12020-10-09T00:00:01  *** dorena has quit IRC
  22020-10-09T00:00:24  *** DeanGuss has quit IRC
  32020-10-09T00:05:43  *** DeanGuss has joined #bitcoin-core-dev
  42020-10-09T00:10:49  *** davterra has joined #bitcoin-core-dev
  52020-10-09T00:16:16  *** AaronvanW has quit IRC
  62020-10-09T00:18:04  *** diogorsergio has quit IRC
  72020-10-09T00:18:30  *** proofofkeags_ has quit IRC
  82020-10-09T00:18:30  *** proofofkeags has quit IRC
  92020-10-09T00:21:03  *** Wolfy87 has joined #bitcoin-core-dev
 102020-10-09T00:21:57  *** molz_ has joined #bitcoin-core-dev
 112020-10-09T00:24:31  *** bjkkkk has joined #bitcoin-core-dev
 122020-10-09T00:25:24  *** mol_ has quit IRC
 132020-10-09T00:27:13  *** rc_423 has quit IRC
 142020-10-09T00:28:20  *** rc_423 has joined #bitcoin-core-dev
 152020-10-09T00:40:35  *** placa has joined #bitcoin-core-dev
 162020-10-09T00:44:40  *** bjkkkk has quit IRC
 172020-10-09T00:46:12  *** S3RK has joined #bitcoin-core-dev
 182020-10-09T00:53:40  *** AaronvanW has joined #bitcoin-core-dev
 192020-10-09T00:58:07  *** AaronvanW has quit IRC
 202020-10-09T01:01:49  *** S3RK has quit IRC
 212020-10-09T01:05:08  *** diogorsergio has joined #bitcoin-core-dev
 222020-10-09T01:11:54  *** nick108 has joined #bitcoin-core-dev
 232020-10-09T01:23:50  *** nick108 has quit IRC
 242020-10-09T01:25:28  *** sr_gi has quit IRC
 252020-10-09T01:26:07  *** sr_gi has joined #bitcoin-core-dev
 262020-10-09T01:26:28  *** pinheadmz has quit IRC
 272020-10-09T01:33:26  *** pinheadmz has joined #bitcoin-core-dev
 282020-10-09T01:53:22  *** pinheadmz has quit IRC
 292020-10-09T01:59:51  *** pinheadmz has joined #bitcoin-core-dev
 302020-10-09T02:07:46  *** proofofkeags_ has joined #bitcoin-core-dev
 312020-10-09T02:07:46  *** proofofkeags has joined #bitcoin-core-dev
 322020-10-09T02:09:58  *** AaronvanW has joined #bitcoin-core-dev
 332020-10-09T02:15:04  *** AaronvanW has quit IRC
 342020-10-09T02:21:47  *** pinheadmz has quit IRC
 352020-10-09T02:43:35  *** AaronvanW has joined #bitcoin-core-dev
 362020-10-09T02:47:47  *** AaronvanW has quit IRC
 372020-10-09T03:00:01  *** Wolfy87 has quit IRC
 382020-10-09T03:15:44  *** proofofkeags_ has quit IRC
 392020-10-09T03:15:44  *** proofofkeags has quit IRC
 402020-10-09T03:16:52  *** AaronvanW has joined #bitcoin-core-dev
 412020-10-09T03:20:59  *** AaronvanW has quit IRC
 422020-10-09T03:21:55  *** jpcw1 has joined #bitcoin-core-dev
 432020-10-09T03:28:04  <sipa> vasild: https://github.com/sipa/bitcoin/commits/202010_compactsize_bound
 442020-10-09T03:28:26  <sipa> use with Using<CompactSizeFormatter<false>>(obj) instead of COMPACTSIZE(obj)
 452020-10-09T03:51:57  *** AaronvanW has joined #bitcoin-core-dev
 462020-10-09T03:56:37  *** AaronvanW has quit IRC
 472020-10-09T04:17:51  *** S3RK has joined #bitcoin-core-dev
 482020-10-09T04:20:42  *** S3RK has quit IRC
 492020-10-09T04:22:25  *** S3RK has joined #bitcoin-core-dev
 502020-10-09T04:25:28  *** AaronvanW has joined #bitcoin-core-dev
 512020-10-09T04:27:15  *** S3RK has quit IRC
 522020-10-09T04:30:30  *** AaronvanW has quit IRC
 532020-10-09T04:56:02  *** S3RK has joined #bitcoin-core-dev
 542020-10-09T05:01:58  *** AaronvanW has joined #bitcoin-core-dev
 552020-10-09T05:03:05  *** mol_ has joined #bitcoin-core-dev
 562020-10-09T05:05:59  *** molz_ has quit IRC
 572020-10-09T05:06:05  *** AaronvanW has quit IRC
 582020-10-09T05:30:52  *** S3RK has quit IRC
 592020-10-09T05:33:21  <hebasto> sipa: re "it's a protection for the case where CompactSize format is used to actually encode a size" in which cases? (`MAX_SIZE = 0x02000000` seems undocumented in the code)
 602020-10-09T05:38:58  <hebasto> it was added with alert system in 401926283a200994ecd7df8eae8ced8e0b067c46
 612020-10-09T05:39:06  <hebasto> right?
 622020-10-09T05:42:24  *** AaronvanW has joined #bitcoin-core-dev
 632020-10-09T05:42:33  <hebasto> as alert system has been removed could we just revert back s/MAX_SIZE/INT_MAX/ ?
 642020-10-09T05:44:26  *** placa has quit IRC
 652020-10-09T05:47:08  *** AaronvanW has quit IRC
 662020-10-09T05:55:25  <sipa> hebasto: most satoshi commits were aggregates of a bunch of unrelated things
 672020-10-09T05:55:41  <sipa> MAX_SIZE is the maximum size of serialized objects, 32 MiB
 682020-10-09T05:56:22  <hebasto> I see. Thanks!
 692020-10-09T06:00:02  *** jpcw1 has quit IRC
 702020-10-09T06:00:39  <hebasto> sipa: do you think that `TxRequestTracker::CountCandidates` is really required to be a part of `TxRequestTracker` public interface?
 712020-10-09T06:02:03  *** kristapsk has quit IRC
 722020-10-09T06:03:36  <sipa> hebasto: it makes testing easier
 732020-10-09T06:03:45  <sipa> but apart from that, it's unneeded - for now at least
 742020-10-09T06:03:55  <sipa> it could be used as part of an overload check later
 752020-10-09T06:11:24  *** brianhoffman has quit IRC
 762020-10-09T06:11:30  *** brianhoffman_ has joined #bitcoin-core-dev
 772020-10-09T06:19:53  *** AaronvanW has joined #bitcoin-core-dev
 782020-10-09T06:22:09  *** BartK has joined #bitcoin-core-dev
 792020-10-09T06:24:10  *** AaronvanW has quit IRC
 802020-10-09T06:40:41  *** molz_ has joined #bitcoin-core-dev
 812020-10-09T06:43:59  *** mol_ has quit IRC
 822020-10-09T06:44:18  *** andreacab has joined #bitcoin-core-dev
 832020-10-09T06:44:58  *** S3RK has joined #bitcoin-core-dev
 842020-10-09T06:46:28  *** bitcoin-git has joined #bitcoin-core-dev
 852020-10-09T06:46:28  <bitcoin-git> [bitcoin] MarcoFalke pushed 2 commits to master: https://github.com/bitcoin/bitcoin/compare/d9de00b3e067...380705ef4f00
 862020-10-09T06:46:28  <bitcoin-git> bitcoin/master faf2999 MarcoFalke: cirrus: Use kvm to avoid spurious CI failures in the default virtualizatio...
 872020-10-09T06:46:29  <bitcoin-git> bitcoin/master 380705e MarcoFalke: Merge #20106: cirrus: Use kvm to avoid spurious CI failures in the default...
 882020-10-09T06:46:30  *** bitcoin-git has left #bitcoin-core-dev
 892020-10-09T06:46:48  *** bitcoin-git has joined #bitcoin-core-dev
 902020-10-09T06:46:48  <bitcoin-git> [bitcoin] MarcoFalke merged pull request #20106: cirrus: Use kvm to avoid spurious CI failures in the default virtualization cluster (master...2010-ciOtherVirt) https://github.com/bitcoin/bitcoin/pull/20106
 912020-10-09T06:46:49  *** bitcoin-git has left #bitcoin-core-dev
 922020-10-09T06:57:34  *** AaronvanW has joined #bitcoin-core-dev
 932020-10-09T06:59:13  *** jonatack has quit IRC
 942020-10-09T07:01:56  *** AaronvanW has quit IRC
 952020-10-09T07:14:52  *** bitcoin-git has joined #bitcoin-core-dev
 962020-10-09T07:14:53  <bitcoin-git> [bitcoin] MarcoFalke closed pull request #20103: test: Enable mocktime RPC for all test chains (master...2010-testMockAllChains) https://github.com/bitcoin/bitcoin/pull/20103
 972020-10-09T07:14:53  <vasild> hi
 982020-10-09T07:15:01  *** bitcoin-git has left #bitcoin-core-dev
 992020-10-09T07:17:08  *** bitcoin-git has joined #bitcoin-core-dev
1002020-10-09T07:17:08  <bitcoin-git> [bitcoin] MarcoFalke opened pull request #20112: test: Speed up wallet_resendwallettransactions with mockscheduler RPC (master...2010-testFasterMock) https://github.com/bitcoin/bitcoin/pull/20112
1012020-10-09T07:17:09  *** bitcoin-git has left #bitcoin-core-dev
1022020-10-09T07:18:10  <sipa> vasild: morning
1032020-10-09T07:18:31  <vasild> evening :)
1042020-10-09T07:18:33  <vasild> (for you)
1052020-10-09T07:19:23  <vasild> is it on purpose that WriteCompactSize() can write big numbers, but ReadCompactSize() cannot read them?
1062020-10-09T07:22:12  <sipa> vasild: i think in the compactsize-is-only-for-sizes setting we used to have, the issue just didn't occur
1072020-10-09T07:22:32  <sipa> as it's not possible to have such a big object to begin with
1082020-10-09T07:22:54  <vasild> yeah, so writecompactsize was never called with big numbers
1092020-10-09T07:23:02  <sipa> indeed
1102020-10-09T07:23:17  <sipa> (and the usage in bip152 is restricted to 16-bit numbers, which obviously won't have any problems)
1112020-10-09T07:23:22  <vasild> I will steal that commit https://github.com/sipa/bitcoin/commit/6edfab492aa67e421af556728c1b8b4d17826b80
1122020-10-09T07:27:34  <vasild> I am just curous - any idea what are those high service flags 0x800043d and 0x800000d that cannot be a result of bitwise-or-ing NODE_* flags?
1132020-10-09T07:27:59  <vasild> of course any peer is free to claim whatever they want in their service flags, including 0xfffff...
1142020-10-09T07:28:14  <sipa> some high bits are used for private extensions
1152020-10-09T07:28:18  <sipa> not sure what this is
1162020-10-09T07:28:41  <vasild> ok
1172020-10-09T07:29:07  <hebasto> sipa: good idea to rename CompactSize
1182020-10-09T07:29:26  <sipa> vasild: what software/user agent reports these?
1192020-10-09T07:29:58  <vasild> I did not record that
1202020-10-09T07:30:30  <vasild> just added printf() in CAddress::unserialize to see what is coming in from the network
1212020-10-09T07:32:25  *** AaronvanW has joined #bitcoin-core-dev
1222020-10-09T07:34:50  <vasild> now I did -addnode=62.107.200.30:8333 to see the user agent
1232020-10-09T07:34:59  <sipa> they're uninteresting
1242020-10-09T07:35:06  <sipa> my dns seed crawler sees them too
1252020-10-09T07:35:20  <sipa> but they're from a variaety of innocently-loking user agents
1262020-10-09T07:35:32  <vasild> but now getnetworkinfo shows
1272020-10-09T07:35:36  <vasild>     "addr": "62.107.200.30:8333",
1282020-10-09T07:35:40  <vasild>     "services": "000000000000040d",
1292020-10-09T07:35:44  <vasild> :-O
1302020-10-09T07:35:47  <sipa> hmm
1312020-10-09T07:35:50  *** jonatack has joined #bitcoin-core-dev
1322020-10-09T07:35:57  <vasild> yeah, and innocent     "subver": "/Satoshi:0.20.1/",
1332020-10-09T07:36:06  <sipa> are they being rumoured with this bit set... but then not reported in the version message?
1342020-10-09T07:36:59  <hebasto> could this hurt the network?
1352020-10-09T07:37:00  *** AaronvanW has quit IRC
1362020-10-09T07:38:11  <vasild> sipa: must be that or something else I am missing
1372020-10-09T07:38:14  <sipa> i don't think so
1382020-10-09T07:38:15  <vasild> I just saw:
1392020-10-09T07:38:16  <vasild> CAddress::unser: time=1602210665, services=67109901, addr=178.128.96.96:8333
1402020-10-09T07:38:19  <vasild> from my printf
1412020-10-09T07:38:40  <vasild> shutdown, restart with -addnode=178.128.96.96:8333
1422020-10-09T07:38:56  <vasild> and getpeerinfo:     "addr": "178.128.96.96:8333",
1432020-10-09T07:39:00  <vasild>     "services": "000000000000000d",
1442020-10-09T07:40:10  <vasild> we don't do any "filter away any unknown service bits before displaying in getpeerinfo, right?
1452020-10-09T07:41:02  *** Pavlenex has joined #bitcoin-core-dev
1462020-10-09T07:41:26  <sipa> vasild: i don't think so
1472020-10-09T07:43:23  <vasild> just for reference, here is my dummy printf() in CAddress unserialize: https://bpa.st/JYYA
1482020-10-09T07:45:45  <sipa> looks right
1492020-10-09T07:48:28  <vasild> ok, assuming they get gossiped with the flags but are later not present in the version message when we actually connect
1502020-10-09T07:49:26  <sipa> perhaps some broken software is relaying them with that bit set
1512020-10-09T07:50:04  <vasild> wumpus: just to confirm - yes, 0.20 would start on a 0.21-modified-peers.dat as if that file was not present, I also think it is ok. Except the obscure/scary deserialize error messages
1522020-10-09T07:50:51  <sipa> perhaps we can give a nicer error message in the future for upgrades
1532020-10-09T07:50:56  <sipa> but for old software it's too late
1542020-10-09T07:50:57  <vasild> given I will be modifying #19954 I will add one line at the start of unserialize to check if the version is too high
1552020-10-09T07:51:00  <gribble> https://github.com/bitcoin/bitcoin/issues/19954 | tor: complete the TORv3 implementation by vasild · Pull Request #19954 · bitcoin/bitcoin · GitHub
1562020-10-09T07:51:02  <vasild> exactly
1572020-10-09T07:51:59  <vasild> and print something like "detected file format of peers.dat from the future: 4, I can only understand up to 3, starting operation as if peers.dat is empty"
1582020-10-09T07:52:40  <hebasto> sipa: if their software is broken should we discourage those peers?
1592020-10-09T07:54:45  *** S3RK has quit IRC
1602020-10-09T07:55:21  *** andreacab has quit IRC
1612020-10-09T07:55:27  <vasild> maybe a honest/legit/unbroken peer is just relaying some shi^H^Hstuff that was given to him by a broken/malicious guy
1622020-10-09T07:55:47  *** andreacab has joined #bitcoin-core-dev
1632020-10-09T07:56:01  <vasild> we don't want to flag the legit peer as misbehaving
1642020-10-09T07:56:07  <hebasto> right
1652020-10-09T07:56:40  *** andreacab has quit IRC
1662020-10-09T07:56:45  <vasild> or also those high service bits could have some purpose, I guess, it is not certain that they are due to broken or malicious software
1672020-10-09T07:56:48  *** andreacab has joined #bitcoin-core-dev
1682020-10-09T07:57:17  * hebasto "hidden service" :D
1692020-10-09T07:58:01  <vasild> did somebody develop a decentralized instant message chat application using bitcoin p2p network, sneaking the data via the service bits?
1702020-10-09T07:59:07  <MarcoFalke> it would be good to log the peer that sent the high bits
1712020-10-09T07:59:24  <sipa> MarcoFalke: unfortunately, they're most likely not the ones who introduced it
1722020-10-09T08:00:39  <vasild> what happens if a node receives via gossip the same address:port but with different service bits?
1732020-10-09T08:01:01  <MarcoFalke> does addrman correct the service bits when a version msg is received?
1742020-10-09T08:01:14  <vasild> that too!
1752020-10-09T08:01:34  <sipa> they get OR'ed together
1762020-10-09T08:01:43  <sipa> but i hope that on an actual connection, it's overwritten
1772020-10-09T08:02:06  *** Pavlenex has quit IRC
1782020-10-09T08:02:46  <vasild> they get OR'ed :( so a node cannot correct his by re-advertising it regularly. Once somebody adds some high bits to his addr:port, it is over.
1792020-10-09T08:02:57  <sipa> we should fix that
1802020-10-09T08:03:09  <sipa> whatever we receive from the node itself should be written exactly into addrman
1812020-10-09T08:03:12  <vasild> change to "newer one overwrites old one"?
1822020-10-09T08:03:28  <sipa> i think OR'ing makes sense for rumoured addr messages
1832020-10-09T08:04:02  <vasild> I agree that if we connect directly then whatever the peer says is to overwrite the other stuff
1842020-10-09T08:04:10  <sipa> but if we receive information from the IP itself (on an outgoing connection), it should overwrite
1852020-10-09T08:04:38  *** AaronvanW has joined #bitcoin-core-dev
1862020-10-09T08:04:38  *** S3RK has joined #bitcoin-core-dev
1872020-10-09T08:09:09  *** AaronvanW has quit IRC
1882020-10-09T08:13:13  <jonatack> catching up with the discussion... yes, the issue I reported is the DB:deserialise error messages and peers.dat being recreated; bitcoind operates fine otherwise. We can't fix the older software, but we can communicate the situation clearly to users (and hopefully fix it henceforth).
1892020-10-09T08:16:35  <wumpus> sipa: it's more of a warning message anyway isn't it, people that don't regaularly check the log won't even notice it
1902020-10-09T08:16:51  <wumpus> vasild: thanks for confirming, I think there's no problem in that case at all
1912020-10-09T08:17:39  <wumpus> if you say "error" I was expecting some kind of emergency shutdown, that *would* be scary, it simply saying that it's unable to load the file and start from scratch isn't scary as such
1922020-10-09T08:18:32  <vasild> y
1932020-10-09T08:18:36  <jonatack> well, anyone who greps the past log for errors will see
1942020-10-09T08:18:37  <jonatack> ERROR: DeserializeDB: Deserialize or I/O error - ReadCompactSize(): size too large: iostream error
1952020-10-09T08:18:53  <jonatack> Invalid or missing peers.dat; recreating
1962020-10-09T08:19:00  <wumpus> yes
1972020-10-09T08:19:09  <vasild> I though jonatack's disk has developed bad sectors :-D
1982020-10-09T08:19:20  <jonatack> that allcaps screaming ERROR is a bit worrisome :D
1992020-10-09T08:19:21  <vasild> thought
2002020-10-09T08:19:24  <wumpus> I agree it could have been avoided with a version marker
2012020-10-09T08:20:16  <wumpus> jonatack: that's nothing, those errors used to appear for malformed incoming *network packets*
2022020-10-09T08:20:21  <wumpus> ERRORS i mean
2032020-10-09T08:21:07  <wumpus> debug.log used to be one big file full of screaming :)
2042020-10-09T08:21:16  <jonatack> hehe
2052020-10-09T08:21:39  <jonatack> the halloween release
2062020-10-09T08:22:40  <wumpus> change the bitcoin logo for a 🎃
2072020-10-09T08:23:28  *** Pavlenex has joined #bitcoin-core-dev
2082020-10-09T08:23:30  <jonatack> fewer color change proposals :)
2092020-10-09T08:25:32  <fanquake> orange.bikeshed.com
2102020-10-09T08:32:49  *** worc3131 has joined #bitcoin-core-dev
2112020-10-09T08:39:06  *** andreacab has quit IRC
2122020-10-09T08:39:33  *** andreacab has joined #bitcoin-core-dev
2132020-10-09T08:42:58  *** AaronvanW has joined #bitcoin-core-dev
2142020-10-09T08:43:06  *** Klox048093 has joined #bitcoin-core-dev
2152020-10-09T08:43:06  *** worc3131 has quit IRC
2162020-10-09T08:44:04  *** Klox04809 has quit IRC
2172020-10-09T08:47:34  *** AaronvanW has quit IRC
2182020-10-09T08:49:25  *** andreacab has quit IRC
2192020-10-09T08:55:26  *** andreacab has joined #bitcoin-core-dev
2202020-10-09T08:55:55  *** andreaca_ has joined #bitcoin-core-dev
2212020-10-09T09:00:02  *** BartK has quit IRC
2222020-10-09T09:00:40  *** AaronvanW has joined #bitcoin-core-dev
2232020-10-09T09:03:47  <vasild> jonatack: https://github.com/bitcoin/bitcoin/pull/19954#discussion_r502280047 "I test if my suggestions compile before making them"... and I did not test that it does not compile before I replied :-/
2242020-10-09T09:04:14  <vasild> I guess it compiles because stream is a pointer which remains unmodified (the pointer itself) after the call to stream->ignore()
2252020-10-09T09:04:25  *** kexkey has quit IRC
2262020-10-09T09:04:58  <vasild> but anyway, semantically we do modify the state of the stream by the ignore() call, so I dont think it should be tagged with const
2272020-10-09T09:06:14  *** andreaca_ has quit IRC
2282020-10-09T09:06:40  *** andreacab has joined #bitcoin-core-dev
2292020-10-09T09:08:04  *** andreaca_ has joined #bitcoin-core-dev
2302020-10-09T09:10:47  *** andreacab has quit IRC
2312020-10-09T09:21:05  *** tw1sted1 has joined #bitcoin-core-dev
2322020-10-09T09:22:35  *** andreaca_ has quit IRC
2332020-10-09T09:24:20  *** andreacab has joined #bitcoin-core-dev
2342020-10-09T09:26:38  *** andreacab has quit IRC
2352020-10-09T09:27:06  *** andreacab has joined #bitcoin-core-dev
2362020-10-09T09:27:13  *** mrostecki has joined #bitcoin-core-dev
2372020-10-09T09:29:23  *** Pavlenex has quit IRC
2382020-10-09T09:31:23  *** andreacab has quit IRC
2392020-10-09T09:33:18  <jonatack> vasild: no worries, it's an edge case question, was just to learn
2402020-10-09T09:35:39  <vasild> jonatack: my reasoning is "if a method changes the object in a way that can be observed from outside, then it should not be const"
2412020-10-09T09:37:20  *** AaronvanW has quit IRC
2422020-10-09T09:37:38  *** AaronvanW has joined #bitcoin-core-dev
2432020-10-09T09:39:18  *** andreacab has joined #bitcoin-core-dev
2442020-10-09T09:39:59  <jonatack> right. looking at it led me to https://isocpp.github.io/CppCoreGuidelines/CppCoreGuidelines#con2-by-default-make-member-functions-const and http://www.cplusplus.com/forum/general/12087/
2452020-10-09T09:40:07  <jonatack> am re-reviewing 19954
2462020-10-09T09:44:57  *** mrostecki has quit IRC
2472020-10-09T09:46:45  *** sr_gi has quit IRC
2482020-10-09T09:47:32  *** sr_gi has joined #bitcoin-core-dev
2492020-10-09T09:52:14  *** tw1sted1 has quit IRC
2502020-10-09T10:00:11  <jonatack> vasild: nice, in my testing the patch has fixed the DB deserialize errors and peers.dat messages
2512020-10-09T10:01:11  <vasild> \o/
2522020-10-09T10:03:07  *** bitcoin-git has joined #bitcoin-core-dev
2532020-10-09T10:03:07  <bitcoin-git> [bitcoin] fanquake pushed 2 commits to master: https://github.com/bitcoin/bitcoin/compare/380705ef4f00...6854dbdc88d7
2542020-10-09T10:03:08  <bitcoin-git> bitcoin/master 2dc79c4 Hennadii Stepanov: doc: Update and improve files.md
2552020-10-09T10:03:08  <bitcoin-git> bitcoin/master 6854dbd fanquake: Merge #20076: doc: Update and improve files.md
2562020-10-09T10:03:19  *** bitcoin-git has left #bitcoin-core-dev
2572020-10-09T10:03:36  *** bitcoin-git has joined #bitcoin-core-dev
2582020-10-09T10:03:37  <bitcoin-git> [bitcoin] fanquake merged pull request #20076: doc: Update and improve files.md (master...201004-files) https://github.com/bitcoin/bitcoin/pull/20076
2592020-10-09T10:03:48  *** bitcoin-git has left #bitcoin-core-dev
2602020-10-09T10:05:41  *** belcher_ has joined #bitcoin-core-dev
2612020-10-09T10:05:53  <jonatack> vasild: restarting on, say, 0.18.1, I see "ERROR: DeserializeDB: Deserialize or I/O error - CAutoFile::read: end of file: iostream error" and "Invalid or missing peers.dat; recreating", as we mention in the new release notes, but not on 19954 after `cp onion_private_key onion_v3_private_key`
2622020-10-09T10:06:26  <jonatack> \o/ indeed :)
2632020-10-09T10:08:07  *** belcher has quit IRC
2642020-10-09T10:08:41  *** bitcoin-git has joined #bitcoin-core-dev
2652020-10-09T10:08:42  <bitcoin-git> [bitcoin] fanquake pushed 3 commits to master: https://github.com/bitcoin/bitcoin/compare/6854dbdc88d7...12a1c3ad1a43
2662020-10-09T10:08:43  <bitcoin-git> bitcoin/master 99992e7 MarcoFalke: doc: Collect release-notes snippets
2672020-10-09T10:08:43  <bitcoin-git> bitcoin/master faa0847 MarcoFalke: doc: Add release notes for #20101
2682020-10-09T10:08:44  <bitcoin-git> bitcoin/master 12a1c3a fanquake: Merge #20107: doc: Collect release-notes snippets
2692020-10-09T10:08:45  *** bitcoin-git has left #bitcoin-core-dev
2702020-10-09T10:09:01  *** bitcoin-git has joined #bitcoin-core-dev
2712020-10-09T10:09:02  <bitcoin-git> [bitcoin] fanquake merged pull request #20107: doc: Collect release-notes snippets (master...2010-docRelSnip) https://github.com/bitcoin/bitcoin/pull/20107
2722020-10-09T10:09:03  *** bitcoin-git has left #bitcoin-core-dev
2732020-10-09T10:10:23  *** vasild has quit IRC
2742020-10-09T10:12:21  *** vasild has joined #bitcoin-core-dev
2752020-10-09T10:13:48  *** shesek has quit IRC
2762020-10-09T10:18:29  *** Imogene64Boyer has joined #bitcoin-core-dev
2772020-10-09T10:20:36  *** andreacab has quit IRC
2782020-10-09T10:21:04  *** andreacab has joined #bitcoin-core-dev
2792020-10-09T10:25:37  *** andreacab has quit IRC
2802020-10-09T10:27:39  *** ericholscher has joined #bitcoin-core-dev
2812020-10-09T10:28:04  *** promag has joined #bitcoin-core-dev
2822020-10-09T10:44:14  *** promag has quit IRC
2832020-10-09T10:45:46  *** Imogene64Boyer has quit IRC
2842020-10-09T10:45:58  *** promag has joined #bitcoin-core-dev
2852020-10-09T10:46:00  *** andreacab has joined #bitcoin-core-dev
2862020-10-09T10:49:33  *** andreacab has quit IRC
2872020-10-09T10:49:59  *** andreacab has joined #bitcoin-core-dev
2882020-10-09T10:55:02  *** mrostecki has joined #bitcoin-core-dev
2892020-10-09T10:56:56  *** andreacab has quit IRC
2902020-10-09T10:58:41  *** andreacab has joined #bitcoin-core-dev
2912020-10-09T11:02:25  *** kyoo[m] has quit IRC
2922020-10-09T11:02:27  *** TheFuzzStone[m] has quit IRC
2932020-10-09T11:02:34  *** sethrogers23[m] has quit IRC
2942020-10-09T11:02:36  *** tianshi[m] has quit IRC
2952020-10-09T11:02:46  *** rcrtn32002[m] has quit IRC
2962020-10-09T11:02:46  *** snowkeld[m] has quit IRC
2972020-10-09T11:02:49  *** RaphalBentgeac[m has quit IRC
2982020-10-09T11:02:50  *** Pasta[m] has quit IRC
2992020-10-09T11:02:51  *** icota[m] has quit IRC
3002020-10-09T11:06:01  *** rh0nj has quit IRC
3012020-10-09T11:07:09  *** rh0nj has joined #bitcoin-core-dev
3022020-10-09T11:09:35  *** kristapsk has joined #bitcoin-core-dev
3032020-10-09T11:09:45  *** rcrtn32002[m] has joined #bitcoin-core-dev
3042020-10-09T11:12:43  *** TallTim has quit IRC
3052020-10-09T11:22:38  *** andreacab has quit IRC
3062020-10-09T11:23:04  *** andreacab has joined #bitcoin-core-dev
3072020-10-09T11:26:51  *** snowkeld[m] has joined #bitcoin-core-dev
3082020-10-09T11:26:51  *** TheFuzzStone[m] has joined #bitcoin-core-dev
3092020-10-09T11:26:52  *** icota[m] has joined #bitcoin-core-dev
3102020-10-09T11:26:52  *** Pasta[m] has joined #bitcoin-core-dev
3112020-10-09T11:26:52  *** kyoo[m] has joined #bitcoin-core-dev
3122020-10-09T11:26:52  *** sethrogers23[m] has joined #bitcoin-core-dev
3132020-10-09T11:26:58  *** tianshi[m] has joined #bitcoin-core-dev
3142020-10-09T11:26:58  *** RaphalBentgeac[m has joined #bitcoin-core-dev
3152020-10-09T11:27:07  *** andreacab has quit IRC
3162020-10-09T11:27:14  *** vincenzopalazzo has joined #bitcoin-core-dev
3172020-10-09T11:30:52  *** jonatack has quit IRC
3182020-10-09T11:32:37  *** andreacab has joined #bitcoin-core-dev
3192020-10-09T11:34:00  *** opsec_x122 has joined #bitcoin-core-dev
3202020-10-09T11:34:26  *** ericholscher has quit IRC
3212020-10-09T11:34:26  *** jeremyrubin has quit IRC
3222020-10-09T11:34:26  *** opsec_x12 has quit IRC
3232020-10-09T11:34:26  *** dergoegge has quit IRC
3242020-10-09T11:34:26  *** hsmiths has quit IRC
3252020-10-09T11:34:26  *** mariorz has quit IRC
3262020-10-09T11:34:26  *** noganoo has quit IRC
3272020-10-09T11:36:53  *** EagleTM has joined #bitcoin-core-dev
3282020-10-09T11:39:55  *** EagleTM has quit IRC
3292020-10-09T11:40:12  *** jeremyrubin has joined #bitcoin-core-dev
3302020-10-09T11:40:13  *** dergoegge has joined #bitcoin-core-dev
3312020-10-09T11:40:13  *** hsmiths has joined #bitcoin-core-dev
3322020-10-09T11:40:13  *** mariorz has joined #bitcoin-core-dev
3332020-10-09T11:40:13  *** noganoo has joined #bitcoin-core-dev
3342020-10-09T11:40:25  *** EagleTM has joined #bitcoin-core-dev
3352020-10-09T11:40:54  *** S3RK has quit IRC
3362020-10-09T11:41:49  *** S3RK has joined #bitcoin-core-dev
3372020-10-09T11:42:38  *** opsec_x122 is now known as opsec_x12
3382020-10-09T11:44:17  *** andreacab has quit IRC
3392020-10-09T11:46:05  *** S3RK has quit IRC
3402020-10-09T11:46:54  *** promag has quit IRC
3412020-10-09T11:47:50  *** promag has joined #bitcoin-core-dev
3422020-10-09T11:49:06  *** EagleTM has quit IRC
3432020-10-09T11:49:32  *** EagleTM has joined #bitcoin-core-dev
3442020-10-09T11:52:05  *** promag has quit IRC
3452020-10-09T11:52:34  *** EagleTM has quit IRC
3462020-10-09T11:52:58  *** EagleTM has joined #bitcoin-core-dev
3472020-10-09T11:53:23  *** mrostecki has quit IRC
3482020-10-09T11:57:15  *** EagleTM has quit IRC
3492020-10-09T11:57:38  *** EagleTM has joined #bitcoin-core-dev
3502020-10-09T11:59:56  *** Victorsueca has joined #bitcoin-core-dev
3512020-10-09T12:03:09  *** EagleTM has quit IRC
3522020-10-09T12:03:33  *** EagleTM has joined #bitcoin-core-dev
3532020-10-09T12:21:05  *** dholbach has joined #bitcoin-core-dev
3542020-10-09T12:21:41  *** pinheadmz has joined #bitcoin-core-dev
3552020-10-09T12:26:24  *** setpill has joined #bitcoin-core-dev
3562020-10-09T12:26:51  *** promag has joined #bitcoin-core-dev
3572020-10-09T12:27:26  *** EagleTM has quit IRC
3582020-10-09T12:27:48  *** EagleTM has joined #bitcoin-core-dev
3592020-10-09T12:31:05  *** promag has quit IRC
3602020-10-09T12:33:07  *** jonatack has joined #bitcoin-core-dev
3612020-10-09T12:36:07  *** EagleTM has quit IRC
3622020-10-09T12:36:33  *** EagleTM has joined #bitcoin-core-dev
3632020-10-09T12:37:40  *** Guyver2 has joined #bitcoin-core-dev
3642020-10-09T12:40:01  *** EagleTM has quit IRC
3652020-10-09T12:40:28  *** EagleTM has joined #bitcoin-core-dev
3662020-10-09T12:45:28  *** gloriazhao has quit IRC
3672020-10-09T12:49:08  *** S3RK has joined #bitcoin-core-dev
3682020-10-09T12:53:23  *** S3RK has quit IRC
3692020-10-09T12:54:11  *** EagleTM has quit IRC
3702020-10-09T12:54:38  *** EagleTM has joined #bitcoin-core-dev
3712020-10-09T12:59:52  *** EagleTM has quit IRC
3722020-10-09T13:00:15  *** EagleTM has joined #bitcoin-core-dev
3732020-10-09T13:07:12  *** molz_ has quit IRC
3742020-10-09T13:09:12  <achow101> #proposedwalletmeetingtopic: wallet.dat vs wallet.sqlite
3752020-10-09T13:11:25  *** Eagle[TM] has joined #bitcoin-core-dev
3762020-10-09T13:11:59  *** EagleTM has quit IRC
3772020-10-09T13:16:11  *** Eagle[TM] has quit IRC
3782020-10-09T13:18:15  *** mol has joined #bitcoin-core-dev
3792020-10-09T13:30:16  *** S3RK has joined #bitcoin-core-dev
3802020-10-09T13:34:50  *** Klox0480931 has joined #bitcoin-core-dev
3812020-10-09T13:35:10  *** S3RK has quit IRC
3822020-10-09T13:35:34  *** Klox048093 has quit IRC
3832020-10-09T13:36:33  <hebasto> gh silently fails to "Load more..." of "52 hidden items" on #19988 (
3842020-10-09T13:36:37  <gribble> https://github.com/bitcoin/bitcoin/issues/19988 | Overhaul transaction request logic by sipa · Pull Request #19988 · bitcoin/bitcoin · GitHub
3852020-10-09T14:01:46  *** setpill has quit IRC
3862020-10-09T14:01:58  *** mol_ has joined #bitcoin-core-dev
3872020-10-09T14:05:27  *** mol has quit IRC
3882020-10-09T14:06:20  *** molz_ has joined #bitcoin-core-dev
3892020-10-09T14:07:58  *** mol has joined #bitcoin-core-dev
3902020-10-09T14:09:19  *** mol_ has quit IRC
3912020-10-09T14:11:16  *** molz_ has quit IRC
3922020-10-09T14:29:34  <jonatack> hebasto: saw the same yesterday, i gave up trying to see the discussion :/
3932020-10-09T14:36:57  *** andreacab has joined #bitcoin-core-dev
3942020-10-09T14:37:39  *** luke-jr has quit IRC
3952020-10-09T14:39:55  *** luke-jr has joined #bitcoin-core-dev
3962020-10-09T14:40:27  *** andreacab has quit IRC
3972020-10-09T14:40:43  *** andreacab has joined #bitcoin-core-dev
3982020-10-09T15:00:02  *** dholbach has quit IRC
3992020-10-09T15:01:33  *** kexkey has joined #bitcoin-core-dev
4002020-10-09T15:03:13  *** andreacab has quit IRC
4012020-10-09T15:03:39  *** andreacab has joined #bitcoin-core-dev
4022020-10-09T15:05:20  *** andreaca_ has joined #bitcoin-core-dev
4032020-10-09T15:06:09  *** jonatack has quit IRC
4042020-10-09T15:07:11  *** Talkless has joined #bitcoin-core-dev
4052020-10-09T15:07:58  *** andreacab has quit IRC
4062020-10-09T15:10:55  *** luke-jr has quit IRC
4072020-10-09T15:10:57  <vasild> gh censorship!
4082020-10-09T15:11:14  *** jonatack has joined #bitcoin-core-dev
4092020-10-09T15:12:20  *** kristapsk has quit IRC
4102020-10-09T15:18:03  <michaelfolkson> I don't see how Core is still managed on GitHub in 10 years personally. But will be a massive headache whenever it does move. Ideally done when things are quiet (not with a potential soft fork on the horizon)
4112020-10-09T15:19:00  *** luke-jr has joined #bitcoin-core-dev
4122020-10-09T15:25:25  *** luke-jr has quit IRC
4132020-10-09T15:26:49  *** luke-jr has joined #bitcoin-core-dev
4142020-10-09T15:33:58  <jonatack> ISTM that GitHub is now starting to hide ACKs and re-ACKs as "one similar comment"
4152020-10-09T15:38:13  *** owowo has quit IRC
4162020-10-09T15:39:10  *** owowo has joined #bitcoin-core-dev
4172020-10-09T15:39:48  *** andreaca_ has quit IRC
4182020-10-09T15:40:16  *** andreacab has joined #bitcoin-core-dev
4192020-10-09T15:43:58  *** Talkless has quit IRC
4202020-10-09T15:44:29  *** Talkless has joined #bitcoin-core-dev
4212020-10-09T15:44:42  *** andreacab has quit IRC
4222020-10-09T15:49:45  *** Pavlenex has joined #bitcoin-core-dev
4232020-10-09T15:55:56  *** kers has joined #bitcoin-core-dev
4242020-10-09T15:56:23  *** vincenzopalazzo has quit IRC
4252020-10-09T15:59:51  <wumpus> I'm not sure moving is that much of a headache, many projects have done this, it's just that the only thing after github that makes sense is self-hosting something like gitlab or gitea, which has quite a lot issues of its own, e.g. who hosts the instance
4262020-10-09T16:02:17  <wumpus> I have a weird issue on my freebsd node: it kept making outgoing connections at a rate of ~2 per second, the whole log is full of it, and it wasn't accepting any new blocks. I hope it's not a problem with the TorV3 PR.
4272020-10-09T16:03:07  *** luke-jr has quit IRC
4282020-10-09T16:03:12  <wumpus> also, it segfaulted on shutdown (I can't reproduce this, unfortunately, so don't have a traceback)
4292020-10-09T16:03:15  *** proofofkeags_ has joined #bitcoin-core-dev
4302020-10-09T16:03:15  *** proofofkeags has joined #bitcoin-core-dev
4312020-10-09T16:03:53  *** Talkless has quit IRC
4322020-10-09T16:06:24  *** luke-jr has joined #bitcoin-core-dev
4332020-10-09T16:09:15  *** mol_ has joined #bitcoin-core-dev
4342020-10-09T16:10:20  *** mol has quit IRC
4352020-10-09T16:12:46  <sipa> wumpus: ugh
4362020-10-09T16:14:00  <wumpus> 2020-10-09T15:06:40Z New outbound peer connected: version: 70015, blocks=651964, peer=1018923 (full-relay)
4372020-10-09T16:14:20  <wumpus> that peer id says enough I think
4382020-10-09T16:15:50  <jonatack> wumpus: huh. i have not been seeing that issue with 19954 (on debian)
4392020-10-09T16:15:54  <wumpus> it's not happening on my other TorV3 node though so there's no clear correlation
4402020-10-09T16:16:19  <wumpus> jonatack: right, it might be a local issue for that node
4412020-10-09T16:16:28  <wumpus> currently building plain master
4422020-10-09T16:19:25  *** luke-jr has quit IRC
4432020-10-09T16:24:10  *** S3RK has joined #bitcoin-core-dev
4442020-10-09T16:25:52  *** luke-jr has joined #bitcoin-core-dev
4452020-10-09T16:28:37  *** S3RK has quit IRC
4462020-10-09T16:31:28  *** roconnor has joined #bitcoin-core-dev
4472020-10-09T16:32:53  <luke-jr> wumpus: I'm not sure self-hosting is really an improvement over GitHub, aside from the platform presumably being free software
4482020-10-09T16:33:08  <luke-jr> the goal is probably a decentralised system, but that doesn't exist yet afaik
4492020-10-09T16:34:13  <vasild> there were some attempts to make decentralized alternative of github
4502020-10-09T16:34:20  <wumpus> luke-jr: it's an improvement in that we'd have direct control over the software, so it can't say, start hiding consecutive ACKs behind our back
4512020-10-09T16:34:29  <vasild> wumpus: I have this sysctl on fbsd: kern.corefile=/coredumps/%U.%N.%P.core
4522020-10-09T16:35:02  <wumpus> vasild: good idea
4532020-10-09T16:35:21  <vasild> and /coredumps has 1777 perms so that anybody can plant files in it
4542020-10-09T16:35:26  <luke-jr> wumpus: whoever is hosting can, whether that's GitHub or ⁇?
4552020-10-09T16:35:59  <wumpus> luke-jr: sure, but if it's one of us they're likely going to be more recentive to these kind of complaints than a big corporation
4562020-10-09T16:36:06  <vasild> ulimit -c is "unlimited" but I don't think I configured that myself, maybe it is the default
4572020-10-09T16:36:11  *** shesek has joined #bitcoin-core-dev
4582020-10-09T16:36:11  *** shesek has joined #bitcoin-core-dev
4592020-10-09T16:37:21  <wumpus> same here, it even said in dmesg that it had dumped core but couldn't find the core file
4602020-10-09T16:37:42  <sipa> it dumped Core
4612020-10-09T16:37:52  <luke-jr> they tend to be big files
4622020-10-09T16:38:09  <vasild> does bitcoind change the working directory at startup? to ~/.bitcoin? or to the root /?
4632020-10-09T16:38:10  <wumpus> sipa: heh
4642020-10-09T16:38:23  <wumpus> vasild: no, it doesn't
4652020-10-09T16:38:41  <wumpus> thinking of it, probably daemonize() does though
4662020-10-09T16:38:48  <vasild> so the core should probably in the current directory where you started it
4672020-10-09T16:38:54  <luke-jr> I know Linux is capable of dumping core into a gdb process directly
4682020-10-09T16:39:55  <wumpus> yes, daemon() changes the working directory to /, it has no write access there, so I guess it just lost it
4692020-10-09T16:43:46  <wumpus> I've set vasild's sysctl parameter now so next time it should go to /coredumps/
4702020-10-09T16:44:41  <vasild> mkdir /coredumps && chmod 1777 /coredumps
4712020-10-09T16:45:08  <sipa> comic relief: https://twitter.com/real_or_random/status/1314596135395262467
4722020-10-09T16:45:16  <vasild> remember to rm /coredumps/* periodically because it grows and grows over time ;)
4732020-10-09T16:46:32  *** andreacab has joined #bitcoin-core-dev
4742020-10-09T16:47:36  *** EagleTM has joined #bitcoin-core-dev
4752020-10-09T16:47:37  <vasild> wumpus: but if dmesg says something like "pid 94776 (conftest), jid 0, uid 0: exited on signal 11 (core dumped)
4762020-10-09T16:47:55  <vasild> with "(core dumped)" at the end then it must be somewhere
4772020-10-09T16:49:37  *** andreacab has quit IRC
4782020-10-09T16:50:03  *** andreacab has joined #bitcoin-core-dev
4792020-10-09T16:52:59  *** EagleTM has quit IRC
4802020-10-09T16:53:07  <wumpus> vasild: I couldn't find it at least, but, I 'make clean'ed so deleted the binary by now so I guess there's no point even if I still found it
4812020-10-09T16:53:25  *** EagleTM has joined #bitcoin-core-dev
4822020-10-09T16:53:33  *** cryptapus has quit IRC
4832020-10-09T16:54:21  <luke-jr> hmm
4842020-10-09T16:54:33  *** andreacab has quit IRC
4852020-10-09T16:54:39  <luke-jr> does C undefined behaviour extend to compile time?
4862020-10-09T16:54:56  <luke-jr> can the compiler explode the system when you try to compile it?
4872020-10-09T16:55:00  <sipa> luke-jr: no
4882020-10-09T16:55:16  <sipa> (afaik)
4892020-10-09T16:56:33  <wumpus> don't give compiler authors any ideas
4902020-10-09T16:58:07  <luke-jr> nuts, could have submitted a patch to GCC to erase everything when it encounters mem*
4912020-10-09T16:58:10  <luke-jr> XD
4922020-10-09T16:58:39  <luke-jr> wumpus: hey, at least I didn't suggest telemetrics on UB?
4932020-10-09T17:00:52  *** EagleTM has quit IRC
4942020-10-09T17:01:20  <wumpus> the most disappointing thing about the whole C UB sitution is how little, historically, C compilers have attempted to detect and reject it, I mean if mem* is only reserved for system functions, fair enough, but make it an error or warning.
4952020-10-09T17:01:56  <wumpus> but no they really like their random explosions :)
4962020-10-09T17:02:17  *** cryptapus has joined #bitcoin-core-dev
4972020-10-09T17:04:04  <sipa> the reason i suspect is that they don't want programs to misbehave if memslartbartfast suddenly becomes a new standard library function in C2043, and your code is already using it
4982020-10-09T17:06:58  <luke-jr> they probably would prefer a C22 compiler is C99-compatible too
4992020-10-09T17:07:20  <luke-jr> by making mem* UB, a C22 memfoo doesn't break C99 code
5002020-10-09T17:08:41  <luke-jr> (same thing, but from another angle)
5012020-10-09T17:09:07  <sipa> i guess a better alternative would be to have a way for the STL to declare "reserved but unused" name patterns to the compiler, so it can error if you use a reserved name, rather than needing to resort to the super-heavy "UB" hammer
5022020-10-09T17:09:41  <wumpus> looks like I have the same problem again after merging 19954: my node keeps making outgoing connections, but never gets any blocks
5032020-10-09T17:10:32  <sipa> hmm
5042020-10-09T17:10:35  <sipa> this looks familiar
5052020-10-09T17:10:36  <wumpus> somehow it's dropping all outgoing connections immediately, it stays at 1-3
5062020-10-09T17:10:55  <sipa> Murch: do you remember someone on bitcoin SE complaining about similar behavior?
5072020-10-09T17:10:56  <wumpus> it does succesfully accept incoming connections but that doesn't help
5082020-10-09T17:11:24  <luke-jr> wumpus: to clarify, you mean without 19954 it was okay?
5092020-10-09T17:11:36  <wumpus> luke-jr: yes, without that it seemed to be okay
5102020-10-09T17:11:43  <wumpus> it did receive a few blocks
5112020-10-09T17:12:23  <wumpus> but no crash at shutdown this time
5122020-10-09T17:13:02  <wumpus> but it does look like 19954 is the problem, will try again without that
5132020-10-09T17:13:35  <sipa> oh, ok
5142020-10-09T17:13:45  <sipa> in that case what i remember seeing must have been unrelated
5152020-10-09T17:13:56  <wumpus> probably, unless this triggers a similar condition
5162020-10-09T17:14:09  <wumpus> I did already try deleting peers.dat
5172020-10-09T17:16:47  <achow101> anyone ever see "corrupted size vs. prev_size in fastbins" on stderr before? it's showing up on a change to sqlite wallet that I'm working on
5182020-10-09T17:18:00  <sipa> achow101: i wouldn't know what prev_size or fastbins are
5192020-10-09T17:18:14  <wumpus> googling it, it's a glibc heap corruption error
5202020-10-09T17:18:22  <achow101> it seems to be coming out of libc
5212020-10-09T17:18:46  *** cryptapus has quit IRC
5222020-10-09T17:18:56  <wumpus> but to answer your questino, no, I've never seen it
5232020-10-09T17:19:02  *** cryptapus has joined #bitcoin-core-dev
5242020-10-09T17:19:11  <achow101> great..
5252020-10-09T17:19:13  <luke-jr> valgrind it
5262020-10-09T17:20:18  <achow101> good idea
5272020-10-09T17:23:48  *** mdunnio has joined #bitcoin-core-dev
5282020-10-09T17:28:34  <roconnor> very early on, the glasgow haskell compiler had an infamous bug where it would delete the source file if you tried to compile it and it had a type error.
5292020-10-09T17:28:51  *** mdunnio has quit IRC
5302020-10-09T17:29:27  *** mdunnio has joined #bitcoin-core-dev
5312020-10-09T17:32:01  <luke-jr> XD
5322020-10-09T17:37:33  <sipa> roconnor: type errors are *really* frowned upon
5332020-10-09T17:41:17  <achow101> ah, I had a double free
5342020-10-09T17:43:01  <sipa> achow101: see youtube link above
5352020-10-09T17:44:21  <achow101> heh
5362020-10-09T17:49:45  *** cryptapus has quit IRC
5372020-10-09T17:51:46  *** cryptapus has joined #bitcoin-core-dev
5382020-10-09T17:51:47  *** cryptapus has quit IRC
5392020-10-09T17:51:47  *** cryptapus has joined #bitcoin-core-dev
5402020-10-09T17:59:56  *** filchef has joined #bitcoin-core-dev
5412020-10-09T18:00:01  *** kers has quit IRC
5422020-10-09T18:04:27  *** jonatack has quit IRC
5432020-10-09T18:05:20  *** jonatack has joined #bitcoin-core-dev
5442020-10-09T18:22:14  *** ButterflyOfFire has joined #bitcoin-core-dev
5452020-10-09T18:27:54  *** GankMove has quit IRC
5462020-10-09T18:31:34  *** owowo has quit IRC
5472020-10-09T18:34:53  <wumpus> gah, the peer keeps disconnectiong outgoing connections without logging the reason at all (even with debug=net)
5482020-10-09T18:35:15  <sipa> wumpus: and just with #19954 ?
5492020-10-09T18:35:18  <gribble> https://github.com/bitcoin/bitcoin/issues/19954 | tor: complete the TORv3 implementation by vasild · Pull Request #19954 · bitcoin/bitcoin · GitHub
5502020-10-09T18:35:50  <wumpus> sipa: yes
5512020-10-09T18:35:59  <wumpus> without it, it makes outgoing connections succesfully
5522020-10-09T18:36:55  <sipa> bizarre, and ungood
5532020-10-09T18:37:26  *** owowo has joined #bitcoin-core-dev
5542020-10-09T18:38:21  *** Pavlenex has quit IRC
5552020-10-09T18:40:39  <Murch> sipa: I don't remember any recent posts to that effect.
5562020-10-09T18:44:57  <wumpus> curiously it even happens with listen=0
5572020-10-09T18:46:47  <wumpus> well maybe not so curiously because it's about outgoing connections but at least it's not the tor bind stuff itself that interferes
5582020-10-09T18:47:44  *** ares_ has quit IRC
5592020-10-09T18:50:39  <jonatack> wumpus: when did the issue start? since the latest 19954 pushes?
5602020-10-09T18:51:42  <wumpus> jonatack: I think so, do we have a previous version of the PR somewhere?
5612020-10-09T18:51:57  <wumpus> I'm pretty sure this node was running with TORv3 succesfully at some point
5622020-10-09T18:52:54  <wumpus> will try to bisect the commits
5632020-10-09T18:53:21  <jonatack> there were a couple of important changes this week: on tuesday and today, iirc
5642020-10-09T18:53:56  <wumpus> first I'll try testing the branch itself instead of the branch merged on top of master
5652020-10-09T18:54:51  <wumpus> to rule out a silent merge conflict
5662020-10-09T18:56:18  *** Pavlenex has joined #bitcoin-core-dev
5672020-10-09T18:57:00  *** bitcoin-git has joined #bitcoin-core-dev
5682020-10-09T18:57:01  <bitcoin-git> [bitcoin] jonatack opened pull request #20115: cli: -netinfo quick updates/fixups and release note (master...netinfo-fixups) https://github.com/bitcoin/bitcoin/pull/20115
5692020-10-09T18:57:02  *** bitcoin-git has left #bitcoin-core-dev
5702020-10-09T19:00:07  <meshcollider> #startmeeting
5712020-10-09T19:00:07  <lightningbot> Meeting started Fri Oct  9 19:00:07 2020 UTC.  The chair is meshcollider. Information about MeetBot at http://wiki.debian.org/MeetBot.
5722020-10-09T19:00:07  <lightningbot> Useful Commands: #action #agreed #help #info #idea #link #topic.
5732020-10-09T19:00:14  <achow101> hi\
5742020-10-09T19:00:24  <meshcollider> #bitcoin-core-dev Wallet Meeting: wumpus sipa gmaxwell jonasschnelli morcos luke-jr sdaftuar jtimon cfields petertodd kanzure bluematt instagibbs phantomcircuit codeshark michagogo marcofalke paveljanik NicolasDorier jl2012 achow101 meshcollider jnewbery maaku fanquake promag provoostenator aj Chris_Stewart_5 dongcarl gwillen jamesob ken281221 ryanofsky gleb moneyball ariard digi_james amiti fjahr
5752020-10-09T19:00:24  <meshcollider> jeremyrubin emilengler jonatack hebasto jb55 kvaciral ariard digi_james amiti fjahr jeremyrubin lightlike emilengler jonatack hebasto jb55 elichai2
5762020-10-09T19:00:31  <jonatack> hi
5772020-10-09T19:00:49  <hebasto> hi
5782020-10-09T19:00:55  <meshcollider> 2:09 AM <achow101> #proposedwalletmeetingtopic: wallet.dat vs wallet.sqlite
5792020-10-09T19:01:26  <meshcollider> #topic wallet.dat vs wallet.sqlite
5802020-10-09T19:01:50  <achow101> for sqlite wallets, there's been an ongoing question of whether the sqlite wallet files should be named wallet.dat or wallet.sqlite
5812020-10-09T19:02:52  <hebasto> what are pros and cons of each approach?
5822020-10-09T19:03:12  <achow101> the PR currently implements wallet.dat. ryanofsky has been arguing for wallet.sqlite in his review comments
5832020-10-09T19:03:19  <achow101> i wanted to hear what everyone else thinks
5842020-10-09T19:03:42  <Murch> hello
5852020-10-09T19:03:47  <hebasto> I've just started to review that pr, so have no opinion
5862020-10-09T19:03:50  <achow101> for wallet.dat, the arguments are to maintain backwards compatibility with external documentation and tooling, as well as not causing a problem with a specific downgrade scenario
5872020-10-09T19:04:44  <meshcollider> Yeah calling it wallet.dat has the advantage that automatic backup scripts, etc. will continue working fine, and also that users are already conditioned to protecting wallet.dat files
5882020-10-09T19:04:56  <achow101> for wallet.sqlite, it's a clearer naming convention, follows sqlite naming convention, and can't be confused with bdb
5892020-10-09T19:05:38  <jonatack> hm, good arguments for both
5902020-10-09T19:05:41  <achow101> wallet.sqlite also avoids a different set of compatibility prolems
5912020-10-09T19:05:47  <sipa> yeah, i'm very much in the middle
5922020-10-09T19:06:08  <sipa> some of the naming conventions and expectations around them were already broken when we moved to per-wallet directories
5932020-10-09T19:06:20  <sipa> and i don't recall that causing many issues for users
5942020-10-09T19:06:35  <achow101> relevant commens. for wallet.dat: https://github.com/bitcoin/bitcoin/pull/19077#issuecomment-705180018 for wallet.sqlite: https://github.com/bitcoin/bitcoin/pull/19077#pullrequestreview-504980287
5952020-10-09T19:06:43  <sipa> though, the specific "wallet.dat must be protected with your life" filename convention remained
5962020-10-09T19:06:50  <luke-jr> meshcollider: it is already wrong and risks corruption to copy wallet.dat directly
5972020-10-09T19:06:57  <fjahr> hi
5982020-10-09T19:07:00  <hebasto> if users adopted per-wallet dir, we could expect such adoption for .sqlite
5992020-10-09T19:07:03  <sipa> luke-jr: it doesn't if you do it while bitcoind is shut down
6002020-10-09T19:07:14  <meshcollider> luke-jr: that doesn't stop users doing it
6012020-10-09T19:07:24  <luke-jr> meshcollider: breaking such scripts would be an advantage to renaming
6022020-10-09T19:07:37  <sipa> luke-jr: i couldn't disagree more with that
6032020-10-09T19:07:39  <luke-jr> a possible issue is restoring though
6042020-10-09T19:07:56  <achow101> luke-jr: but breaking such scripts would probably result in backups not being made, which is dangerous
6052020-10-09T19:07:58  *** ares_ has joined #bitcoin-core-dev
6062020-10-09T19:08:11  <luke-jr> achow101: most likely would result in errors instead of possibly-corrupt backups
6072020-10-09T19:08:17  <achow101> I would be surprised if said scripts failed in a way that was obvious to the person running it
6082020-10-09T19:08:21  <luke-jr> O.o
6092020-10-09T19:08:26  <sipa> luke-jr: i agree that we should discourage bad practice, but (a) not by making decisions that can actually cause people to lose money and (b) i disagree this is unsupported - it's only supported when bitcoind is not running though
6102020-10-09T19:08:28  <meshcollider> And there are also other tools I imagine, not just backup scripts, which look for wallet.dat by default
6112020-10-09T19:08:34  <luke-jr> if a cronjob fails, typically you get an email
6122020-10-09T19:09:00  <luke-jr> sipa: at least some versions would reuslt in corruption even if bitcoind exited cleanly
6132020-10-09T19:09:07  <achow101> luke-jr: fwiw I have a system backup cronjob and I don't know when/if it fails until I check the logs, and that happens maybe once every 6 months
6142020-10-09T19:09:09  <sipa> how so?
6152020-10-09T19:09:17  <luke-jr> sipa: we used to not flush/close the db
6162020-10-09T19:09:32  <luke-jr> achow101: you should fix that :p
6172020-10-09T19:09:47  <sipa> luke-jr: i believe that was very briefly the case, in ancient times
6182020-10-09T19:09:50  <achow101> luke-jr: right, but that's an example of a backup script failing and the user not knowing
6192020-10-09T19:10:05  <luke-jr> I suppose people doing backups wrong, are also likely to do error notifications wrong
6202020-10-09T19:10:33  <sipa> people will do lots of things wrong
6212020-10-09T19:10:43  <sipa> doesn't mean we shouldn't do a best effort to avoid them losing money
6222020-10-09T19:10:55  <luke-jr> but wallet.dat are currently in a dedicated directory
6232020-10-09T19:11:00  <luke-jr> there's no need for that for sqlite, right?
6242020-10-09T19:11:10  <sipa> that's a good question
6252020-10-09T19:11:34  <achow101> luke-jr: yes, but I think it would be more confusing to users if we stopped doing that
6262020-10-09T19:11:42  <sipa> hmm
6272020-10-09T19:11:49  <jonatack> modulo the risk of users losing money if renamed, a risk i don't feel competent to evaluate, i tend to agree with ryanofsky's arguments
6282020-10-09T19:12:09  <sipa> to me, that'd be one of the advantages of sqlite... not needing a directory for every wallet anymore
6292020-10-09T19:12:51  <luke-jr> btw, even if it's wallet.sqlite, it's not like we're renaming without the user knowing
6302020-10-09T19:13:15  <luke-jr> wouldn't you expect anyone setting up a backup script to check that it works when they create the wallet, at least once? :P
6312020-10-09T19:13:20  <achow101> luke-jr: not necessarily. they'd need to read the release notes, and who the hell does that?
6322020-10-09T19:13:47  <luke-jr> achow101: you're seriously suggesting automatically transforming BDB to sqlite?
6332020-10-09T19:14:04  <luke-jr> without user interaction?
6342020-10-09T19:14:15  <achow101> luke-jr: there's no transformation
6352020-10-09T19:14:31  <achow101> what I mean is that sqlite would be default for descriptor wallets, but the only way you would know that is to read the release notes
6362020-10-09T19:14:32  <luke-jr> ok, so wallet.dat would remain wallet.dat even if new wallets are wallet.sqlite…
6372020-10-09T19:14:51  <achow101> existing wallets are unaffected
6382020-10-09T19:15:17  <luke-jr> so the only way someone should lose data is if they never check for a successful backup ever..
6392020-10-09T19:15:31  <achow101> sipa: I suppose that getting rid of the wallet directory thing would solve both of these problems
6402020-10-09T19:15:33  <luke-jr> or  maybe are backing up numerous wallets and expect newly created ones automatically included
6412020-10-09T19:16:57  <achow101> luke-jr: when we get around to implementing bdb to sqlite migration, there could be problems there with the rename
6422020-10-09T19:17:18  <luke-jr> achow101: but we get the chance to tell users when they opt into it
6432020-10-09T19:17:23  <achow101> true
6442020-10-09T19:17:46  <luke-jr> a reason not to rename: acting on file extensions has been kindof deprecated for a long time?
6452020-10-09T19:18:46  <achow101> there's also the problems with restoring, and that one downgrade case where a new wallet.dat is made
6462020-10-09T19:19:43  <meshcollider> IMO we should get rid of individual directories for sqlite, I don't think that would be confusing
6472020-10-09T19:21:29  <achow101> meshcollider: that still has the backup and restore problems, although not the downgrade one if we name the file as the wallet name
6482020-10-09T19:26:25  <achow101> any other comments on this topic?
6492020-10-09T19:27:13  <meshcollider> Did this help make a decision ;)
6502020-10-09T19:27:28  <achow101> not really
6512020-10-09T19:27:33  <fjahr> I'm undecided as well, sorry
6522020-10-09T19:28:14  <achow101> I'll experiment with a no wallet directory approach and see how big the diff is
6532020-10-09T19:28:25  <meshcollider> Yeah that sounds good
6542020-10-09T19:28:27  <meshcollider> Any other topics then?
6552020-10-09T19:29:56  <jonatack> fjahr: at some point, sometime, we should maybe discuss #18418
6562020-10-09T19:29:58  <gribble> https://github.com/bitcoin/bitcoin/issues/18418 | wallet: Increase OUTPUT_GROUP_MAX_ENTRIES to 100 by fjahr · Pull Request #18418 · bitcoin/bitcoin · GitHub
6572020-10-09T19:30:15  <jonatack> perhaps Murch can look at it
6582020-10-09T19:30:43  <jonatack> (just a thought, no need to duscuss now)
6592020-10-09T19:30:55  <fjahr> Yeah, thanks, I guess at the moment nobody has time but maybe in 2 weeks
6602020-10-09T19:32:30  <meshcollider> Ok let's discuss it next time then :)
6612020-10-09T19:32:38  <meshcollider> #endmeeting
6622020-10-09T19:32:38  <lightningbot> Meeting ended Fri Oct  9 19:32:38 2020 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)
6632020-10-09T19:32:38  <lightningbot> Minutes:        http://www.erisian.com.au/meetbot/bitcoin-core-dev/2020/bitcoin-core-dev.2020-10-09-19.00.html
6642020-10-09T19:32:38  <lightningbot> Minutes (text): http://www.erisian.com.au/meetbot/bitcoin-core-dev/2020/bitcoin-core-dev.2020-10-09-19.00.txt
6652020-10-09T19:32:38  <lightningbot> Log:            http://www.erisian.com.au/meetbot/bitcoin-core-dev/2020/bitcoin-core-dev.2020-10-09-19.00.log.html
6662020-10-09T19:32:44  <jonatack> o/
6672020-10-09T19:32:45  <meshcollider> Thanks all
6682020-10-09T19:33:06  <meshcollider> \o
6692020-10-09T19:37:57  *** Pavlenex has quit IRC
6702020-10-09T19:58:17  <luke-jr> btw, I wonder if creating a wallet is really the right response to not finding one named :P
6712020-10-09T20:02:10  <roconnor> In a different program I complained where running an obviously read-only operation would create a new file if the required file didn't exist.
6722020-10-09T20:02:26  <achow101> luke-jr: it definitely isn't and we should remove that behavior
6732020-10-09T20:06:08  *** tripleslash has quit IRC
6742020-10-09T20:06:17  *** [\\\] has joined #bitcoin-core-dev
6752020-10-09T20:25:12  *** kristapsk has joined #bitcoin-core-dev
6762020-10-09T20:25:57  *** S3RK has joined #bitcoin-core-dev
6772020-10-09T20:30:35  *** S3RK has quit IRC
6782020-10-09T20:31:47  *** Guyver2_ has joined #bitcoin-core-dev
6792020-10-09T20:34:08  *** Guyver2 has quit IRC
6802020-10-09T21:00:02  *** ButterflyOfFire has quit IRC
6812020-10-09T21:02:19  *** filchef has quit IRC
6822020-10-09T21:03:02  *** Guyver2_ has quit IRC
6832020-10-09T21:05:05  *** Victorsueca has quit IRC
6842020-10-09T21:21:25  *** wright has joined #bitcoin-core-dev
6852020-10-09T21:31:53  *** brianhoffman has joined #bitcoin-core-dev
6862020-10-09T21:32:30  *** brianhoffman_ has quit IRC
6872020-10-09T21:37:23  *** k3tan has quit IRC
6882020-10-09T21:41:42  *** k3tan has joined #bitcoin-core-dev
6892020-10-09T21:54:43  *** Talkless has joined #bitcoin-core-dev
6902020-10-09T22:10:43  *** vasild has quit IRC
6912020-10-09T22:12:17  *** vasild has joined #bitcoin-core-dev
6922020-10-09T22:20:11  *** tryphe has quit IRC
6932020-10-09T22:20:57  *** tryphe has joined #bitcoin-core-dev
6942020-10-09T22:21:35  <sipa> wumpus: any progress?
6952020-10-09T22:51:45  *** mdunnio has quit IRC
6962020-10-09T22:56:42  *** mdunnio has joined #bitcoin-core-dev
6972020-10-09T22:59:22  *** S3RK has joined #bitcoin-core-dev
6982020-10-09T23:02:01  *** mdunnio has quit IRC
6992020-10-09T23:04:19  *** S3RK has quit IRC
7002020-10-09T23:05:38  *** Talkless has quit IRC
7012020-10-09T23:09:49  *** mol_ has quit IRC
7022020-10-09T23:23:41  *** mol has joined #bitcoin-core-dev
7032020-10-09T23:34:04  *** mdunnio has joined #bitcoin-core-dev
7042020-10-09T23:34:13  *** mol has quit IRC
7052020-10-09T23:39:22  *** mdunnio has quit IRC
7062020-10-09T23:47:14  *** mol has joined #bitcoin-core-dev
7072020-10-09T23:47:46  *** pinheadmz has quit IRC
7082020-10-09T23:52:03  *** pinheadmz has joined #bitcoin-core-dev
7092020-10-09T23:58:36  *** pinheadmz has quit IRC