12016-09-20T00:34:55  *** mrkent has quit IRC
  22016-09-20T00:43:04  *** dcousens has quit IRC
  32016-09-20T00:43:48  *** AaronvanW has quit IRC
  42016-09-20T00:44:47  *** dcousens has joined #bitcoin-core-dev
  52016-09-20T00:45:05  <dcousens> jonasschnelli: 8757 tag as wallet?
  62016-09-20T00:45:20  *** mrkent has joined #bitcoin-core-dev
  72016-09-20T00:46:17  *** mrkent has joined #bitcoin-core-dev
  82016-09-20T00:52:05  *** Guest8798 has quit IRC
  92016-09-20T00:52:05  *** Guest8798 has joined #bitcoin-core-dev
 102016-09-20T00:57:28  <dcousens> also, just a crash after a long time running, only non RPC-related debug I have that was 'in the vicinity' is https://gist.githubusercontent.com/dcousens/be26e196b308fb3a8aa6c085a093d1d1/raw/f3e48e9f0283699727f4c2a4fff9f4bb2542a16f/gistfile1.txt
 112016-09-20T00:57:48  <dcousens> not sure if there is enough info worth reporting as an issue
 122016-09-20T00:59:05  <gmaxwell> dcousens: what code is that node running?
 132016-09-20T00:59:36  <dcousens> gmaxwell: master + a ~5 custom RPC calls, though haven't had an issues with them for months
 142016-09-20T01:00:20  <dcousens> will get commit hash, probably a few PRs behind
 152016-09-20T01:01:00  <gmaxwell> I see no crashes on any of my 0.13 or master nodes at least.
 162016-09-20T01:01:06  <dcousens> ddc308068d69c6c9aa629ee3c4ce75e1d1cf08b5 (Sep 8)
 172016-09-20T01:01:44  <dcousens> gmaxwell: I'll keep watch, any debug options worth throwing on to catch if happens again?
 182016-09-20T01:02:00  <gmaxwell> if you have space for it... debug=1 is most helpful.
 192016-09-20T01:02:26  <gmaxwell> also run with daemon=0 in screen so if it crashes and spits anyhting on the console you can catch it.
 202016-09-20T01:02:45  <dcousens> gmaxwell: sure,  I run it in tmux
 212016-09-20T01:02:55  <dcousens> never as daemon anyway
 222016-09-20T01:03:20  <gmaxwell> (running in gdb is an option too, though I believe it will slow it down some)
 232016-09-20T01:03:52  <dcousens> its being used heavily, but not mission critical, so probably will avoid the slow down, for now anyway
 242016-09-20T01:04:46  <dcousens> gmaxwell: cheers, will let you know if it happens again
 252016-09-20T01:05:11  <dcousens> (and isn't an RPC issue)
 262016-09-20T01:55:49  *** jtimon has quit IRC
 272016-09-20T02:00:20  *** mrkent has quit IRC
 282016-09-20T02:10:01  *** jtimon has joined #bitcoin-core-dev
 292016-09-20T02:20:14  <GitHub27> [bitcoin] murtyjones opened pull request #8762: Trivial: Fix typo (master...patch-1) https://github.com/bitcoin/bitcoin/pull/8762
 302016-09-20T02:27:26  *** btcdrak has quit IRC
 312016-09-20T02:49:11  *** moli has quit IRC
 322016-09-20T02:49:38  *** moli has joined #bitcoin-core-dev
 332016-09-20T03:05:16  *** Guest8798 is now known as amiller
 342016-09-20T03:34:14  *** jtimon has quit IRC
 352016-09-20T03:36:25  *** Atomic_zEU0b has joined #bitcoin-core-dev
 362016-09-20T03:38:53  *** DigiByteDev has joined #bitcoin-core-dev
 372016-09-20T03:40:47  *** jl2012 has quit IRC
 382016-09-20T03:58:47  *** Chris_Stewart_5 has quit IRC
 392016-09-20T04:13:02  *** jl2012 has joined #bitcoin-core-dev
 402016-09-20T04:17:16  *** Alopex has quit IRC
 412016-09-20T04:18:22  *** Alopex has joined #bitcoin-core-dev
 422016-09-20T04:37:00  *** kadoban has quit IRC
 432016-09-20T04:42:01  *** Alopex has quit IRC
 442016-09-20T04:43:07  *** Alopex has joined #bitcoin-core-dev
 452016-09-20T04:57:33  *** Giszmo has quit IRC
 462016-09-20T05:18:08  *** harrymm has quit IRC
 472016-09-20T05:23:16  *** Alopex has quit IRC
 482016-09-20T05:24:22  *** Alopex has joined #bitcoin-core-dev
 492016-09-20T05:32:50  *** harrymm has joined #bitcoin-core-dev
 502016-09-20T05:43:52  *** AtashiCon has quit IRC
 512016-09-20T05:43:53  *** paveljanik has quit IRC
 522016-09-20T05:43:53  <GitHub2> [bitcoin] CryptAxe opened pull request #8763: Optionally sweep funds from private key. (master...2751PullRequest) https://github.com/bitcoin/bitcoin/pull/8763
 532016-09-20T06:03:05  *** AtashiCon has joined #bitcoin-core-dev
 542016-09-20T06:05:59  *** Evel-Knievel has quit IRC
 552016-09-20T06:08:39  <gmaxwell> TD-Linux: ::sigh:: yea, 64kB is insanely small. I wonder why its so low.
 562016-09-20T06:12:42  <TD-Linux> gmaxwell, dunno, it doesn't make much sense with a single-user system. probably just very few users of mlock() to complain
 572016-09-20T06:15:15  <gmaxwell> TD-Linux: well it messes up all the realtime multimedia apps, but those just tell you to go radically increase it.
 582016-09-20T06:20:09  <TD-Linux> is fedora the only distro that does it? maybe just post on fedora-devel. what are you trying to mlock, wallet private keys?
 592016-09-20T06:20:41  <gmaxwell> TD-Linux: yes.
 602016-09-20T06:21:28  <TD-Linux> I can also post... what's the max you'd expect to use?
 612016-09-20T06:22:19  <gmaxwell> A large wallet will use about 300kB apparently.
 622016-09-20T06:23:36  <gmaxwell> might want to look at what the ardour people ask users to set.
 632016-09-20T06:32:45  *** btcdrak has joined #bitcoin-core-dev
 642016-09-20T06:33:07  <luke-jr> Gentoo seems to be 64k default also FWIW
 652016-09-20T06:35:00  <gmaxwell> $ ulimit -l
 662016-09-20T06:35:00  <gmaxwell> 64
 672016-09-20T06:35:05  <gmaxwell> ^ sure enough
 682016-09-20T06:43:23  <TD-Linux> looks like fedora has special rules for jackd that give it 4GB
 692016-09-20T06:51:08  *** rubensayshi has joined #bitcoin-core-dev
 702016-09-20T07:37:21  *** GreenIsMyPepper has quit IRC
 712016-09-20T07:37:37  *** GreenIsMyPepper has joined #bitcoin-core-dev
 722016-09-20T07:44:43  *** go1111111 has quit IRC
 732016-09-20T07:50:39  *** Squidicuz has quit IRC
 742016-09-20T07:51:23  *** Squidicuz has joined #bitcoin-core-dev
 752016-09-20T07:53:34  *** DigiByteDev has quit IRC
 762016-09-20T07:58:32  <wumpus> gmaxwell: that's useful to know but IMO an issue for another time and should not hold up progress on 8753. If only 32kb is available the first arena will be 32kb, which means it will at least use everything allotted to it. We coudl add a warning to the log in that case I guess but the only solution is to a) use a distribution with sane defaults b) change the ulimits
 772016-09-20T07:58:46  *** go1111111 has joined #bitcoin-core-dev
 782016-09-20T07:59:37  <jonasschnelli> Anyone interested in reviewing/acking https://github.com/bitcoin/bitcoin/pull/7783 (nested commands for Qt), and https://github.com/bitcoin/bitcoin/pull/8371 (out-of-sync info layer)?
 792016-09-20T07:59:55  <jonasschnelli> Getting them in asap leave enough time to polish them for 0.14
 802016-09-20T08:00:10  <wumpus> I don't know exactly why it uses so much locked memory, I do know that the usage is about 50% less with 8753, so it's on the right track to improving things
 812016-09-20T08:01:02  <gmaxwell> wumpus: I wouldn't consider suggesting delaying that PR.
 822016-09-20T08:01:08  <wumpus> this makes it easier to figure out though, at least more statistics could be addede to getmemoryinfo
 832016-09-20T08:01:19  <wumpus> e.g. number of allocated chunks may be useful
 842016-09-20T08:01:47  <wumpus> could even use standard heap profiling techniques now such as storing the traceback for every alloc
 852016-09-20T08:03:11  <GitHub102> [bitcoin] laanwj pushed 2 new commits to master: https://github.com/bitcoin/bitcoin/compare/d8b4b631c5be...82eacc786d8d
 862016-09-20T08:03:11  <GitHub102> bitcoin/master 0766d1c isle2983: [copyright] add MIT license headers to .sh scripts where missing...
 872016-09-20T08:03:12  <GitHub102> bitcoin/master 82eacc7 Wladimir J. van der Laan: Merge #8700: [copyright] add MIT license headers to .sh scripts where missing...
 882016-09-20T08:03:21  <GitHub147> [bitcoin] laanwj closed pull request #8700: [copyright] add MIT license headers to .sh scripts where missing (master...sh-copyright) https://github.com/bitcoin/bitcoin/pull/8700
 892016-09-20T08:03:22  <gmaxwell> how many keys did your 300k-mlockedmemory wallet have?
 902016-09-20T08:05:07  <wumpus> let me see. gah getwalletinfo is hanging
 912016-09-20T08:05:19  <wumpus> "keypoolsize": 10000,
 922016-09-20T08:05:40  <gmaxwell> well.. 300k ~= 10000*32
 932016-09-20T08:06:18  <wumpus> yes :)
 942016-09-20T08:06:54  <gmaxwell> It would be sad if we successfully get N bytes of locked ram, use it to store X encrypted keys, then when decrypting the decrypted keys end up in non-locked ram.
 952016-09-20T08:07:40  <wumpus> it's not an encrypted wallet
 962016-09-20T08:07:57  <wumpus> but yes that'd be sad, I don't think it's necessary to use locked ram for encrypted keys?
 972016-09-20T08:08:25  <wumpus> I mean it's the same data as is in wallet.dat anyhow
 982016-09-20T08:08:28  <gmaxwell> I'd guess we should only put the master key and unencrypted keys in locked ram. And if the wallet is encrypted then we should need very little locked ram, since we really only need to have one key actually decrypted at a time.
 992016-09-20T08:08:50  <wumpus> this would only be an issue if you have an encrypted disk but not encrypted swap
1002016-09-20T08:08:57  <wumpus> in which case your setup is stupid
1012016-09-20T08:09:23  <gmaxwell> wumpus: I agree. (though also discovered that somehow my swap wasn't encrypted.. just recently)
1022016-09-20T08:09:24  <wumpus> so yes, encrypted keys do not need to be stored in lcoked ram
1032016-09-20T08:09:55  <wumpus> openbsd does that by default since 2005 *ducks*
1042016-09-20T08:10:19  <gmaxwell> wumpus: I've done it by default since before then. :P
1052016-09-20T08:12:21  <wumpus> I was surprised openbsd gives 5MiB or so of locked ram by default to processes, it's not usually known for its lenient ulimits (  https://github.com/bitcoin/bitcoin/blob/master/doc/build-openbsd.md#resource-limits :p)
1062016-09-20T08:12:57  *** GreenIsMyPepper has quit IRC
1072016-09-20T08:14:09  *** GreenIsMyPepper has joined #bitcoin-core-dev
1082016-09-20T08:17:59  *** MarcoFalke has joined #bitcoin-core-dev
1092016-09-20T08:34:50  *** mn3monic has quit IRC
1102016-09-20T08:44:43  *** AaronvanW has joined #bitcoin-core-dev
1112016-09-20T08:44:52  *** DigiByteDev has joined #bitcoin-core-dev
1122016-09-20T08:45:10  *** AaronvanW has quit IRC
1132016-09-20T08:45:10  *** AaronvanW has joined #bitcoin-core-dev
1142016-09-20T08:48:41  *** mrkent has joined #bitcoin-core-dev
1152016-09-20T08:49:32  *** tucenaber has joined #bitcoin-core-dev
1162016-09-20T08:53:18  *** mrkent has quit IRC
1172016-09-20T08:57:34  <wumpus> "chunks_used": 10190,    yes, the steady state seems a chunk for each key
1182016-09-20T09:01:08  *** DigiByteDev has quit IRC
1192016-09-20T09:01:15  <wumpus> which makes sense. We should change the wallet to not use locked memory for encrypted keys, and it will go to around ~0
1202016-09-20T09:02:20  <wumpus> (well, for encrypted wallets, but that's when you'd expect increased security. For the unencrypted wallet the attacker can grab wallet.dat instead of the swap file)
1212016-09-20T09:03:17  <wumpus> also on platforms such as windows (I think), osx, ubuntu, openbsd it's no problem to use a lot of locked memory. It's mostly just fine
1222016-09-20T09:03:48  <wumpus> (to be clear here I call 512kB 'a lot')
1232016-09-20T09:06:13  *** rebroad has joined #bitcoin-core-dev
1242016-09-20T09:15:04  *** JackH has joined #bitcoin-core-dev
1252016-09-20T09:16:48  <phantomcircuit> wumpus: can you take a look at 8696 ?
1262016-09-20T09:17:04  <phantomcircuit> im blocked on other changes to the CWalletDB stuff
1272016-09-20T09:35:34  *** jannes has joined #bitcoin-core-dev
1282016-09-20T09:40:20  *** Atomic_zEU0b has quit IRC
1292016-09-20T09:48:30  *** rubensayshi has quit IRC
1302016-09-20T10:01:59  *** rubensayshi has joined #bitcoin-core-dev
1312016-09-20T10:02:16  <wumpus> jonasschnelli, phantomcircuit will take a look at those
1322016-09-20T10:03:49  *** mang0_ has joined #bitcoin-core-dev
1332016-09-20T10:04:14  <mang0_> Does anyone know the best place to get a small fast bitcoin loan?
1342016-09-20T10:04:38  <mang0_> whoops wrong channel
1352016-09-20T10:04:41  *** mang0_ has quit IRC
1362016-09-20T10:25:59  *** mrkent has joined #bitcoin-core-dev
1372016-09-20T10:37:22  *** laurentmt has joined #bitcoin-core-dev
1382016-09-20T10:40:45  *** JackH has quit IRC
1392016-09-20T10:46:14  <GitHub13> [bitcoin] laanwj pushed 2 new commits to master: https://github.com/bitcoin/bitcoin/compare/82eacc786d8d...02ac669730c0
1402016-09-20T10:46:15  <GitHub13> bitcoin/master faddd62 MarcoFalke: init: Get rid of some ENABLE_WALLET
1412016-09-20T10:46:15  <GitHub13> bitcoin/master 02ac669 Wladimir J. van der Laan: Merge #8760: [init] Get rid of some ENABLE_WALLET...
1422016-09-20T10:46:28  <GitHub78> [bitcoin] laanwj closed pull request #8760: [init] Get rid of some ENABLE_WALLET (master...Mf1609-walletInitGuard) https://github.com/bitcoin/bitcoin/pull/8760
1432016-09-20T10:47:51  <GitHub55> [bitcoin] laanwj pushed 5 new commits to master: https://github.com/bitcoin/bitcoin/compare/02ac669730c0...a1f8d3ed9569
1442016-09-20T10:47:52  <GitHub55> bitcoin/master 59adc86 Patrick Strateman: Add CWallet::ListAccountCreditDebit...
1452016-09-20T10:47:52  <GitHub55> bitcoin/master d2e678d Patrick Strateman: Add CWallet::ReorderTransactions and use in accounting_tests.cpp
1462016-09-20T10:47:53  <GitHub55> bitcoin/master 02e2a81 Patrick Strateman: Remove pwalletdb parameter from CWallet::AddAccountingEntry
1472016-09-20T10:48:02  <GitHub182> [bitcoin] laanwj closed pull request #8696: [Wallet] Remove last external reference to CWalletDB (master...2016-09-09-cwallet-listaccountcreditdebit) https://github.com/bitcoin/bitcoin/pull/8696
1482016-09-20T10:50:02  <GitHub169> [bitcoin] laanwj pushed 4 new commits to 0.13: https://github.com/bitcoin/bitcoin/compare/4731623777ab...8d9e8adc05f4
1492016-09-20T10:50:03  <GitHub21> [bitcoin] laanwj closed pull request #8744: [0.13.1] qa Backports (0.13...Mf1609-qaBackports) https://github.com/bitcoin/bitcoin/pull/8744
1502016-09-20T10:50:03  <GitHub169> bitcoin/0.13 63462c2 whythat: [qa] remove root test directory for RPC tests...
1512016-09-20T10:50:04  <GitHub169> bitcoin/0.13 ae8c7df MarcoFalke: [qa] create_cache: Delete temp dir when done...
1522016-09-20T10:50:04  <GitHub169> bitcoin/0.13 d6ebe13 MarcoFalke: [qa] Refactor RPCTestHandler to prevent TimeoutExpired...
1532016-09-20T10:50:54  <phantomcircuit> wumpus: ty
1542016-09-20T10:52:45  *** dermoth has joined #bitcoin-core-dev
1552016-09-20T10:56:07  *** Chris_Stewart_5 has joined #bitcoin-core-dev
1562016-09-20T10:56:52  *** mrkent has quit IRC
1572016-09-20T10:57:33  *** mrkent has joined #bitcoin-core-dev
1582016-09-20T11:03:00  *** cryptapus has joined #bitcoin-core-dev
1592016-09-20T11:03:00  *** cryptapus has joined #bitcoin-core-dev
1602016-09-20T11:09:21  *** JackH has joined #bitcoin-core-dev
1612016-09-20T11:20:42  *** rebroad has quit IRC
1622016-09-20T11:33:46  *** murch has joined #bitcoin-core-dev
1632016-09-20T11:44:24  *** cdecker has quit IRC
1642016-09-20T11:56:39  *** cryptapus_ has joined #bitcoin-core-dev
1652016-09-20T12:00:55  *** cryptapus has quit IRC
1662016-09-20T12:00:57  <wumpus> jonasschnelli: #7783 seems to work great, can't seem to break it on first try at least :)
1672016-09-20T12:01:21  <jonasschnelli> wumpus: heh. Yes. It's surprisingly stable. :)
1682016-09-20T12:03:14  <wumpus> created some monster expressions using echo/echon and it works
1692016-09-20T12:03:21  *** cryptapus_ is now known as cryptapus
1702016-09-20T12:04:03  *** laurentmt has quit IRC
1712016-09-20T12:05:49  <jonasschnelli> wumpus: It would also be not-very-complicate to add some expressions (basic math ops). Like sendtoaddress(getnewaddress(), getbalance()/2)
1722016-09-20T12:05:57  <jonasschnelli> but meh
1732016-09-20T12:06:03  *** JackH has quit IRC
1742016-09-20T12:06:25  <wumpus> nah, that's probably going too far
1752016-09-20T12:06:42  <wumpus> may as well add a complete js interpreter then and ne done with it *ducks*
1762016-09-20T12:06:49  <jonasschnelli> heh
1772016-09-20T12:07:06  <jonasschnelli> we could include node.js ...
1782016-09-20T12:07:18  <jonasschnelli> and migrate the whole Qt stack to CSS/HTML with node.js.
1792016-09-20T12:07:38  <jonasschnelli> And.. while we'r at it,... include an app store
1802016-09-20T12:09:09  <wumpus> hahahah there have actually been people proposing that last thing
1812016-09-20T12:09:38  <wumpus> 'wallet applications' that run inside bitcoin core
1822016-09-20T12:09:45  <wumpus> absolutely crazy from a security perspective
1832016-09-20T12:10:35  <jonasschnelli> heh. Yes. Full access to privat keys... the plugin may want to do a sweep operation. :)
1842016-09-20T12:10:37  <wumpus> even having a semi-capable interpreter in the address space makes exploitation easier, but ok, that's pretty far-out. But explicitly including a sandbox for third-party code in your wallet is plain sick.
1852016-09-20T12:10:59  *** Chris_Stewart_5 has quit IRC
1862016-09-20T12:11:19  <wumpus> yes :)
1872016-09-20T12:12:26  *** arowser has quit IRC
1882016-09-20T12:12:40  *** Chris_Stewart_5 has joined #bitcoin-core-dev
1892016-09-20T12:12:41  *** arowser has joined #bitcoin-core-dev
1902016-09-20T12:13:17  <wumpus> from a technical point of view it'd be really easy to do, just link qtscript/qjsengine
1912016-09-20T12:15:13  <wumpus> anyhow ACK on #7783
1922016-09-20T12:15:57  <GitHub67> [bitcoin] laanwj pushed 2 new commits to master: https://github.com/bitcoin/bitcoin/compare/a1f8d3ed9569...4335d5a41bc2
1932016-09-20T12:15:58  <GitHub67> bitcoin/master 1586044 Jonas Schnelli: [Qt] RPC-Console: support nested commands and simple value queries...
1942016-09-20T12:15:58  <GitHub67> bitcoin/master 4335d5a Wladimir J. van der Laan: Merge #7783: [Qt] RPC-Console: support nested commands and simple value queries...
1952016-09-20T12:16:02  <GitHub141> [bitcoin] laanwj closed pull request #7783: [Qt] RPC-Console: support nested commands and simple value queries (master...2016/04/qt_console_nested) https://github.com/bitcoin/bitcoin/pull/7783
1962016-09-20T12:29:20  *** mrkent has quit IRC
1972016-09-20T12:30:15  *** mrkent has joined #bitcoin-core-dev
1982016-09-20T12:35:37  *** rubensayshi has quit IRC
1992016-09-20T12:44:13  *** To7 has quit IRC
2002016-09-20T12:51:40  *** achow101 has quit IRC
2012016-09-20T12:54:23  *** YOU-JI has joined #bitcoin-core-dev
2022016-09-20T12:54:45  *** rubensayshi has joined #bitcoin-core-dev
2032016-09-20T12:57:30  <GitHub129> [bitcoin] jonasschnelli opened pull request #8764: [Wallet] get rid of pwalletMain, add simple CWallets infrastructure (master...2016/09/wallet_pointer) https://github.com/bitcoin/bitcoin/pull/8764
2042016-09-20T13:03:08  *** cryptapus has quit IRC
2052016-09-20T13:03:34  *** cryptapus has joined #bitcoin-core-dev
2062016-09-20T13:09:05  *** mrkent has quit IRC
2072016-09-20T13:14:45  *** mrkent has joined #bitcoin-core-dev
2082016-09-20T13:19:06  *** user_8488 has joined #bitcoin-core-dev
2092016-09-20T13:19:38  *** jyap has quit IRC
2102016-09-20T13:19:54  *** user_8488 is now known as Guest_49853
2112016-09-20T13:22:02  *** jyap has joined #bitcoin-core-dev
2122016-09-20T13:22:03  *** jyap has joined #bitcoin-core-dev
2132016-09-20T13:27:21  *** tadasv has quit IRC
2142016-09-20T13:34:53  <GitHub161> [bitcoin] jonasschnelli pushed 3 new commits to master: https://github.com/bitcoin/bitcoin/compare/4335d5a41bc2...6052d5091057
2152016-09-20T13:34:54  <GitHub161> bitcoin/master fdf82fb Hampus Sjöberg: Adding method GetTotalSize() to CTransaction...
2162016-09-20T13:34:54  <GitHub161> bitcoin/master c015634 Hampus Sjöberg: qt: Adding transaction size to transaction details window
2172016-09-20T13:34:55  <GitHub161> bitcoin/master 6052d50 Jonas Schnelli: Merge #8672: Qt: Show transaction size in transaction details window...
2182016-09-20T13:35:03  <GitHub165> [bitcoin] jonasschnelli closed pull request #8672: Qt: Show transaction size in transaction details window (master...qttxsizeindetails) https://github.com/bitcoin/bitcoin/pull/8672
2192016-09-20T13:38:33  *** To7 has joined #bitcoin-core-dev
2202016-09-20T13:42:26  *** Guest_49853 has left #bitcoin-core-dev
2212016-09-20T13:45:30  <GitHub175> [bitcoin] laanwj pushed 2 new commits to master: https://github.com/bitcoin/bitcoin/compare/6052d5091057...aff69278063c
2222016-09-20T13:45:30  <GitHub175> bitcoin/master 12a721b Marty Jones: Trivial: Fix typo
2232016-09-20T13:45:31  <GitHub175> bitcoin/master aff6927 Wladimir J. van der Laan: Merge #8762: Trivial: Fix typo...
2242016-09-20T13:45:41  <GitHub164> [bitcoin] laanwj closed pull request #8762: Trivial: Fix typo (master...patch-1) https://github.com/bitcoin/bitcoin/pull/8762
2252016-09-20T13:48:09  <jonasschnelli> https://github.com/bitcoin/bitcoin/blob/master/src/wallet/rpcdump.cpp#L347
2262016-09-20T13:48:24  <jonasschnelli> (doesn't this spin off another wallet flush thread that runs endless)?
2272016-09-20T13:49:06  *** Chris_Stewart_5 has quit IRC
2282016-09-20T13:49:20  <wumpus> eh ...
2292016-09-20T13:49:47  <wumpus> it changes a RPC thread to another wallet flush thread
2302016-09-20T13:49:54  <wumpus> good catch, that's weird
2312016-09-20T13:50:11  *** Chris_Stewart_5 has joined #bitcoin-core-dev
2322016-09-20T13:50:27  <wumpus> (to actually spawn a thread it'd need to do create_thread, but wtf why would it call a thread main function from aRPC call?!?)
2332016-09-20T13:50:44  <wumpus> does that RPC ever return? I'd guess not
2342016-09-20T13:51:39  <jonasschnelli> wumpus: Yes. It grabs the RPC thread and runs endless... right?
2352016-09-20T13:53:04  <jonasschnelli> I guess this occupies a rpc-thread per removeprunedfunds call
2362016-09-20T13:53:49  <wumpus> how the hell can the RPC test 'importprunedfunds.py' pass
2372016-09-20T13:53:55  <wumpus> it exercises that RPC
2382016-09-20T13:54:25  <wumpus> the same is done in IMportPrunedFunds btw
2392016-09-20T13:54:51  <jonasschnelli> I think it can pass because the test does not exceed the amount of parallel PRC threads...
2402016-09-20T13:55:00  <wumpus> but it should never return
2412016-09-20T13:55:01  <jonasschnelli> But... right.. the calls never returns?!
2422016-09-20T13:55:26  <wumpus> anyhow if the author wast rying to flush the wallet database, that's not the right way to do it
2432016-09-20T13:55:28  <jonasschnelli> wumpus: timeout?
2442016-09-20T13:55:37  <wumpus> not sure if explicit flushes are even needed?
2452016-09-20T13:55:43  <wumpus> timeout would be an error
2462016-09-20T13:56:54  <wumpus> it never gets to loopin, see the fOneThread check in the beginning
2472016-09-20T13:56:57  <jonasschnelli> I guess a CWallet::Flush() should also do it.
2482016-09-20T13:57:15  <wumpus> the whole thing is NOP, *apart* from renaming the RPC thread to 'bitcoin-wallet' :')
2492016-09-20T13:57:19  <jonasschnelli> wumpus: Ah. Right. the fOneThread
2502016-09-20T13:57:49  <wumpus> we should delete these, replace with CWallet::Flush if it needs explicit flush
2512016-09-20T13:58:02  <wumpus> although I don't see the logic here, why do some wallet calls need explicit flush and others don't?
2522016-09-20T13:58:23  <sipa> importing a key should flush
2532016-09-20T13:58:50  <sipa> sending a transaction maybe not
2542016-09-20T13:58:51  <jonasschnelli> A flush in pruning funds could be done for privacy ... but also on lots of other locations..
2552016-09-20T13:59:14  <sipa> what does privacy have to do with it?
2562016-09-20T13:59:31  <wumpus> apparently the importprunedfunds doesn't flush, only the removepruned funds tries to (in this weird way)
2572016-09-20T13:59:42  <wumpus> I think this requirement needs to be well-documented
2582016-09-20T13:59:55  <wumpus> when does the wallet need explicit flushes and how to do it
2592016-09-20T14:00:03  <wumpus> this easily confuses me at least
2602016-09-20T14:00:27  <sipa> well whenever you could otherwise lose data, i would say
2612016-09-20T14:00:42  <sipa> *lose funds
2622016-09-20T14:00:54  <wumpus> isn't that on all calls that mutate wallet state?
2632016-09-20T14:01:09  <wumpus> okay, only those that add or generate keys then
2642016-09-20T14:02:05  <sipa> getnewaddress no, but yes when the keypool was topped up
2652016-09-20T14:02:28  <wumpus> right
2662016-09-20T14:02:56  <wumpus> and we should rename THreadFlushWalletDB. It doesn't just flush, it periodically closes and checkpoints (compacts) the database.
2672016-09-20T14:03:32  <sipa> phantomcircuit: are you working on thatm
2682016-09-20T14:03:33  <wumpus> so it's different from cwallet->Flush which (I assume) does a flush
2692016-09-20T14:05:29  <jonasschnelli> ThreadFlushWalletDB directly operates with a file to the walletdb and with the global bitdb
2702016-09-20T14:05:49  <wumpus> yes. It's pretty serious magic.
2712016-09-20T14:06:14  <jonasschnelli> heh
2722016-09-20T14:06:17  <wumpus> we shoud be happy the the fOneTHread check is at the top of it, or I could imagine two of those threads shredding the wallet together
2732016-09-20T14:07:17  <jonasschnelli> I really wonder how a repetitive call `bitdb.CloseDb(strFile);` without a follow up "Open" does work
2742016-09-20T14:07:34  <wumpus> it's automagically reopened on next use
2752016-09-20T14:07:43  <jonasschnelli> hahaha
2762016-09-20T14:07:59  <jonasschnelli> automagically ... lol
2772016-09-20T14:08:21  <wumpus> but yes it's things like this that make people afraid to touch, maybe even look at the wallet code :)
2782016-09-20T14:09:05  <wumpus> I really wonder if it's necessary to be so hacky there, maybe it simply is, but it need 100% more comments
2792016-09-20T14:09:30  <jonasschnelli> Yes. I really dislike the time based periodical flushing...
2802016-09-20T14:10:11  <wumpus> well periodical flushing in itself makes sense I think
2812016-09-20T14:10:33  <wumpus> e.g. important changes flush and sync immediately, like adding new keys to the keypool
2822016-09-20T14:10:45  <wumpus> other changes can potentially wait until shutdown or the next timer
2832016-09-20T14:12:02  <jonasschnelli> I'm not opposed against periodical flushing.. but why time based in a separate thread.
2842016-09-20T14:12:35  <jonasschnelli> Hmm.. maybe it makes sense if the flushing takes a couple of seconds...
2852016-09-20T14:13:12  <wumpus> what is being done is not completely insane, it's just heavily underdocumented, and function names and variable names don't match what is being done, which is deceptive
2862016-09-20T14:13:31  <wumpus> yes, right, for big wallets this periodical compaction can take a while
2872016-09-20T14:13:40  <wumpus> so it makes sense to do it in a thread in that regard
2882016-09-20T14:14:41  <wumpus> but anyhow we should have caught the ThreadFlushWalletDB() in review :)
2892016-09-20T14:15:14  <wumpus> calling into another thread's main function is 100% of times a mistake
2902016-09-20T14:16:02  <wumpus> are you going to submit a pull to remove it or shoudl I?
2912016-09-20T14:16:33  <jonasschnelli> I think removing the line in rpcdump.cpp would probably the best solution for now (should not change the behavior).
2922016-09-20T14:16:36  <jonasschnelli> I can to a PR
2932016-09-20T14:16:37  <wumpus> it's in 0.12 too
2942016-09-20T14:16:39  <jonasschnelli> can do
2952016-09-20T14:16:42  <wumpus> yes
2962016-09-20T14:18:40  *** cryptapus_ has joined #bitcoin-core-dev
2972016-09-20T14:18:40  *** cryptapus_ has joined #bitcoin-core-dev
2982016-09-20T14:22:03  <btcdrak> https://github.com/bitcoin-core/bitcoincore.org/pull/214
2992016-09-20T14:22:05  <GitHub50> [bitcoin] jonasschnelli opened pull request #8765: [Wallet] remove "unused" ThreadFlushWalletDB from removeprunedfunds (master...2016/09/flush_wallet_dump) https://github.com/bitcoin/bitcoin/pull/8765
3002016-09-20T14:22:35  *** cryptapus has quit IRC
3012016-09-20T14:22:42  <GitHub189> [bitcoin] laanwj pushed 2 new commits to master: https://github.com/bitcoin/bitcoin/compare/aff69278063c...1c24d5f63720
3022016-09-20T14:22:43  <GitHub189> bitcoin/master 157254a Suhas Daftuar: Fix broken sendcmpct test in p2p-compactblocks.py...
3032016-09-20T14:22:43  <GitHub189> bitcoin/master 1c24d5f Wladimir J. van der Laan: Merge #8739: [qa] Fix broken sendcmpct test in p2p-compactblocks.py...
3042016-09-20T14:22:56  <GitHub94> [bitcoin] laanwj closed pull request #8739: [qa] Fix broken sendcmpct test in p2p-compactblocks.py (master...fix-cb-test) https://github.com/bitcoin/bitcoin/pull/8739
3052016-09-20T14:24:59  *** tadasv has joined #bitcoin-core-dev
3062016-09-20T14:25:54  *** Guyver2 has joined #bitcoin-core-dev
3072016-09-20T14:27:42  <GitHub110> [bitcoin] unsystemizer opened pull request #8766: Fix URL for Debian 8.5.0 ISO (master...patch-3) https://github.com/bitcoin/bitcoin/pull/8766
3082016-09-20T14:31:47  *** Giszmo has joined #bitcoin-core-dev
3092016-09-20T14:53:51  *** YOU-JI has quit IRC
3102016-09-20T14:57:44  *** achow101 has joined #bitcoin-core-dev
3112016-09-20T15:03:43  *** molz has joined #bitcoin-core-dev
3122016-09-20T15:05:45  *** rubensayshi has quit IRC
3132016-09-20T15:06:40  *** moli has quit IRC
3142016-09-20T15:32:45  <GitHub147> [bitcoin] MarcoFalke opened pull request #8768: init: Get rid of fDisableWallet (master...Mf1609-initDisableWallet) https://github.com/bitcoin/bitcoin/pull/8768
3152016-09-20T15:37:21  <gmaxwell> jonasschnelli: luke-jr submitted a fix for removeprunedfunds a week or so ago, #8687
3162016-09-20T15:39:38  *** Evel-Knievel has joined #bitcoin-core-dev
3172016-09-20T15:40:11  <wumpus> that doesn't seem to be the right solution
3182016-09-20T15:40:19  <wumpus> there's no need to call the compaction there
3192016-09-20T15:53:01  *** cryptapus_ is now known as cryptapus
3202016-09-20T16:48:40  *** paveljanik has joined #bitcoin-core-dev
3212016-09-20T16:48:41  *** paveljanik has joined #bitcoin-core-dev
3222016-09-20T16:57:56  *** laurentmt has joined #bitcoin-core-dev
3232016-09-20T16:58:16  *** laurentmt has quit IRC
3242016-09-20T17:07:50  *** mol has joined #bitcoin-core-dev
3252016-09-20T17:10:50  *** molz has quit IRC
3262016-09-20T17:10:53  *** rogerwilco has joined #bitcoin-core-dev
3272016-09-20T17:20:08  *** mrkent has quit IRC
3282016-09-20T17:21:39  *** Atomic_zEU0b has joined #bitcoin-core-dev
3292016-09-20T17:35:54  *** To7 has quit IRC
3302016-09-20T17:38:49  *** Atomic_zEU0b has quit IRC
3312016-09-20T17:38:59  *** mrkent has joined #bitcoin-core-dev
3322016-09-20T17:42:07  *** Alopex has quit IRC
3332016-09-20T17:43:12  *** Alopex has joined #bitcoin-core-dev
3342016-09-20T17:46:27  *** tucker111 has joined #bitcoin-core-dev
3352016-09-20T17:47:53  <tucker111> I recently updated btc core to 0.13.0 and my btc and transactions history is gone. pls help I ve re scan .dat to make sure and nothing happened
3362016-09-20T17:49:10  <sipa> what did you update from?
3372016-09-20T17:49:43  <tucker111> 0.12
3382016-09-20T17:50:02  *** dcousens has quit IRC
3392016-09-20T17:50:50  *** dcousens has joined #bitcoin-core-dev
3402016-09-20T17:51:59  <tucker111> I have all addresses that I ve generated from the core and the amount I ve sent and received to authenticate...could it help?
3412016-09-20T17:53:15  <sipa> go to the debug window, and type "validateaddress <address>"
3422016-09-20T17:53:31  <sipa> without the "", and with <address> replaced by an address you know is yours
3432016-09-20T17:53:41  <sipa> that you've received funds on
3442016-09-20T18:01:12  <tucker111> do I type it in ''debug log file'' or ''console'' in debug window?
3452016-09-20T18:01:32  <achow101> the console tab on the textbox at the bottom
3462016-09-20T18:04:10  <sipa> console
3472016-09-20T18:06:54  <sipa> does it say ismine true or false?
3482016-09-20T18:11:54  <tucker111> false
3492016-09-20T18:13:21  <tucker111> ''isvalid'' false
3502016-09-20T18:14:42  <sipa> that means you're not pasting an address
3512016-09-20T18:15:02  <sipa> maybe you left off some characters from the beginning or the end?
3522016-09-20T18:16:49  <tucker111> i ve checked the address 3 times
3532016-09-20T18:17:22  <gmaxwell> are you confusing addresses for different cryptocurrencies?
3542016-09-20T18:17:36  <tucker111> all in btcv
3552016-09-20T18:17:45  <tucker111> btc
3562016-09-20T18:18:14  <sipa> it should be
3572016-09-20T18:18:30  <sipa> validateaddress 1bcdefhdhduabandkdbsvag
3582016-09-20T18:18:33  <sipa> for example
3592016-09-20T18:18:39  <sipa> no other characters
3602016-09-20T18:19:15  <tucker111> without  < >?
3612016-09-20T18:19:28  <sipa> yes
3622016-09-20T18:19:53  <sipa> <> just means "replace something here"
3632016-09-20T18:21:11  <gmaxwell> sipa: you should have given a valid example.
3642016-09-20T18:21:32  <tucker111> sometimes theyre important tho
3652016-09-20T18:22:11  <tucker111> ismine false
3662016-09-20T18:26:48  *** To7 has joined #bitcoin-core-dev
3672016-09-20T18:29:34  <tucker111> what does ismine false means?
3682016-09-20T18:30:58  <achow101> it means that the address is not in your wallet
3692016-09-20T18:31:55  <tucker111> can I recover the btc that were on it?
3702016-09-20T18:31:57  <btcdrak> well it means you dont have the private key for it.
3712016-09-20T18:32:15  <tucker111> i do
3722016-09-20T18:32:23  <achow101> you don't have the private key so you can't get any of the Bitcoin that was associated with it
3732016-09-20T18:32:36  <achow101> if it isn't in your wallet, you don't have the key
3742016-09-20T18:33:12  <gmaxwell> tucker111: how did you upgrade?
3752016-09-20T18:40:23  <tucker111> actually I did a tails usb driver and then logged back into win10 and then plane mode was stuck, so I had to do a factory reset after trying millions solutions to get rid of plane mode
3762016-09-20T18:40:33  *** Evel-Knievel has quit IRC
3772016-09-20T18:40:46  <tucker111> and then i download the core again...
3782016-09-20T18:41:12  *** Evel-Knievel has joined #bitcoin-core-dev
3792016-09-20T18:41:19  <sipa> factory reset of what?
3802016-09-20T18:41:22  <achow101> doesn't the factory reset wipe everything?
3812016-09-20T18:41:37  <tucker111> couldnt login to btc core at all to do the transfer before hand
3822016-09-20T18:41:41  <achow101> did Core have to resync when you first started it after reinstall
3832016-09-20T18:41:50  <tucker111> it should
3842016-09-20T18:41:52  <tucker111> no it didnt
3852016-09-20T18:42:17  <tucker111> i m still reinstalling it so i dont know yet
3862016-09-20T18:42:45  <achow101> well if it wiped your wallet.dat file you're screwed
3872016-09-20T18:43:08  <sipa> if you type
3882016-09-20T18:43:12  <sipa> getblockcount
3892016-09-20T18:43:20  <sipa> in the debug console, what does it say?
3902016-09-20T18:44:49  <tucker111> 278384
3912016-09-20T18:45:30  <achow101> well there's your problem
3922016-09-20T18:45:48  <tucker111> i not synced yet thats why?
3932016-09-20T18:46:12  <achow101> yes
3942016-09-20T18:46:25  <tucker111> should i get it afterward?
3952016-09-20T18:46:38  <achow101> right now you are still syncing. If your factory reset wiped the drive, then your wallet.dat and the blockchain got wiped to
3962016-09-20T18:46:49  <achow101> so if your wallet.dat is gone (as it likely is) then you have lost your bitcoin
3972016-09-20T18:47:12  <tucker111> i know but is there a way to go back to the same wallet since i cant use it offline?
3982016-09-20T18:47:24  <tucker111> ok
3992016-09-20T18:47:27  <gmaxwell> tucker111: do you have a backup of your wallet?
4002016-09-20T18:47:40  <tucker111> i couldnt do it!
4012016-09-20T18:51:07  <arubi> tucker111, did you use windows 10 "reset this pc" when you did factory reset?
4022016-09-20T18:51:10  <jonasschnelli> How would it be if we change __rpcfunc(const UniValue& params, bool fHelp) to __rpcfunc(const RPCContext& ctx), where ctx would contain const UniValue& params, bool fHelp along with info about the request
4032016-09-20T18:51:21  <MarcoFalke> Arg, the review divs won't collapse when they are outdated
4042016-09-20T18:51:32  <jonasschnelli> MarcoFalke: Yes. Thats bad.
4052016-09-20T18:51:43  <arubi> tucker111, -> #bitcoin
4062016-09-20T18:51:45  <jonasschnelli> Maybe they get removed when you do another github "review"?
4072016-09-20T18:51:59  <gmaxwell> tucker111: arubi: achow101: etc. please move to #bitcoin for further discussion. :)
4082016-09-20T18:52:22  <jonasschnelli> If we change it to __rpcfunc(const RPCContext& ctx), we could add the request URI, etc. which would allow accessing multiple wallets
4092016-09-20T19:01:28  *** mrkent has quit IRC
4102016-09-20T19:03:07  <gmaxwell> wumpus: if your locked page ulimit is (say) 64k  and you try to lock 256k... will it fail to lock anything at all?  Might be useful to make the mlocks run in smaller increments to at least get some locked.
4112016-09-20T19:03:39  *** achow101 has quit IRC
4122016-09-20T19:04:01  *** achow101 has joined #bitcoin-core-dev
4132016-09-20T19:10:39  <gmaxwell> someone in #bitcoin with a gdb backtrace of a node that keeps crashing during sync running 0.13 ppa... crashing at assert(coins) in CheckTxInputs.
4142016-09-20T19:10:50  <gmaxwell> about to hop a flight, but someone might want to look at it.
4152016-09-20T19:10:58  <gmaxwell> user reports 0.12 was running fine before.
4162016-09-20T19:11:20  <gmaxwell> (user is nou1)
4172016-09-20T19:11:20  <sipa> gmaxwell: his PR checks the locked page limit, and if it nonzero, it first allocates that amount immediately
4182016-09-20T19:11:34  <gmaxwell> oh great!
4192016-09-20T19:11:41  <sipa> so if you have 64kb it will still allocate 64, even if the default is 256
4202016-09-20T19:13:25  *** cryptapus has quit IRC
4212016-09-20T19:17:42  *** tucker111 has quit IRC
4222016-09-20T19:30:45  <GitHub4> [bitcoin] unsystemizer closed pull request #8766: Fix URL for Debian 8.5.0 ISO (master...patch-3) https://github.com/bitcoin/bitcoin/pull/8766
4232016-09-20T19:39:09  <GitHub79> [bitcoin] unsystemizer opened pull request #8769: Trivial: Fix ISO URL, capitalization (master...patch-4) https://github.com/bitcoin/bitcoin/pull/8769
4242016-09-20T19:39:50  <phantomcircuit> sipa: yes that's something i am working on
4252016-09-20T19:40:08  <phantomcircuit> before trying to fix the flush logic im cleaning things up in general
4262016-09-20T20:03:32  *** laurentmt has joined #bitcoin-core-dev
4272016-09-20T20:03:34  *** laurentmt has quit IRC
4282016-09-20T20:10:17  *** xiangfu has quit IRC
4292016-09-20T20:18:44  *** paveljanik has quit IRC
4302016-09-20T20:37:37  *** Guyver2 has quit IRC
4312016-09-20T20:42:31  *** xiangfu has joined #bitcoin-core-dev
4322016-09-20T20:50:03  *** droark has quit IRC
4332016-09-20T21:00:11  *** jtimon has joined #bitcoin-core-dev
4342016-09-20T21:04:07  <GitHub150> [bitcoin] MarcoFalke closed pull request #8286: [doc] Docs and copyright headers bump (master...Mf1607-trivialPre13) https://github.com/bitcoin/bitcoin/pull/8286
4352016-09-20T21:12:08  *** mrkent has joined #bitcoin-core-dev
4362016-09-20T21:47:41  *** echonaut has quit IRC
4372016-09-20T21:47:57  *** echonaut has joined #bitcoin-core-dev
4382016-09-20T21:48:32  *** mrkent has quit IRC
4392016-09-20T21:52:48  *** mrkent has joined #bitcoin-core-dev
4402016-09-20T21:57:19  *** kadoban has joined #bitcoin-core-dev
4412016-09-20T22:01:58  <jtimon> is "functions start with uppercase" still part of our style guidelines? I can't find them (maybe they're now reduced to src/.clang-format )?
4422016-09-20T22:07:56  <sipa> yes it is
4432016-09-20T22:08:16  *** dcousens has quit IRC
4442016-09-20T22:14:53  *** droark has joined #bitcoin-core-dev
4452016-09-20T22:18:20  *** ___tau___ has joined #bitcoin-core-dev
4462016-09-20T22:22:52  <jtimon> sipa: I was asking because ___tau___ and I are discussing the namin in https://github.com/bitcoin/bitcoin/pull/8493/commits/ca1f6622f49cf5653ba074a9392102ef515c9d3c#diff-646fe6309aabacfd041a20dcad10be05R11 and https://github.com/bitcoin/bitcoin/pull/8493/commits/3165f95cc2f60ebf37d073da10640fbfa141b765#diff-c2a099d775bac1dccc5f146a3cda81b8R16
4472016-09-20T22:24:03  <jtimon> he suggested get_ancestor_fn get_ancestor; is clearer, but since we have uppercase for functions as a style rule maybe GetAncestor_fn  GetAncestor; is a good compromise ?
4482016-09-20T22:26:20  <jtimon> nothing urgent, but yeah, AncestorGetter Ancestor; seems less clear
4492016-09-20T22:26:20  <___tau___> Ah, the point was to not drop the verbs from the function names rather than use underscores (that was just an example for explaining)
4502016-09-20T22:28:29  <jtimon> yeah, just saying that it will be free for me to change the names to anything else while I do that, so if there's any other bike-shedding to put in or reject (ie the _fn ending), now it's better than any other time
4512016-09-20T22:28:46  <jtimon> personally I like the _fn convention
4522016-09-20T22:32:22  *** mrkent has quit IRC
4532016-09-20T22:32:51  *** MarcoFalke has quit IRC
4542016-09-20T22:42:29  *** mrkent has joined #bitcoin-core-dev
4552016-09-20T23:01:12  *** Alopex has quit IRC
4562016-09-20T23:02:17  *** Alopex has joined #bitcoin-core-dev
4572016-09-20T23:15:05  *** rogerwilco has quit IRC
4582016-09-20T23:15:19  *** rogerwilco has joined #bitcoin-core-dev
4592016-09-20T23:27:33  *** murch has quit IRC