--- Day changed Wed Sep 11 2019
02:09 -!- MaxSan [~four@185.156.175.35] has joined #joinmarket     
02:10 -!- jdm [~jdm@unaffiliated/user1138] has joined #joinmarket     
02:26 -!- Giszmo [~leo@2407:7000:9d28:5100:a868:5a35:748a:4bfa] has quit [Read error: Connection reset by peer]     
02:27 -!- Giszmo [~leo@2407:7000:9d28:5100:542b:7e2:8abc:d3f2] has joined #joinmarket     
03:49 -!- reallll [~belcher@unaffiliated/belcher] has joined #joinmarket     
03:52 -!- belcher [~belcher@unaffiliated/belcher] has quit [Ping timeout: 240 seconds]     
06:35 -!- technonerd_ is now known as technonerd     
06:35 -!- Giszmo [~leo@2407:7000:9d28:5100:542b:7e2:8abc:d3f2] has quit [Read error: Connection reset by peer]     
06:36 -!- Giszmo [~leo@2407:7000:9d28:5100:542b:7e2:8abc:d3f2] has joined #joinmarket     
08:33 -!- dopplergange [~dop@213.183.57.27] has quit [Ping timeout: 258 seconds]     
10:28 -!- davex__ [~user@45.74.60.130] has quit [Quit: Leaving]     
10:28 -!- davex_ [~user@45.74.60.130] has quit [Remote host closed the connection]     
10:29 -!- davex__ [~user@45.74.60.130] has joined #joinmarket     
10:29 -!- deafboy [quasselcor@cicolina.org] has quit [Quit: deafboy]     
10:54 < takinbo> arubi: i updated the test harness in #391 to use the simplified install method. happy to report that all the platforms (including osx via homebrew) have libsodium. all dockers are building and running tests successfully. 
10:56 < takinbo> i did notice that the tests are not being run on osx though. i wonder why that is but i plan on fixing that. 
11:02 < arubi> awesome takinbo, I'll take a look over the weekend (gotta push some code at work).  iirc the reason osx doesn't run tests is because for some reason miniircd doesn't seem to be launching correctly on tracis
11:02 < arubi> er, travis
11:04 < takinbo> i'll investigate if there's a way to fix that (might have something to do with how the process is spawned). i have an osx host i can test that on.
11:04 < arubi> it would be a huge improvement to fix that.  for a while I was trying to debug it on hackintosh, but it was so slow I wasn't getting anywhere.  eventually I borrowed a mac from work and when I tried it was all just working fine
11:05 < takinbo> so it might have been a travis specific problem?
11:06 < arubi> that's what I figured.  maybe some exotic network setup that makes the miniircd on localhost somehow unreachable, or maybe it's a permissions issue.  would be worth to try spawning it on a different port in case 6667 is blocked specifically
11:09 < takinbo> would be worth investigating
11:10 < takinbo> also, the caches are very aggressive, could this have something to do with the build time? maybe if we didn't have to build all those packages anymore, we wouldn't need it?
11:13 < takinbo> i'm only concerned that it skips running tests because nothing "changed" when there might be cases where something changed but not where it was looking at?
11:22 < arubi> I'm not exactly following, sorry, can you point me to it>
11:24 < takinbo> https://travis-ci.org/JoinMarket-Org/joinmarket-clientserver/jobs/583441111 is one of the docker builds
11:25 < arubi> ah yes, they're disabled by default because they do take a while to complete.  they should be running in a cron job and for every tagged commit
11:25 < takinbo> if you look at line 229, it took 0.01s to run `./test/Dockerfiles/build_docker.sh`
11:26 < arubi> yea, see the config
11:26 < arubi> there's a should_run_dockers function
11:27 < takinbo> ah. it should be in the commit message to set `TRAVIS_RUN_DOCKERS`
11:27 < fiatjaf> I've put some coins on mixdepth 0, then started yg
11:27 < arubi> yep     
11:27 < takinbo> okay I'll need to do that. thanks.
11:27 < fiatjaf> two transactions happened, the logs say I've potentially earned x and then potentially earned y 
11:27 < arubi> np     
11:28 < fiatjaf> my 'history' shows only one transaction extra with x + y
11:28 < fiatjaf> and now I have 2 new utxos as 'internal addresses' on mixdepth 1
11:29 < fiatjaf> then my yg offer is reduced by a big amount. why? because I can't mix coins from mixdepth 0 and 1? then how should I proceed to get my mixdepth 1 coins to also be their own yg offer?
11:30 < waxwing> it offers the largest amount you have in any mixdepth. and yes, after you start getting joins that offer maximum reduces correspondingly.
11:30 < waxwing> this is needed to keep the mixdepths (accounts) isolated.
12:15 < fiatjaf> can I have one offer from mixdepth0 and another from mixdepth1?
12:39 -!- davex__ [~user@45.74.60.130] has quit [Ping timeout: 276 seconds]     
12:39 -!- davex_ [~user@45.74.60.130] has joined #joinmarket     
13:26 < fiatjaf> just run two yg scripts, maybe?
13:26 < CgRelayBot> [cgan/AlexCato1] if you have, just as an example, 1 BTC in mixdepth0 and 2 BTC in mixdepth1, it will offer 2 BTC.
13:27 < CgRelayBot> [cgan/AlexCato1] What would the point of additionally offering the other 1 BTC be?
13:27 < CgRelayBot> [cgan/AlexCato1] it always offers the maximum that any of your mixdepths can support
13:28 < CgRelayBot> [cgan/AlexCato1] you cant add those amounts in any way, because if someone would want to coinjoin 2.5 BTC, it would need to take coins from more than 1 mixdepth. That would break the privacy of coinjoins, and this is what the takers are paying the makers for; so this cannot be allowed
13:29 < kristapsk> fiatjaf, it offers higher possible from single of your mixdepths, but when actually cj happen, it can take from any single one of your mixdepths
13:35 < fiatjaf> ok, makes sense
13:35 < fiatjaf> what if I offered coins from both mixdepths, but didn't allow both to be used in the same coinjoin?
13:36 < fiatjaf> I guess there's no way of knowing that in the protocol without revealing things that shouldn't be revealed?
13:41 < kristapsk> you don't offer exact coins, you offer min max range for coinjoins
13:49 < waxwing> fiatjaf, "what if I offered coins from both mixdepths" - you effectively do. if you offer say 2btc publically, but have 2btc in one depth and 1btc in a second, and someone requests 0.5, you may well use the 1btc mixdepth for the join.
13:50 < waxwing> the exact choices can be varied in the algorithm. but as a general concept, all mixdepths are being offered at once, just the maximum size (with randomness) is being published.
13:56 < fiatjaf> oh, right, that makes perfect sense
14:18 < fiatjaf> now I think I've finally got it
14:18 < fiatjaf> thank you.
14:51 -!- deafboy [quasselcor@cicolina.org] has joined #joinmarket     
14:52 -!- Giszmo [~leo@2407:7000:9d28:5100:542b:7e2:8abc:d3f2] has quit [Read error: Connection reset by peer]     
14:53 -!- Giszmo [~leo@2407:7000:9d28:5100:542b:7e2:8abc:d3f2] has joined #joinmarket     
16:03 < kristapsk> was doing some testing with testnet, looks like there could be a bug in txfee estimation? 2019-09-12 02:02:07,941 [DEBUG]  error pushing = -26 min relay fee not met, 133 < 134 (code 66)
16:05 < kristapsk> it was 1 input, 1 output tx sending from qt to external wallet, tx_fees = 3, bitcoin-cli -testnet estimatesmartfee 3 returns 0.00001000
16:10 < kristapsk> and send was from P2SH-SegWit to P2PKH address
17:04 < kristapsk> ahh, "we assume the output types are the same as the input (to be fair, outputs only contribute a little to the overall total). This combined with a few bytes variation in signature sizes means we will expect, say, 10% inaccuracy here."
17:04 < kristapsk> kinda works, until you decide to do 1 UTXO in, 1 UTXO out non-cj send :)
17:55 -!- AgoraRelay [~jmrelayfn@p5DE4A75B.dip0.t-ipconnect.de] has quit [Ping timeout: 268 seconds]     
17:56 -!- CgRelayBot [~CgRelayBo@p5DE4A75B.dip0.t-ipconnect.de] has quit [Ping timeout: 246 seconds]     
18:09 -!- CgRelayBot [~CgRelayBo@p5DE4A8AB.dip0.t-ipconnect.de] has joined #joinmarket     
18:11 -!- AgoraRelay [~jmrelayfn@p5DE4A8AB.dip0.t-ipconnect.de] has joined #joinmarket     
19:05 -!- StopAndDecrypt [~StopAndDe@unaffiliated/stopanddecrypt] has quit [Remote host closed the connection]     
23:12 -!- azizLIGHT [~azizLIGHT@unaffiliated/azizlight] has quit [Ping timeout: 246 seconds]     
23:17 -!- azizLIGHT [~azizLIGHT@unaffiliated/azizlight] has joined #joinmarket     
23:44 -!- Sentineo [~Undefined@unaffiliated/sentineo] has quit [Ping timeout: 245 seconds]