1 2015-10-23T00:05:01  *** dcousens has joined #bitcoin-core-dev
  2 2015-10-23T00:10:27  *** challisto has quit IRC
  3 2015-10-23T00:36:02  *** PRab has quit IRC
  4 2015-10-23T00:54:20  *** Ylbam has quit IRC
  5 2015-10-23T01:18:37  *** jgarzik has quit IRC
  6 2015-10-23T01:18:57  *** jgarzik has joined #bitcoin-core-dev
  7 2015-10-23T01:25:26  *** dcousens has quit IRC
  8 2015-10-23T01:37:50  *** dcousens has joined #bitcoin-core-dev
  9 2015-10-23T01:38:07  <dcousens> petertodd: maybe change the PR title to `Full-RBF with nSequence opt-out`?
 10 2015-10-23T01:38:41  <dcousens> petertodd: it reads as though its an opt-out for the node (until you read the PR itself, obviously)
 11 2015-10-23T01:38:50  <dcousens> (but even still, I had to check the code to be sure)
 12 2015-10-23T01:41:07  <GitHub159> [bitcoin] jgarzik opened pull request #6873: [cleanup] leveldbwrapper becomes more generic 'dbwrapper' module (master...2015_dbwrapper) https://github.com/bitcoin/bitcoin/pull/6873
 13 2015-10-23T01:42:23  <gmaxwell> It would also be less confusing if it were called opt-in. (since transactions are currently opt out under the design)
 14 2015-10-23T01:42:41  <dcousens> gmaxwell: agreed
 15 2015-10-23T01:42:55  <dcousens> though it is called opt-in in the commit
 16 2015-10-23T01:42:59  <dcousens> So, maybe consistency there
 17 2015-10-23T01:48:19  <dcousens> jgarzik: in 6873
 18 2015-10-23T01:48:28  <dcousens> what actually makes CDBWrapper agnostic of leveldb?
 19 2015-10-23T01:48:34  <dcousens> It seems to still use all the statuscode types
 20 2015-10-23T01:48:43  <dcousens> And the wrapper can still be instantiated by itself
 21 2015-10-23T01:49:23  <dcousens> Or is this just a transitionary step?
 22 2015-10-23T01:49:52  <sipa> dcousens: it's API-agnostic... code using it doesn't need to know that it's backed by leveldb
 23 2015-10-23T01:50:01  <sipa> even though you can't compile it any other way
 24 2015-10-23T01:50:13  <jgarzik> nod.  The interface is backend agnostic
 25 2015-10-23T01:50:59  <dcousens> jgarzik: then maybe the PR should be about removal of leveldb types in DbWrapper
 26 2015-10-23T01:51:20  <jgarzik> dcousens, did you look at the sqlite commit?
 27 2015-10-23T01:51:34  <dcousens> sqlite commit?
 28 2015-10-23T01:51:47  <dcousens> nvm
 29 2015-10-23T01:51:54  <dcousens> right
 30 2015-10-23T01:52:02  <dcousens> cheers
 31 2015-10-23T01:52:39  <dcousens> jgarzik: my bad for not reading the PR before reading the changes
 32 2015-10-23T01:53:25  <dcousens> this just reads weird: https://github.com/bitcoin/bitcoin/pull/6873/files#diff-ac09ab7f710fc2fb7850f59601545636R137
 33 2015-10-23T01:53:59  <dcousens> Given its called a 'wrapper', but its now generic,  seems a bit odd IMHO
 34 2015-10-23T01:54:14  <sipa> call it DbInterface then
 35 2015-10-23T01:54:31  <phantomcircuit> jgarzik, :)
 36 2015-10-23T01:54:55  <dcousens> sipa: but its not just an interface no,  it instantiates and manages the entire thing right?
 37 2015-10-23T01:55:04  <dcousens> its more like 'Db', ha
 38 2015-10-23T01:55:54  <jgarzik> it is a class that wraps the backend database in a nice snuggly generic interface
 39 2015-10-23T01:56:39  <gmaxwell> (1) dear god, are people really arguing about this, (2) dbwrapper says what it does!
 40 2015-10-23T01:56:48  <jcorgan> jgarzik: compile ack on the 2015_sqlite branch, but looks like retains all the leveldb naming in the build system, checking now
 41 2015-10-23T01:57:36  <dcousens> gmaxwell: wasn't arguing, just seemed better to ask here than ask about it on GH seeing as its probably me just missing context
 42 2015-10-23T01:57:58  <jgarzik> jcorgan, naming?  2015_sqlite branch should not include any leveldb naming in the bitcoin->sqlite path.  it _does_ build leveldb.  it _does not_ link leveldb to bitcoind.
 43 2015-10-23T01:58:11  <jcorgan> oh, that's what i'm seeing then
 44 2015-10-23T01:58:51  <sipa> dcousens: then DbWrapper is a better name! it's something that wraps a db, but you don't know or care what kind of database that is
 45 2015-10-23T01:59:02  <dcousens> gmaxwell: in any case, a wrapper typically wraps an existing object no?
 46 2015-10-23T01:59:23  <dcousens> in this case, the object is entirely owned by the wrapper
 47 2015-10-23T01:59:43  <sipa> grrrrr, EncapsulatedDb then?
 48 2015-10-23T01:59:50  *** gmaxwell has left #bitcoin-core-dev
 49 2015-10-23T01:59:57  <dcousens> haha, sipa this isn't important, I'm just curious
 50 2015-10-23T02:01:16  <jcorgan> jgarzik: there are some stray leveldb name issues
 51 2015-10-23T02:01:35  <jgarzik> jcorgan, where?
 52 2015-10-23T02:01:42  <jcorgan> like using 'leveldb_error' in dbwrapper.cpp
 53 2015-10-23T02:01:58  <jgarzik> hrm, I thought I got that one :(
 54 2015-10-23T02:02:22  <jgarzik> jcorgan, no, wait, I did.  git pull.  :)
 55 2015-10-23T02:02:25  <jcorgan> it's still defined in dbwrapper.h, try commenting that and see where the errors
 56 2015-10-23T02:02:38  <dcousens> sipa: yeah, quick investigation of the wrapper/adapter pattern is that it is typically applied as a glue code, aka, between two things with their own ownership.
 57 2015-10-23T02:02:50  <dcousens> In this case, its not that, so, anyway.  Just food for though
 58 2015-10-23T02:02:52  <dcousens> thought*
 59 2015-10-23T02:03:21  <dcousens> tl;dr EncapsulatedDb would probably be better
 60 2015-10-23T02:03:24  <jcorgan> i'm using b778481
 61 2015-10-23T02:03:39  <jgarzik> jcorgan, git pull
 62 2015-10-23T02:03:56  <CodeShark> HighlyVerboseButGenericNameForDbInterface
 63 2015-10-23T02:04:05  <jgarzik> dbwrapper is just fine, let's move on.
 64 2015-10-23T02:04:10  <sipa> yes, please
 65 2015-10-23T02:04:56  <jcorgan> jgarzik: ok, that fixed it
 66 2015-10-23T02:04:59  <dcousens> jgarzik: well, it causes confusion in anyone who hasn't read through that part of the source.  So, IMHO it matters enough to be discussed.  But not to block your PR.  Moving on :)
 67 2015-10-23T02:05:11  <jcorgan> but to be pedantic, there are a bunch of 'leveldb' strings in the comments :-)
 68 2015-10-23T02:06:03  *** arowser has joined #bitcoin-core-dev
 69 2015-10-23T02:06:22  * jgarzik is focusing on operating stuff, like why this unit test doesn't want to compile my sql
 70 2015-10-23T02:06:30  <jgarzik> *operational
 71 2015-10-23T02:06:54  <jcorgan> no worries.  i haven't tested anything, but compiles fine on ubuntu trusty
 72 2015-10-23T02:07:08  <jcorgan> and 6873 is fine
 73 2015-10-23T02:07:34  *** belcher has quit IRC
 74 2015-10-23T02:32:06  <GitHub80> [bitcoin] pstratem opened pull request #6874: Net: Cork socket send writes with MSG_MORE (master...msg_more) https://github.com/bitcoin/bitcoin/pull/6874
 75 2015-10-23T02:32:31  <phantomcircuit> jgarzik, it would be nice if there was a virtual base class
 76 2015-10-23T02:32:48  <phantomcircuit> DBWrapperInterface
 77 2015-10-23T02:33:01  * phantomcircuit goes to look for some InterfaceFactories
 78 2015-10-23T02:37:36  <dcousens> phantomcircuit: if the default is being hard-coded in at each entry anyway
 79 2015-10-23T02:37:40  <dcousens> probably not much point
 80 2015-10-23T02:38:29  <dcousens> unless the wrapper is instantiated at a much earlier point in the program, and passed through.  At which point then the generic interface could be used
 81 2015-10-23T02:38:47  <dcousens> s/earlier/higher
 82 2015-10-23T02:38:55  * jgarzik reminds channel: we've moved on
 83 2015-10-23T02:41:30  <phantomcircuit> dcousens, well it's only actually named in one place i think so it doesn't much matter
 84 2015-10-23T02:52:35  <GitHub95> [bitcoin] TheBlueMatt opened pull request #6875: Fix pre-push-hook regexes (master...verify-commits-fixes) https://github.com/bitcoin/bitcoin/pull/6875
 85 2015-10-23T02:54:12  <GitHub11> [bitcoin] TheBlueMatt closed pull request #6875: Fix pre-push-hook regexes (master...verify-commits-fixes) https://github.com/bitcoin/bitcoin/pull/6875
 86 2015-10-23T02:58:47  *** PRab has joined #bitcoin-core-dev
 87 2015-10-23T03:37:35  <phantomcircuit> jgarzik, updated 6874
 88 2015-10-23T04:24:18  *** jgarzik has quit IRC
 89 2015-10-23T04:24:46  *** jgarzik has joined #bitcoin-core-dev
 90 2015-10-23T04:26:28  <phantomcircuit> <phantomcircuit> jgarzik, updated 6874
 91 2015-10-23T04:51:39  *** jl2012 has joined #bitcoin-core-dev
 92 2015-10-23T05:13:54  *** AtashiCon has quit IRC
 93 2015-10-23T05:18:41  *** AtashiCon has joined #bitcoin-core-dev
 94 2015-10-23T05:21:04  <phantomcircuit> possibly we should at least try to reconnect to disconnected peers...
 95 2015-10-23T05:26:28  *** maaku has quit IRC
 96 2015-10-23T05:28:07  *** maaku has joined #bitcoin-core-dev
 97 2015-10-23T05:29:38  *** evoskuil has joined #bitcoin-core-dev
 98 2015-10-23T06:23:50  *** MarcoFalke has joined #bitcoin-core-dev
 99 2015-10-23T06:28:31  *** Thireus has quit IRC
100 2015-10-23T06:33:16  *** Ylbam has joined #bitcoin-core-dev
101 2015-10-23T06:34:29  *** Thireus has joined #bitcoin-core-dev
102 2015-10-23T06:35:14  *** Thireus has quit IRC
103 2015-10-23T06:37:59  *** MarcoFalke has quit IRC
104 2015-10-23T06:54:34  *** MarcoFalke has joined #bitcoin-core-dev
105 2015-10-23T07:26:31  <GitHub136> [bitcoin] laanwj opened pull request #6877: rpc: Add maxmempool and effective min fee to getmempoolinfo (master...2015_10_mempool_effective_fee) https://github.com/bitcoin/bitcoin/pull/6877
106 2015-10-23T07:36:27  *** dcousens has quit IRC
107 2015-10-23T07:40:48  *** ParadoxSpiral has joined #bitcoin-core-dev
108 2015-10-23T07:43:51  *** ParadoxSpiral_ has quit IRC
109 2015-10-23T07:54:11  *** CodeShark has quit IRC
110 2015-10-23T07:57:50  *** Guest17781 has joined #bitcoin-core-dev
111 2015-10-23T07:59:02  *** d_t has quit IRC
112 2015-10-23T08:04:40  <GitHub27> [bitcoin] laanwj pushed 1 new commit to 0.11: https://github.com/bitcoin/bitcoin/commit/95a50390e1052a0a501eb446f87d63f58d95b7e7
113 2015-10-23T08:04:41  <GitHub27> bitcoin/0.11 95a5039 Gregory Maxwell: Set TCP_NODELAY on P2P sockets....
114 2015-10-23T08:04:58  *** BashCo has quit IRC
115 2015-10-23T08:11:37  *** Guest17781 has quit IRC
116 2015-10-23T08:12:33  <GitHub187> [bitcoin] laanwj pushed 1 new commit to 0.10: https://github.com/bitcoin/bitcoin/commit/5297194bbd6e0d6730515567248caf9c135e296c
117 2015-10-23T08:12:33  <GitHub187> bitcoin/0.10 5297194 Gregory Maxwell: Set TCP_NODELAY on P2P sockets....
118 2015-10-23T08:22:10  *** BashCo has joined #bitcoin-core-dev
119 2015-10-23T08:25:06  *** CodeShark has joined #bitcoin-core-dev
120 2015-10-23T08:43:37  *** matsjj_ has joined #bitcoin-core-dev
121 2015-10-23T08:46:11  <GitHub44> [bitcoin] laanwj pushed 3 new commits to master: https://github.com/bitcoin/bitcoin/compare/b2b173acabea...fa1d252e82a8
122 2015-10-23T08:46:11  <GitHub44> bitcoin/master 338f62f MarcoFalke: [devtools] add clang-format.py
123 2015-10-23T08:46:12  <GitHub44> bitcoin/master 8c15f33 MarcoFalke: [trivial] Update contrib/devtools/README.md
124 2015-10-23T08:46:13  <GitHub44> bitcoin/master fa1d252 Wladimir J. van der Laan: Merge pull request #6790...
125 2015-10-23T08:46:16  <GitHub153> [bitcoin] laanwj closed pull request #6790: [devtools] add clang-format.py (master...MarcoFalke-2015-clangFormatWrapper) https://github.com/bitcoin/bitcoin/pull/6790
126 2015-10-23T08:53:13  *** lightningbot` has quit IRC
127 2015-10-23T10:49:43  *** lightningbot has joined #bitcoin-core-dev
128 2015-10-23T10:59:30  *** dcousens has quit IRC
129 2015-10-23T11:03:27  *** wump is now known as wumpus
130 2015-10-23T11:33:36  <GitHub151> [bitcoin] laanwj pushed 4 new commits to master: https://github.com/bitcoin/bitcoin/compare/bf7c1958d199...2a1090d4f55c
131 2015-10-23T11:33:37  <GitHub151> bitcoin/master 287f54f Peter Todd: Add CHECKLOCKTIMEVERIFY (BIP65) soft-fork logic...
132 2015-10-23T11:33:38  <GitHub151> bitcoin/master cde7ab2 Peter Todd: Add RPC tests for the CHECKLOCKTIMEVERIFY (BIP65) soft-fork...
133 2015-10-23T11:33:38  <GitHub151> bitcoin/master 65ef372 Peter Todd: Add BIP65 to getblockchaininfo softforks list
134 2015-10-23T11:34:21  <GitHub0> [bitcoin] laanwj closed pull request #6706: CLTV IsSuperMajority() soft-fork, rebased against v0.10.2 (0.10...cltv-soft-fork-v0.10) https://github.com/bitcoin/bitcoin/pull/6706
135 2015-10-23T11:34:30  * btcdrak jumps for joy
136 2015-10-23T11:57:25  <GitHub121> [bitcoin] laanwj opened pull request #6879: doc: mention BIP65 softfork in bips.md (master...2015_10_bip65) https://github.com/bitcoin/bitcoin/pull/6879
137 2015-10-23T12:36:17  *** danielsocials has joined #bitcoin-core-dev
138 2015-10-23T12:41:52  <GitHub155> [bitcoin] laanwj pushed 2 new commits to master: https://github.com/bitcoin/bitcoin/compare/2a1090d4f55c...c719cefc417c
139 2015-10-23T12:41:52  <GitHub155> bitcoin/master d3b09f6 Alex Morcos: Do not allow blockfile pruning during reindex....
140 2015-10-23T12:41:53  <GitHub155> bitcoin/master c719cef Wladimir J. van der Laan: Merge pull request #6856...
141 2015-10-23T12:41:57  <GitHub81> [bitcoin] laanwj closed pull request #6856: Do not allow block file pruning during reindex. (master...noPruneDuringReindex) https://github.com/bitcoin/bitcoin/pull/6856
142 2015-10-23T12:42:35  <GitHub177> [bitcoin] laanwj pushed 1 new commit to 0.11: https://github.com/bitcoin/bitcoin/commit/dfe55bdc32b5333dcce1a7f2c74628f64028d1fe
143 2015-10-23T12:42:35  <GitHub177> bitcoin/0.11 dfe55bd Alex Morcos: Do not allow blockfile pruning during reindex....
144 2015-10-23T12:51:48  *** treehug8_ has joined #bitcoin-core-dev
145 2015-10-23T13:09:10  <GitHub89> [bitcoin] jgarzik pushed 4 new commits to master: https://github.com/bitcoin/bitcoin/compare/c719cefc417c...46f74379b86b
146 2015-10-23T13:09:11  <GitHub89> bitcoin/master 6ec4b7e Jeff Garzik: leveldbwrapper: Remove unused .Prev(), .SeekToLast() methods...
147 2015-10-23T13:09:11  <GitHub89> bitcoin/master 8587b23 Jeff Garzik: leveldbwrapper symbol rename: Remove "Level" from class, etc. names
148 2015-10-23T13:09:12  <GitHub89> bitcoin/master 3795e81 Jeff Garzik: leveldbwrapper file rename to dbwrapper.*
149 2015-10-23T13:09:17  <GitHub179> [bitcoin] jgarzik closed pull request #6873: [cleanup] leveldbwrapper becomes more generic 'dbwrapper' module (master...2015_dbwrapper) https://github.com/bitcoin/bitcoin/pull/6873
150 2015-10-23T13:20:09  *** harding has joined #bitcoin-core-dev
151 2015-10-23T13:31:09  *** treehug8_ has quit IRC
152 2015-10-23T13:35:17  *** treehug8_ has joined #bitcoin-core-dev
153 2015-10-23T13:49:46  *** rubensayshi has quit IRC
154 2015-10-23T15:06:29  *** treehug8_ has quit IRC
155 2015-10-23T15:08:00  *** dgenr8 has joined #bitcoin-core-dev
156 2015-10-23T15:12:36  *** treehug8_ has joined #bitcoin-core-dev
157 2015-10-23T15:23:51  <GitHub199> [bitcoin] gavinandresen opened pull request #6880: New -logtimerelative option to do millisec debug.log timestamps (master...millisec_debuglog) https://github.com/bitcoin/bitcoin/pull/6880
158 2015-10-23T15:30:30  *** sdaftuar has quit IRC
159 2015-10-23T15:30:42  *** morcos has quit IRC
160 2015-10-23T15:31:01  *** zxzzt_ has quit IRC
161 2015-10-23T15:32:27  *** morcos has joined #bitcoin-core-dev
162 2015-10-23T15:32:33  *** sdaftuar has joined #bitcoin-core-dev
163 2015-10-23T15:33:15  *** zxzzt has joined #bitcoin-core-dev
164 2015-10-23T15:40:50  *** randy-waterhouse has quit IRC
165 2015-10-23T15:41:27  *** d_t has joined #bitcoin-core-dev
166 2015-10-23T15:49:08  *** maaku__ is now known as maaku
167 2015-10-23T15:58:36  *** danielsocials has quit IRC
168 2015-10-23T16:22:38  <GitHub54> [bitcoin] gavinandresen closed pull request #6880: New -logtimerelative option to do millisec debug.log timestamps (master...millisec_debuglog) https://github.com/bitcoin/bitcoin/pull/6880
169 2015-10-23T16:54:54  *** BashCo has quit IRC
170 2015-10-23T16:55:23  *** matsjj_ has quit IRC
171 2015-10-23T16:59:15  *** gmaxwell has joined #bitcoin-core-dev
172 2015-10-23T17:17:10  *** BashCo has joined #bitcoin-core-dev
173 2015-10-23T18:07:02  *** fkhan has quit IRC
174 2015-10-23T18:21:19  *** fkhan has joined #bitcoin-core-dev
175 2015-10-23T18:23:47  <GitHub0> [bitcoin] sdaftuar opened pull request #6881: Debug: Add option for microsecond precision in debug.log (master...add-microsecond-timestamps) https://github.com/bitcoin/bitcoin/pull/6881
176 2015-10-23T18:25:20  *** Dyanisus has joined #bitcoin-core-dev
177 2015-10-23T18:26:06  <maaku> I am leaving bitcoin core development for a while, not sure how long. If someone wants to take over pushing for #6312, #6564, and #6566 please contact me.
178 2015-10-23T18:26:52  <Dyanisus> why you leaving?
179 2015-10-23T18:27:04  <btcdrak> maaku: why are you leaing?
180 2015-10-23T18:27:19  <btcdrak> maaku: we're literally right at the end!
181 2015-10-23T18:28:35  *** randy-waterhouse has joined #bitcoin-core-dev
182 2015-10-23T18:28:44  <Dyanisus> maaku: did you get some sort of subpoena or something?
183 2015-10-23T18:30:23  *** CodeShark_ has joined #bitcoin-core-dev
184 2015-10-23T18:35:55  <btcdrak> maaku: I've sent you an email about those PRs.
185 2015-10-23T18:52:30  *** molly has quit IRC
186 2015-10-23T19:09:07  *** fkhan has quit IRC
187 2015-10-23T19:13:23  *** CodeShark_ has quit IRC
188 2015-10-23T19:26:18  *** fkhan has joined #bitcoin-core-dev
189 2015-10-23T19:47:22  *** Guest89666 is now known as pigeons
190 2015-10-23T19:49:36  *** CodeShark has quit IRC
191 2015-10-23T20:02:56  *** randy-waterhouse has quit IRC
192 2015-10-23T20:16:35  *** treehug8_ has quit IRC
193 2015-10-23T20:17:12  <GitHub65> [bitcoin] petertodd opened pull request #6883: Add BIP65 CHECKLOCKTIMEVERIFY to release notes (master...cltv-release-notes-v0.12.0) https://github.com/bitcoin/bitcoin/pull/6883
194 2015-10-23T20:21:09  <GitHub95> [bitcoin] btcdrak opened pull request #6884: Backport #6566, median-past locktime, rebased against 0.11 (0.11...mpl-0.11) https://github.com/bitcoin/bitcoin/pull/6884
195 2015-10-23T20:28:44  *** dhill has joined #bitcoin-core-dev
196 2015-10-23T20:28:47  <dhill> maaku: hi
197 2015-10-23T20:29:56  <dhill> maaku: regarding checksequenceverify, i ran your tests against btcd, and they all pass
198 2015-10-23T20:29:57  <dhill> but
199 2015-10-23T20:30:03  <dhill> +    if (txToSequence & CTxIn::SEQUENCE_LOCKTIME_DISABLED_FLAG)
200 2015-10-23T20:30:03  <dhill> +        return false;
201 2015-10-23T20:30:09  <dhill> was not hit
202 2015-10-23T20:30:19  <dhill> i think there is a dup test in tx_invalid.json
203 2015-10-23T21:21:03  *** belcher has joined #bitcoin-core-dev
204 2015-10-23T21:43:25  *** davec has joined #bitcoin-core-dev
205 2015-10-23T21:50:36  *** molly has joined #bitcoin-core-dev
206 2015-10-23T21:59:55  <maaku> I added btcdrak as a collaborator for my personal repo, so he should be able to update the 'sequencenumbers' 'checksequenceverify' and 'medianpastlocktime' branches
207 2015-10-23T22:32:22  <btcdrak> maaku: thanks!
208 2015-10-23T23:11:07  *** belcher has quit IRC
209 2015-10-23T23:14:21  *** belcher has joined #bitcoin-core-dev
210 2015-10-23T23:30:18  <cfields> sipa: ping. I'm trying to understand the current dnsseed thread behavior. It looks like in the non-proxied case, we resolve all seed nodes and add them to addrman without testing them. In the proxied case, we connect to each resolved entry as a oneshot...
211 2015-10-23T23:30:51  <cfields> i assume the goal is to query them all once if possible in order to never have to touch 'em again?
212 2015-10-23T23:32:23  <cfields> just seems a bit unbalanced between the proxy/non-proxy cases, want to be sure i'm not missing anything in the big picture
213 2015-10-23T23:39:41  * jgarzik wonders about the proxy case behavior
214 2015-10-23T23:40:34  <jgarzik> cfields, The add-untested logic follows the getaddr logic, where it is unknown possibly garbage data coming in from unknown sources
215 2015-10-23T23:40:56  <jgarzik> cfields, overall I do think addrman must do a much better job of testing and expiring bad addresses
216 2015-10-23T23:41:49  <jgarzik> cfields, notably we engage in the poor network behaviors of (1) propagating known bad addresses and (2) repeatedly trying the same dead nodes, in some near-empty-bucket edge cases
217 2015-10-23T23:44:34  <gmaxwell> The proxy behavior is simply because we cannot do the dns lookup via the proxy. So connecting to the addresses returned (by proxy connecting to the names) was the best thing we could do.
218 2015-10-23T23:44:40  <cfields> jgarzik: ok, that lines up with how i read it. Looks like in the proxied case where we can't run a query, we just connect to all reachable nodes (query done through the proxy), getaddr on them, and disconnect. seems to me like that'd yield less garbage, but i suppose you're right about letting addrman do it.
219 2015-10-23T23:44:45  <gmaxwell> It does a less good job of initilizing the state for sure.
220 2015-10-23T23:45:25  <gmaxwell> not to all reachable nodes, we connect 'to' each dnsseed by name (meaning connect to one of the results of it).
221 2015-10-23T23:45:41  <cfields> gmaxwell: less good because the dns seeds tend to be much fresher, i assume?
222 2015-10-23T23:46:05  <jgarzik> dns seeds are fresh and quick and easy and wonderfully centralizing ;p
223 2015-10-23T23:46:11  <gmaxwell> cfields: no, just less good because it any one of the dnsseed results could be broken or malicious.
224 2015-10-23T23:46:39  <cfields> gmaxwell: ah right, only one-per-seed that way.
225 2015-10-23T23:46:45  <jgarzik> notably there is now code that "waits a bit, before checking seeds" to attempt to P2P without DNS seeds
226 2015-10-23T23:46:47  <gmaxwell> e.g. dns seed returns nodes a,b,c,d,e,f,g  ... and we only end up connecting to g (out of that dns seed's results) and perhaps g gives us useless answers.
227 2015-10-23T23:46:59  <jgarzik> so DNS seeding is intentionally not immediate
228 2015-10-23T23:47:13  <cfields> ok, all clear now. thanks.
229 2015-10-23T23:49:36  <cfields> ok, along those lines, are there concerns (fingerprinting) about hitting the seeds too quickly or uniformly? I don't see any active approach in the current code to avoid that
230 2015-10-23T23:50:40  <jgarzik> cfields, sure those concerns exist
231 2015-10-23T23:50:55  <jgarzik> cfields, that's part of the reason why the current code tries to avoid DNS lookups
232 2015-10-23T23:51:10  <jgarzik> cfields, a DNS seed can build a picture of bitcoin users etc.
233 2015-10-23T23:52:02  *** Luke-Jr has quit IRC
234 2015-10-23T23:52:31  <jgarzik> cfields, Current logic is:  (1) if addrman.size() > 0, Attempt to connect to network given current peer database  (2) If "too few" P2P connections after 11 seconds, perform DNS lookup.
235 2015-10-23T23:52:34  <cfields> obviously not worried about the seeds themselves, since they're already trusted. more about someone trying to get 2+ entries in a seed's current entries in order to distinguish individuals
236 2015-10-23T23:52:43  <gmaxwell> cfields: DNS seeds are prohibited by policy from having a TTL less than 60 seconds, so recursive resolution provides some increase in privacy.
237 2015-10-23T23:53:12  <jgarzik> I think it's quite likely that a single DNS seed might spit out two addresses which are in fact controlled by one person/org
238 2015-10-23T23:53:32  <gmaxwell> it's almost certan. well plus a substantial number of nodes on the network are sybil attackers in any case.
239 2015-10-23T23:53:35  <gmaxwell> :P
240 2015-10-23T23:53:52  <cfields> jgarzik: sure, i'm speaking of the procedure when we do fall into the dns lookup logic.
241 2015-10-23T23:54:34  <jgarzik> sure.  the logic is dumb:  query all DNS seeds, let addrman sort it out
242 2015-10-23T23:55:03  <gmaxwell> cfields: addrman knows where addresses came from to e.g. prevent any one source from dominating the table. There is only so much it can do.
243 2015-10-23T23:55:10  <cfields> ok. no problem with that, just want to be sure i'm understanding the logic that's there
244 2015-10-23T23:55:22  <gmaxwell> Lack of actual users running nodes provides a very limiting upper bound on how much protection is possible.
245 2015-10-23T23:57:30  <cfields> got it. thanks jgarzik/gmaxwell.
246 2015-10-23T23:58:40  *** Luke-Jr has joined #bitcoin-core-dev