00:50:50 first time using the new release and when i hit enter in monero-wallet-cli it just prints ^M to the screen instead of what i would expect...anyone have this problem? OSX 00:52:37 seems the issue is zsh, nevermind 04:13:23 ruh roh, saddam's loading wallets up... 04:13:45 so what bitcoin wallets do ppl use these days 04:14:33 ive seen it recommended to not use the core wallet, so .... i guess no one runs bitcoin nodes or connects their wallets to nodes? 04:14:50 i tried the CLI a while back, and woof. 04:15:28 or do everyday users just use phone wallets because ..... 04:30:34 I started with electrum long ago because I was clueless 04:30:48 I am still clueless 04:44:43 i only use core. enabling pruning if needed 04:45:40 was heck of a learning curve doing monero cli but figured it out 04:45:51 i never di 04:46:03 do gui for any wallet 04:47:11 ended up making bash scripts for starting the daemon and rpc since such a bitch remembering all the command line options 06:34:44 hello 09:01:44 hello 09:02:04 sorry for adding some noise on this channel but i have no choice 09:02:45 tx version can not be specified anywhere 09:03:09 tried on basic.h tx_utils but nothing 09:03:24 for some reasons allways tx version is choosed 09:03:35 for some reasons allways tx version one is choosed 09:04:37 hf version has no issues 11:22:27 do you fear Bitcoin's add-ons privacy features may render xmr obsolete? 11:29:15 KrishnaPedo[m], no 11:29:29 why 11:30:14 depends on what privacy features you're actually referring to 11:30:46 and I don't really have time to discuss this at length, several issues with bitcoin's "privacy features" though: 11:31:25 I have a feeling it'll be like zcash. Advertise optional privacy but in the end ensure nobody uses them because convenience. 11:32:07 Many people (and I'm not talking about governments) seem to actively not want privacy in bitcoin. 11:32:31 coinjoins are sybililable, privacy is opt-in and therefore there's no economic incentive (quite the contrary) for you to taint your coins on purpose, coinjoin models don't work on bitcoin high fees scenarios, etc 11:33:20 So the worst that could happen is some optional stuff, people as a whole think bitcoin's private, but in practice it isn't. That'd please a lot of people. 11:33:25 I was thinking about Wasabi wallet and ln 11:33:42 Even if you have amazing opsec, Bitcoin's privacy is a nightmare, even using all the best practices 11:34:03 ok 11:34:03 And then we'll be fucked. But then I suppose when you have a government and you like privacy, you'll always end up fucked anyway. 11:34:12 you have to have an amazing utxo hygiene,all it takes is one mistake 11:34:43 plus, values are always transparent so there could be a way to correlate because of that 11:34:49 are you pessimistic about moneros future moneromooo? 11:37:06 I'm pessimistic on the future, and individual's privacy and liberty 11:37:28 Monero is one of the best tools we have to fight back imo. 11:39:33 it's not perfect, faces a lot of challenges but it has been reassuring to see all the improvements made to make it more resilient over the years 11:56:26 (-(-(--)_-)-) 11:56:42 (-(-(--)_-)-) 12:04:55 KrishnaPedo[m]: please change your nickname 12:50:51 why 12:51:49 please change your underwear 12:52:02 fascism level 10k 12:52:03 wtf 12:52:18 u change your shit nerd geek binaryFate 12:52:37 fuck off jack 12:53:01 no need of this just block 12:53:17 chat chief where are u 12:53:31 guernsey change your nick pleade 12:53:32 I mean yes, but not everyone does it, and he has already been banned before 12:53:40 please 12:54:22 yanmaani change your nick please 12:54:45 change your skin color please 12:54:58 wtf 12:56:20 people like binaryFate are the reason why we need monero 12:57:51 What just happend? 12:58:09 jack showed up again 14:08:35 binaryFate what is a quality discussion, can you please give examples of not-quality discussion. i honestly have no idea what that could even begin to mean 14:15:06 And now we've got the time wasters... 14:16:20 ive been thinking about this for about a week now, since last time i was here. 14:18:10 [2020-06-05 18:42:24] <@binaryFate> A heads up: in the last couple of weeks the quantity of spam / unrelated posts in this channel has been high, and the quality low 14:18:29 2020-06-05 18:44:46] <@binaryFate> Some slightly tighter moderation policy will be warranted 15:04:19 anicow: https://en.wikipedia.org/wiki/Brandolini%27s_law 15:25:25 that nick though was pretty good clue eh 15:26:24 hm interesting 15:56:17 We need Twitter Whale Alerts like BTC has when large amounts are moved. 15:56:19 :P 16:00:19 ill let you know when i make my transfers 16:34:36 is this nick better dear master binaryFate ? 16:35:36 please authorize my to keep it 16:35:39 dear lord 16:36:01 antifa....damn 16:36:32 I'm a capitalist antiFa 16:36:57 what are you Spaceguide 16:37:37 we see... what antifa does worldwide...destroying 16:38:54 destroying fascism 16:38:56 yeah 17:08:59 We need Twitter Whale Alerts like BTC has when large amounts are moved <--- it sorts of exist... https://twitter.com/xmr_whale_alert 17:14:43 lol 18:28:36 hehehe binaryFate ;) 18:32:53 IMHO, if we change the channel topic, Changelly should nowadays get the top spot in the "DON'T USE" list in there. Can't remember Hitbtc.com ever comming up in our subreddit, but always weekly there are posts with somebody complaining about Changelly 18:38:14 I have a question about an error I have in monero-cli-wallet. Is it appropriate to ask it here, or another channel? 18:38:33 appropriate 18:39:11 Trying to transfer some monero. I get this error: 18:39:13 Warning: Some input keys being spent are from blocks that are temporally very close, which can break the anonymity of ring signatures. Make sure this is intentional! 18:39:35 How I fix this so I don't break my anonymity? 18:40:11 you should probably do a sweep_single of one of the outputs 18:40:27 (or of each them, if you wait and do each one slowly over time) 18:41:21 If I were move the outputs to another account, how long would I have to wait in between sweeping the single outputs? 18:43:28 they don't have to go to another account. you can just sweep them to yourself 18:43:41 I don't know what the minimum time interval is 18:44:00 how close in time are those outputs right now? 18:49:53 I think the problematic outputs are ~30 blocks in between each other. 18:51:38 That's probably OK. The check is really tailored for pre-rct outs, where you could get huge split txes, so outputs in neighbouring blocks could be linked. This is much less true nowadays. 18:52:00 I thought the message tested for 10 blocks though. I'll check. 18:53:33 Actually it checks for < 5. This is not the test I was thinking about. 18:53:56 Updated topic based on rbrunner remark, and based on the agreed list of things to avoid on the monero subreddit. 18:54:05 If you set print-rings (IIRC) you will see which outputs it's using. 18:54:23 print-ring-members 18:58:02 hrm, yah know what it is with the solo mining... there's little that tells you that things are working fine 18:58:28 i remember this is why ive bounced back and forth with this over the years. i mean, yeah, with the daemon doing the mining, the daemon will tell you 18:58:50 but you point xmrig at the daemon, and daemon doesn't tell you your mining, so your reliant on the xmrig telling you something. 18:58:59 and all xmrig tells you is that you got a new job from the daemon 18:59:37 IIRC show_hr will tell you about it regularly if that's your thing. 19:00:02 And if you're really, really lucky, monerod *will* tell you regularly too :) 19:00:13 yeah, show_hr always works 19:00:20 Error: Unsuccessful-- rpc_request: NOT MINING 19:00:26 because im pointing xmrig at it 19:00:30 oh right 19:00:33 xmrig is doing the mining 19:00:38 moneromooo If I understand correctly then I can ignore this warning then? 19:00:41 Oh. Soery, misunderstood. 19:00:42 because its like 3% more :/ 19:00:44 well, xmrig will tell you when it submits a block 19:00:51 and most of the time, it won't have a block to submit 19:00:59 so there's really not much to report 19:01:00 What does "mining_status" show in this case ? 19:01:09 Nothing I guess ? 19:01:25 Not currently mining 19:01:29 right 19:01:36 the daemon isn't doing the mining, so it has nothing to tell you 19:01:46 well right. 19:01:55 user13456789: I think you can ignore it, but doing what hyc suggested also works if you're paranoid. 19:02:29 im just saying the UX of solo mining with multiple miners and one daemon is wanting, for me at least. i like to be able to see that things are doing stuff. 19:02:37 might be enough to sweep every-other output. that would widen the gap between them all 19:02:41 Then after at least 20 blocks, or up to a day or two, make the tx you intended, though that will also be subject to heuristics. 19:03:01 You could sweep both outs, then make the tx. Also heuristics, but less strong. 19:03:23 Especially if you don't sweep the first two at roughly the same time. 19:03:48 gingeropolous: you could probably use xmrig-proxy in front of monerod instead 19:03:49 yeah in cases like that i usually just sweep to a new output, let the chain bake a while, and then send 19:03:58 I think I'll sweep all the inputs and give ample (hours) of time between sweeps. 19:04:06 if you have multiple miners running, that's probably the better approach 19:04:30 ah jesus another thing to run 19:04:55 If Linus intended us to run several programs, he would have given a multithreaded OS. 19:05:10 Multi... process. 19:05:26 yeah.. i guess this is a more advanced usecase 19:05:44 most endusers would probably only have 1 PC they're worried abouyt 19:05:55 their 19:05:58 bah 19:08:23 all right i'll learn this proxy thing 19:09:27 man i wanna get the 128 thread epyc 19:14:53 gingeropolous Threadripper does very well, more or less even better 19:47:56 Spaceguide: not quite. not enough memory channels on 3990x (for rx) 20:19:21 @Inge- Threadripper get's up^to 44KH, surpassing Epyc with 4KH. counting in price difference , tr is still better 20:21:53 Someone on Reddit claims they get `mdb_bad_txn` when running monerod without sudo. Is this permission related or corrupted db? 20:29:20 meaning, it works *with* sudo? if so then it's got to be a permission thing 20:30:41 find ~/.bitmonero \! -user $(whoami) -ls 20:30:53 that’s what someone reports, yes 20:31:46 probably ran it as root once by mistake, it created a new lmdb file, and now it's owned by root 20:34:12 ndorf: cd ~/.bitmonero/ && ls -l 20:34:18 should work too to see the owner? 20:34:23 your command does not work for me 21:02:23 selsta: you probably want ls -lR, since the offending files are likely under lmdb... 21:02:29 do you get an error when you run my find command? 21:03:57 if it prints nothing, that means your permissions are fine 21:08:13 ah yes, it prints nothing 21:09:05 ty will ask the user to test this 21:50:48 selsta: it's informational only, anyway. if it prints something, the fix will be `sudo chown -R username:groupname .bitmonero` 21:51:04 it does not print anything for the user but I suspect the problem is somewhere else 21:51:25 indeed, it's probably not ownership then 21:51:34 assuming .bitmonero is the dir it's using 21:57:26 oh, hmm. when he runs it as root, it probably looks in ~root/.bitmonero, not ~regularuser/.bitmonero 21:57:40 so probably there are two blockchains, the former being ok and the latter one being corrupt 21:58:21 yep, that could be the case too