-07-04 03:52:18 <plato> offering 0.05 BTC if someone gives me a complete solution to my win7 build problems of Moneychanger:
 22 2014-07-04 03:52:23 <plato> RBecker: http://bitcoin.stackexchange.com/questions/28293/building-win7-moneychanger-where-to-put-ot-dlls-how-to-fix-was-unexpected
 23 2014-07-04 03:52:41 <lewellyn> plato: i guess we need to move here ;)
 24 2014-07-04 03:52:45 <plato> mhm
 25 2014-07-04 03:53:00 <RBecker> plato: i'm not a dev, i'm not gonna be of much help :)
 26 2014-07-04 03:53:16 <plato> yeah sry your name was still on the scrollback when i pasted it in here
 27 2014-07-04 03:53:25 <RBecker> np
 28 2014-07-04 03:53:27 <lewellyn> and i've never touched that particular software. but i do plenty of qt (though not directly using .pro files on windows...)
 29 2014-07-04 03:54:05 <lewellyn> plato: have you built the code on another platform, in a virtual machine or otherwise? just to rule out the code being broken in general?
 30 2014-07-04 03:54:48 <plato> no, i'm installing Qt on a new linux vm, i'll test it out there
 31 2014-07-04 03:54:58 <plato> https://github.com/Open-Transactions/Moneychanger/tree/develop/project there's the .pro file and sub-projects
 32 2014-07-04 03:56:05 <lewellyn> ok. i'm having dinner but i'll look it over in a few minutes. :)
 33 2014-07-04 03:56:53 <wumpus> this is not the appropriate channel for open-transactions build problems
 34 2014-07-04 03:57:07 <plato> wumpus: i'm paying bitcoin for someone to fix it
 35 2014-07-04 03:57:34 <lewellyn> plato: it honestly looks like it's minorly broken out the gate, from a glance.
 36 2014-07-04 03:57:43 <plato> and #opentransactions hasnt been able to help me for a week
 37 2014-07-04 04:06:51 <shesek> plato, the fact that you're paying with bitcoin doesn't make it any more relevant here
 38 2014-07-04 04:07:11 <shesek> this channel is meant for bitcoin development related discussions
 39 2014-07-04 04:11:31 <lewellyn> shesek: it's also where RBecker sent the discussion :P
 40 2014-07-04 04:12:40 <lewellyn> plato: see privmsg with an alternate channel suggestion.
 41 2014-07-04 05:04:04 <wumpus> cherry-picked the getNumBlocksOfPeers removal to 0.9.2 branch just in case we do a future 0.9.x
 42 2014-07-04 05:16:37 <warren> wumpus: that related to the banning?
 43 2014-07-04 05:16:48 <wumpus> no
 44 2014-07-04 05:17:03 <wumpus> just an UI glitch, nothing important
 45 2014-07-04 05:18:10 <gmaxwell> Just dug through the debug log a little, I believe this is the node/peer in question broadcasting incorrect block height.
 46 2014-07-04 05:18:13 <gmaxwell> 2014-07-04 02:42:44 receive version message: /getaddr.bitnodes.io:0.1/: version 70001, blocks=959328314, us=xx.xxx.xxx.xx:8333, them=0.0.0.0:0, peer=148.251.238.178:18702
 47 2014-07-04 05:18:17 <gmaxwell> ^ reply on reddit.
 48 2014-07-04 05:18:44 <wumpus> dazzlepod.com
 49 2014-07-04 05:18:45 <gmaxwell> uninitilied memory in their crawler? or perhaps it halariously pulls the max of the nodes its been polling?
 50 2014-07-04 05:19:22 <gmaxwell> wumpus: host bitnodes.io = that ip
 51 2014-07-04 05:19:29 <gmaxwell> so its the bitnode.io crawler doing it…
 52 2014-07-04 05:20:04 <gmaxwell> also explains how it manages to trigger it: it's an agressive reconnecter.
 53 2014-07-04 05:20:14 <wumpus> hah, well at least it's no one malicious
 54 2014-07-04 05:20:35 <wumpus> let's contact them
 55 2014-07-04 05:21:49 <gmaxwell> they should probably report 0 always, presuming they never serve blocks.
 56 2014-07-04 05:22:44 <wumpus> right, reporting 0 would be harmless
 57 2014-07-04 05:27:20 <wumpus> does anyone know a contact for bitnodes.io?
 58 2014-07-04 05:28:55 <SomeoneWeird> is anyone at nodeconf?
 59 2014-07-04 05:29:09 <gmaxwell> ayeowch@gmail.com
 60 2014-07-04 05:33:50 <wumpus> ok, mail sent
 61 2014-07-04 06:10:56 <michagogo> hmm
 62 2014-07-04 06:11:11 <michagogo> What's libtool?
 63 2014-07-04 06:11:58 <michagogo> I have a bunch of untracked files, one called libtool, one called src/m4/libtool.m4, and then a few src/m4/lt* files
 64 2014-07-04 06:12:11 <wumpus> need to be added to .gitignore
 65 2014-07-04 06:12:41 <wumpus> eh, at least if they're autogenerated
 66 2014-07-04 06:13:07 <michagogo> I suspect they are -- Ic ertainly didn't create them
 67 2014-07-04 06:13:40 <wumpus> ok going to try with a new, empty repo
 68 2014-07-04 06:14:39 <michagogo> Looks like autogen creates them
 69 2014-07-04 06:15:06 <michagogo> or rather, it creates the src/m4/* files
 70 2014-07-04 06:15:27 <michagogo> configure creates the libtool file in the root
 71 2014-07-04 06:15:42 <wumpus> I see
 72 2014-07-04 06:16:48 <wumpus> voila https://github.com/bitcoin/bitcoin/commit/81efcd76e644d52e2296fd95c594ab49b7b3fca5
 73 2014-07-04 06:17:02 <michagogo> :D
 74 2014-07-04 06:18:29 <lewellyn> yay removing junk :D
 75 2014-07-04 06:18:45 <michagogo> BTW, what's the current status of libsecp256k1?
 76 2014-07-04 06:19:01 <michagogo> I saw some commits mentioning it the other day, but it doesn't seem to be in yet
 77 2014-07-04 06:21:44 <wumpus> the build system part of the support has been merged
 78 2014-07-04 06:21:47 <wumpus> not the library itself yet
 79 2014-07-04 06:22:23 <wumpus> not entirely sure why Cory decided to do it in two parts, but being careful is good
 80 2014-07-04 07:56:55 <michagogo> wumpus: welcome to Reddit!
 81 2014-07-04 08:25:14 <wumpus> michagogo: thanks :)
 82 2014-07-04 11:20:02 <aburan428> i just ran across this error in my bitcoind client... "*** buffer overflow detected ***: ./bitcoind terminated"
 83 2014-07-04 11:21:24 <aburan428> is this a potential security vulnerability and if so who do I contact with the details to replicate this overflow
 84 2014-07-04 11:44:03 <gmaxwell> aburan428: do you know how you triggered it?
 85 2014-07-04 11:44:44 <aburan428> my client reached 800 connections and it occurred
 86 2014-07-04 11:45:28 <gmaxwell> Oh, you almost certantly did that to yourself then.
 87 2014-07-04 11:46:08 <aburan428> yeah i thought so
 88 2014-07-04 11:46:42 <gmaxwell> The software cannot have more than 1024 file descriptors open at once due to the design of select(). It normally has measures to prevent that from happening, but since you had to modify the software to end up with 800 connections, you probably accidentally (or not accidentally) bypassed them.
 89 2014-07-04 11:47:21 <gmaxwell> There is no advantage in having a great many connections— no improvement in forwarding speed or anything like that, and if they are outbound its harmful to the network.
 90 2014-07-04 11:51:39 <michagogo> Eh? Didn't someone come in the other day and say that the maximum is 873?
 91 2014-07-04 11:52:09 <gmaxwell> michagogo: depends on your system.
 92 2014-07-04 11:52:16 <michagogo> ah
 93 2014-07-04 11:53:15 <gmaxwell> We might have bugs right around the boundary too, e.g. undercounting our non-connection file descriptor usage some, but since — I assume— what aburan428 is running is all hacked up I'm not going to troubleshoot further without more information.
 94 2014-07-04 14:19:13 <otila> in bitcoin-qt, is it necessary to call TransactionTableModel::updateConfirmations() for every transaction, it triggers call to QSortFilterProxyModelPrivate::build_source_to_proxy_mapping?  it has used 12 mins of CPU time since starting wallet, and I am still waiting
 95 2014-07-04 14:20:18 <otila> or at least less updates to qt5
 96 2014-07-04 14:22:14 <otila> finished! 15 min CPU time to startup
 97 2014-07-04 14:28:03 <wumpus> otila: no, it's not necessary, I think there's an issue for it open somewhere
 98 2014-07-04 14:33:16 <otila> ok good, but I didn't find the issue with google
 99 2014-07-04 14:34:17 <wumpus> maybe it's in my private todo then, anyhow, it's known that the GUI currently scales badly to lots of transactions
100 2014-07-04 14:34:35 <wumpus> feel free to profile/optimize things
101 2014-07-04 14:43:53 <sirius-m> Easy 0.5 BTC for someone who's familiar with bitcoin core's CDataStream: https://github.com/identifi/identifi/issues/3
102 2014-07-04 14:51:59 <wumpus> sirius-m: looks like a send buffer is growing unbounded; try the same with -O0 optimization to get a more useful traceback
103 2014-07-04 14:52:59 <wumpus> it's very doubtful that the serialization function is leaking memory, probably it's what is calling it
104 2014-07-04 14:58:16 <wumpus> keep an eye on the nSendSize of cnodes
105 2014-07-04 14:59:17 <sirius-m> thanks, I'll try that
106 2014-07-04 18:40:11 <Alina-malina> hmmm
107 2014-07-04 18:40:38 <Alina-malina> how to make the json-rpc bitcoin protocol so when the expecting amount arrive on particular address it notify or do something?
108 2014-07-04 20:30:08 <davec_> interesting.  Haven't seen this UA before
109 2014-07-04 20:30:10 <davec_> 15:23:58 2014-07-04 [INF] BMGR: New valid peer 192.33.90.253:44509 (inbound) (/Snoopy:0.1/)
110 2014-07-04 20:31:26 <kazcw> really? I see Snoopys all the time
111 2014-07-04 20:31:30 <dhill> me too :)
112 2014-07-04 20:31:41 <dhill> dunno what it is
113 2014-07-04 20:31:43 <dhill> or Dain
114 2014-07-04 20:32:19 <davec_> perhaps because my node is up and down so much while developing
115 2014-07-04 20:32:36 <dhill> two dns seeds down
116 2014-07-04 20:33:30 <dhill> dnsseed.bitcoin.dashjr.org: no address associated with name
117 2014-07-04 20:34:18 <kazcw> davec_: I keep an  "experimental node" that is :8333 on my external IP and a "stable node", 'cause I like to try git master but every restart wipes out all my hapless inbound peers
118 2014-07-04 20:34:37 <kazcw> er, swap stable/experimental ^
119 2014-07-04 20:49:21 <agorist001> petertodd, ping
120 2014-07-04 21:13:34 <mr_burdell> Alina-malina: not quite the same, but you can use "walletnotify" to run a script when a transaction happens related to your wallet
121 2014-07-04 23:03:01 <petertodd> agorist001: travelling rigjt now, better to email me
122 2014-07-04 23:09:08 <ronkrt> what would the easiest way of creating a bitcoin clone be, with little c++ history?
123 2014-07-04 23:09:37 <andytoshi> https://download.wpsoftware.net/bitcoin/alts.pdf dev experience is the least of your worries
124 2014-07-04 23:22:52 <SomeoneWeird> ronkrt: i thought you said you weren't writing a clone? :P