--- Day changed Wed Feb 08 2017
00:05 -!- wump is now known as wumpus     
00:22 -!- HostFat__ [~HostFat@host234-246-dynamic.16-87-r.retail.telecomitalia.it] has joined #joinmarket     
00:24 -!- dserrano5 [~9dsd9s9da@dserrano5.es] has quit [Ping timeout: 260 seconds]     
00:25 -!- HostFat_ [~HostFat@host234-246-dynamic.16-87-r.retail.telecomitalia.it] has quit [Ping timeout: 256 seconds]     
00:26 -!- dserrano5 [~9dsd9s9da@dserrano5.es] has joined #joinmarket     
01:40 -!- espen____ [~espen@160.121.9.46.customer.cdi.no] has joined #joinmarket     
01:46 -!- Netsplit *.net <-> *.split quits: roasbeef, espen___, waxwing
01:46 -!- Netsplit over, joins: roasbeef
01:52 -!- waxwing [~waxwing@185.65.132.137] has joined #joinmarket     
04:31 < waxwing> <smuggle_r> I cant free the channel for you. we dont do that here
04:37 < waxwing> assuming no one else is going to do it, i'll fix 707 today and plan to merge it quickly after testing. because if we can't boot the people posting thousands of orders, we at least need to prevent sendpayment crashing.
04:37 < waxwing> (i guess the bug is not universal though; had a few txs today)
04:51 < waxwing> since that agora channel is now 'adrift' without a topic, i guess we'll have to deprecate it. options are another channel on that server, or a new optional server.
04:51 < waxwing> whatever change we do, it'll take ages for people to update though.
04:52 < waxwing> i was told the channel was lost because i didn't "op" on it for > 21 days and it expired.
05:41 -!- cluelessperson [~cluelessp@unaffiliated/cluelessperson] has quit [Ping timeout: 276 seconds]     
06:03 -!- Giszmo [~leo@ip-176-233.219.201.nextelmovil.cl] has joined #joinmarket     
06:46 -!- lnostdal [~lnostdal@62.90-149-73.nextgentel.com] has quit [Ping timeout: 240 seconds]     
07:27 < GithubBot5678> [joinmarket] AdamISZ opened pull request #708: lock db access from taker scripts to avoid cursor exceptions (develop...dblock_txcreate) https://git.io/vD0q2
08:31 -!- mrkent [~textual@unaffiliated/mrkent] has joined #joinmarket     
08:44 -!- Guest86034 is now known as mr_burdell     
08:44 -!- mr_burdell is now known as Guest56969     
08:48 -!- Guest56969 [~mr_burdel@bounce.cryptolabs.net] has quit [Quit: ZNC - 1.6.0 - http://znc.in]     
08:48 -!- mr_burdell_ [~mr_burdel@bounce.cryptolabs.net] has joined #joinmarket     
08:48 -!- mr_burdell_ is now known as mr_burdell__     
08:56 -!- mrkent [~textual@unaffiliated/mrkent] has quit [Ping timeout: 245 seconds]     
09:07 -!- Giszmo [~leo@ip-176-233.219.201.nextelmovil.cl] has quit [Remote host closed the connection]     
09:45 -!- mr_burdell__ is now known as mr_burdell     
09:45 -!- mr_burdell is now known as Guest74039     
09:53 -!- juscamarena [~justin@47.148.176.74] has quit [Ping timeout: 245 seconds]     
10:00 -!- juscamarena [~justin@47.148.176.74] has joined #joinmarket     
10:03 < waxwing> i'm experimenting with kicking some of the bots that are posting tons of orders (but only on cgan, see the above)
10:03 < waxwing> restricted to those posting ovelapping orders and huge numbers (like 100 and so on)
10:04 < waxwing> if you happen to run a custom bot with a few tens of orders and respect the rule of not using overlapping amounts, no worries
10:04 < waxwing> (and anyway you can use agora atm, heh)
10:23 < waxwing> kicked a bunch, see https://joinmarket.me/ob/ , but this is deliberately only reading from cgan. i presume they'll all come back fairly soon anyway.
10:24 < waxwing> reduced orders from ~3600 to ~ 400. one thing i noticed, they are probably all using the same "algo", probably real coins, just massively duplicating same amount ranges.
10:24 < waxwing> completely pointless, just extremely annoying.
10:35 -!- weex [sid152634@fsf/member/weex] has quit [Remote host closed the connection]     
10:35 -!- iinaj [sid110431@gateway/web/irccloud.com/x-exrhtkzexxeiextw] has quit [Remote host closed the connection]     
10:45 -!- btcdrak [uid165369@gateway/web/irccloud.com/x-nopucefuxoylojer] has quit [Remote host closed the connection]     
10:46 < JM-IRCRelay> [AlexCato] nice. Thanks.
11:34 -!- Buckyy [buckowski@gateway/shell/elitebnc/x-xtpqaxwmidydxseb] has quit [Excess Flood]     
11:34 -!- iinaj [sid110431@gateway/web/irccloud.com/x-kbevkunkgyapvhqk] has joined #joinmarket     
11:35 < JM-IRCRelay> [mrnice] Am still busy "finding" the missing coins. Recently had to reload the bitcoin blockchain in core, then did a -rescan.
11:36 < JM-IRCRelay> [mrnice] Been testing  with gap sizes of up to 10000.
11:37 < JM-IRCRelay> [mrnice] have a basic understanding of a HD wallet.would like to
11:37 -!- buckowski [buckowski@gateway/shell/elitebnc/x-xedihllgdllhknpr] has joined #joinmarket     
11:37 < JM-IRCRelay> [mrnice] better understand the numbers in the wallet file?
11:38 < JM-IRCRelay> [mrnice] Anybody care to explain these?
11:44 < waxwing> mrnice: this gives an overview of the format: https://github.com/JoinMarket-Org/JoinMarket-Docs/blob/master/High-level-design.md#wallet-persistence
11:45 < JM-IRCRelay> [mrnice] waxwing: ah, you are here. Thanks for the pointer. Will have a read.
11:50 < JM-IRCRelay> [mrnice] waxwing: so there are actually 10 branches in total by default. 5 levels and each level with a external and internal branch?
11:51 < waxwing> yep     
11:51 < JM-IRCRelay> [mrnice] copied
11:51 < JM-IRCRelay> [mrnice] ok, that helps. thanks.
11:52 -!- buckowski [buckowski@gateway/shell/elitebnc/x-xedihllgdllhknpr] has quit [Excess Flood]     
11:55 -!- buckowski [buckowski@gateway/shell/elitebnc/x-bjpjqipbfhsiqoxf] has joined #joinmarket     
11:55 < JM-IRCRelay> [mrnice] I really appreciate the JM-IRCRelay for the fact that it is possible to use the jm channel from tor through CG. It would be nice if it would also show the list of nics from the proxied irc server for the benefit of the nics on the other side. How could this be done?
11:58 < JM-IRCRelay> [mrnice] waxwing: just be sure - The eg wallet-tool.py script will scan up to those addresses during say a history query?
11:58 < belcher> the bot would need a new feature added where you pm it and it sends you the names or something
11:59 < JM-IRCRelay> [mrnice] waxwing: Assuming I increase the number that is...
12:00 < JM-IRCRelay> [mrnice] belcher: simple solution. right. You wrote the bot? The source is available?
12:00 < waxwing> mrnice, should be up to those indexes + gap limit, iirc
12:01 < belcher> i didnt write it, just got it from github somewhere
12:02 < JM-IRCRelay> [mrnice] waxwing: so to be sure, I could arbitrarily increase the numbers to search a larger space.
12:02 < waxwing> mrnice, btw, you could try PR #706 when doing large re-syncs, may speed it up. most people should just use --fast
12:02 < JM-IRCRelay> [mrnice] then a simple history query would be enough, or would I need to -rescan core first?
12:02 < waxwing> it was an obvious speedup we somehow missed.
12:03 < waxwing> i believe you only need to -rescan once. i always forget the logic of it though.
12:04 < JM-IRCRelay> [mrnice] I am also blurry on the logic - still dont fully understand the read only addresses (and have not invested much effort yet either)
12:04 -!- btcdrak [uid165369@gateway/web/irccloud.com/x-pjwuluufeiymcfea] has joined #joinmarket     
12:05 < JM-IRCRelay> [mrnice] waxwing: but the wallet script informs core of which addresses to "watch" right?
12:05 < waxwing> yes, see 'importaddress'
12:05 < JM-IRCRelay> [mrnice] yes, i read about that.
12:05 -!- weex [sid152634@gateway/web/irccloud.com/x-bklnmofgyqmkzcnz] has joined #joinmarket     
12:06 -!- weex is now known as Guest40251     
12:06 < JM-IRCRelay> [mrnice] can the wallet script query arbitrary addresses or do they first require an "importaddress" command to core?
12:09 < belcher> it first requires importaddress
12:10 < JM-IRCRelay> [mrnice] the script does this prior to query I guess, else it would not bring back any meaningful result, yes?
12:12 < JM-IRCRelay> [mrnice] I guess my question really is if I can switch JM wallets about arbitrarily, or if they first need to be -rescan ned.
12:12 < belcher> you can switch wallets yes
12:13 < belcher> -rescan is only needed if restoring a wallet from backup, in all other cases importaddress is used before any transactions happen involving it so -rescan isnt needed
12:19 < JM-IRCRelay> [mrnice] belcher: on restoring - because core has not yet got the previously used addresses imported?
12:20 < JM-IRCRelay> [mrnice] not quite clear on this...
12:22 < belcher> transaction information comes from the blockchain, which contains transaction, not addresses
12:22 < belcher> the best way to use importaddress is to import addresses you know to be new and unused, then as new blocks come in core will watch if they have transactions involving your imported address
12:23 < belcher> but if you import addresses that are used, which happens if you're restoring from backup, you must do -rescan where core scans the entire blockchain for transactions involving your addresses
12:25 < JM-IRCRelay> [mrnice] belcher: ok, got it. But why would core "forget" these used addresses from previously? Are you assuming that the wallet was never used on this node?
12:25 < belcher> for example if you're restoring a wallet backup from your mnemonic seed
12:26 < JM-IRCRelay> [mrnice] belcher: yes, but the same wallet was used previously on the same node. Would a -rescan still be necessary?
12:26 < belcher> depends what you mean by used previously, in general no, you only need rescan when restoring from backup
12:30 < JM-IRCRelay> [mrnice] belcher: scenario - JM in use with core and a wallet A. wallet inadvertedly deleted and restored from seed. wallet-tool runs with -g option to find the right index numbers. This is where the script tells you to -rescan if recovering from seed. Is this really required, since core should still have been watching this same addresses previously. Yes?
12:34 < belcher> when you say the wallet was deleted, what exactly was deleted? the wallet.json file?
12:35 < JM-IRCRelay> [mrnice] yes, the current wallet.json file
12:35 < JM-IRCRelay> [mrnice] (remember the full fs problem bug)
12:37 < waxwing> yeah when it says 'if recovering from seed' it means on a new Core instance. if you recover from seed on the same Core instance, nothing has changed as far as Core is concerned (same wallet name, same txs and addresses)
12:46 < belcher> hmm so the wiki could be edited then
12:52 < belcher> waxwing you've misunderstood how locks work :p
12:53 < belcher> ill try to find the old online book where i learned about concurrency if you like
12:53 < belcher> you have to use put locks everywhere where the resource is accessed
12:54 < waxwing> well for sure in general, but are you sure it's necessary in this case?
12:54 < belcher> wait just reading the code, self.dblock already exists
12:54 < belcher> looks like we've been using it wrong all this time
12:56 < belcher> i dont know if using locks will fix that problem, but if you use locks then they dont do anything unless you put them around every resource access place
12:57 < belcher> just doing some grepping..
12:57 < waxwing> it depends when the access occurs, sure in general the correct thing is to do what you just said
12:58 < belcher> yes, i suppose if the access happens its certain that theres only one thread existing then you dont have to put a lock
12:59 < belcher> okay so looks like dblock is already used everywhere in taker.py except for one place
12:59 < belcher> and used nowhere in sendpayment.py tumbler.py etc
13:00 < waxwing> yes, i put it there to deal with having orderbook updates simultaneously from 2 message channels.
13:00 < waxwing> global lock makes more sense, you won't find me disagreeing there.
13:01 < waxwing> it had never cropped up presumably because we never had ridiculously huge orderbooks
13:01 < waxwing> well to be fair to myself, i already explicitly said the above in the PR comments :)
13:01 < belcher> yep, concurrency bugs are the worst
13:01 < waxwing> not sure it's fair to say i don't understand it...
13:01 < belcher> yep, i take that back sorry, didnt realize dblock already existed
13:05 < belcher> its my fault then actually, i think i coded the dblock thing and didnt add it to the sendpayment, tumbler, etc scripts
13:06 < waxwing> no, that was me. when adding multi-MC. i don't think there was any locking before on the db.
13:06 < waxwing> i mean, bright side is, you can only get this cursor exception. not like it's going to be corrupted or something.
13:06 < waxwing> well, i think it was me .. got me doubting now.
13:07 < waxwing> but for sure i remember adding dblock in taker
13:07 < belcher> git blame will tell us, if we care enough to find out
13:07 < belcher> iv been thinking of this idea https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2016-May/012636.html and i think i know an attack on it
13:08 < belcher> writing an email to explain, it can be mitigated but it has to be known about
13:08 < belcher> its a good idea, committed bloom filters
15:10 -!- owowo [~ovovo@unaffiliated/ovovo] has quit [Ping timeout: 256 seconds]     
15:25 -!- owowo [~ovovo@31.7.59.226] has joined #joinmarket     
15:25 -!- owowo [~ovovo@31.7.59.226] has quit [Changing host]     
15:25 -!- owowo [~ovovo@unaffiliated/ovovo] has joined #joinmarket     
15:39 < puddinpop> I've been getting "irc ping timed out" on cgan a lot.  It seems to happen around every 6 minutes.
15:41 < puddinpop> When I eventually kill the maker, I in the debug output I see a lot of "joinmarket.irc.IRCMessageChannel object at " and each has a different IRC nick "J5XXXXXXXXXXXXXX" that I assume was used for a cgan connection.  Why aren't these objects getting cleaned up?
15:43 < puddinpop> Also connected to agore and don't see the same frequent timeouts there.
15:43 < puddinpop> *agora
15:51 -!- instagibbs [~instagibb@pool-100-15-114-3.washdc.fios.verizon.net] has quit [Ping timeout: 240 seconds]     
15:52 -!- Giszmo [~leo@ip-64-233.219.201.nextelmovil.cl] has joined #joinmarket     
15:53 < belcher> just checked my yieldgen and i dont get anything like that puddinpop 
15:53 < belcher> dunno what to suggest
15:55 -!- instagibbs [~instagibb@pool-100-15-114-3.washdc.fios.verizon.net] has joined #joinmarket     
16:08 -!- Guest40251 is now known as weex     
16:08 -!- weex [sid152634@gateway/web/irccloud.com/x-bklnmofgyqmkzcnz] has quit [Changing host]     
16:08 -!- weex [sid152634@fsf/member/weex] has joined #joinmarket     
16:40 -!- Giszmo [~leo@ip-64-233.219.201.nextelmovil.cl] has quit [Quit: Leaving.]     
16:46 -!- goregrind [~goregrind@unaffiliated/goregrind] has joined #joinmarket     
16:48 -!- cluelessperson [~cluelessp@unaffiliated/cluelessperson] has joined #joinmarket     
16:49 -!- goregrin1 [~goregrind@unaffiliated/goregrind] has quit [Ping timeout: 258 seconds]     
18:21 -!- Giszmo [~leo@ip-141-233.219.201.nextelmovil.cl] has joined #joinmarket     
18:54 -!- HostFat_ [~HostFat@host79-9-dynamic.25-79-r.retail.telecomitalia.it] has joined #joinmarket     
18:58 -!- HostFat__ [~HostFat@host234-246-dynamic.16-87-r.retail.telecomitalia.it] has quit [Ping timeout: 252 seconds]     
18:58 -!- instagibbs [~instagibb@pool-100-15-114-3.washdc.fios.verizon.net] has quit [Ping timeout: 252 seconds]     
19:00 -!- core [~core@unaffiliated/core] has quit [Ping timeout: 252 seconds]     
19:05 -!- instagibbs [~instagibb@pool-100-15-114-3.washdc.fios.verizon.net] has joined #joinmarket     
19:05 -!- core [~core@unaffiliated/core] has joined #joinmarket     
19:09 -!- xcvvcx [53e42f33@gateway/web/freenode/ip.83.228.47.51] has quit [Ping timeout: 260 seconds]     
19:24 -!- Giszmo [~leo@ip-141-233.219.201.nextelmovil.cl] has quit [Remote host closed the connection]     
20:00 -!- coins123 [~coins123@unaffiliated/coins123] has quit [Read error: Connection reset by peer]     
20:00 -!- coins123 [~coins123@unaffiliated/coins123] has joined #joinmarket     
20:07 -!- coins123 [~coins123@unaffiliated/coins123] has quit [Ping timeout: 276 seconds]     
20:13 -!- coins123 [~coins123@ip-244-225.sn1.clouditalia.com] has joined #joinmarket     
20:13 -!- coins123 [~coins123@ip-244-225.sn1.clouditalia.com] has quit [Changing host]     
20:13 -!- coins123 [~coins123@unaffiliated/coins123] has joined #joinmarket     
20:18 -!- coins123 [~coins123@unaffiliated/coins123] has quit [Ping timeout: 240 seconds]     
22:16 -!- coins123 [~coins123@unaffiliated/coins123] has joined #joinmarket     
22:17 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Ping timeout: 276 seconds]     
22:21 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket     
22:29 -!- iinaj_ [sid110431@gateway/web/irccloud.com/x-btrtmfbohsuhlsma] has joined #joinmarket     
22:33 -!- iinaj [sid110431@gateway/web/irccloud.com/x-kbevkunkgyapvhqk] has quit [Ping timeout: 240 seconds]     
22:33 -!- waxwing [~waxwing@185.65.132.137] has quit [Ping timeout: 240 seconds]     
22:33 -!- windsok [~windsok@45.63.59.8] has quit [Ping timeout: 240 seconds]     
22:33 -!- windsok [~windsok@45.63.59.8] has joined #joinmarket     
22:33 -!- iinaj_ is now known as iinaj     
22:33 -!- waxwing [~waxwing@185.65.132.137] has joined #joinmarket     
22:40 -!- coins123 [~coins123@unaffiliated/coins123] has quit [Ping timeout: 258 seconds]     
22:47 -!- coins123 [~coins123@unaffiliated/coins123] has joined #joinmarket     
22:50 -!- coins123 [~coins123@unaffiliated/coins123] has quit [Read error: Connection reset by peer]     
22:51 -!- coins123 [~coins123@ip-244-225.sn1.clouditalia.com] has joined #joinmarket     
22:51 -!- coins123 [~coins123@ip-244-225.sn1.clouditalia.com] has quit [Changing host]     
22:51 -!- coins123 [~coins123@unaffiliated/coins123] has joined #joinmarket