--- Day changed Wed Oct 26 2016
00:27 -!- mkarrer_ [~mkarrer@7.red-83-47-85.dynamicip.rima-tde.net] has quit [Remote host closed the connection]     
00:28 -!- mkarrer [~mkarrer@7.red-83-47-85.dynamicip.rima-tde.net] has joined #joinmarket     
00:59 < waxwing> alexcato yeah thanks, that cleared it up; by "N+x" you mean the same as what i called "adlai's idea", "taker sophistication", need better names for things :) and yes, agree with comment.
01:55 -!- Yohkii [~IceChat9@dhcp2372-stud.wifi.uit.no] has joined #joinmarket     
02:52 -!- Yohkii [~IceChat9@dhcp2372-stud.wifi.uit.no] has quit [Quit: ASCII a stupid question, get a stupid ANSI!]     
04:22 < JM-IRCRelay> [AlexCato] @waxwing in PR #647 in this line: https://github.com/JoinMarket-Org/joinmarket/pull/647/files#diff-e5ebbf3237752f0c78e5834760582020R438 shouldnt that crash in case the taker upgraded from an earlier version and doesnt have the config setting in his joinmarket.cfg?
04:24 < waxwing> alexcato no, don't think so, as the load_program_config has an "overlay" approach, where the hardcoded global_singleton values remain if they're not overwritten by joinmarket.cfg
04:25 < waxwing> there might be another corner case though, must look at it now: what if they set minimum_makers = 4 and then use -N 3 :) would be dumb but have to check what happens.
04:26 < waxwing> alexcato: ah no, i'm wrong, sorry
04:27 < waxwing> i forgot the difference between config and singleton, that was dumb .. yes, need to sanity check it then, it load_program_config
04:28 -!- sturles [~sturles@unaffiliated/sturles] has quit [Remote host closed the connection]     
04:28 < waxwing> mm, no again :)
04:28 < waxwing> it does use overlay, where it reads default config from hardcoded, then overlays joinmarket.cfg, and minimum_makers is in the hardcoded bit.
04:29 < waxwing> so my error is that i put global_singleton.minimum_makers which is irrelevant, will remove. sanity checking now.
04:32 < JM-IRCRelay> [AlexCato] hm. But if you remove the global_singleton.minimum_makers in configure.py, there's no value that can be read if the setting is not in the .cfg; or I still didnt get it quite right
04:33 < waxwing> it's also in the hardcoded default joinmarket.cfg in the same configure.py
04:34 < waxwing> that's the right one, the global_singleton.minimum_makers i will just remove. sanity checked ok.
04:34 < JM-IRCRelay> [AlexCato] wouldnt the 'usual' way be to read in that minimum_config option in load_program_config(), have a check whether the settings exists in the cfg and then set the global_singleton.minimum_makers . From the outside scripts it should only ever access the singleton (architecturally easier)
04:34 < waxwing> (because we're actually using global_singleton.config.POLICY.minimum_makers, not global_singleton.minimum_makers
04:34 < JM-IRCRelay> [AlexCato] yes, but wont the new cfg setting only be written if no prior .cfg file exists?
04:35 < waxwing> no, it reads from the default config in the file first, then reads again the joinmarket.cfg of the user; that effectively "overlays"
04:35 < waxwing> so if setting X was in the default but not in the joinmarket.cfg, it is still kept.
04:36 < waxwing> i can't remember who did that PR for us, i wasn't originally aware of that, but it's obviously a useful feature for configs :)
04:36 < JM-IRCRelay> [AlexCato] then i've added a few too many lines for the logging configuration :P
04:36 < waxwing> yes i think that's correct. i wasn't interested enough to go into it though. a couple extra sanity check lines don't really matter
04:37 < JM-IRCRelay> [AlexCato] ic, got it
05:03 < GithubBot5678> [joinmarket] AdamISZ pushed 1 new commit to develop: https://git.io/vXe1E
05:03 < GithubBot5678> joinmarket/develop ebf28ed Adam Gibson: revert to (4,6) default for -N sendpayment; better in line with #647
05:06 -!- gmaxwell [greg@wikimedia/KatWalsh/x-0001] has joined #joinmarket     
05:54 -!- mkarrer [~mkarrer@7.red-83-47-85.dynamicip.rima-tde.net] has quit []     
06:02 < waxwing> relevant to yesterday's joke :) https://twitter.com/Kuwaddo/status/790902104835690496
06:05 < adlai> belcher: looks like we're just slightly over 100 sat/byte for next-block confirmation
06:06 < adlai> (btc:estimatesmartfee 1) -> ((:FEERATE . 100543/100000000) (:BLOCKS . 2))
06:06 < waxwing> ah; estimatesmartfee? is that a new one?
06:07 < adlai> yes. (btc:estimatefee 1) returns -1 but this new one does give a value
06:07 < waxwing> cool, is that .13.0 or .13.1? i still haven't upgraded 0.12.1 (lazy)
06:08 < adlai> probably 13.0
06:08  * adlai checks
06:08 < waxwing> (btw presumably i don't have to rescan if i do that?)
06:08 < adlai> why would it be related? iirc it's based off the last few blocks, and current mempool
06:08 < waxwing> maybe we could add it in to JM, i guess then we'd need to check the Core version
06:09 < adlai> bitcoin-daemon 0.13.0-1 (archlinux)
06:09 < waxwing> oh no i just wasn't sure whether upgrading your Core node impacts in such a way that one might need to rescan; presumably not, i guess a dense question :)
06:09 < JM-IRCRelay> [AlexCato] smartfee works with 13.0 (and i rushed that update, wanted lean blocks to remove the spikiness of new blocks from the tor network)
06:10 < adlai> ah. this specific node runs with disablewallet, joinmarket is on a computer with less of a bleeding-edge update cycle :)
06:11 < waxwing> adlai: dust threshold? my specific question is, do you agree that because it's an edge-case protocol break, it should be reverted, or do you think the benefits outweigh the few cases where people will have errors from it
06:11 < waxwing> magic numbers are always a pain
06:11 < adlai> i'm not satisfied yet with my understanding of it, but reverting it seems the best approach atm
06:12 < adlai> ideally makers would advertise their dust threshold, along with the minimal fee/kb for the final transaction, so we wouldn't have half our coins stuck in limbo now that the blocks are all full
06:12  * adlai bbl lunch
06:13 < waxwing> yeah that's a valid point about makers advertising their req. fees, needs thinking through. probably 0.3 i guess :)
06:13 < waxwing> "req. fees" = "required minimum bitcoin transaction fee"
06:14 < waxwing> i think i'll just squash 647 now, wait a day or two, merge and then it'll be release time i think. opinions? is there something else pressing?
06:16 < waxwing> yeah i think i see 1 tx stuck in limbo from yesterday, it'll confirm eventually no doubt, it's low but not amazingly low
06:29 < waxwing> but not 6000 sat/kB as belcher said yesterday, that's pretty crazy. i don't think i ever saw something that low for many months.
06:33 < JM-IRCRelay> [AlexCato] thats mainly because most fee predictions kind of suck, though. There were several hour periods this weekend with half-full blocks, yet even the core estimation for 25 target block confirmation still suggested 40 sat/byte; this also creates a vicious circle. Fees not going down on half-empty blocks = fees rising faster when blocks are full
06:33 < waxwing> re dust threshold adlai, it's just that the old version makers are proposing change outputs too small for the new limit; i also think reverting is better on balance, favouring takers not having transaction failure-to-complete problems over other priorities. if you agree feel free to make a revert commit.
07:06 -!- Giszmo [~leo@pc-40-227-45-190.cm.vtr.net] has joined #joinmarket     
07:10 -!- King_Rex [~King_Rex@unaffiliated/king-rex/x-3258444] has joined #joinmarket     
07:11 -!- King_Rex [~King_Rex@unaffiliated/king-rex/x-3258444] has quit [Client Quit]     
09:07 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has joined #joinmarket     
09:11 < Numin0us> hey I deposited some BTC in my joinmarket wallet when the blockchain_source was set to blockr, I wanted to switch to bitcoin-rpc so that I could run a yield generator with a full node but by bitcoin doesn't appear in the wallet addresses with that configuration
09:12 < Numin0us> I'm a bit of a noob.. how can I go about fixing this
09:12 < waxwing> Numin0us: you probably need to restart bitcoind with -rescan
09:13 < waxwing> otherwise bitcoind can't find coins in the joinmarket addresses that have been imported
09:18 < waxwing> Numin0us: this is mentioned in this section: https://github.com/JoinMarket-Org/joinmarket/wiki/Running-JoinMarket-with-Bitcoin-Core-full-node#running--checking-it-works
09:18 < waxwing> but do also read the whole of that page, if you haven't
09:37 -!- owowo [~ovovo@unaffiliated/ovovo] has quit [Ping timeout: 244 seconds]     
09:41 -!- mkarrer [~mkarrer@7.red-83-47-85.dynamicip.rima-tde.net] has joined #joinmarket     
09:42 -!- owowo [~ovovo@31.169.51.195] has joined #joinmarket     
09:42 -!- owowo [~ovovo@31.169.51.195] has quit [Changing host]     
09:42 -!- owowo [~ovovo@unaffiliated/ovovo] has joined #joinmarket     
11:04 < GithubBot5678> [joinmarket] AlexCato opened pull request #648: Suggestion for 0.2.2 release notes and 0.2.2 README. Feel free to add… (develop...develop) https://git.io/vXvrT
11:12 < JM-IRCRelay> [waxwing] AlexCato: thanks, much appreciated. will add the extra bits at the end, but that should be the the bulk of it. should try to include every commit, but that is offset in cases where there are trivial fixups. main thing is not to forget to credit any external contributors, i think you've done that.
11:12 < JM-IRCRelay> [waxwing] i'll also give more prominence to the new things as 'features' rather than just bugfixes (logging, direct send, minimum makers)
11:13 < JM-IRCRelay> [AlexCato] sure, as said, it's meant to take the menial work off you. Feel free to adjust any way you see fit :)
11:17 < waxwing> the biggest channel apart from #joinmarket-pit on cgan has 19 nicks. we have 138. i wonder if it'll fall over at some point.
11:18 < JM-IRCRelay> [AlexCato] hope not... overall, it should still be just a few KBs of traffic, potentially even easier to host on a homeserver than a full node. But with 0.2.2 agora is in the default config as well, so there's at least some redundancy now
11:19 < waxwing> yeah it's not the number of nicks for sure, and 'fall over' is a bit of a silly concern. i guess it's more about the burstiness. it would be cool to have to worry about > 500 or something :)
11:20 < waxwing> i think smuggler might get upset if we suddenly plonk > 100 on agora, i warned him about 'maybe 50' a month or two back; for now it's only 17 there and very little activity
13:06 < waxwing> https://www.reddit.com/r/joinmarket/comments/59gnot/joinmarket_going_nuts/ <- any other ideas feel free to chime in
13:55 -!- Giszmo [~leo@pc-40-227-45-190.cm.vtr.net] has quit [Quit: Leaving.]     
13:57 -!- Giszmo [~leo@pc-40-227-45-190.cm.vtr.net] has joined #joinmarket     
14:19 < belcher> read all of today's chatting
14:19 < belcher> noticing that half my yieldgen's coins are in limbo, investigating
14:21 < belcher> okay this could be accident, ~30sat/kb tx is holding up about 30btc
14:21 < belcher> likely is a borked fee estimator somewhere
14:22 < belcher> if this site had an API it would be amazing https://bitcoinfees.github.io/ it does a cute monte carlo solver to estimate the fee rate
14:23 < midnightmagic> belcher: if you PM me a tx, I can guarantee confirmation "eventually".
14:26 < Numin0us> thanks @waxwing I can see my BTC from Bitcoin-QT but now my simple-wallet tool isn't working, when I try to open my wallet.json file I'm getting an error like this
14:26 < Numin0us> @waxwing joinmarket.jsonrpc.JsonRpcConnectionError: authentication for JSON-RPC failed
14:27 < belcher> Numin0us did you put your correct username and password into joinmarket.cfg? the same one thats in bitcoin.conf
14:27 < Numin0us> I'm using bitcoin RPC and all the credentials are correct 
14:27 < belcher> midnightmagic thanks, how would it work?
14:27 < belcher> like, wouldnt that cost money to whatever miner did it
14:27 < Numin0us> it was working prior to me rerunning bitcoind with the rescan option
14:27 < Numin0us> @belcher
14:28 < belcher> that is weird, rescan shouldnt affect it
14:28 < JM-IRCRelay> [LastManStanding] Hello folks. Just popped in. New to this and just snooping
14:28 < belcher> hello LastManStanding
14:28 < JM-IRCRelay> [LastManStanding] Hi there.
14:29 < belcher> thought, how would RBF on joinmarket work?
14:32 < midnightmagic> belcher: A miner has graciously given me access to their mempool.
14:32 < midnightmagic> (in limited fashion anyway)
14:32 < midnightmagic> belcher: It would cost nothing additional. It could be 0-fee and I could push it through.
14:36 < JM-IRCRelay> [LastManStanding] Okay guys, enjoy your day. Thanks for having me.
14:36 < midnightmagic> \o     
14:36 < midnightmagic> later dude
14:36 < JM-IRCRelay> [LastManStanding] tada!
15:58 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has quit [Ping timeout: 245 seconds]     
16:58 -!- Numin0us [~Numin0us@cpe-74-64-92-78.hvc.res.rr.com] has joined #joinmarket     
16:58 -!- Numin0us [~Numin0us@cpe-74-64-92-78.hvc.res.rr.com] has quit [Changing host]     
16:58 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has joined #joinmarket     
17:37 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has quit [Ping timeout: 256 seconds]     
18:17 -!- Socket_0x03 [~Socket_0x@c-66-176-87-156.hsd1.fl.comcast.net] has joined #joinmarket     
18:29 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has joined #joinmarket     
19:19 -!- Giszmo [~leo@pc-40-227-45-190.cm.vtr.net] has quit [Quit: Leaving.]     
19:41 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has quit [Ping timeout: 260 seconds]     
20:26 -!- midnightmagic [~midnightm@unaffiliated/midnightmagic] has quit [Ping timeout: 250 seconds]     
20:45 -!- midnightmagic [~midnightm@unaffiliated/midnightmagic] has joined #joinmarket     
20:53 -!- Socket_0x03 [~Socket_0x@c-66-176-87-156.hsd1.fl.comcast.net] has quit [Remote host closed the connection]     
21:02 -!- Socket_0x03 [~Socket_0x@c-66-176-87-156.hsd1.fl.comcast.net] has joined #joinmarket     
21:15 -!- Socket_0x03 [~Socket_0x@c-66-176-87-156.hsd1.fl.comcast.net] has quit [Remote host closed the connection]     
21:32 -!- Socket_0x03 [~Socket_0x@c-66-176-87-156.hsd1.fl.comcast.net] has joined #joinmarket