--- Day changed Fri May 11 2018
00:40 -!- puddinpop [~puddinpop@unaffiliated/puddinpop] has quit [Ping timeout: 246 seconds]     
00:51 < arubi> ok, so bitcoind is definitely starting up on travis, on the expected port and does respond to rpc from curl
00:55 -!- asymptotically [~asymptoti@gateway/tor-sasl/asymptotically] has joined #joinmarket     
00:55 -!- asymptotically [~asymptoti@gateway/tor-sasl/asymptotically] has quit [Remote host closed the connection]     
00:56 -!- asymptotically [~asymptoti@gateway/tor-sasl/asymptotically] has joined #joinmarket     
01:12 < arubi> waxwing, just a shot in the dark, maybe this : https://github.com/JoinMarket-Org/joinmarket-clientserver/blob/master/jmclient/test/test_configure.py#L28-L29 ?
01:14 < arubi> hmm then again, why doesn't it fail locally if it's related :\
01:18 < waxwing> hmm maybe the assert message there. but yeah, that doesn't explain a travis v local distinction
01:19 < waxwing> you could ignore that test module as a sanity check, i forget now, --ignore? no it's not that. however you do that on pytest.
01:56 -!- belcher_ [~user@unaffiliated/belcher] has quit [Ping timeout: 264 seconds]     
03:04 -!- GitHub121 [GitHub121@gateway/service/github.com/x-jsaradqhdsyrikau] has joined #joinmarket     
03:04 < GitHub121> [joinmarket-clientserver] kristapsk opened pull request #145: Specify format for "total profit" output (master...total-profit) https://git.io/vpD21
03:04 -!- GitHub121 [GitHub121@gateway/service/github.com/x-jsaradqhdsyrikau] has left #joinmarket []     
05:04 -!- diogorsergio [~diogorser@81.92.202.8] has joined #joinmarket     
05:04 -!- diogorsergio [~diogorser@81.92.202.8] has quit [Client Quit]     
05:07 -!- diogorsergio [~diogorser@81.92.202.8] has joined #joinmarket     
06:17 -!- diogorsergio [~diogorser@81.92.202.8] has quit [Quit: diogorsergio]     
07:00 -!- Sentineo_ is now known as Sentineo     
07:15 < waxwing> arubi, if i'm reading it right, the travis builds are on trusty/1404 LTS right?
07:15 < waxwing> i'm on 1604 here. maybe i can fire up a 1404 VM and see if that's the delta? sound worth it?
07:18 < waxwing> another thing i noticed is that Lightsword 's commit uses the socket package, which we don't really use any more elsewhere. maybe that's it. i mean i'm grasping at straws here,but ...
07:21 -!- quitobro [~quitobro@pool-108-41-0-186.nycmny.fios.verizon.net] has joined #joinmarket     
07:24 -!- quitobro_ [quitobro@gateway/vpn/privateinternetaccess/quitobro] has joined #joinmarket     
07:27 -!- quitobro [~quitobro@pool-108-41-0-186.nycmny.fios.verizon.net] has quit [Ping timeout: 265 seconds]     
07:27 -!- quitobro_ is now known as quitobro     
07:35 < waxwing> arubi, ran install script from scratch on fresh 1404 VM and got this gpg error: pastebin.com/rB0Mi3Hf 
07:36 < asymptotically> waxwing: i got that too on debian sid, so i just installed libsodium through the package manager and commented the libsodium parts out of install.sh
07:38 < waxwing> right; i think it's trying to download keys or something, and failing? i see a download thing that's stuck now. thanks, i'll do that.
07:40 < waxwing> asymptotically, good to report stuff like that. doesn't mean anyone will necessarily fix it quickly, but still.
07:43 < asymptotically> oh, well another thing i noticed was that SegwitOrder wasn't imported in tumbler.py
07:43 < waxwing> SegwitOrder? or SegwitWallet perhaps?
07:45 < asymptotically> sorry, SegwitWallet
07:49 < waxwing> yeah it uses get_wallet_cls() now and hooks off the setting in the config file
07:57 < waxwing> ah it appears there's a --no-gpg-validation option to the install script
07:57 < waxwing> not that that's necessarily something to be advised :) just for my tests security is irrelevant.
08:02 -!- Giszmo [~leo@pc-37-38-86-200.cm.vtr.net] has quit [Remote host closed the connection]     
08:04 < waxwing> ok error pattern replicates exactly as per travis on 1404
08:04 < waxwing> so it's definitely a ubuntu version delta
08:04 < waxwing> arubi, ^
08:10 < waxwing> further, also confirmed: if i get it to ignore test_configure, the rest of the tests pass. so the problem is isolated.
08:11 < arubi> amazing.  I actually tested locally on a travis docker image and did was not able to recreate the error
08:11 < waxwing> arubi, was it 1404?
08:11 < arubi> yes, just one minor version behind what travis is
08:11 < waxwing> i mean basically it seems from this you were right, good spot: that hack in test_configure breaks because of relying on the jsonrpc connection error.
08:11 < waxwing> huh. that does seem amazing.
08:12 < waxwing> i'm on a Virtual Box VM here with a 1404 LTS iso that i grabbed off the ubuntu site.
08:12 < waxwing> i believe it's 14.04.5
08:12 < arubi> yea, that's what I have here
08:13 < waxwing> btw this syntax for ignoring works: -k "not configure"
08:13 < waxwing> py.test runs all the rest but skips test_configure then
08:13 < arubi> ah cool.  let's see on travis now
08:14 < waxwing> not being a shell script person i wasn't sure if you could put double quotes in a script like that, but it seemed to work fine
08:15 < waxwing> oh wow. second time without the "ignore" part, i still get *one* fail in the test_configure, but all the succeeding tests pass. hmm that might not be so strange, depends on exactly what's happening here.
08:15 < waxwing> the tests aren't perfectly stateless, it's probably that kind of thing (a file dangling?)
08:15 < arubi> test_configure fails for me also, something about the dummydirconfig already existing
08:16 < waxwing> honestly that was a complete screwup. my fault.
08:16 < waxwing> i've avoided worrying about it by: tests run from scratch with an empty regtest pass. but needs fixing up.
08:18 < arubi> brb getting coffee.  travis running here : https://travis-ci.org/fivepiece/joinmarket-clientserver/jobs/377776557 if you wanna follow it
08:20 < waxwing> yeah i think: if you re-run tests without removing dummydirforconfig, it errors on that specifically, and everything else passes. if you do remove that directory, then the full set of "connection errors" is reproducible.
08:20 < waxwing> and lastly if you -k "not configure" to remove that set of tests, then it works normally/correctly.
08:21 < waxwing> best thing is to remove the test_configure module for now, as it's clearly completely broken, and run travis without it.
08:21 < waxwing> then i'll fix it up.
08:21 -!- puddinpop [~puddinpop@unaffiliated/puddinpop] has joined #joinmarket     
08:26 < waxwing> yeah that passed. feel free to PR it i guess it's just that one edit in run_tests
08:29 < arubi> alright
08:29 -!- GitHub181 [GitHub181@gateway/service/github.com/x-eaurnqrrgyskpjbe] has joined #joinmarket     
08:29 < GitHub181> [joinmarket-clientserver] fivepiece opened pull request #146: Ignore test_configure (master...tests_disable_test_configure) https://git.io/vpytq
08:29 -!- GitHub181 [GitHub181@gateway/service/github.com/x-eaurnqrrgyskpjbe] has left #joinmarket []     
08:36 < arubi> oh btw re. gpg pubkeys not downloading, it's sporadic.  sometimes the keyserver just doesn't answer (also why we added the -no-gpg.. flag to travis)
08:37 < arubi> now that commits are gpg signed, it's probably fine to include the packages' pubkeys in the jm repo
08:45 -!- akrmn [~akrmn@79.red-83-54-170.dynamicip.rima-tde.net] has quit [Ping timeout: 255 seconds]     
08:48 < waxwing> arubi, dammit that build is taking so long. the tests already passed i think (the first part, right)
08:48 < waxwing> iirc you said the second one just tests the environment for mac or something?
08:48 < arubi> yea, it's osx that hasn't started
08:49 < arubi> linux built and tested
08:58 < arubi> waxwing, I think you mean "see #147"
09:13 < waxwing> arubi, oh no! i accidentally clicked restart
09:13 < waxwing> lol. on the 2nd (osx) one. it failed.
09:13 < waxwing> i wanted to read the log
09:14 < arubi> failed on my repo too.  timeout
09:14 < waxwing> yeah 147 thanks.
09:14 < waxwing> arubi, ok timeout; shoudl we be concerned about it or is it just normal?
09:14 < arubi> sometimes that happens with osx.  they don't have so many instances
09:15 < arubi> mine is stuck on "booting" now.  osx doesn't even run the tests so it won't reach that pytest line
09:23 < waxwing> k thx
09:23 -!- GitHub165 [GitHub165@gateway/service/github.com/x-ojrkcogputbjinbo] has joined #joinmarket     
09:23 < GitHub165> [joinmarket-clientserver] AdamISZ pushed 2 new commits to master: https://git.io/vpysq
09:23 < GitHub165> joinmarket-clientserver/master cd3b31d fivepiece: ignore configure
09:23 < GitHub165> joinmarket-clientserver/master df2a4f9 AdamISZ: Merge #146: Ignore test_configure...
09:23 -!- GitHub165 [GitHub165@gateway/service/github.com/x-ojrkcogputbjinbo] has left #joinmarket []     
09:23 -!- GitHub173 [GitHub173@gateway/service/github.com/x-giafoufjgedpxnnz] has joined #joinmarket     
09:23 < GitHub173> [joinmarket-clientserver] AdamISZ closed pull request #146: Ignore test_configure (master...tests_disable_test_configure) https://git.io/vpytq
09:23 -!- GitHub173 [GitHub173@gateway/service/github.com/x-giafoufjgedpxnnz] has left #joinmarket []     
10:05 -!- d3spwn [53a19dc8@gateway/web/freenode/ip.83.161.157.200] has joined #joinmarket     
10:17 -!- akrmn [~akrmn@79.red-83-54-170.dynamicip.rima-tde.net] has joined #joinmarket     
10:25 < waxwing> arubi, so it just timed out again. is something wrong? or should i just restart it, say tomorrow.
10:26 < arubi> seems that their osx instances are working overtime.  restart tomorrow I think
10:26 < arubi> see https://www.traviscistatus.com/ "Backlog macOS Builds for Open Source Projects"
10:26 < arubi> nothing is advancing
10:30 < waxwing> arubi, can we RBF our builds? :)
10:31 < arubi> yes, if you increase the fee you pay to travis ;)
10:32 < arubi> wish I knew how they're managing their osx build envs..  I mean, are they just running mac computers\servers over there?  weird
10:53 -!- Giszmo [~leo@pc-37-38-86-200.cm.vtr.net] has joined #joinmarket     
11:51 -!- bsm1175321 [~mcelrath@173-9-124-61-NewEngland.hfc.comcastbusiness.net] has quit [Ping timeout: 256 seconds]     
11:58 -!- belcher_ [~user@unaffiliated/belcher] has joined #joinmarket     
14:38 -!- asymptotically [~asymptoti@gateway/tor-sasl/asymptotically] has quit [Quit: Leaving]     
16:55 -!- StopAndDecrypt [~StopAndDe@unaffiliated/stopanddecrypt] has joined #joinmarket     
20:51 -!- AgoraRelay [~jmrelayfn@p5DE4A55C.dip0.t-ipconnect.de] has quit [Ping timeout: 268 seconds]     
21:08 -!- AgoraRelay [~jmrelayfn@p54866067.dip0.t-ipconnect.de] has joined #joinmarket     
22:07 < arubi> ugh.  finally found the issue with the jm-qt install.sh PR where osx on travis somehow kept on nesting past the "if".  basically 's/unkown/unknown/' .  the script doesn't need the 60 seconds timeout when asking if the user wants to install qt libs or not
22:17 -!- quitobro [quitobro@gateway/vpn/privateinternetaccess/quitobro] has quit [Quit: quitobro]