12017-02-15T00:09:31 *** kadoban_ is now known as kadoban
22017-02-15T00:35:30 *** Guyver2 has quit IRC
32017-02-15T00:52:36 <BlueMatt> hmmmmmm...i asked to use the MTP of the current block in the default importmulti timestamps because I thought that would be sufficient (+/- reorgs)...I assume #9761 is pretty much just for reorgs, then?
42017-02-15T00:52:37 <gribble> https://github.com/bitcoin/bitcoin/issues/9761 | Use 2 hour grace period for key timestamps in importmulti rescans by ryanofsky · Pull Request #9761 · bitcoin/bitcoin · GitHub
52017-02-15T00:52:38 <BlueMatt> gmaxwell: maybe?
62017-02-15T01:34:35 <morcos_> BlueMatt: the motivation of 9761 was a hypothetical situation that i didn't bother to test the reasonability of
72017-02-15T01:34:46 <morcos_> how do we think someone is going to get these birthdates
82017-02-15T01:35:21 <morcos_> in particular is it likely that you'll get it from a dumpprivkey or a validateaddress call or something else from a node where the key was created in the wallet
92017-02-15T01:35:37 <morcos_> if that wallet created timestamp is exposed, then thats not created with MTP
102017-02-15T01:35:51 <morcos_> so you need the buffer
112017-02-15T01:36:41 *** morcos_ is now known as morcos
122017-02-15T01:40:59 <morcos> i guess looking now before the recent change to validateaddress, there was no way to expose nCreateTime? but in any case there is now i think.
132017-02-15T01:43:09 <morcos> kind of a weird feature then when it was created. what were people expected to put in for the timestamp? how would they know? i guess maybe they know separately the time of the first block that was relevant?
142017-02-15T01:46:10 *** goksinen has quit IRC
152017-02-15T02:06:55 *** goksinen has joined #bitcoin-core-dev
162017-02-15T02:12:05 *** goksinen has quit IRC
172017-02-15T02:32:03 *** chjj has joined #bitcoin-core-dev
182017-02-15T02:38:58 *** goksinen has joined #bitcoin-core-dev
192017-02-15T02:40:01 *** chjj has quit IRC
202017-02-15T02:41:59 *** Arvidt_ has joined #bitcoin-core-dev
212017-02-15T02:43:27 *** goksinen has quit IRC
222017-02-15T02:45:45 <Arvidt_> When running bitcoind with onlynet=onion, proxy=1.2.3.4:9050, listen=1 and externalip=xxx.onion (Tor with bitcoind hidden service xxx.onion on 1.2.3.4) can I still connect directly to bitcoind on 127.0.0.1:8333 ? How could I test that?
232017-02-15T02:46:10 <sipa> yes
242017-02-15T02:46:26 <sipa> you can test by running a second bitcoind and -connect=127.0.0.1:8333 to it
252017-02-15T02:46:38 <sipa> (with a different datadir, -port, and -rpcport)
262017-02-15T02:49:31 <Arvidt_> Thanks a lot for answer. Hm I thought more of a little telnet test, second instance is a little bit too much for me for test setup.
272017-02-15T02:55:20 *** Giszmo has quit IRC
282017-02-15T02:56:17 <sipa> a telnet test would work too, but you'd need to construct a handshake
292017-02-15T02:56:17 <sipa> just a version message should be enough, which you could send with netcat
302017-02-15T03:03:10 *** echonaut1 has joined #bitcoin-core-dev
312017-02-15T03:03:10 *** Aaronvan_ has joined #bitcoin-core-dev
322017-02-15T03:03:10 *** juscamarena_ has joined #bitcoin-core-dev
332017-02-15T03:03:59 *** Ylbam has quit IRC
342017-02-15T03:04:13 *** rubensayshi_ has joined #bitcoin-core-dev
352017-02-15T03:04:55 *** paracyst_ has joined #bitcoin-core-dev
362017-02-15T03:06:01 *** wolfspraul has joined #bitcoin-core-dev
372017-02-15T03:06:26 *** jouke_ has joined #bitcoin-core-dev
382017-02-15T03:06:33 *** emzy_ has joined #bitcoin-core-dev
392017-02-15T03:06:38 *** gwollon has joined #bitcoin-core-dev
402017-02-15T03:06:43 *** wump has joined #bitcoin-core-dev
412017-02-15T03:07:08 *** thestrin1puller has joined #bitcoin-core-dev
422017-02-15T03:07:08 *** Alina-malina_ has joined #bitcoin-core-dev
432017-02-15T03:07:10 *** crescendo has joined #bitcoin-core-dev
442017-02-15T03:09:08 *** rubensayshi has quit IRC
452017-02-15T03:09:08 *** pindarhk has quit IRC
462017-02-15T03:09:08 *** AaronvanW has quit IRC
472017-02-15T03:09:08 *** profall has quit IRC
482017-02-15T03:09:08 *** jouke has quit IRC
492017-02-15T03:09:08 *** isis has quit IRC
502017-02-15T03:09:08 *** lejitz has quit IRC
512017-02-15T03:09:08 *** juscamarena has quit IRC
522017-02-15T03:09:08 *** Alina-malina has quit IRC
532017-02-15T03:09:08 *** crescend1 has quit IRC
542017-02-15T03:09:08 *** ibrightly has quit IRC
552017-02-15T03:09:08 *** gwillen has quit IRC
562017-02-15T03:09:08 *** emzy has quit IRC
572017-02-15T03:09:08 *** belcher has quit IRC
582017-02-15T03:09:08 *** echonaut has quit IRC
592017-02-15T03:09:08 *** face has quit IRC
602017-02-15T03:09:08 *** thestringpuller has quit IRC
612017-02-15T03:09:08 *** wolfspra1l has quit IRC
622017-02-15T03:09:08 *** wumpus has quit IRC
632017-02-15T03:09:21 *** paracyst has quit IRC
642017-02-15T03:09:21 *** rubensayshi_ is now known as rubensayshi
652017-02-15T03:09:21 *** face has joined #bitcoin-core-dev
662017-02-15T03:09:21 *** belcher has joined #bitcoin-core-dev
672017-02-15T03:09:36 *** isis has joined #bitcoin-core-dev
682017-02-15T03:10:53 *** goksinen has joined #bitcoin-core-dev
692017-02-15T03:13:10 *** pindarhk has joined #bitcoin-core-dev
702017-02-15T03:13:10 *** ibrightly has joined #bitcoin-core-dev
712017-02-15T03:15:47 *** goksinen has quit IRC
722017-02-15T03:16:33 *** gwollon is now known as gwillen
732017-02-15T03:30:44 <Arvidt_> @sipa thanks for the tip. I took the second example from https://en.bitcoin.it/wiki/Protocol_documentation#version to bitcoind and got a VERACK :-)
742017-02-15T03:30:55 <sipa> cool
752017-02-15T03:41:45 *** goksinen has joined #bitcoin-core-dev
762017-02-15T03:45:57 *** goksinen has quit IRC
772017-02-15T04:07:50 *** PaulCape_ has quit IRC
782017-02-15T04:12:01 *** truquini___ has quit IRC
792017-02-15T04:13:31 *** goksinen has joined #bitcoin-core-dev
802017-02-15T04:15:46 *** PRab has quit IRC
812017-02-15T04:17:57 *** goksinen has quit IRC
822017-02-15T04:30:06 *** kadoban has quit IRC
832017-02-15T04:45:36 *** goksinen has joined #bitcoin-core-dev
842017-02-15T04:49:57 *** goksinen has quit IRC
852017-02-15T05:01:18 *** lejitz has joined #bitcoin-core-dev
862017-02-15T05:01:25 *** profall has joined #bitcoin-core-dev
872017-02-15T05:14:53 *** tunafizz has joined #bitcoin-core-dev
882017-02-15T05:16:34 *** goksinen has joined #bitcoin-core-dev
892017-02-15T05:21:05 *** goksinen has quit IRC
902017-02-15T05:32:38 *** goksinen has joined #bitcoin-core-dev
912017-02-15T05:36:57 *** goksinen has quit IRC
922017-02-15T05:39:47 *** Alina-malina_ has quit IRC
932017-02-15T05:39:47 *** Alina-malina_ has joined #bitcoin-core-dev
942017-02-15T05:40:07 *** Alina-malina_ is now known as Alina-malina
952017-02-15T06:51:55 *** lclc has joined #bitcoin-core-dev
962017-02-15T07:09:58 <bitcoin-git> [bitcoin] CryptAxe opened pull request #9763: [Trivial] Update comments referencing main.cpp (master...comments) https://github.com/bitcoin/bitcoin/pull/9763
972017-02-15T07:21:28 *** jtimon has quit IRC
982017-02-15T07:25:59 *** jonasschnelli has quit IRC
992017-02-15T07:25:59 *** jonasschnelli has joined #bitcoin-core-dev
1002017-02-15T07:35:07 *** BashCo has quit IRC
1012017-02-15T07:36:22 *** Chris_Stewart_5 has quit IRC
1022017-02-15T07:44:40 *** Ylbam has joined #bitcoin-core-dev
1032017-02-15T08:01:43 *** BashCo has joined #bitcoin-core-dev
1042017-02-15T08:02:50 *** BashCo_ has joined #bitcoin-core-dev
1052017-02-15T08:06:58 *** BashCo has quit IRC
1062017-02-15T08:09:17 *** bakunin has joined #bitcoin-core-dev
1072017-02-15T08:13:10 <bakunin> hiho, just wanted to say thank you, bitcoin core developers, you are doing an outstanding job.-Thank you very much.
1082017-02-15T08:14:44 *** LeMiner has quit IRC
1092017-02-15T08:16:52 *** LeMiner has joined #bitcoin-core-dev
1102017-02-15T08:22:10 *** windsok has quit IRC
1112017-02-15T08:23:08 *** Guyver2 has joined #bitcoin-core-dev
1122017-02-15T08:23:39 *** windsok has joined #bitcoin-core-dev
1132017-02-15T08:24:26 <wump> good to hear that bakunin
1142017-02-15T08:24:31 *** wump is now known as wumpus
1152017-02-15T08:28:17 <bitcoin-git> [bitcoin] laanwj pushed 2 new commits to master: https://github.com/bitcoin/bitcoin/compare/a441db01b527...4c69d683f22a
1162017-02-15T08:28:18 <bitcoin-git> bitcoin/master a47da4b practicalswift: Use z = std::max(x - y, 0); instead of z = x - y; if (z < 0) z = 0;
1172017-02-15T08:28:19 <bitcoin-git> bitcoin/master 4c69d68 Wladimir J. van der Laan: Merge #9553: Use z = std::max(x - y, 0) instead of z = x - y; if (z < 0) z = 0;...
1182017-02-15T08:28:32 <bitcoin-git> [bitcoin] laanwj closed pull request #9553: Use z = std::max(x - y, 0) instead of z = x - y; if (z < 0) z = 0; (master...std-max) https://github.com/bitcoin/bitcoin/pull/9553
1192017-02-15T08:46:25 *** bakunin has quit IRC
1202017-02-15T08:47:25 *** Guyver2 has quit IRC
1212017-02-15T09:38:03 *** whphhg has quit IRC
1222017-02-15T09:49:04 *** MarcoFalke has joined #bitcoin-core-dev
1232017-02-15T10:12:11 *** MarcoFalke has quit IRC
1242017-02-15T10:14:08 <bitcoin-git> [bitcoin] laanwj pushed 3 new commits to master: https://github.com/bitcoin/bitcoin/compare/4c69d683f22a...d8e8b06bd065
1252017-02-15T10:14:09 <bitcoin-git> bitcoin/master a58370e Russell Yanofsky: Dedup nTimeFirstKey update logic...
1262017-02-15T10:14:09 <bitcoin-git> bitcoin/master a80f98b Russell Yanofsky: Use importmulti timestamp when importing watch only keys...
1272017-02-15T10:14:10 <bitcoin-git> bitcoin/master d8e8b06 Wladimir J. van der Laan: Merge #9108: Use importmulti timestamp when importing watch only keys (on top of #9682)...
1282017-02-15T10:14:16 <bitcoin-git> [bitcoin] laanwj closed pull request #9108: Use importmulti timestamp when importing watch only keys (on top of #9682) (master...watchtime) https://github.com/bitcoin/bitcoin/pull/9108
1292017-02-15T10:16:38 <Victorsueca> ^ about that, is the function to remove watch-only addresses from your wallet coming in the near future?
1302017-02-15T10:35:44 *** chjj has joined #bitcoin-core-dev
1312017-02-15T10:38:45 <bitcoin-git> [bitcoin] laanwj opened pull request #9764: wallet: Prevent "overrides a member function but is not marked 'override'" warnings (master...2017_02_wallet_inconsistent_missing_override) https://github.com/bitcoin/bitcoin/pull/9764
1322017-02-15T10:46:56 *** chjj has quit IRC
1332017-02-15T10:56:11 *** justanotheruser has quit IRC
1342017-02-15T10:58:20 <wumpus> Victorsueca: depends on whether someone picks up https://github.com/bitcoin/bitcoin/pull/5525
1352017-02-15T10:58:58 <wumpus> Victorsueca: it needs rebase and needs tests
1362017-02-15T11:24:44 *** BashCo has joined #bitcoin-core-dev
1372017-02-15T11:24:47 *** Anduck has quit IRC
1382017-02-15T11:24:57 *** Anduck has joined #bitcoin-core-dev
1392017-02-15T11:27:08 *** BashCo_ has quit IRC
1402017-02-15T11:30:16 *** BashCo has quit IRC
1412017-02-15T11:54:03 *** BashCo has joined #bitcoin-core-dev
1422017-02-15T11:58:49 *** BashCo_ has joined #bitcoin-core-dev
1432017-02-15T12:01:38 *** BashCo has quit IRC
1442017-02-15T12:20:17 *** Kexkey has joined #bitcoin-core-dev
1452017-02-15T12:35:55 *** chjj has joined #bitcoin-core-dev
1462017-02-15T12:39:40 <BlueMatt> morcos: ahh, so you're saying people will import with dates that dont match block time, yea, ok, that sucks
1472017-02-15T12:40:03 <BlueMatt> would be nice if we had been using some mtp-or-hours-back rule to generate the birthdays to begin with :/
1482017-02-15T12:40:04 <BlueMatt> oh well
1492017-02-15T12:52:07 *** BashCo has joined #bitcoin-core-dev
1502017-02-15T12:54:12 *** BashCo_ has quit IRC
1512017-02-15T12:55:52 *** BashCo_ has joined #bitcoin-core-dev
1522017-02-15T12:56:09 <morcos> BlueMatt: ryanofsky suggested changing the way the wallet generates the key birthday to be MTP... i couldn't immediately see why that would be a problem.., but regardless i think we need the buffer for now
1532017-02-15T12:56:40 <BlueMatt> yes, well if users are going to get key birthdays from generated birthdays from old wallets then we're gonna need a buffer as long as they do that
1542017-02-15T12:57:18 <wumpus> in retrospect we should have used birth block number instead of birthdate
1552017-02-15T12:57:30 <BlueMatt> yes
1562017-02-15T12:57:42 <BlueMatt> usability issues with that if you're not syned, but, yes
1572017-02-15T12:57:49 <wumpus> the drawback is that then it's no longer possible to generate keys on hardware that has no block chain access
1582017-02-15T12:58:02 <wumpus> well no big one in that case, it'd just make the birth block a bit earlier
1592017-02-15T12:58:15 <jonasschnelli> but keys have timestamps as birthdays
1602017-02-15T12:58:28 *** BashCo has quit IRC
1612017-02-15T12:58:33 <wumpus> yes, it's no longer possible to do that, that's why I said in retrospect
1622017-02-15T12:59:12 <wumpus> keeping a safety margin of 2 hours seems prudent to me
1632017-02-15T12:59:23 <BlueMatt> indeed
1642017-02-15T12:59:27 <jonasschnelli> Yes.
1652017-02-15T12:59:48 <jonasschnelli> I also mentioned this in the initial importmulti PR IIRC
1662017-02-15T13:00:33 *** BashCo has joined #bitcoin-core-dev
1672017-02-15T13:00:47 <jonasschnelli> https://github.com/bitcoin/bitcoin/pull/7551#discussion_r80929574
1682017-02-15T13:02:57 *** BashCo_ has quit IRC
1692017-02-15T13:09:45 *** BashCo_ has joined #bitcoin-core-dev
1702017-02-15T13:13:08 *** BashCo has quit IRC
1712017-02-15T13:18:09 *** BashCo has joined #bitcoin-core-dev
1722017-02-15T13:18:15 <bitcoin-git> [bitcoin] NicolasDorier closed pull request #8460: Parametrize buried soft fork in regtest and refactor (master...buriedsf) https://github.com/bitcoin/bitcoin/pull/8460
1732017-02-15T13:21:23 *** BashCo_ has quit IRC
1742017-02-15T13:23:56 <bitcoin-git> [bitcoin] sdaftuar opened pull request #9765: Harden against mistakes handling invalid blocks (master...fix-checkblock-call) https://github.com/bitcoin/bitcoin/pull/9765
1752017-02-15T13:33:22 *** kadoban has joined #bitcoin-core-dev
1762017-02-15T13:42:59 *** justanotheruser has joined #bitcoin-core-dev
1772017-02-15T13:52:36 *** chjj has quit IRC
1782017-02-15T14:01:38 *** justanotheruser has quit IRC
1792017-02-15T14:07:10 *** lclc has quit IRC
1802017-02-15T14:09:00 *** isle2983 has quit IRC
1812017-02-15T14:09:28 <bitcoin-git> [bitcoin] laanwj pushed 2 new commits to master: https://github.com/bitcoin/bitcoin/compare/d8e8b06bd065...476cc47da084
1822017-02-15T14:09:28 <bitcoin-git> bitcoin/master 4b6cccc Jonas Schnelli: Selectively suppress deprecation warnings
1832017-02-15T14:09:29 <bitcoin-git> bitcoin/master 476cc47 Wladimir J. van der Laan: Merge #9758: Selectively suppress deprecation warnings...
1842017-02-15T14:09:51 <bitcoin-git> [bitcoin] laanwj closed pull request #9758: Selectively suppress deprecation warnings (master...2017/02/deprac_warns) https://github.com/bitcoin/bitcoin/pull/9758
1852017-02-15T14:10:58 *** goksinen has joined #bitcoin-core-dev
1862017-02-15T14:11:13 *** isle2983 has joined #bitcoin-core-dev
1872017-02-15T14:12:27 *** BashCo_ has joined #bitcoin-core-dev
1882017-02-15T14:15:28 *** BashCo has quit IRC
1892017-02-15T14:19:43 *** chjj has joined #bitcoin-core-dev
1902017-02-15T14:26:37 *** laurentmt has joined #bitcoin-core-dev
1912017-02-15T14:47:47 *** Giszmo has joined #bitcoin-core-dev
1922017-02-15T14:48:38 *** eminogrande has joined #bitcoin-core-dev
1932017-02-15T15:00:44 *** Kexkey has quit IRC
1942017-02-15T15:02:49 *** jtimon has joined #bitcoin-core-dev
1952017-02-15T15:05:12 *** eminogrande has quit IRC
1962017-02-15T15:05:24 *** eminogrande has joined #bitcoin-core-dev
1972017-02-15T15:06:15 <bitcoin-git> [bitcoin] jnewbery opened pull request #9766: Add --exclude option to rpc-tests.py (master...rpctestexclude) https://github.com/bitcoin/bitcoin/pull/9766
1982017-02-15T15:10:17 *** lclc has joined #bitcoin-core-dev
1992017-02-15T15:13:00 *** lejit has quit IRC
2002017-02-15T15:21:51 *** emzy_ is now known as emzy
2012017-02-15T15:21:57 *** emzy has joined #bitcoin-core-dev
2022017-02-15T15:25:28 *** Guyver2 has joined #bitcoin-core-dev
2032017-02-15T15:28:44 *** laurentmt1 has joined #bitcoin-core-dev
2042017-02-15T15:29:38 <bitcoin-git> [bitcoin] laanwj pushed 2 new commits to master: https://github.com/bitcoin/bitcoin/compare/476cc47da084...7a93af8340d9
2052017-02-15T15:29:38 <bitcoin-git> bitcoin/master 9acf25c Russell Yanofsky: Return error when importmulti called with invalid address....
2062017-02-15T15:29:39 <bitcoin-git> bitcoin/master 7a93af8 Wladimir J. van der Laan: Merge #9756: Return error when importmulti called with invalid address....
2072017-02-15T15:30:00 <bitcoin-git> [bitcoin] laanwj closed pull request #9756: Return error when importmulti called with invalid address. (master...pr/multiaddr) https://github.com/bitcoin/bitcoin/pull/9756
2082017-02-15T15:31:48 *** laurentmt has quit IRC
2092017-02-15T15:31:48 *** laurentmt1 is now known as laurentmt
2102017-02-15T15:40:12 *** Chris_Stewart_5 has joined #bitcoin-core-dev
2112017-02-15T15:44:58 *** tan1k has joined #bitcoin-core-dev
2122017-02-15T15:50:50 *** Chris_Stewart_5 has quit IRC
2132017-02-15T15:51:15 *** Chris_Stewart_5 has joined #bitcoin-core-dev
2142017-02-15T15:53:03 *** Guest74962 has joined #bitcoin-core-dev
2152017-02-15T15:53:05 <Guest74962> hey
2162017-02-15T15:53:19 *** Aaronvan_ has quit IRC
2172017-02-15T15:55:27 *** Guest74962 has left #bitcoin-core-dev
2182017-02-15T15:57:10 *** AaronvanW has joined #bitcoin-core-dev
2192017-02-15T15:59:23 *** senior1 has joined #bitcoin-core-dev
2202017-02-15T15:59:42 *** senior1 has joined #bitcoin-core-dev
2212017-02-15T15:59:49 *** senior2 has joined #bitcoin-core-dev
2222017-02-15T15:59:54 *** senior1 has quit IRC
2232017-02-15T15:59:56 *** senior2 has quit IRC
2242017-02-15T16:00:07 *** senior1 has joined #bitcoin-core-dev
2252017-02-15T16:00:13 *** senior1 has left #bitcoin-core-dev
2262017-02-15T16:00:32 *** senior1 has joined #bitcoin-core-dev
2272017-02-15T16:01:36 *** lclc has quit IRC
2282017-02-15T16:06:52 *** laurentmt1 has joined #bitcoin-core-dev
2292017-02-15T16:08:00 *** laurentmt has quit IRC
2302017-02-15T16:08:00 *** laurentmt1 is now known as laurentmt
2312017-02-15T16:08:19 *** wasi_ has quit IRC
2322017-02-15T16:10:07 *** senior1 has joined #bitcoin-core-dev
2332017-02-15T16:17:29 <morcos> wumpus: can we powwow on what we want to do about importmulti and pruned nodes?
2342017-02-15T16:17:55 <morcos> you mentioned its the same issues as importwallet, but that is just disabled for pruned nodes, importmulti isn't
2352017-02-15T16:18:11 *** senior1 has left #bitcoin-core-dev
2362017-02-15T16:18:26 *** senior1 has joined #bitcoin-core-dev
2372017-02-15T16:18:32 <wumpus> importmulti should ideally work when using timestamps more recent than what is pruned
2382017-02-15T16:18:46 *** wasi_ has joined #bitcoin-core-dev
2392017-02-15T16:18:51 <morcos> i think right now it'll mostly silently fail if you importmulti with a key timestamp before your earliest on disk block... (looks to me like ReadBlockFromDisk will fail and error to debug log, but rescan should keep chugging along until it finds blocks)
2402017-02-15T16:19:00 <morcos> right, agreed.
2412017-02-15T16:19:08 <wumpus> if that is broken then I'd say disable it for 0.14 and worry about it later
2422017-02-15T16:19:09 <morcos> but important enough to fix that for 0.14?
2432017-02-15T16:19:16 <morcos> ah ok, that was the question
2442017-02-15T16:19:29 <morcos> so just disable importmulti for pruned nodes altogether for 0.14, then fix properly later
2452017-02-15T16:19:42 *** senior1 has quit IRC
2462017-02-15T16:19:46 <morcos> that seems reasonable to me.. gmaxwell or sipa any serious objections?
2472017-02-15T16:20:07 <wumpus> my mention about the problem being the same as for importwallet was about the 2-hour grace period, we had that for importwallet and I think it should be the same for importmulti
2482017-02-15T16:21:14 <wumpus> no matter how that interacts with nodes that have just pruned those two hours, it should just be documented that the grace period is there and blocks should exist for its duration until now
2492017-02-15T16:21:38 <wumpus> but anyhow if importmulti doesn't even throw an error when there's not enough block data then disabling it would be better
2502017-02-15T16:21:48 <morcos> right.. it's a separate edge case to be solved later if your last non-pruned block falls in the grace period.
2512017-02-15T16:21:49 <wumpus> I'd assume the ReadBlockFromDisk error would not be ignored
2522017-02-15T16:22:21 <morcos> just from a quick read my guess is it does not error, but i have not tried anything...
2532017-02-15T16:22:32 <morcos> yeah i agree thats a problem, b/c you could think you have all the funds and not
2542017-02-15T16:24:32 *** thestrin1puller is now known as thestringpuller
2552017-02-15T16:24:45 *** thestringpuller has joined #bitcoin-core-dev
2562017-02-15T16:25:02 <wumpus> yes, silently missing blocks is a very bad
2572017-02-15T16:28:14 <wumpus> important disctinction: importprivkey is only disabled in pruning mode when a rescan is requested
2582017-02-15T16:28:46 <wumpus> probably should be the same for importmulti then, it should bark when any keys are not 'now', otherwise it'd lose functionality compared to importprivkey
2592017-02-15T16:34:19 *** arubi has quit IRC
2602017-02-15T16:40:03 *** arubi has joined #bitcoin-core-dev
2612017-02-15T16:46:08 *** abpa has joined #bitcoin-core-dev
2622017-02-15T16:52:33 <bitcoin-git> [bitcoin] jaladin1222 opened pull request #9767: 0.9 (master...0.9) https://github.com/bitcoin/bitcoin/pull/9767
2632017-02-15T16:58:41 *** lclc has joined #bitcoin-core-dev
2642017-02-15T17:03:37 *** laurentmt1 has joined #bitcoin-core-dev
2652017-02-15T17:05:10 *** laurentmt has quit IRC
2662017-02-15T17:05:11 *** laurentmt1 is now known as laurentmt
2672017-02-15T17:31:37 <bitcoin-git> [bitcoin] sipa closed pull request #9767: 0.9 (master...0.9) https://github.com/bitcoin/bitcoin/pull/9767
2682017-02-15T17:48:26 <bitcoin-git> [bitcoin] jnewbery opened pull request #9768: [qa] [WIP] Add logging to test_framework.py (master...rpctestlogging) https://github.com/bitcoin/bitcoin/pull/9768
2692017-02-15T17:50:34 *** laurentmt has quit IRC
2702017-02-15T17:57:05 *** Chris_Stewart_5 has quit IRC
2712017-02-15T17:57:32 *** Arvidt_ has left #bitcoin-core-dev
2722017-02-15T17:58:58 <gmaxwell> wumpus: it should only be disabled where the rescan would scan pruned blocks... that is really the only utility in having the timestamps at all.
2732017-02-15T17:59:33 *** Chris_Stewart_5 has joined #bitcoin-core-dev
2742017-02-15T18:00:18 *** BashCo_ has quit IRC
2752017-02-15T18:00:55 *** BashCo has joined #bitcoin-core-dev
2762017-02-15T18:05:18 *** BashCo has quit IRC
2772017-02-15T18:19:05 *** BashCo has joined #bitcoin-core-dev
2782017-02-15T18:24:40 *** Kexkey has joined #bitcoin-core-dev
2792017-02-15T18:31:07 *** eminogrande has quit IRC
2802017-02-15T18:52:54 *** Chris_Stewart_5 has quit IRC
2812017-02-15T18:56:44 <ryanofsky> gmaxwell, are you talking about a preemptive check? would there be any problem with importing the keys then throwing an exception saying import succeeded, but some blocks were not present and some transactions might be missing?
2822017-02-15T18:58:29 <ryanofsky> also i don't understand what's wrong with the suggestion from wumpus to just forbid rescan=True on pruned nodes
2832017-02-15T18:59:15 <gmaxwell> Because it guts the utility of pruning, when you lose major functionality outright just because you were so foolish as to not waste 100GB of diskspace.
2842017-02-15T18:59:38 <gmaxwell> Esp because rescans of the whole chain take hours, so it's not like they were even all that usable (except in emergencies) in any case.
2852017-02-15T18:59:56 <gmaxwell> As far as after the fact, I think that would be way less bad than denying it completely.
2862017-02-15T19:00:39 <gmaxwell> it would be best to deny it first but the way importmulti is constructed makes that harder to implement.
2872017-02-15T19:00:58 <ryanofsky> oh ok, i got that part, i didn't understand what you were saying about the defeating the "utility of timestamps"
2882017-02-15T19:03:27 *** goksinen has quit IRC
2892017-02-15T19:06:44 *** goksinen has joined #bitcoin-core-dev
2902017-02-15T19:10:29 <gmaxwell> Bascially the only good timestamps do is that they let you import a key without reading all the blocks. (scanning extra blocks is something no one would care about.) If pruning is prevented then you do save some rescan time, but still take on the 100GB of cost, -- this is a major barrier for people doing their own payment processing on VPSes.
2912017-02-15T19:11:32 <gmaxwell> The workflow there is that they have a payment processing front end which generates keys using BIP32 public derrivation, and then imports the addresses. Because of contingencies around multiple nodes and restarts they may need to rescan a bit. The imported keys are used to watch for payments so they can display status to the user.
2922017-02-15T19:11:49 <gmaxwell> The actual private keys are not put onto the VPS systems, for obvious reasons. :)
2932017-02-15T19:13:13 <gmaxwell> The alternative to this setup is to use a third party payment processor. But this has the downside of introducing trust where it could be avoided, and all the major processors are known to capricious shut down merchants or block customers on the basis of flimsy risk (that they piss off banks or regulators) analysis.
2942017-02-15T19:13:34 <gmaxwell> (and, of course, the systemic risk created by those processors holding large amounts of customer funds).
2952017-02-15T19:16:29 *** AaronvanW has quit IRC
2962017-02-15T19:17:33 <bitcoin-git> [bitcoin] jnewbery opened pull request #9770: Allow maxsigcachesize to be zero (master...sigcachemaxsize) https://github.com/bitcoin/bitcoin/pull/9770
2972017-02-15T19:17:43 <bitcoin-git> [bitcoin] jnewbery closed pull request #9770: Allow maxsigcachesize to be zero (master...sigcachemaxsize) https://github.com/bitcoin/bitcoin/pull/9770
2982017-02-15T19:18:06 <ryanofsky> ok this use both requires import timestamps and pruning, makes sense
2992017-02-15T19:18:16 *** AaronvanW has joined #bitcoin-core-dev
3002017-02-15T19:18:16 *** AaronvanW has joined #bitcoin-core-dev
3012017-02-15T19:19:56 <gmaxwell> the other somewhat related bug is that even when you don't import a private key, it requires that the wallet be unlocked.
3022017-02-15T19:20:13 <gmaxwell> the same design make it difficult to reject the rpc based on the arguments.
3032017-02-15T19:21:12 <bitcoin-git> [bitcoin] jnewbery reopened pull request #9770: Allow maxsigcachesize to be zero (master...sigcachemaxsize) https://github.com/bitcoin/bitcoin/pull/9770
3042017-02-15T19:38:49 *** goksinen has quit IRC
3052017-02-15T19:40:04 *** goksinen has joined #bitcoin-core-dev
3062017-02-15T19:49:45 *** wasi_ is now known as wasi
3072017-02-15T20:46:08 <cfields> gmaxwell: I'm going to add a little blurb about the net speedup in the changelog. I think i remember you mentioning you'd prefer to see some sort of "validation improvements" or "ibd speedups" section, which may be a more useful way to describe it?
3082017-02-15T21:14:41 <morcos> Should we be making it so POTENTIAL DEADLOCK DETECTED's are elimintated or is it impossible to eliminate them all
3092017-02-15T21:15:21 <morcos> I was just trackign down what caused one and I think its not an issue, but is there somehwere I should comment what caused it, or should we actually change the code to not lock incorrectly
3102017-02-15T21:15:33 <morcos> i'm just not very familar with DEBUG_LOCKORDER
3112017-02-15T21:16:01 <sipa> i think we can replace it with tsan soon
3122017-02-15T21:16:07 <sipa> and/or helgrind
3132017-02-15T21:16:38 <morcos> This particular case is the loading of a wallet can call Mark Conflicted, which causes cs_wallet then cs_main which is the opposite order it happens elsewhere
3142017-02-15T21:17:11 <sipa> what is the other case?
3152017-02-15T21:17:45 <morcos> we have tons of LOCK2(cs_main, cs_wallet)'s
3162017-02-15T21:18:14 <morcos> i was assuming this isn't a problem b/c the wallet loading is done before other threads are spun up, but maybe thats actually not right?
3172017-02-15T21:18:38 <sipa> well the only question is whether those 2 orders can occur simultaneously
3182017-02-15T21:19:27 <morcos> yes thats what i mean, can anything else happen while the wallet is still loading?
3192017-02-15T21:21:16 <morcos> dpesm
3202017-02-15T21:21:19 <morcos> oops
3212017-02-15T21:21:40 <morcos> doesn't look like it i suppose...
3222017-02-15T21:24:23 <sipa> regardless, i think we should have consistent lock orders everywhere
3232017-02-15T21:24:49 <sipa> i'd say all those LOCK2(cs_main, cs_wallet)s can be swapped
3242017-02-15T21:25:03 <morcos> yikes!!
3252017-02-15T21:25:10 <sipa> wallet should call main, not the other way around IMHO
3262017-02-15T21:25:13 <morcos> i bet several of them are reentrant
3272017-02-15T21:25:20 <morcos> main is already held before we get there
3282017-02-15T21:25:38 <sipa> well, that's the issue then
3292017-02-15T21:25:48 <sipa> main shouldn't be held while calling wallet code
3302017-02-15T21:26:21 <morcos> :) we're working on it! see #9725
3312017-02-15T21:26:24 <gribble> https://github.com/bitcoin/bitcoin/issues/9725 | CValidationInterface Cleanups by TheBlueMatt · Pull Request #9725 · bitcoin/bitcoin · GitHub
3322017-02-15T21:27:08 <sipa> great
3332017-02-15T21:29:42 <morcos> hmm
3342017-02-15T21:29:59 <morcos> it assert fails when you have a potential deadlock?
3352017-02-15T21:30:33 <morcos> so this must have been somehow recently introduced?
3362017-02-15T21:41:27 <cfields> mm, I'm going to go ahead and PR some mutex cleanups I've been waiting on for a while. I'm going to start moving the net code towards non-recursive locks. Would you guys prefer a generic LOCK(cs) that takes any mutex, or NON_RECURSIVE_LOCK(cs) as a form of self-documentation?
3372017-02-15T22:00:10 <morcos> so... what should i do about this stupid deadlock warnign for now? seems like it might annoy someone else.. but only obvious fix i could see would be holding cs_main for all of loadwallet. its not immediately apparent to me whether that is bad?
3382017-02-15T22:01:31 <morcos> eh.. i'll move on and make an issue.. if you have a clean wallet, i think the problem goes away
3392017-02-15T22:08:09 <bitcoin-git> [bitcoin] ryanofsky opened pull request #9771: Add missing cs_wallet lock that triggers new lock held assertion (master...pr/loadlock) https://github.com/bitcoin/bitcoin/pull/9771
3402017-02-15T22:19:25 <bitcoin-git> [bitcoin] ryanofsky opened pull request #9773: WIP: Return errors from importmulti if complete rescans are not successful (master...pr/multicheck) https://github.com/bitcoin/bitcoin/pull/9773
3412017-02-15T22:21:41 *** chjj_ has joined #bitcoin-core-dev
3422017-02-15T22:24:53 *** chjj has quit IRC
3432017-02-15T22:31:06 *** schmidty has joined #bitcoin-core-dev
3442017-02-15T22:45:38 <bitcoin-git> [bitcoin] pstratem closed pull request #9762: Add txdetails parameter to getblock. (master...2017-02-14-getblock-includetxs) https://github.com/bitcoin/bitcoin/pull/9762
3452017-02-15T23:14:45 *** Guyver2 has quit IRC
3462017-02-15T23:17:07 *** jnewbery has quit IRC
3472017-02-15T23:19:22 *** chjj_ is now known as chjj
3482017-02-15T23:26:24 *** Chris_Stewart_5 has joined #bitcoin-core-dev