12018-10-22T00:18:49  *** josephnicholas has joined #bitcoin-core-dev
  22018-10-22T00:23:50  *** jarthur has joined #bitcoin-core-dev
  32018-10-22T00:32:08  *** josephnicholas has quit IRC
  42018-10-22T00:34:01  *** rh0nj has quit IRC
  52018-10-22T00:35:08  *** rh0nj has joined #bitcoin-core-dev
  62018-10-22T00:39:51  *** mistergo1d has quit IRC
  72018-10-22T00:44:24  *** justanotheruser has joined #bitcoin-core-dev
  82018-10-22T00:44:46  *** bitcoin-git has joined #bitcoin-core-dev
  92018-10-22T00:44:46  <bitcoin-git> [bitcoin] MarcoFalke closed pull request #14540: Enable flake8 rule E231. (master...flake8-fix-E231) https://github.com/bitcoin/bitcoin/pull/14540
 102018-10-22T00:44:46  *** bitcoin-git has left #bitcoin-core-dev
 112018-10-22T00:46:00  *** justan0theruser has quit IRC
 122018-10-22T00:46:07  *** bitconner has joined #bitcoin-core-dev
 132018-10-22T00:51:04  *** bitconner has quit IRC
 142018-10-22T00:54:28  *** opdenkamp has quit IRC
 152018-10-22T00:58:57  *** opdenkamp has joined #bitcoin-core-dev
 162018-10-22T01:14:59  *** josephnicholas has joined #bitcoin-core-dev
 172018-10-22T01:19:07  *** josephnicholas has quit IRC
 182018-10-22T01:23:58  *** robo7 has joined #bitcoin-core-dev
 192018-10-22T01:44:01  *** josephnicholas has joined #bitcoin-core-dev
 202018-10-22T01:48:42  *** josephnicholas has quit IRC
 212018-10-22T01:49:34  *** josephnicholas has joined #bitcoin-core-dev
 222018-10-22T01:55:44  *** rhavar has joined #bitcoin-core-dev
 232018-10-22T02:03:46  *** josephnicholas has quit IRC
 242018-10-22T02:05:15  *** jarthur has quit IRC
 252018-10-22T02:05:55  *** jarthur has joined #bitcoin-core-dev
 262018-10-22T02:13:41  *** josephnicholas has joined #bitcoin-core-dev
 272018-10-22T02:16:53  *** justan0theruser has joined #bitcoin-core-dev
 282018-10-22T02:17:47  *** justanotheruser has quit IRC
 292018-10-22T02:18:24  *** josephnicholas has quit IRC
 302018-10-22T02:28:45  *** josephnicholas has joined #bitcoin-core-dev
 312018-10-22T02:33:22  *** josephnicholas has quit IRC
 322018-10-22T02:41:27  *** Victorsueca has quit IRC
 332018-10-22T02:42:46  *** Victorsueca has joined #bitcoin-core-dev
 342018-10-22T02:49:45  *** ken2812221_ has joined #bitcoin-core-dev
 352018-10-22T02:51:07  *** ken2812221 has quit IRC
 362018-10-22T02:51:14  *** ken2812221_ is now known as ken2812221
 372018-10-22T03:00:37  *** michaelsdunn1 has joined #bitcoin-core-dev
 382018-10-22T03:07:20  *** michaelsdunn1 has quit IRC
 392018-10-22T03:22:59  *** fanquake has joined #bitcoin-core-dev
 402018-10-22T03:23:30  <fanquake> MarcoFalke Not sure if your reply was automated, but it's been posted 5 times on #14540
 412018-10-22T03:23:32  <gribble> https://github.com/bitcoin/bitcoin/issues/14540 | Enable flake8 rule E231. by jbampton · Pull Request #14540 · bitcoin/bitcoin · GitHub
 422018-10-22T03:25:36  <gmaxwell> where time becomes a loop.
 432018-10-22T03:25:43  <fanquake> I'm also seeing lots of GH weirdness, duplicated emails etc.
 442018-10-22T03:25:51  <kallewoof> am I the only one who keeps getting 'You can't comment at this time' on github?
 452018-10-22T03:25:59  <phantomcircuit> sipa, hmm i guess if we're careful it can be edge triggered and then just set a recvable flag
 462018-10-22T03:26:24  <phantomcircuit> i should note that it seems a bit weird to me to be waiting for the write buffer to empty before calling recv though
 472018-10-22T03:26:41  <phantomcircuit> the comment says something about tcp congestion control but im not sure how that's related exactly?
 482018-10-22T03:26:46  *** josephnicholas has joined #bitcoin-core-dev
 492018-10-22T03:28:16  *** Krellan has quit IRC
 502018-10-22T03:31:07  *** josephnicholas has quit IRC
 512018-10-22T03:35:25  <echeveria> kallewoof: github is down basically.
 522018-10-22T03:35:51  <echeveria> https://status.github.com/
 532018-10-22T03:36:26  <echeveria> “We continue working to repair a data storage system for GitHub.com. You may see inconsistent results during this process.”
 542018-10-22T03:39:01  <fanquake> echeveria good to know. Was getting emailed about comments that didn't seem to exist on the site, thought I was going crazy.
 552018-10-22T03:43:44  *** bralyclo_ has quit IRC
 562018-10-22T03:44:30  *** Bullitje has quit IRC
 572018-10-22T03:45:22  *** bralyclow has joined #bitcoin-core-dev
 582018-10-22T03:50:15  <kallewoof> echeveria: Thanks! I didn&t realize that
 592018-10-22T03:50:25  *** Bullit has joined #bitcoin-core-dev
 602018-10-22T03:50:53  *** Krellan has joined #bitcoin-core-dev
 612018-10-22T03:51:29  <kallewoof> I'm seeing "wallet/walletutil.cpp:57:78: error: ‘end’ was not declared in this scope" aside from "wallet/walletutil.cpp:60:32: error: ‘relative’ is not a member of ‘fs’" on debian jessie (ppc bigendian) btw (@promag). Not sure if that was addressed in follow-up PR
 622018-10-22T03:52:20  <kallewoof> Where would this non-namespaced 'end()' come from, anyway? Very confused.
 632018-10-22T03:53:39  *** schnerch_ has joined #bitcoin-core-dev
 642018-10-22T03:56:51  *** schnerchi has quit IRC
 652018-10-22T04:15:04  <sipa> phantomcircuit: i think the reasoning is that most incoming messages are things we need to respond to, so if our send buffer is still (sufficiently) full, we can delay processing it (and prefer working on stuff given by other peers instead) until the send buffer empties
 662018-10-22T04:18:46  <fanquake>  kallewoof should be fixed in #14531. I assume your compiling with Boost < 1.64.0 ?
 672018-10-22T04:18:49  <gribble> https://github.com/bitcoin/bitcoin/issues/14531 | Replace fs::relative call with custom GetRelativePath by promag · Pull Request #14531 · bitcoin/bitcoin · GitHub
 682018-10-22T04:19:14  <kallewoof> fanquake: I don't see a fix for the undeclared 'end' but will look again
 692018-10-22T04:20:03  <kallewoof> fanquake: Yeah, that mysterious end() is still there..
 702018-10-22T04:20:29  <kallewoof> boost is 1.55, it apperas.
 712018-10-22T04:45:06  *** jarthur has quit IRC
 722018-10-22T04:46:05  *** jarthur has joined #bitcoin-core-dev
 732018-10-22T04:57:10  <phantomcircuit> sipa, except during ibd where we're asking for lots of things we wont respond to
 742018-10-22T04:57:41  <phantomcircuit> and we're asking them of peers on potentially asymetric links, ie asking is actually slower than sending for them
 752018-10-22T05:27:51  *** sanket1729 has joined #bitcoin-core-dev
 762018-10-22T05:33:56  <sanket1729> Hello, can anyone comment on the following claim. "If there is a clock drifts more than 2 hours and and there are 2 chains similar PoW, then we can have a fork where both chains think they are longest in their own view. So, bitcoin network operates on some synchronous assumptions."
 772018-10-22T05:35:17  <sanket1729> By synchronous I mean, all network  participants are assumed to have max drift of 2 hours.
 782018-10-22T05:35:41  <echeveria> sanket1729: wall time isn't used in block validation like that.
 792018-10-22T05:37:56  *** sanket1729 has quit IRC
 802018-10-22T05:38:17  *** sanket1729 has joined #bitcoin-core-dev
 812018-10-22T05:41:47  *** ken2812221 has quit IRC
 822018-10-22T05:42:10  *** ken2812221 has joined #bitcoin-core-dev
 832018-10-22T05:42:54  *** hebasto has joined #bitcoin-core-dev
 842018-10-22T05:44:28  *** Zenton has quit IRC
 852018-10-22T05:45:58  <sanket1729> This code appears to reject blocks 2 hours into future based on local time. https://github.com/bitcoin/bitcoin/blob/5c25409d6851182c5e351720cee36812c229b77a/src/validation.cpp#L3245 . What am I missing?
 862018-10-22T05:49:37  *** sanket1729 has quit IRC
 872018-10-22T05:49:58  *** sanket1729 has joined #bitcoin-core-dev
 882018-10-22T05:52:27  <kallewoof> sanket1729: It is true that a node will reject a block that is too far in the future *now*, but once time passes, the block ends up being acceptable. It's just delayed. So nodes building blocks will ignore it for a bit, which probably means it ends up being orphaned.
 892018-10-22T05:52:35  *** robo7 has quit IRC
 902018-10-22T05:52:47  <kallewoof> "nodes building blocks" = miners
 912018-10-22T05:55:29  <echeveria> realistically you either have an accurate clock, or yours is days out very quickly. real time clocks in computers are especially bad.
 922018-10-22T05:56:02  <echeveria> to the scale of losing seconds a day. they aren't designed to be free running.
 932018-10-22T05:56:22  <sanket1729> Thanks, but theoretically if we have 2 chains with similar PoW, then we can have 2 forks which are longest in their own view. For ex, chain c1 is in 2 hour lead with respect to chain c2. c1 and c2 have similar PoW, then we can have a scenario where 2 forks exist both of which are longest in their own view
 942018-10-22T05:57:16  <sanket1729> or atleast there could be weird oscillatory type reorgs back and forth
 952018-10-22T05:57:24  <wumpus> "you can't comment at this time." ahh thanks github...
 962018-10-22T05:57:35  <wumpus> yep
 972018-10-22T05:57:35  <echeveria> wumpus: status.github.com
 982018-10-22T05:57:44  <sipa> sanket1729: the chain with the earlier timestamps is acceptable to both sets of nodes
 992018-10-22T05:58:28  <sipa> but yes, there is an assumptions that clocks's aren't off too much
1002018-10-22T05:59:20  <sanket1729> I am trying to figure out what worst scenario that can occur because of these clock assumptions.
1012018-10-22T05:59:32  <echeveria> realistically I've crawled the p2p network and found that almost all nodes use NTP.
1022018-10-22T05:59:58  <echeveria> they either use NTP, or are so out of sync with the real world that they would be better using a sundial. there's only tens of reachable peers like this.
1032018-10-22T06:06:12  *** josephnicholas has joined #bitcoin-core-dev
1042018-10-22T06:06:42  <sipa> i see a number of PRs closed when in their own pages, but they're listed as open on the pr overview page
1052018-10-22T06:07:00  <sipa> 22:05 PDT
1062018-10-22T06:07:01  <sipa> We continue working to repair a data storage system for GitHub.com. You may see inconsistent results during this process.
1072018-10-22T06:07:06  <sipa> ^ explains
1082018-10-22T06:10:09  <kallewoof> sanket1729: if 51% of miners magically have a clock set 1 day in the future, they can build a chain that, for awhile, is considered invalid to the rest of the network. if they keep going at 1 day ahead of the rest, they will be building a delayed, and considered invalid (despite having slightly more work) chain for awhile. eventually the valid part of their chain will outgrow the alternative chain. I believe this could
1092018-10-22T06:10:09  <kallewoof> result in a case where new nodes will accept the future-built chain while old nodes will not, but I'm not sure. That'd be bad, but very drawn out and expensive (and meaningless, I think) to do..
1102018-10-22T06:10:52  *** josephnicholas has quit IRC
1112018-10-22T06:13:27  *** ken2812221 has quit IRC
1122018-10-22T06:24:00  <sanket1729> Maybe 51% miners can mess around with difficulty adjustment. Let's say 51% miner can afford to do the timestamp attack for k blocks, then he would want the last k blocks in difficulty period to be placed at (2 hours + k*10 mins), right?
1132018-10-22T06:24:19  *** jarthur has quit IRC
1142018-10-22T06:25:54  <echeveria> you'd exploit the difficulty adjustment off by one, rather than that.
1152018-10-22T06:26:06  <sanket1729> All the timewrap bugs only considered 2016th block to have +2 hours, maybe we can do more with a little more risk.
1162018-10-22T06:26:14  <sanket1729> difficulty bug + this thing
1172018-10-22T06:26:56  <echeveria> there's much lower hanging fruit than that even.
1182018-10-22T06:28:11  <echeveria> with 10% of the hash rate you can reverse a 1 confirmation transaction 21% of the time, and most exchanges seem to accept 1 confirmation deposits. so realistically if anybody felt like it there's no need to do anything novel. this has happened, and will likely happen again.
1192018-10-22T06:28:46  <sanket1729> I am trying to explore some theoretical things, I know this will never happen realistically :)
1202018-10-22T06:28:59  <echeveria> might be best to continue this in #bitcoin then.
1212018-10-22T06:30:33  *** bitconner has joined #bitcoin-core-dev
1222018-10-22T06:34:47  *** bitconner has quit IRC
1232018-10-22T06:35:30  *** sanket1729 has quit IRC
1242018-10-22T06:38:25  <echeveria> it'll be an interesting github post mortem. pretty long downtime.
1252018-10-22T07:00:37  *** sanket1729 has joined #bitcoin-core-dev
1262018-10-22T07:01:49  *** klot has joined #bitcoin-core-dev
1272018-10-22T07:01:56  *** sanket1729 has quit IRC
1282018-10-22T07:02:22  *** klot has quit IRC
1292018-10-22T07:02:48  *** klot has joined #bitcoin-core-dev
1302018-10-22T07:03:52  *** klot has quit IRC
1312018-10-22T07:04:19  *** klot has joined #bitcoin-core-dev
1322018-10-22T07:05:22  *** klot has quit IRC
1332018-10-22T07:05:44  <provoostenator> Github is still in a bad mood. Seeing "You can't comment at this time", unicorns all over the place and "Start Review" doesn't work.
1342018-10-22T07:05:52  *** klot has joined #bitcoin-core-dev
1352018-10-22T07:06:38  <sipa> phantomcircuit: status.github.com
1362018-10-22T07:06:41  <sipa> eh provoostenator ^
1372018-10-22T07:07:15  *** ken2812221 has joined #bitcoin-core-dev
1382018-10-22T07:08:09  <provoostenator> Accurate status page, that's new :-)
1392018-10-22T07:17:27  *** promag has quit IRC
1402018-10-22T07:20:15  *** phwalkr has joined #bitcoin-core-dev
1412018-10-22T07:36:01  *** rh0nj has quit IRC
1422018-10-22T07:37:07  *** rh0nj has joined #bitcoin-core-dev
1432018-10-22T07:45:14  *** rhavar has quit IRC
1442018-10-22T07:49:22  *** luke-jr has quit IRC
1452018-10-22T07:49:32  *** luke-jr has joined #bitcoin-core-dev
1462018-10-22T07:55:37  *** jungly_ has joined #bitcoin-core-dev
1472018-10-22T07:57:19  <wumpus> hehe yes that's a new one,typically no one even bothers to check status pages because they tend to be only updated when a problem is already fixed
1482018-10-22T08:03:14  *** setpill has joined #bitcoin-core-dev
1492018-10-22T08:04:45  *** setpill has quit IRC
1502018-10-22T08:07:20  *** setpill has joined #bitcoin-core-dev
1512018-10-22T08:08:32  <wumpus> it's taking long though must be a serious issue
1522018-10-22T08:09:53  <jonasschnelli> sipa: with the CNetMessageSerializer, that would be an instance per peer?
1532018-10-22T08:10:38  <jonasschnelli> Maybe even a global instance for v1 / v2(enc.) protocol?
1542018-10-22T08:13:41  *** setpill has quit IRC
1552018-10-22T08:14:23  <kallewoof> https://blog.github.com/2018-10-21-october21-incident-report/
1562018-10-22T08:15:39  *** setpill has joined #bitcoin-core-dev
1572018-10-22T08:15:54  <kallewoof> "multiple services on GitHub.com were affected by a network partition and subsequent database failure resulting in inconsistent information being presented on our website"
1582018-10-22T08:25:16  *** timothy has joined #bitcoin-core-dev
1592018-10-22T08:29:48  <echeveria> a little unclear what happened there. smart tools being dumb?
1602018-10-22T08:30:48  *** bitconner has joined #bitcoin-core-dev
1612018-10-22T08:31:58  *** martinus_ has joined #bitcoin-core-dev
1622018-10-22T08:34:15  *** ken2812221 has quit IRC
1632018-10-22T08:35:33  *** josephnicholas has joined #bitcoin-core-dev
1642018-10-22T08:35:47  *** bitconner has quit IRC
1652018-10-22T08:40:07  *** josephnicholas has quit IRC
1662018-10-22T08:44:56  <hebasto> how can I get blocks/ and chainstate/ folders as if the node was offline some blocks (for testnet)?
1672018-10-22T08:46:58  *** ken2812221 has joined #bitcoin-core-dev
1682018-10-22T08:50:31  *** promag has joined #bitcoin-core-dev
1692018-10-22T08:51:37  *** josephnicholas has joined #bitcoin-core-dev
1702018-10-22T08:51:46  <promag> kallewoof: thanks
1712018-10-22T08:53:42  <kallewoof> promag: NP! What is end()? :o
1722018-10-22T08:53:51  <kallewoof> Or, where, rather.
1732018-10-22T08:54:17  *** luke-jr has quit IRC
1742018-10-22T08:54:32  *** luke-jr has joined #bitcoin-core-dev
1752018-10-22T08:54:40  <promag> std::end
1762018-10-22T08:58:48  <promag> i think for (auto it : fs::recursive_directory_iterator(wallet_dir)) should work
1772018-10-22T08:58:53  <promag> I'll try in a bit
1782018-10-22T09:06:29  <promag> kallewoof: actually must be fs::end()
1792018-10-22T09:07:01  <kallewoof> But you're not "using" any namespaces and it doesn't have a namespace prefix in the code. I'm so confused why this compiles.
1802018-10-22T09:12:14  *** berndj-blackout has joined #bitcoin-core-dev
1812018-10-22T09:12:21  *** berndj has quit IRC
1822018-10-22T09:14:01  *** berndj-blackout is now known as berndj
1832018-10-22T09:15:31  *** Victorsueca has quit IRC
1842018-10-22T09:16:46  *** Victorsueca has joined #bitcoin-core-dev
1852018-10-22T09:21:40  *** Guyver2 has joined #bitcoin-core-dev
1862018-10-22T09:21:54  *** tryphe has quit IRC
1872018-10-22T09:23:39  *** tryphe has joined #bitcoin-core-dev
1882018-10-22T09:30:07  <promag> kallewoof: can you try replacing with "    for (auto it = fs::recursive_directory_iterator(wallet_dir); it != fs::recursive_directory_iterator(); ++it) { " ?
1892018-10-22T09:30:39  <kallewoof> sure thing, one sec
1902018-10-22T09:32:10  <kallewoof> That compiles (but still having the relative not a part of fs errors)
1912018-10-22T09:32:11  <promag> ty
1922018-10-22T09:32:20  <promag> right
1932018-10-22T09:32:29  <promag> that was identified earlier
1942018-10-22T09:32:41  <kallewoof> *nod*
1952018-10-22T09:33:20  <promag> it turns out that fs::relative is doing something that we don't want
1962018-10-22T09:33:36  <promag> see this comment https://github.com/bitcoin/bitcoin/pull/14291/files#r226843422
1972018-10-22T09:42:33  *** afsd has joined #bitcoin-core-dev
1982018-10-22T09:47:09  *** afsd has quit IRC
1992018-10-22T09:52:09  *** Victorsueca has quit IRC
2002018-10-22T09:53:15  *** fanquake has quit IRC
2012018-10-22T09:53:36  <promag> I guess better not comment on gh for now
2022018-10-22T09:58:47  <echeveria> yeah they backed off on their estimates. back to an hour and a half.
2032018-10-22T10:01:41  *** davec has quit IRC
2042018-10-22T10:02:57  *** davec has joined #bitcoin-core-dev
2052018-10-22T10:09:40  *** Victorsueca has joined #bitcoin-core-dev
2062018-10-22T10:12:04  *** owowo has quit IRC
2072018-10-22T10:17:17  *** owowo has joined #bitcoin-core-dev
2082018-10-22T10:18:27  *** promag has quit IRC
2092018-10-22T10:36:44  *** spinza has quit IRC
2102018-10-22T10:38:48  *** promag has joined #bitcoin-core-dev
2112018-10-22T10:42:04  *** phwalkr has quit IRC
2122018-10-22T10:45:18  <promag> kallewoof: can you try https://github.com/promag/bitcoin/tree/2018-10-getrelativepath?
2132018-10-22T10:46:03  <promag> it should fix listwalletdir behavior and building with boost 1.47
2142018-10-22T10:54:27  *** josephnicholas has quit IRC
2152018-10-22T10:54:41  *** spinza has joined #bitcoin-core-dev
2162018-10-22T10:55:36  *** josephnicholas has joined #bitcoin-core-dev
2172018-10-22T10:56:20  *** AaronvanW has joined #bitcoin-core-dev
2182018-10-22T11:05:48  <wumpus> huh looks like my comment did go through a zillion times, wtf
2192018-10-22T11:06:02  *** rh0nj has quit IRC
2202018-10-22T11:07:07  *** rh0nj has joined #bitcoin-core-dev
2212018-10-22T11:07:25  *** infernix has quit IRC
2222018-10-22T11:08:44  <promag> wumpus: I thought you wrote it multiple times :P
2232018-10-22T11:08:48  *** promag has quit IRC
2242018-10-22T11:09:29  <wumpus> I only wrote it two times
2252018-10-22T11:09:36  <wumpus> first time I thought I'd lost it
2262018-10-22T11:09:49  * wumpus really unhappy with github right now
2272018-10-22T11:13:03  <echeveria> they cooked something really well, and sadly the way they didn't take down the site means lots of things got massive duplicate submissions (see the number of issue emails in my inbox, derp)
2282018-10-22T11:25:06  *** josephnicholas has quit IRC
2292018-10-22T11:35:05  *** klot_ has joined #bitcoin-core-dev
2302018-10-22T11:36:20  *** klot_ has quit IRC
2312018-10-22T11:36:48  *** klot_ has joined #bitcoin-core-dev
2322018-10-22T11:37:50  *** klot_ has quit IRC
2332018-10-22T11:38:17  *** klot_ has joined #bitcoin-core-dev
2342018-10-22T11:38:51  *** klot has quit IRC
2352018-10-22T11:42:36  *** bitconner has joined #bitcoin-core-dev
2362018-10-22T11:43:13  *** ken2812221_ has joined #bitcoin-core-dev
2372018-10-22T11:44:02  *** josephnicholas has joined #bitcoin-core-dev
2382018-10-22T11:47:08  *** bitconner has quit IRC
2392018-10-22T11:55:00  *** ken2812221 has quit IRC
2402018-10-22T12:01:27  *** schmidty has joined #bitcoin-core-dev
2412018-10-22T12:01:49  *** schmidty is now known as Guest19314
2422018-10-22T12:03:59  *** SopaXorzTaker has joined #bitcoin-core-dev
2432018-10-22T12:06:38  *** infernix has joined #bitcoin-core-dev
2442018-10-22T12:07:52  *** josephnicholas has quit IRC
2452018-10-22T12:10:07  <wumpus> yess sorry for the spams
2462018-10-22T12:21:34  <luke-jr> >_<
2472018-10-22T12:29:04  *** promag has joined #bitcoin-core-dev
2482018-10-22T12:49:19  *** OzPac has joined #bitcoin-core-dev
2492018-10-22T12:52:02  *** promag has quit IRC
2502018-10-22T12:52:17  *** promag has joined #bitcoin-core-dev
2512018-10-22T12:56:40  <promag> are we there yet?
2522018-10-22T12:59:21  *** esotericnonsense has quit IRC
2532018-10-22T13:00:45  *** esotericnonsense has joined #bitcoin-core-dev
2542018-10-22T13:34:10  *** Zenton has joined #bitcoin-core-dev
2552018-10-22T14:02:34  *** ExtraCrispy has quit IRC
2562018-10-22T14:03:05  *** ExtraCrispy has joined #bitcoin-core-dev
2572018-10-22T14:08:30  *** timothy has quit IRC
2582018-10-22T14:10:24  *** setpill has quit IRC
2592018-10-22T14:12:14  *** timothy has joined #bitcoin-core-dev
2602018-10-22T14:18:28  *** timothy has quit IRC
2612018-10-22T14:18:49  *** timothy has joined #bitcoin-core-dev
2622018-10-22T14:21:01  <cdecker> ccccccfnijgituclltkdgdvkejjhrbdeevrhrkefibun
2632018-10-22T14:26:11  <BlueMatt> cdecker: ffs
2642018-10-22T14:26:24  <instagibbs> grief or yubikey press, you decide
2652018-10-22T14:26:31  <BlueMatt> why do people use that yubikey print shit anyway
2662018-10-22T14:29:06  *** michaelsdunn1 has joined #bitcoin-core-dev
2672018-10-22T14:31:28  *** Wadexs has joined #bitcoin-core-dev
2682018-10-22T14:32:02  <luke-jr> instagibbs: IMO cat on keyboard
2692018-10-22T14:32:47  <luke-jr> (should be an option()
2702018-10-22T14:33:55  <Wadexs> Bitcoin testnet exchange
2712018-10-22T14:37:34  *** jarthur has joined #bitcoin-core-dev
2722018-10-22T14:38:47  *** Wadexs has quit IRC
2732018-10-22T14:46:45  *** jcorgan has quit IRC
2742018-10-22T14:46:46  *** jcorgan has joined #bitcoin-core-dev
2752018-10-22T14:48:39  *** jungly_ has quit IRC
2762018-10-22T14:50:53  *** AaronvanW has quit IRC
2772018-10-22T14:52:01  *** AaronvanW has joined #bitcoin-core-dev
2782018-10-22T14:57:02  *** AaronvanW has quit IRC
2792018-10-22T14:59:45  *** Victorsueca has quit IRC
2802018-10-22T15:01:03  *** Victorsueca has joined #bitcoin-core-dev
2812018-10-22T15:07:14  *** AaronvanW has joined #bitcoin-core-dev
2822018-10-22T15:10:52  *** michagogo has joined #bitcoin-core-dev
2832018-10-22T15:25:05  *** booyah has joined #bitcoin-core-dev
2842018-10-22T15:29:03  *** bralyclo_ has joined #bitcoin-core-dev
2852018-10-22T15:33:08  *** bralyclow has quit IRC
2862018-10-22T15:34:51  <phantomcircuit> BlueMatt, because it was the only thing that actually worked for a while
2872018-10-22T15:35:01  <phantomcircuit> now you can do the totp stuff over nfc
2882018-10-22T15:40:25  *** AaronvanW has quit IRC
2892018-10-22T16:02:06  *** spinza has quit IRC
2902018-10-22T16:02:25  <gwillen> also because it's on by default even if you only use your yubikey for other stuff
2912018-10-22T16:02:37  <gwillen> and at least on my usb C yubikeys there's some kind of problem that prevents me from turning it off
2922018-10-22T16:02:51  <gwillen> they both came in a state where my copy of the yubikey personalization tool claims they are locked and cannot be reconfigured
2932018-10-22T16:06:54  *** Krellan has quit IRC
2942018-10-22T16:12:13  *** jungly has joined #bitcoin-core-dev
2952018-10-22T16:12:23  *** michaelfolkson has joined #bitcoin-core-dev
2962018-10-22T16:14:42  <phantomcircuit> gwillen, iirc there's a default pin
2972018-10-22T16:14:58  <phantomcircuit> gwillen, if not then that's mega suspicious
2982018-10-22T16:15:57  *** michaelfolkson has quit IRC
2992018-10-22T16:16:54  <gwillen> oh, is there? What's the default pin? My USB-A ones came with no PIN set.
3002018-10-22T16:17:31  <gwillen> in some sense it's mega-suspicious, I guess, but ... the subsystem that uses the PIN does not interact in any way with the U2F subsystem as far as I know
3012018-10-22T16:17:51  <gwillen> so if someone were going to mess with my yubikeys, which I use for U2F, it would be unimaginably sloppy to mess with the PIJ
3022018-10-22T16:17:55  <gwillen> PIN*
3032018-10-22T16:19:37  *** michaelfolkson has joined #bitcoin-core-dev
3042018-10-22T16:19:41  <phantomcircuit> gwillen, and yet...
3052018-10-22T16:19:54  <phantomcircuit> 123456 maybe?
3062018-10-22T16:26:55  *** spinza has joined #bitcoin-core-dev
3072018-10-22T16:30:11  *** Guyver2 has quit IRC
3082018-10-22T16:49:45  *** Krellan has joined #bitcoin-core-dev
3092018-10-22T16:49:51  *** bralyclow has joined #bitcoin-core-dev
3102018-10-22T16:50:01  *** rh0nj has quit IRC
3112018-10-22T16:51:08  *** rh0nj has joined #bitcoin-core-dev
3122018-10-22T16:53:10  *** bralyclo_ has quit IRC
3132018-10-22T16:53:56  *** Murch has joined #bitcoin-core-dev
3142018-10-22T17:06:06  *** AaronvanW has joined #bitcoin-core-dev
3152018-10-22T17:07:24  *** pope__ has quit IRC
3162018-10-22T17:15:09  *** jungly has quit IRC
3172018-10-22T17:25:25  <andytoshi> BlueMatt: on a yubikey 4 you can't turn it off if you want to use both ssh and u2f
3182018-10-22T17:25:35  *** Zorbak has joined #bitcoin-core-dev
3192018-10-22T17:26:09  <andytoshi> because the apparently did not have time to test their new closed-source software on all possible enable/disable combinations. i guess 8 is a pretty big number.
3202018-10-22T17:26:17  <BlueMatt> andytoshi: uhhh, its definitely off for my yubikey 4
3212018-10-22T17:26:23  <BlueMatt> ohoh, i guess i dont have u2f on my 4
3222018-10-22T17:26:25  <BlueMatt> only ssh
3232018-10-22T17:26:29  <BlueMatt> well, pgp
3242018-10-22T17:26:36  *** Zorbak has quit IRC
3252018-10-22T17:26:38  <andytoshi> yep, if you try to turn on u2f it'll just fail with a mysterious error
3262018-10-22T17:26:45  <andytoshi> or succeed but not actually turn on? i forget now
3272018-10-22T17:26:49  <BlueMatt> i have u2f off
3282018-10-22T17:26:57  <BlueMatt> i only have pgp on, though
3292018-10-22T17:27:04  <BlueMatt> not u2f + the native yubikey bullshit
3302018-10-22T17:29:02  <andytoshi> well, if you want u2f the native bullshit is gonna come along for the ride.
3312018-10-22T17:29:10  <andytoshi> so try to avoid wanting u2f
3322018-10-22T17:30:40  <BlueMatt> well I have the native bullshit turned off on my non-4 yubikey
3332018-10-22T17:30:44  <BlueMatt> the one that only does u2f+native
3342018-10-22T17:30:48  <BlueMatt> so just get you one of those
3352018-10-22T17:36:07  *** Zenton has quit IRC
3362018-10-22T17:37:56  <phantomcircuit> sipa, cant say i understand the logic for the send/recv stuff really
3372018-10-22T17:38:03  <phantomcircuit> but whatever
3382018-10-22T17:38:11  <sipa> phantomcircuit: i think part of it is historical
3392018-10-22T17:39:03  <sipa> at some point there was 'eager sending', where we'd send directly from the message handler if the send queue was empty, rather than putting it in the queue and waiting for the send handler
3402018-10-22T17:39:33  <sipa> i think the 'don't processing incoming when send buffer full' also dates from that time
3412018-10-22T17:42:06  <sipa> seems totally reasonable to get rid of that logic, imho
3422018-10-22T17:43:06  <gmaxwell> So what happens if the send buffer is full, and you process some more messages and thus need to send more?
3432018-10-22T17:43:59  <sipa> ah right, it's when it's *full*, not just when it's non-empty
3442018-10-22T17:44:44  <phantomcircuit> sipa, we still do eager sending
3452018-10-22T17:45:04  *** Guest19314 has quit IRC
3462018-10-22T17:45:36  <sipa> oh, right!
3472018-10-22T17:45:45  <sipa> seems i'm out of touch with the network logic
3482018-10-22T17:45:50  <phantomcircuit> indeed not calling recv() when the send buffer is full makes sense
3492018-10-22T17:45:55  <sipa> it's called optimistic send
3502018-10-22T17:45:59  <phantomcircuit> but currently we dont call it unless it's empty
3512018-10-22T17:46:04  <phantomcircuit> which doesn't seem to make sense
3522018-10-22T17:46:32  <sipa> gmaxwell: ah, there's a distinction between not calling recv when the send buffer is full, and not processing messages in the recv buffer when the send buffer is full
3532018-10-22T17:46:38  <phantomcircuit> otoh we disconnect peers when either buffer is full iirc
3542018-10-22T17:48:04  <sipa> phantomcircuit: glad to have someone look at this stuff again
3552018-10-22T17:49:43  <phantomcircuit> it seems like we're probably just wasting time not processing stuff when we have an asymetric connection
3562018-10-22T17:50:14  *** Krellan has quit IRC
3572018-10-22T17:50:25  <phantomcircuit> (and making the logic here way more complicted than it needs to be)
3582018-10-22T17:50:52  *** michaelfolkson has joined #bitcoin-core-dev
3592018-10-22T17:51:43  <sipa> phantomcircuit: i think we should (a) always listen for receives unless the recc buffer is full (b) always process messages in the recv buffer unless the send buffer is full
3602018-10-22T17:51:43  *** schmidty has joined #bitcoin-core-dev
3612018-10-22T17:51:54  <sipa> (where full means over threshold, not just nonempty)
3622018-10-22T17:52:05  <gmaxwell> what sipa says
3632018-10-22T17:52:07  *** schmidty is now known as Guest18904
3642018-10-22T17:52:41  <gmaxwell> we can't process a recv message if the send is actually full, since virtually all recieves msg will require us to send.
3652018-10-22T17:53:37  *** Krellan has joined #bitcoin-core-dev
3662018-10-22T17:53:57  *** michaelfolkson has quit IRC
3672018-10-22T17:54:08  *** AaronvanW has quit IRC
3682018-10-22T17:56:27  *** michaelfolkson has joined #bitcoin-core-dev
3692018-10-22T17:57:24  <phantomcircuit> gmaxwell, but we disconnect peers when we're actually full right?
3702018-10-22T17:58:34  *** dviola has joined #bitcoin-core-dev
3712018-10-22T17:59:16  <sipa> phantomcircuit: i think we shouldn't
3722018-10-22T17:59:32  <sipa> and i think it's also very hard currently to trigger that logic
3732018-10-22T18:02:26  *** michaelfolkson has quit IRC
3742018-10-22T18:03:05  <gwillen> 
3752018-10-22T18:03:52  <phantomcircuit> sipa, i think it's basically impossible because of the select logic
3762018-10-22T18:06:25  *** AaronvanW has joined #bitcoin-core-dev
3772018-10-22T18:08:32  *** michaelfolkson has joined #bitcoin-core-dev
3782018-10-22T18:10:16  *** SopaXorzTaker has quit IRC
3792018-10-22T18:13:14  <phantomcircuit> sipa, you'd need to fill the queue with inv messages i think
3802018-10-22T18:13:22  <phantomcircuit> which is pretty unlikely unless the remote is actually gone
3812018-10-22T18:15:32  *** timothy has quit IRC
3822018-10-22T18:17:36  <phantomcircuit> BlueMatt, MATTTT
3832018-10-22T18:17:37  *** Zorbak has joined #bitcoin-core-dev
3842018-10-22T18:19:11  <phantomcircuit> part of the issue is that the send queue is relatively small (for good reason) versus the largest possible message
3852018-10-22T18:19:33  <phantomcircuit> asking for even a single block could potentially be most of the queue iirc
3862018-10-22T18:19:34  <gmaxwell> well for not that good a reason, :P
3872018-10-22T18:20:10  <sipa> i think the logic should be "the send buffer can contain one message worth in excess of the limit"
3882018-10-22T18:20:54  <phantomcircuit> gmaxwell, if it's too small then it's a dos risk
3892018-10-22T18:21:01  <phantomcircuit> it could be shared but that's the same thing
3902018-10-22T18:22:33  *** Victorsueca has quit IRC
3912018-10-22T18:23:48  *** Victorsueca has joined #bitcoin-core-dev
3922018-10-22T18:25:07  <gmaxwell> phantomcircuit: there are more options than that... e.g. every peer could be allowed to buffer one message on its own, plus access to one of a couple shared queues.
3932018-10-22T18:25:22  <gmaxwell> just maing a point that we don't have to have really small send buffers.
3942018-10-22T18:35:22  *** Zorbak has quit IRC
3952018-10-22T18:35:50  *** Krellan has quit IRC
3962018-10-22T18:36:21  <phantomcircuit> gmaxwell, hmm
3972018-10-22T18:36:30  <phantomcircuit> we have sort of two classes of traffic really
3982018-10-22T18:37:07  <phantomcircuit> theres (vaguely) broadcast traffic and there's req/res type traffic
3992018-10-22T18:37:19  <phantomcircuit> which have different priority really
4002018-10-22T18:42:16  <phantomcircuit> actually iono
4012018-10-22T18:52:37  *** Guest18904 has quit IRC
4022018-10-22T18:52:37  *** Guest18904 has joined #bitcoin-core-dev
4032018-10-22T18:52:43  *** Guest18904 is now known as schmidty
4042018-10-22T18:55:12  *** bitconner has joined #bitcoin-core-dev
4052018-10-22T18:56:50  *** csknk has joined #bitcoin-core-dev
4062018-10-22T19:03:36  *** Murch has quit IRC
4072018-10-22T19:15:36  *** belcher_ has joined #bitcoin-core-dev
4082018-10-22T19:18:52  *** belcher has quit IRC
4092018-10-22T19:24:09  *** Zenton has joined #bitcoin-core-dev
4102018-10-22T19:30:07  *** ken2812221_ has quit IRC
4112018-10-22T19:33:38  *** AaronvanW has quit IRC
4122018-10-22T19:34:15  *** AaronvanW has joined #bitcoin-core-dev
4132018-10-22T19:34:22  *** csknk has quit IRC
4142018-10-22T19:38:27  *** AaronvanW has quit IRC
4152018-10-22T19:39:30  *** bitcoin-git has joined #bitcoin-core-dev
4162018-10-22T19:39:31  <bitcoin-git> [bitcoin] practicalswift closed pull request #12984: logs: Make the columns in subsequent UpdateTip log entries horizontally aligned (master...attention-to-detail) https://github.com/bitcoin/bitcoin/pull/12984
4172018-10-22T19:39:31  *** bitcoin-git has left #bitcoin-core-dev
4182018-10-22T19:40:06  *** bitcoin-git has joined #bitcoin-core-dev
4192018-10-22T19:40:06  <bitcoin-git> [bitcoin] practicalswift closed pull request #13766: Prefer initialization to assignment in constructors. Prefer in-class initializers to member initializers in constructors for constant initializers. (master...initialize-members-in-initialization-list) https://github.com/bitcoin/bitcoin/pull/13766
4202018-10-22T19:40:06  *** bitcoin-git has left #bitcoin-core-dev
4212018-10-22T19:41:44  *** bralyclow has quit IRC
4222018-10-22T19:43:30  *** bralyclow has joined #bitcoin-core-dev
4232018-10-22T19:45:48  *** ponyo has joined #bitcoin-core-dev
4242018-10-22T19:48:45  *** qrestlove has quit IRC
4252018-10-22T19:52:21  *** ponyo has quit IRC
4262018-10-22T19:54:43  *** klot_ has quit IRC
4272018-10-22T19:59:22  *** ExtraCrispy has quit IRC
4282018-10-22T20:01:55  *** qrestlove has joined #bitcoin-core-dev
4292018-10-22T20:03:20  *** Krellan has joined #bitcoin-core-dev
4302018-10-22T20:13:33  *** molz has quit IRC
4312018-10-22T20:35:16  *** molz has joined #bitcoin-core-dev
4322018-10-22T20:35:58  *** Murch has joined #bitcoin-core-dev
4332018-10-22T20:43:15  *** iphelix has quit IRC
4342018-10-22T20:47:16  *** iphelix has joined #bitcoin-core-dev
4352018-10-22T20:47:39  *** iphelix is now known as Guest37734
4362018-10-22T20:52:29  *** iphelix- has joined #bitcoin-core-dev
4372018-10-22T21:00:45  *** hebasto has quit IRC
4382018-10-22T21:00:45  *** hebasto_ has joined #bitcoin-core-dev
4392018-10-22T21:07:45  *** lnostdal has quit IRC
4402018-10-22T21:10:06  *** hebasto__ has joined #bitcoin-core-dev
4412018-10-22T21:10:14  *** hebasto_ has quit IRC
4422018-10-22T21:10:32  *** lnostdal has joined #bitcoin-core-dev
4432018-10-22T21:11:31  *** Victorsueca has quit IRC
4442018-10-22T21:12:52  *** Victorsueca has joined #bitcoin-core-dev
4452018-10-22T21:19:09  *** hebasto_ has joined #bitcoin-core-dev
4462018-10-22T21:19:15  *** hebasto__ has quit IRC
4472018-10-22T21:26:07  *** michaelfolkson has joined #bitcoin-core-dev
4482018-10-22T21:27:42  *** bitcoin-git has joined #bitcoin-core-dev
4492018-10-22T21:27:42  <bitcoin-git> [bitcoin] instagibbs opened pull request #14543: [QA] minor p2p_sendheaders fix of height in coinbase (master...p2p_sendheaders_height) https://github.com/bitcoin/bitcoin/pull/14543
4502018-10-22T21:27:42  *** bitcoin-git has left #bitcoin-core-dev
4512018-10-22T21:37:25  *** OzPac has quit IRC
4522018-10-22T21:37:56  *** iphelix- is now known as iphelix
4532018-10-22T21:38:26  *** iphelix is now known as Guest74867
4542018-10-22T21:42:50  *** dviola has quit IRC
4552018-10-22T21:58:49  *** michaelfolkson has quit IRC
4562018-10-22T22:00:05  *** michaelfolkson has joined #bitcoin-core-dev
4572018-10-22T22:12:26  *** michaelfolkson has quit IRC
4582018-10-22T22:14:26  *** michaelsdunn1 has quit IRC
4592018-10-22T22:15:57  *** michaelfolkson has joined #bitcoin-core-dev
4602018-10-22T22:19:27  *** spinza has quit IRC
4612018-10-22T22:23:44  *** bitcoin-git has joined #bitcoin-core-dev
4622018-10-22T22:23:44  <bitcoin-git> [bitcoin] sipa pushed 4 new commits to master: https://github.com/bitcoin/bitcoin/compare/5c25409d6851...0a8f519a0626
4632018-10-22T22:23:45  <bitcoin-git> bitcoin/master 2c6281f Pieter Wuille: Add key origin support to descriptors
4642018-10-22T22:23:45  <bitcoin-git> bitcoin/master ff37459 Pieter Wuille: Add tests for key origin support
4652018-10-22T22:23:46  <bitcoin-git> bitcoin/master 8afb166 Pieter Wuille: Update documentation to incude origin information
4662018-10-22T22:23:46  *** bitcoin-git has left #bitcoin-core-dev
4672018-10-22T22:24:38  *** bitcoin-git has joined #bitcoin-core-dev
4682018-10-22T22:24:38  <bitcoin-git> [bitcoin] sipa closed pull request #14150: Add key origin support to descriptors (master...201807_minedesc_origin) https://github.com/bitcoin/bitcoin/pull/14150
4692018-10-22T22:24:38  *** bitcoin-git has left #bitcoin-core-dev
4702018-10-22T22:30:23  *** michaelfolkson has quit IRC
4712018-10-22T22:30:42  *** michaelfolkson has joined #bitcoin-core-dev
4722018-10-22T22:31:28  *** michaelfolkson has quit IRC
4732018-10-22T22:32:20  *** spinza has joined #bitcoin-core-dev
4742018-10-22T22:33:19  *** laurentmt has joined #bitcoin-core-dev
4752018-10-22T22:51:41  *** jarthur has quit IRC
4762018-10-22T22:51:54  *** fanquake has joined #bitcoin-core-dev
4772018-10-22T23:00:35  *** Krellan has quit IRC
4782018-10-22T23:01:13  *** Krellan has joined #bitcoin-core-dev
4792018-10-22T23:29:10  *** laurentmt has quit IRC
4802018-10-22T23:29:17  *** anonkun has joined #bitcoin-core-dev
4812018-10-22T23:33:32  *** anonkun has quit IRC
4822018-10-22T23:49:40  *** Krellan has quit IRC
4832018-10-22T23:54:23  *** fanquake has quit IRC