1 2015-09-17 00:05:18 <midnightmagic> bedeho: There's an interesting script builder for experimentation but it doesn't, as far as I know, hook into real transactions or help you build tx for your own wallet. Well, I guess except circuitously.
 2 2015-09-17 00:06:28 <belcher> i saw a tool like this i beileve bedeho
 3 2015-09-17 00:06:37 <belcher> somewhere on the subreddit in the last few weeks
 4 2015-09-17 00:06:47 <belcher> actually it might've been unfinished
 5 2015-09-17 00:11:11 <midnightmagic> bedeho: But technically this is a question for a channel like #bitcoin, not #bitcoin-dev.
 6 2015-09-17 04:26:57 <TrevinHofmann> Is this channel archived anywhere?
 7 2015-09-17 04:42:11 <maaku> TrevinHofmann: see the topic
 8 2015-09-17 04:43:27 <maaku> TrevinHofmann: soft-fork for tree-sigs is not likely to hit bitcoin anytime soon. no one is working on it and there are other soft forks in the pipeline first
 9 2015-09-17 04:43:54 <maaku> but it would be nice, and there are probably other cool things that could be done with the required soft-fork copcode
10 2015-09-17 04:52:53 <TrevinHofmann> maaku: Thank you. I'll explore other options in the meantime.
11 2015-09-17 05:08:53 <amiller> i need hella testnet coins who wants to give me some
12 2015-09-17 05:09:03 <midnightmagic> amiller: How many do you need?
13 2015-09-17 05:09:05 <amiller> im gonna do some invasive probing
14 2015-09-17 05:10:11 <amiller> i dunno, a hundred
15 2015-09-17 05:10:23 <midnightmagic> amiller: see PM :)
16 2015-09-17 06:42:50 <jonasschnelli> amiller: google for testnet faucet. At least there are two faucet that will give you ~1BTC.
17 2015-09-17 07:12:51 <btcdrak> amiller: I've got a ton of testnet coins if anyone wants. pm me an address.
18 2015-09-17 10:30:37 <shesek> did testnet just have a huge reorg, or is my btcd node acting crazy?
19 2015-09-17 10:44:00 <gmaxwell> shesek: my testnet node hasn't seen one AFAICT.
20 2015-09-17 13:53:23 <jgarzik> jonasschnelli, thanks!
21 2015-09-17 13:53:33 <jonasschnelli> jgarzik: np
22 2015-09-17 13:56:03 <jonasschnelli> https://github.com/bitcoin/bitcoin/pull/6315 <- seeks final ACKs or merge.
23 2015-09-17 14:02:26 <jgarzik> jonasschnelli, Ideally I would like to get wumpus's ACK on that one, maybe one more ACK
24 2015-09-17 14:02:43 <jgarzik> er, one more ACK (wumpus inclusive), I meant
25 2015-09-17 14:03:28 <jonasschnelli> jgarzik: Agreed. I don't want to push... i just hope i can escape the rebase-hamsterwheel for this particular PR. :)
26 2015-09-17 14:04:44 <jonasschnelli> But most QT PR's are trivial to rebase (not much is going on on the QT side)
27 2015-09-17 14:09:24 <jgarzik> File "/home/travis/build/bitcoin/bitcoin/bitcoin-i686-pc-linux-gnu/qa/rpc-tests/zmq_test.py", line 12, in <module>
28 2015-09-17 14:09:24 <jgarzik> ImportError: No module named zmq
29 2015-09-17 14:09:24 <jgarzik> import zmq
30 2015-09-17 14:09:24 <jgarzik> Running testscript zmq_test.py...
31 2015-09-17 14:09:24 <jgarzik> Traceback (most recent call last):
32 2015-09-17 14:09:49 <jgarzik> We got further though :)
33 2015-09-17 14:10:56 <davec> shesek: I don't see anything strange here.  My testnet nodes are all at 556291 and agree with each other
34 2015-09-17 14:11:48 <jonasschnelli> jgarzik: meh... python zmq module is missing...
35 2015-09-17 14:11:55 <jgarzik> nod
36 2015-09-17 14:12:32 <jgarzik> Might just turn off zmq testing temporarily until deps are sorted
37 2015-09-17 14:14:27 <phantomcircuit> shesek, do their logs show where it reorgd?
38 2015-09-17 14:14:57 <jonasschnelli> jgarzik: yes. Maybe more easy. Or add "before_install: - sudo apt-get install python-zmq" to travis.yml
39 2015-09-17 14:56:27 <morcos> I've been thinking about jonasschnelli and sipa conversation yesterday regarding locking with ReadBlockFile.  There are some locks missing with regards to this.
40 2015-09-17 14:57:52 <morcos> In addition I think I've discovered another lock that is missing which doesn't involve pruning.  LoadExternalBlockFile accesses mapBlockIndex without any locks.  But mapBlockIndex could be in the process of being modified.
41 2015-09-17 14:59:37 <morcos> The fImporting flag I think should protect from p2p causing a modification, but I believe submitblock can be modifying mapBlockIndex at the same time as LEBF is reading it
42 2015-09-17 15:00:41 <morcos> I was hoping someone with a bit more experience might suggest the right way to protect against this.  I'm going to go track down the pruning related locks that are needed, and then I guess file issues for these things if I can't see an obvious way to fiix them
43 2015-09-17 15:02:20 <xeon-enouf> meh, i recently just tossed some older full debian 5.x or so ISOs to make up for some much needed space
44 2015-09-17 15:02:44 <sipa> i think we need separate locks for: cs_nodestate, cs_mapblockindex (for adding/finding entries in the map, but not their context), and node for the block position in the entries
45 2015-09-17 15:02:45 <xeon-enouf> (maybe 6.x versions)
46 2015-09-17 15:03:13 <sipa> which is pretty neat
47 2015-09-17 15:03:43 <sipa> validation flags probably several
48 2015-09-17 15:05:57 <xeon-enouf> aha! found one - debian-504-i386-xfce+lxde-CD-1.iso
49 2015-09-17 15:08:20 <xeon-enouf> i have a 25mb/sec upload too... though not sure where i could upload a 677812224 sized file to :-)
50 2015-09-17 15:08:53 <xeon-enouf> jgarzik: ^  (if you need something like that)
51 2015-09-17 15:23:23 <jgarzik> morcos, LoadExternalBlockFile should not be called during operation
52 2015-09-17 15:23:28 <jgarzik> morcos, just init
53 2015-09-17 15:23:52 <morcos> jgarzik: yes but during init, submitblock can still be called
54 2015-09-17 15:23:56 <sipa> jgarzik: it runs simultaneously with full node operation
55 2015-09-17 15:24:09 <sipa> though we ignore block download during that phase
56 2015-09-17 15:24:21 <sipa> but otherwise, it's a fully functional environment
57 2015-09-17 15:25:02 <jgarzik> sipa, hmm, that's not how I remember it.  Looking
58 2015-09-17 15:25:30 <jgarzik> ah, thread import
59 2015-09-17 15:27:00 <phantomcircuit> sipa, we should change the download logic so that it will get headers btw
60 2015-09-17 15:28:45 <sipa> ph
61 2015-09-17 15:29:07 <phantomcircuit> sipa, checkpoints dont work when reindexing or using loadblock
62 2015-09-17 15:31:04 <sipa> i think i need nsome more context
63 2015-09-17 16:18:49 <phantomcircuit> sipa, the fScriptCheck flag is always true if you dont have the headers
64 2015-09-17 16:21:12 <jgarzik> jonasschnelli, I consider it a much greater sin that global tests are broken for any length of time  :)  Pushed the disable-zmq-test quick fix to get the tree going again.  Next step is a "turn on ZMQ tests" PR that reverts my change, imports the zmq module and anything Travis needs.
65 2015-09-17 16:21:30 <jgarzik> iterate on that PR until it's working
66 2015-09-17 16:25:29 <wumpus> LoadExternalBlockSize runs in a thread, it should certainly not only run in init, this would hold up the initialization until everything is reindexed!
67 2015-09-17 16:26:35 <wumpus> (it's launched from init, but so are all threads)
68 2015-09-17 16:30:01 <wumpus> BlueMatt recently added code to wait for exiting AppInit2 until the genesis block is indexed, maybe that's where the confusion comes from
69 2015-09-17 16:30:40 <wumpus> was the zmq stuff stable enough for merging?
70 2015-09-17 16:37:06 <jonasschnelli> wumpus: a difficult question (if the ZMQ stuff was stable enough). It was a relatively large PR with new dependencies. I'm fine with the merge, even if there are some cleanup PR to do now.
71 2015-09-17 16:38:34 <jonasschnelli> To bad that the depends and Travis issue where not detected in advance (I take that on me, I gave an ACK).
72 2015-09-17 16:39:22 <wumpus> if someone commits to solving the issues now that it's merged I'm OK with it