--- Day changed Thu Jul 05 2018
00:05 -!- lnostdal [~lnostdal@109.160.62.107] has quit [Read error: Connection reset by peer]     
00:06 -!- lnostdal [~lnostdal@109.160.62.107] has joined #joinmarket     
00:20 -!- lnostdal [~lnostdal@109.160.62.107] has quit [Ping timeout: 245 seconds]     
00:54 -!- undeath [~undeath@unaffiliated/undeath] has joined #joinmarket     
01:41 -!- undeath [~undeath@unaffiliated/undeath] has quit [Quit: WeeChat 2.1]     
02:56 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Ping timeout: 240 seconds]     
03:08 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket     
04:53 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Ping timeout: 264 seconds]     
04:53 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket     
05:30 -!- belcher_ [~belcher@unaffiliated/belcher] has joined #joinmarket     
05:55 -!- undeath [~undeath@unaffiliated/undeath] has joined #joinmarket     
06:57 -!- lnostdal [~lnostdal@85-118-68-230.mtel.net] has joined #joinmarket     
07:01 < waxwing> catoalex: do your logs for the cases you saw show the 'nickname collision....' message?
07:08 < waxwing> see: https://github.com/JoinMarket-Org/joinmarket-clientserver/blob/master/jmdaemon/jmdaemon/irc.py#L278 and
07:08 < waxwing> https://github.com/JoinMarket-Org/joinmarket-clientserver/blob/master/jmdaemon/jmdaemon/irc.py#L38-L39
07:09 < waxwing> the last line of that second link was supposed to pick up the exact right number of characters, because we pad to NICK_MAX_ENCODED with repeated 'O'
07:09 < waxwing> so any '_' that were appended should be truncated.
07:09 < waxwing> it's a pain to test though.
07:34 < AgoraRelay> [agora-irc/CatoAlex] yes, i see that nick collision message a lot. Next time a maker has a changed name, i'll try to do a sendpayment and see if/where it fails
07:35 < AgoraRelay> [agora-irc/CatoAlex] a simple periodic try to rename it back to its original name might fix it, if theres an issue
07:36 < waxwing> ok thanks. i'm delving in to write a test to sanity check the nick-creation and nick-extraction code.
07:36 < waxwing> i've personally never seen that message, maybe it's just luck or something. but it def wouldn't surprise me that there's an error here, important if so!
07:38 < AgoraRelay> [agora-irc/CatoAlex] no need to do anything just yet, let me confirm if theres actually a problem first... and even if there is, its a minor issue imho
07:44 -!- lnostdal [~lnostdal@85-118-68-230.mtel.net] has quit [Ping timeout: 256 seconds]     
08:27 < AgoraRelay> [agora-irc/CatoAlex] btw, thanks @belcher for electrum private server, i really love it. Works great and makes working with hardware wallets so much better (privacy-wise)
09:06 < AgoraRelay> [agora-irc/belcher] thanks for saying so
09:06 < AgoraRelay> [agora-irc/belcher] electrum personal server is basically joinmarket's BitcoinCoreInterface with the electrum protocol
10:06 -!- belcher [~user@unaffiliated/belcher] has quit [Ping timeout: 240 seconds]     
10:10 -!- belcher [~user@unaffiliated/belcher] has joined #joinmarket     
10:12 -!- undeath [~undeath@unaffiliated/undeath] has quit [Quit: WeeChat 2.1]     
10:47 -!- lnostdal [~lnostdal@vlan-153-aitos-70.comnet.bg] has joined #joinmarket     
11:10 -!- belcher_ [~belcher@unaffiliated/belcher] has quit [Ping timeout: 256 seconds]     
12:21 -!- belcher_ [~belcher@unaffiliated/belcher] has joined #joinmarket     
12:49 -!- undeath [~undeath@unaffiliated/undeath] has joined #joinmarket     
13:45 < waxwing> ok think i see the problem (and it is indeed a screwup): '_'-appended nicks are not receiving their privmsgs because their counterparties are sending to '_'-not-appended version .. facepalm
13:48 -!- rdymac [uid31665@gateway/web/irccloud.com/x-yvybzcrfquhfjtun] has joined #joinmarket     
14:12 -!- reallll [~belcher@unaffiliated/belcher] has joined #joinmarket     
14:14 -!- belcher_ [~belcher@unaffiliated/belcher] has quit [Ping timeout: 256 seconds]     
14:17 -!- reallll [~belcher@unaffiliated/belcher] has quit [Ping timeout: 256 seconds]     
14:43 -!- lnostdal [~lnostdal@vlan-153-aitos-70.comnet.bg] has quit [Ping timeout: 240 seconds]     
14:46 -!- dx25_ [~dx25@97-119-117-177.omah.qwest.net] has joined #joinmarket     
14:54 < AgoraRelay> [agora-irc/CatoAlex] :D
14:57 < waxwing> fixing (it's only a bit non-trivial because there are multiple message channels so they have to manage this individually, at MCCollection level it has to not know about this)
15:01 < waxwing> meh, not that bad, just a few lines now i've tidied it up, just keep a table of where the nicks have been collided in irc.py. i'll post it as a PR shortly after a few more sanity checks.
15:06 -!- dx25_ [~dx25@97-119-117-177.omah.qwest.net] has quit [Quit: ZNC 1.7.0 - https://znc.in]     
15:16 -!- GitHub61 [GitHub61@gateway/service/github.com/x-yonibeaygjwtzdku] has joined #joinmarket     
15:16 < GitHub61> [joinmarket-clientserver] AdamISZ opened pull request #157: Fix bug in IRC nick collision handling (master...fix-irc-collisions) https://git.io/fbjYF
15:16 -!- GitHub61 [GitHub61@gateway/service/github.com/x-yonibeaygjwtzdku] has left #joinmarket []     
15:21 < undeath> does this handle the case that someone re-claims the original nick later on again?
15:22 < undeath> also, memleak
15:22 < undeath> can you not monitor joins/parts?
15:22 < waxwing> no, it doesn't (at least at first thought)
15:23 < waxwing> yes that sounds more logical
15:23 < waxwing> arubi, i got this one: https://travis-ci.org/JoinMarket-Org/joinmarket-clientserver/jobs/400642846#L1100
15:23 < waxwing> libsodium not built, is it a case of just 'try again'?
15:23 < waxwing> undeath, thanks, i will try again, feel free to pass more comments if you have them
15:24 < undeath> re build: looks like it failed to dl libsodium
15:24 < undeath> probably a temporary problem
15:25 < waxwing> yes     
15:25 < waxwing> so we have userJoined, should be possible to hook off that ... i guess monitoring parts isn't needed for this?
15:26 < undeath> depends on the implementation I guess
15:27 < undeath> we would want a way to prevent memleaks, else the collision table will just keep growing
15:27 < waxwing> yes, got that
15:31 < waxwing> by which i mean 'understand the problem' not necessarily 'know the right solution' ...
15:32 < undeath> there is no hook for parts? :(
15:33 < waxwing> there is
15:33 < waxwing> oh, so you mean, that's a natural solution?
15:33 < undeath> I think so
15:33 < waxwing> ok, yeah, makes sense. thanks.
15:42 -!- undeath [~undeath@unaffiliated/undeath] has quit [Quit: WeeChat 2.1]     
15:48 < AgoraRelay> [agora-irc/CatoAlex] might this be done without a collision table? e.g. always use the full, untruncated nickname for everything except the steps in which the nickname is used for authentication, in which case the namestring could just remove any "_" on the fly?
15:50 < waxwing> i think so. the difficulty arises in the fact that the nick is being used across N message channels, usually only 1 of which has the collision
15:51 < waxwing> so it seemed more logical to me as per comments to isolate the handling to the specific message channel itself
16:20 < AgoraRelay> [agora-irc/CatoAlex] or even easier idea: let the makers check periodically if their nick is too long and rename themselves back. Tolerate the few minutes in which they might not be able to get joins. Rationale: its usually just a few seconds of overlap, when the server does not know about the disconnection yet, but the maker already rejoined
16:24 < waxwing> hmm yeah possibly. i'm off now (and will be all over the place next day or two) but keep thinking about limitations of various solutions. worth remembering that the dynamic switch between message channels already probably solves a good chunk of these problems (if you change nick on one channel but not the other the switch should happen when you disconnect, the MCC notices and changes the channel on which you're active, iirc)
20:47 -!- AgoraRelay [~jmrelayfn@p5DE4A326.dip0.t-ipconnect.de] has quit [Ping timeout: 260 seconds]     
20:59 -!- AgoraRelay [~jmrelayfn@p5DE4A329.dip0.t-ipconnect.de] has joined #joinmarket     
21:17 < AgoraRelay> [agora-irc/paradox] Is anyone here?
21:54 -!- belcher_ [~user@unaffiliated/belcher] has joined #joinmarket     
21:56 -!- belcher [~user@unaffiliated/belcher] has quit [Ping timeout: 255 seconds]     
23:37 < arubi> waxwing, oh I was positive curl would retry the download, guess not.  I'll add that in.
23:39 < arubi> paradox (not sure if you're still around), it's usually a slow channel.  best to ask a question if you have one and lurk around