--- Day changed Sat Jul 02 2016
00:18 -!- fqtw_ is now known as fqtw
02:05 -!- p15 [~p15@140.91.145.64.unassigned.bringover.net] has quit [Ping timeout: 250 seconds]
04:15 <+JM-IRCRelay> [AlexCato] currently working on more documentation for JM, this time: step-by-step guide to configure joinmarket to use bitcoin-rpc on windows. Two questions about that:
04:16 <+JM-IRCRelay> [AlexCato] 1. which cUrl version is needed? I assume the simplest win32 generic version (without ssl, ssh and all the other possible options) is sufficient?
04:16 <+JM-IRCRelay> [AlexCato] 2. Should go into %windir% to work seemlessly with JM?
05:23 <+JM-IRCRelay> [AlexCato] 2. %windir%\system32 works and seems to be a better choice
06:05 <+JM-IRCRelay> [AlexCato] @waxwing: testing this with joinmarket-qt (windows binary), so a full bitcoin core node is synced and configured to work with it. Created new wallet, sent funds to it. After 1 confirmation and a rightclick/"resync wallet from blockchain" it did show the funds. After a restart of jm-qt it took quite long to sync (>3 minutes), now shows "wallet synced
06:05 <+JM-IRCRelay> successfully", but with 0 balance and shows the external address as unused
06:06 <+JM-IRCRelay> [AlexCato] blockexplorer shows the funds are in the adress. There's no logs in jm-qt (none i can find at least) and the bitcoin-core-log shows its synced just fine, all blocks accounted for. Any pointers?
06:08 <+JM-IRCRelay> [AlexCato] changing the jm-qt from bitcoin-rpc to blockr makes it work. So it must have something to do with the rpc connection
06:12 -!- malte [QxJL4CBES2@alkaid.uberspace.de] has quit [Ping timeout: 250 seconds]
06:26 -!- malte [WRpcFcUkiG@alkaid.uberspace.de] has joined #joinmarket
06:37 -!- malte [WRpcFcUkiG@alkaid.uberspace.de] has quit [Read error: Connection timed out]
06:38 -!- malte [tVpNoJejMW@alkaid.uberspace.de] has joined #joinmarket
06:43 -!- coins123 [~coins123@unaffiliated/coins123] has quit [Ping timeout: 272 seconds]
06:49 <+JM-IRCRelay> [AlexCato] hm. Does bitcoin-qt ignore the bitcoin.conf file and only use the (reduced) configuration options available under settings->options?
06:58 < waxwing> alexcato bitcoin.conf or joinmarket.cfg?
06:59 < waxwing> settings->options ought to show all the options in joinmarket.cfg if i remember, but best to just edit joinmarket.cfg directly i guess
07:03 <+JM-IRCRelay> [AlexCato] bitcoin.conf in that case. The jm.cfg works as intended, but i cant get bitcoin-qt to accept rpc connections. And for newbie windows users, using bitcoind might be too complicated
07:05 < waxwing> right. i have a feeling someone looked into this.
07:05 < waxwing> hmm, i think i managed to set it up myself at least once or twice.
07:05 <+JM-IRCRelay> [AlexCato] alright, so there is a way. I'll dig into it some more then. Thanks :)
07:05 < waxwing> maybe search the reddit for threads about windows
07:05 < waxwing> not promising you'll find anything, just a thought
07:06 < waxwing> so you're saying bitcoin-cli can't connect on the command line?
07:11 <+JM-IRCRelay> [AlexCato] ah, good way to check. This gives:
07:11 <+JM-IRCRelay> [AlexCato] C:\Program Files\Bitcoin\daemon>bitcoin-cli -rpcconnect=localhost -rpcport=8332 -rpcuser=AlexCato -rpcpassword 8Xd04Wsm8347fYG8FEO6 estimatefee 3
07:11 <+JM-IRCRelay> [AlexCato] error: Could not locate RPC credentials. No authentication cookie could be found, and no rpcpassword is set in the configuration file (C:\Users\alex\AppData\Roaming\Bitcoin\bitcoin.conf)
07:11 <+JM-IRCRelay> [AlexCato] except: there is a bitcoin.conf file at that location specifying exactly the credentials
07:12 <+JM-IRCRelay> [AlexCato] (and i've changed the pasted password. Nothing to gain here, hackers ;) )
07:16 < waxwing> bitcoin.conf.txt? :)
07:16 < waxwing> i really doubt it, that's a dumb error, but i remember seeing someone have that once
07:16 <+JM-IRCRelay> [AlexCato] nah :) That was step 2 of the tutorial: disable file extension hiding
07:17 <+JM-IRCRelay> [AlexCato] but i just see a typo, theres a = missing at the password parameter
07:18 <+JM-IRCRelay> [AlexCato] bitcoind on the windows machine works as expected. Trying again without the typo in qt
07:24 <+JM-IRCRelay> [AlexCato] interesting: the bitcoin-cli command now also works with bitcoin-qt. But joinmarket-qt still doesnt seem to be able to get the transaction history of its wallet
07:24 < waxwing> did you change the pwd?
07:24 <+JM-IRCRelay> [AlexCato] it "finishes" (=timeout) after 4 minutes, giving a "synced successfully" and shows 0 balance.
07:25 <+JM-IRCRelay> [AlexCato] the pwd was generated by me, yes. Changed in both joinmarket.cfg and bitcoin.conf
07:25 < waxwing> hmm, did you do any command line version first i wonder? wondering if you might need to rescan? is it a new wallet or a heavily used one?
07:26 < waxwing> just musing, no idea really
07:26 <+JM-IRCRelay> [AlexCato] new wallet, just has 1 transaction into an external address so far
07:29 <+JM-IRCRelay> [AlexCato] anyways, continuing later. Hopefully have some doc updates PRs ready soonish
07:32 <+JM-IRCRelay> [jony] hey. what is joinmarket ? how is the best way to start?
07:33 -!- coins123 [~coins123@unaffiliated/coins123] has joined #joinmarket
07:33 <+JM-IRCRelay> [AlexCato] both questions are answered here: https://github.com/JoinMarket-Org/joinmarket/wiki , right at the top :)
07:33 <+JM-IRCRelay> [jony] thank you...
07:33 <+JM-IRCRelay> [jony] :]
07:52 -!- mec_ [~mec@209.95.50.138] has joined #joinmarket
08:27 -!- puddinpop [~puddinpop@unaffiliated/puddinpop] has joined #joinmarket
08:49 <+JM-IRCRelay> [AlexCato] resync indeed fixed it. This is a usability nightmare though ;( Might as well do the tutorial for the python scripts on windows instead. Same hassle, much more gain, because then at least the users can tumble and yieldgen too
08:49 <+JM-IRCRelay> [AlexCato] all this trouble just for sendpayment probably isnt worth it
08:55 < waxwing> yes, totally agree. a gui is only really for "trying it out" or maybe doing a few retail payments. at least, as it is now; nothing saying it can't be improved! (although i'd prefer to make improvements on the electrum plugin style approach)
08:56 < waxwing> (and that's why blockr is default)
08:57 < waxwing> but, def arguable, i.e. even for "retail usage" the Core backend has such better privacy properties ... even if we mix and match backends like APIs and electrum servers, it never reaches the level of a local node (or even close)
08:58 < waxwing> and, hey, if we ever get the whole thing into a better state, then something like bitcoin-qt can be a front end at some point.
08:58 <+JM-IRCRelay> [AlexCato] yeah, agreed to all of that.
09:00 <+JM-IRCRelay> [AlexCato] though there still seems to be some kind of problem. Initial opening of a wallet with joinmarket-qt is lightning fast, shows the balance correctly. Sending via sendpayment is frozen since ~3 minutes now, status bar says "syncing wallet...", last debug-message is "starting sendpayment"
09:01 <+JM-IRCRelay> [AlexCato] now it continues. Frozen for ~4-6 minutes i'd estimate
09:01 <+JM-IRCRelay> [AlexCato] anyways, the single biggest boost to joinmarket probably would indeed be making it easily usable with some kind of wallet integration
09:01 <+JM-IRCRelay> [AlexCato] plus fees should stop rising, this could kill it as well
09:02 < waxwing> the combination of symptoms you describe baffles me.
09:02 < waxwing> i never saw that kind of thing. there is a resync delay on starting a transaction, but why it would be ultra-slow whereas fine the first time, i have no idea.
09:04 <+JM-IRCRelay> [AlexCato] yup, makes no sense. Same symptoms when i try to right-click the balance and choose "resync from blockchain"
09:05 < waxwing> sorry, last debug is "starting sendpayment" *while* there is a 5 minute delay?
09:05 <+JM-IRCRelay> [AlexCato] sendpayment went through now, though
09:05 < waxwing> or was it showing "syncing wallet" during that time?
09:05 <+JM-IRCRelay> [AlexCato] yes
09:05 <+JM-IRCRelay> [AlexCato] debug window sayd "starting sendpayment"
09:05 <+JM-IRCRelay> [AlexCato] status said "syncing wallet"
09:05 <+JM-IRCRelay> [AlexCato] both at the same time
09:05 < waxwing> ok, i see
09:06 < waxwing> above you said "resync indeed fixed it". i think you meant 'rescan'?
09:06 <+JM-IRCRelay> [AlexCato] aw. Yes, rescan
09:06 <+JM-IRCRelay> [AlexCato] resync didnt do anything before, except freeze in the same way
09:07 <+JM-IRCRelay> [AlexCato] anyways, gotta run now. Will think about doing the tutorial with python-scripts instead
09:24 -!- Giszmo [~leo@pc-122-14-46-190.cm.vtr.net] has joined #joinmarket
10:06 -!- molz [~molly@unaffiliated/molly] has joined #joinmarket
10:09 -!- moli [~molly@unaffiliated/molly] has quit [Ping timeout: 240 seconds]
10:20 < jm_user> I am stillhaving issues using the core backend with JM.
10:21 < jm_user> "bitcoin-cli listtransactions joinmarket-wallet-xxxxxx 1000 0 true" wirks and gives a sensible output showing transactions.
10:21 < jm_user> 2016-07-02 17:16:28,047 [MainThread  ] [DEBUG]  rpc: listtransactions ['joinmarket-wallet-xxxxxx', 1000, 0, True]
10:21 < jm_user> 2016-07-02 17:16:34,907 [MainThread  ] [DEBUG]  too few addresses in [(0, 1), (1, 1), (2, 1), (3, 1), (4, 1)]
10:21 < jm_user> 2016-07-02 17:16:38,756 [MainThread  ] [DEBUG]  importing 300 addresses into account joinmarket-wallet-7ac2f5
10:21 < jm_user> restart Bitcoin Core with -rescan if you're recovering an existing wallet from backup seed otherwise just restart this joinmarket script
10:22 < K1773R> just restart the script, it has to import some addresses first
10:22 < jm_user> "python wallet-tool.py wallet.json" ?
10:24 < jm_user> the second part was output from the python script, which seems to fail...
10:25 < K1773R> you can run yigen or any other script, each will import addresses on the start
10:25 < K1773R> wallet tool works too therefore
10:26 < jm_user> right. just finihsed, with same message and no output.
10:26 < K1773R> you will have to do it several times, at least that was the case for me
10:27 < jm_user> ok. thanks. Trying that now...
10:28 < jm_user> can someone give a pointer on how to use the JM-IRCRelay (via tor if possible)?
10:28 -!- megaddin [aladdin@bnc.fnordserver.net] has quit [Remote host closed the connection]
10:28 < K1773R> you go to the cyberguerilla network and chat there
10:29 <+JM-IRCRelay> [K1773R] like this, for example
10:29 < K1773R> all it does is relay message from one channel to the other, because freenode dosnt allow tor
10:30 < jm_user> is that 6dvj6v5imhny3anf.onion or irc.cyberguerrilla.org?
10:31 < K1773R> let me check the onion for cyberguerilla...
10:32 < K1773R> https://cyberguerrilla.info/ways-to-connect-to-cyberguerrilla-irc/
10:32 < K1773R> choose for yourself
10:33 < jm_user> t#
10:33 < jm_user> thanks for the link.
10:34 < jm_user> any recommendation ofr a windows irc client? It has been years for me. mirc still ok? Last I remember there was a change in licencing?
10:37 <+JM-IRCRelay> [jm_loser] ah yes, here we go...
10:39 < K1773R> dunno, dont use winblows...
10:41 <+JM-IRCRelay> [jm_loser] good for you!
10:52 < waxwing> jm_user do let us know if multiple retries as suggested doesn't work. it can be sped up by increasing addr_req_count in the code (but it isn't necessary). we've been looking into ways to improve this recently (it has been made more of an issue by the spammer).
10:56 <+JM-IRCRelay> [jm_loser] sofar no joy. This is number 5 run now in progress.
10:56 <+JM-IRCRelay> [jm_loser] What was the spammer issue?
10:57 < waxwing> someone starting but not finishing transactions, which marks addresses in your wallet as used (which it has to, to prevent you reusing them), even though you didn't use them in transactions.
10:57 <+JM-IRCRelay> [jm_loser] I did notice numerous JM attempts with "user left" before the final signatures could be completed...
10:57 < waxwing> it results in lots of large gaps in your wallet
10:58 <+JM-IRCRelay> [jm_loser] Gaps?
10:58 <+JM-IRCRelay> [jm_loser] And what does the spammer gain? DOS?
10:58 < waxwing> HD wallets have "branches". on each branch there is a sequence (effectively infinite) of addresses. you see those sequences when you run wallet-tool.
10:59 < waxwing> if you use address 0 and then don't use address 1-9 and then use address 10, then 1-9 is a gap
10:59 <+JM-IRCRelay> [jm_loser] ...which bloats the wallet?
10:59 < waxwing> gains: we don't think the aim is DOS, we think the aim is to collect utxos of yield generator bots in an attempt to deanonymise. but it's not entirely clear, but that's the main issue. read issue 156 on the repo.
11:00 < waxwing> bloat, well, not really, perhaps it uses more data in your bitcoin core instance, and it makes wallet syncing slower for sure.
11:00 < waxwing> forgive me if i don't write every detail about this, it is a very big subject which has been discussed at length in many places, prob best to start at issue 156 on the repo
11:01 < waxwing> https://github.com/JoinMarket-Org/joinmarket/issues/156
11:01 <+JM-IRCRelay> [jm_loser] that is fine. It gives me a pointer and already a good overview.
11:02 <+JM-IRCRelay> [jm_loser] I have been wondering what was going on there... are there any leads / theories  who might be behind this?
11:06 -!- megaddin [aladdin@gateway/shell/fnordserver.eu/x-ykjdbbkqrkzkjxnd] has joined #joinmarket
11:29 <+JM-IRCRelay> [jm_loser] running the script the 10th time now and still no sensible output.
11:29 <+JM-IRCRelay> [jm_loser] still giving me the "restart Bitcoin Core with -rescan if you're recovering an existing wallet from backup seed  otherwise just restart this joinmarket script".
11:30 < waxwing> what numbers do you see in index_cache in the wallet.json file? just for context
11:31 < waxwing> oh, and is it still saying "too few addresses in: " etc?
11:33 <+JM-IRCRelay> [jm_loser] the lowest number is over 200 and the highest over 300 in "index_cache". Yes, still telling me too "few addresses in..." So I keep going?
11:33 < waxwing> ah, that sounds strange. let me think a minute (but yes, keep going, there's no harm in doing so)
11:34 <+JM-IRCRelay> [jm_loser] ok, going on. At about 13now. have to manually enter the wallet password at each restart...
11:35 < waxwing> how did you start this process? are you starting with a previously-existing wallet.json file on a new bitcoin core?
11:35 -!- mkarrer [~mkarrer@48.red-83-47-111.dynamicip.rima-tde.net] has quit []
11:35 < waxwing> or is it just, you stopped the script/bot and are just restarting with no change?
11:36 -!- mkarrer [~mkarrer@48.red-83-47-111.dynamicip.rima-tde.net] has joined #joinmarket
11:37 < waxwing> and the last question is, what version are you running, i.e. is it the release/master or develop (if so which commit)?
11:40 <+JM-IRCRelay> [jm_loser] started using JM with blockr and did a number of joins for a few weeks. Once the blockchain was finally synced I am trying to move away from blockr and use core. standby for version numbers.
11:40 < GithubBot5678> [joinmarket] adlai pushed 1 new commit to develop: https://git.io/vKvQj
11:40 < GithubBot5678> joinmarket/develop 984626b Adlai: Add index logging to wallet-tool import loop...
11:40 < adlai> jm_luser: pull develop
11:41 < waxwing> OK. well, apart from that, i strongly suggest editing this line: https://github.com/JoinMarket-Org/joinmarket/blob/master/joinmarket/blockchaininterface.py#L590
11:41 < waxwing> increase to 200
11:41 < adlai> this should alleviate the "running the script the 10th time now and still no sensible output." issue
11:41 <+JM-IRCRelay> [jm_loser] Bitcoin Core Daemon version v0.12.1.0-e7ec24e
11:41 < waxwing> adlai: hmm, i'd really rather you didn't do that
11:42 < waxwing> the PR i've just produced should kill this problem in principle
11:42 < adlai> "that" = commit directly, or the change itself?
11:42 < waxwing> yeah the commit. can be reverted of course.
11:43 <+JM-IRCRelay> [jm_loser] dumb question but how do I check which version I am running release/master or develop? I think I switched to develop some time ago...
11:44 < waxwing> aside from what i set in the notes to #586, it ought to mean that syncing happens in 2 runs, i think, in most cases.
11:44 <+JM-IRCRelay> [jm_loser] looks like the develop branch.
11:44 < waxwing> jm_loser well you can do git branch
11:44 <+JM-IRCRelay> [jm_loser] yes, develop !
11:44 < waxwing> ok, you can check which commit you're on with git log. but can i suggest just making the single edit i isolated above first, and try running it twice more, first.
11:45 < adlai> waxwing: as penance, I shall now test your PR
11:45 < waxwing> the problem is basically that it's not importing enough addresses, at least i (we) think so
11:46 < waxwing> adlai: well i'm just being a bit annoying really; it turns out it doesn't even conflict :)
11:46 < waxwing> more logging of that stuff is a good idea generally
11:46  * adlai agrees, and suspects there may be a problem with GNA giving you new addresses to import and skipping the ones you needed too
11:46 < adlai> the current code calls GNA again as though it'll give the same addr_req_count addresses, but it gives the ~next~ ones
11:46 < waxwing> each run always starts the import process from the beginning
11:46 < waxwing> at what step adlai ?
11:47 < waxwing> or where in the code rather
11:47 < adlai> line 12 of your 0bin
11:47 < adlai> https://0bin.net/paste/gCSoYSCCEcOsCYmo#3ZjlEicKTeFAK+y9Br1ZH40GWgmMqgR2YAHe1wNbVef
11:48 <+JM-IRCRelay> [jm_loser] Running the latest version from "Sat Jul 2 21:40:28 2016 +0300"
11:49 < waxwing> adlai: ah, see what you mean. thinking.
11:49 <+JM-IRCRelay> [jm_loser] just upgrade. nu actually run it yet.
11:49 < adlai> this is my explanation for how i got stuck in an infinite loop, although i didn't add enough log.debug statements to fully confirm it
11:50 < adlai> jm_loser: "nu" in hebrew means "get on with it", so... nu!
11:52 < adlai> waxwing: https://github.com/JoinMarket-Org/joinmarket/pull/586/files#r69381513
11:55 <+JM-IRCRelay> [jm_loser] "importing 2000 addresses into account..."
11:57 <+JM-IRCRelay> [jm_loser] "too few addresses..." has dissapeared.
12:00 < waxwing> adlai: yes i believe you're right, but i'm looking at the latest code on develop. ie. i agree that it's setting index to next_avail_idx even if TFAMC is set and so it'll start importing only from there.
12:00 < waxwing> jm_loser, and what happened then?
12:01 <+JM-IRCRelay> [jm_loser] still processing. It seems niot to respond to "nu". Its a RPi.
12:05 < waxwing> i think that's an error on my part adlai ; while we do want to set next_avail_idx to the index_cache value if there are *not* too few addresses imported, if there are, then the index should be set to the start of that batch before doing the add_watchonly_addresses.
12:06 < waxwing> i wasn't able to reproduce or figure out a way to get an infinite loop, but maybe that is it.
12:06 < waxwing> re: comment adlai thanks, of course i forgot that sorry. good catch.
12:10 < waxwing> adlai: do you agree with what i wrote in that sentence above adlai ? if so, we can fold it into #586 i guess. or maybe do it separately.
12:13 <+JM-IRCRelay> [jm_loser] not looking too great:
12:13 <+JM-IRCRelay> [jm_loser] "2016-07-02 18:56:28,563 [MainThread  ] [DEBUG]  hello joinmarket Enter wallet decryption passphrase: 2016-07-02 18:56:34,170 [MainThread  ] [DEBUG]  requesting wallet history 2016-07-02 18:56:58,521 [MainThread  ] [DEBUG]  rpc: getaddressesbyaccount ['joinmarket-wallet-xxxxxx'] 2016-07-02 18:56:58,809 [MainThread  ] [DEBUG]  importing 2000 addresses
12:13 <+JM-IRCRelay> into account joinmarket-wallet-xxxxxx restart Bitcoin Core with -rescan if you're recovering
12:13 < waxwing> so you've done it once after the edit?
12:14 <+JM-IRCRelay> [jm_loser] once, yes. Second run is on it way now. give it another few minutes...
12:15 < waxwing> so before this you were using blockr, and this happened when you switched to Core? and you rescanned, did you? once?
12:27 < adlai> waxwing: sorry, planning for dinner (and leaving now). my infinite loop was very bizarre, i was more pressed to get the coins than investigate.
12:27 < waxwing> adlai: np, speak again later
12:28 < adlai> it "should" be preproducible, i'm still engaging in my heinous practice of keeping the same wallets on multiple computers (although never running multiple bots on the same wallet, because that reduces UX for takers)
12:30  * adlai heads out, ttyl
12:32 -!- akoko [~u@77-46-245-49.dynamic.isp.telekom.rs] has joined #joinmarket
12:42 <+JM-IRCRelay> [jm_loser] no joy after second run. 3rd underway now. What next?
12:44 <+JM-IRCRelay> [jm_loser] hm. "importing 3000 addresses" in secopnd run. 3rd run is now " importing 1800 addresses" and has "too few addresses".
12:44 <+JM-IRCRelay> [jm_loser] maybe there is some progress. monitoring.
12:47 < waxwing> it seems unlikely; if you have max 300 in the index cache then it should need to import less than 3000 total. did you rescan once? you might need to rescan again if you are using Core for the first time with an existing wallet.
12:49 < waxwing> a shame 586 is not ready, it was intended to fix exactly that scenario.
13:06 -!- malte [tVpNoJejMW@alkaid.uberspace.de] has quit [Remote host closed the connection]
13:08 -!- malte [MbqKe9tZ9M@alkaid.uberspace.de] has joined #joinmarket
13:14 <+JM-IRCRelay> [jm_loser] i must have -rescanned 4 or more times already.
13:16 -!- Giszmo [~leo@pc-122-14-46-190.cm.vtr.net] has quit [Quit: Leaving.]
13:26 -!- moli [~molly@unaffiliated/molly] has joined #joinmarket
13:28 -!- molz [~molly@unaffiliated/molly] has quit [Ping timeout: 250 seconds]
13:44 <+JM-IRCRelay> [jm_loser] si i need to move the coins to a new wallet, or is there a fix for me?
15:01 -!- Einherjer [~einherjer@69.64.40.177] has joined #joinmarket
15:31 -!- moli [~molly@unaffiliated/molly] has quit [Ping timeout: 244 seconds]
15:31 -!- molz [~molly@unaffiliated/molly] has joined #joinmarket
17:02 -!- belcher [~user@unaffiliated/belcher] has joined #joinmarket
18:55 -!- fqtw_ [~me@x5d801ff7.dyn.telefonica.de] has joined #joinmarket
18:58 -!- fqtw [~me@x5d804c40.dyn.telefonica.de] has quit [Ping timeout: 240 seconds]
19:37 -!- Giszmo [~leo@pc-122-14-46-190.cm.vtr.net] has joined #joinmarket
19:54 -!- belcher [~user@unaffiliated/belcher] has quit [Quit: Leaving]
20:54 -!- foreigneye [~quassel@2a02:898:162:1::2] has quit [Read error: Connection reset by peer]
21:00 -!- eyeforeigneye [~quassel@2a02:898:162:1::2] has joined #joinmarket
21:09 -!- AdrianG [~User@unaffiliated/amphetamine] has joined #joinmarket
21:23 -!- Giszmo [~leo@pc-122-14-46-190.cm.vtr.net] has quit [Quit: Leaving.]