03:09:42 <xmrmatterbridge> <xmrhaelan> moneromoo I think I may have figured out the Qubes daemon/wallet isolation issue
03:09:53 <xmrmatterbridge> <xmrhaelan> Well, a guy on Reddit helped me.
03:10:47 <xmrmatterbridge> <xmrhaelan> Have you updated the monerod.service file to remove the local folder from the line starting with ExecStart?
03:11:25 <xmrmatterbridge> <xmrhaelan> It should read /usr/bin/monerod instead of /usr/local/bin/monerod
03:13:48 <xmrmatterbridge> <xmrhaelan> Also to avoid problems you have to start monerod-ws first and let it fully sync, then start monero-wallet-ws
04:23:03 <ndorf> stop this ride please, i want to get off. Brock Pierce, co-founder of Tether, is officially on my state's ballot for the (US) presidential election
04:27:24 <riceandbeans> ndorf: Could be worse. Could be Biden or Trump.
04:27:49 <ndorf> touché
04:27:56 <ndorf> Kanye West is also on the ballot FWIW
04:28:04 <riceandbeans> Under what party?
04:28:22 <ndorf> both Pierce and West are unaffiliated
04:30:00 <riceandbeans> I'm voting for COVID-19
04:30:26 <riceandbeans> I do have a monero question though...
04:31:04 <ndorf> Monero is also unaffiliated with any political party.
04:31:16 <riceandbeans> I can get hugepages to say that they're supported but not to be enabled, not sure what I'm doing wrong
04:31:23 <ndorf> with xmrig?
04:31:26 <riceandbeans> Yeah
04:31:32 <ndorf> what OS?
04:31:39 <riceandbeans> Take your pick.
04:31:48 <riceandbeans> CentOS 8, Debian 10, FreeBSD 12
04:31:55 <riceandbeans> Same thing on all 3
04:32:05 <ndorf> for Linux, there's an included shell script that will enable both types of huge pages.
04:32:24 <ndorf> https://github.com/xmrig/xmrig/blob/master/scripts/enable_1gb_pages.sh
04:32:28 <ndorf> run that as root before starting xmrig
04:33:05 <ndorf> (Linux supports "regular" huge pages, which are 2MB, as well as 1GB pages ("really huge"). this script enables the optimal mix of both)
04:33:40 <riceandbeans> I mean, I wrap xmrig in a script so I might as well just shove that on top of it
04:33:52 <ndorf> yup
04:34:15 <ndorf> this assumes that xmrig is the only thing that will be using hugepages, you'd want to increase the numbers if that's not the case
04:41:24 <riceandbeans> Still doesn't say enabled, only supported :\
04:42:38 <riceandbeans> forgive me cjs77 user who is no longer here but going on ignore to not drive me nuts with the flapping
04:59:42 <ndorf> riceandbeans: did you run the hugepages script as root? did it print any errors?
05:00:00 <riceandbeans> I had to tweak it, but did it as sudo
05:00:05 <riceandbeans> I'll pastebin
05:00:09 <ndorf> ok
05:00:43 <ndorf> also, what's the output of `sysctl vm.nr_hugepages` and `head /sys/devices/system/node/node*/hugepages/hugepages-1048576kB/nr_hugepages` ?
05:01:17 <riceandbeans> Yeah that path doesn't exist.
05:01:25 <riceandbeans> I had to change it for this
05:01:47 <ndorf> really? that's very strange, what's the kernel version?
05:01:52 <riceandbeans> https://pastebin.com/Evx1p7zs
05:02:09 <riceandbeans> 4.18
05:02:13 <riceandbeans> CentOS 8
05:02:37 <ndorf> so first off, the change you made actually made it worse, hehe
05:02:54 <riceandbeans> That path actually exists, the other did not.
05:02:59 <ndorf> hugepages-2048kB are normal, not 1GB pages. so you're now reducing those even further, from $(nproc) to 3
05:03:22 <ndorf> (IOW it's the same kind of hugepages that are set by the sysctl command at the beginning of the script)
05:03:43 <ndorf> is that the only subdir under /sys/devices/system/node/node0/hugepages/ ?
05:04:42 <riceandbeans> $ find /sys/devices/system/node/node* -maxdepth 0 -type d
05:04:43 <riceandbeans> /sys/devices/system/node/node0
05:04:49 <riceandbeans> There's only one there
05:05:34 <riceandbeans> $ ls /sys/devices/system/node/node0/hugepages/
05:05:34 <riceandbeans> hugepages-2048kB
05:05:38 <riceandbeans> That's all she wrote
05:06:35 <ndorf> hm, i guess that CentOS kernel has them disabled somehow? seems very strange
05:06:40 <ndorf> Debian 10 with kernel 4.19 definitely has it
05:07:25 <ndorf> anyway, you can get by just fine with just the normal huge pages, without the 1GB ones. you will just need a lot more of them
05:08:30 <ndorf> `sudo sysctl -w vm.nr_hugepages=1280` should be enough, i think
05:08:51 <ndorf> for this setup you don't actually need a script, you can just add vm.nr_hugepages=1280 to /etc/sysctl.conf
05:09:37 <ndorf> there is a hashrate penalty as compared to using 1GB pages, but it's very slight, only 1% or so. still much better than no huge pages at all
05:21:29 <riceandbeans> Why would a hashrate penalty be a good thing
05:21:53 <ndorf> it's not a good thing, the good thing is that it's only a slight penalty
05:22:06 <ndorf> and will work immediately on your system
05:22:09 <riceandbeans> I have 32G RAM
05:22:21 <riceandbeans> This box is dedicated to nothing but monero right now
05:23:20 <ndorf> that should definitely be enough. i have no idea as to why your CentOS kernel isn't showing the 1GB knob, if you figure that out then you can have a ~1% higher hashrate.
05:23:38 <ndorf> you said you had a Debian 10 box around also? that file exists there, right?
05:23:52 <riceandbeans> The debian one is different hardware so it's hard to compare
05:24:03 <riceandbeans> But that CentOS 8 box cranks out 630H/s
05:24:16 <riceandbeans> My Debian one cranks out 1100H/s
05:25:42 <ndorf> ah
05:25:53 <ndorf> i forgot about older processors. probably, the CentOS box CPU is too old to support 1GB pages
05:26:33 <ndorf> you can verify that with `fgrep -c pdpe1gb /proc/cpuinfo`, if it returns 0 then your CPU doesn't support it
05:26:57 <riceandbeans> Yeah 0
05:27:40 <ndorf> yeah. so, just enable a lot of normal 2MB hugepages, and it will be almost as good (-1% or so)
05:27:56 <riceandbeans> Like how many
05:27:56 <ndorf> 23:08 < ndorf> `sudo sysctl -w vm.nr_hugepages=1280` should be enough, i think
05:27:59 <ndorf> 23:08 < ndorf> for this setup you don't actually need a script, you can just add
05:28:02 <ndorf>                vm.nr_hugepages=1280 to /etc/sysctl.conf
05:28:06 <riceandbeans> What if I go way higher?
05:28:12 <riceandbeans> like 12800
05:28:16 <ndorf> i think 1280 should be enough, once xmrig starts you'll be able to see how many it is using
05:28:37 <ndorf> you can go higher, but it won't help anything, and that memory won't be available for any other use
05:29:03 <ndorf> you need 1) around 2.3GB worth of huge pages for the randomx dataset, plus 2) one 2MB huge page per thread
05:29:42 <ndorf> xmrig will print status for both things, if they are both 100% hugepages then adding more won't do anything except waste memory
05:30:10 <riceandbeans> Thanks man
05:30:34 <ndorf> np
05:31:18 <riceandbeans> oh no
05:31:23 <riceandbeans> The Debian box returned 0 too
05:31:23 <ndorf> what happened?
05:31:26 <riceandbeans> weak
05:31:28 <ndorf> ah
05:31:39 <ndorf> well like i said, you will still see a massive increase over not using hugepages at all
05:31:48 <ndorf> it's only the last 1% you're missing out on
05:32:05 <riceandbeans> I don't suppose you'd know how to get that going on FreeBSD too
05:34:06 <ndorf> FreeBSD calls them superpages, also set via sysctl somewhere
05:34:16 <ndorf> i've never run xmrig on it, though, so i have no idea if it even supports them
05:34:37 <ndorf> also IIRC FreeBSD does a better job at "transparent" huge pages, so they might not even be necessary to get the best performance. i don't really know for usre
05:37:30 <ndorf> i think you might just need to ensure that sysctl vm.pmap.pg_ps_enabled is set to 1
11:59:27 <Inge-> 2 days in a row with > 16K transactions
12:00:07 <p3rL> only^
12:00:15 <Inge-> first time in XMr history
12:00:24 <p3rL> still in #17
12:11:35 <ErCiccione[m]> p3rL: 17 for market capitaliazion i guess. Which is a meaningless metric. if you want to see if a currency it's used you have to look at the amount of daily transactions on the network
12:17:57 <Inge-> Then BSV must be really good, averaging some 500K+ tx/day
12:21:26 <ErCiccione[m]> Ok, let's add that said currency shouldn't be lead by a scammer :P
12:21:44 <moneromooo> Oh my. That strikes a lot of them down does it not.
12:25:08 <ErCiccione[m]> Nah, all thousands of coins + ICOs + ERC tokens are all legit and equally contribute to the space
12:37:00 <alexanarcho[m]> hey everybody, we have our CCS Propsal for tipxmr.live on GitLab - check it out and give a thumbs up if you like it https://ccs.getmonero.org/ideas/
13:03:21 <Inge-> what is happening with LTC transactions going up and up over the past 8 days
13:14:38 <nioc> alexanarcho[m]: I see no info on that think.  You upvoted your own proposal o_O
13:15:12 <nioc> #monero-community is a good channel to post info about CCS proposals
13:22:40 <Inge-> ndorf: never had issue with 1250 per numa node.
14:29:05 <d4ndo[m]> Inge-: A new RPG released based on the Litecoin blockchain. See https://www.litebringer.com/
14:31:10 <d4ndo[m]> You can buy any kind of digital item on litebringer using Litecoin. e.G. swords, gloves etc
14:31:36 <Inge-> heh
14:32:21 <Inge-> 40K -> 150K transactions due to a game launch?
14:32:24 <alexanarcho[m]> <nioc "alexanarcho: I see no info on th"> well, i m a big fan of my propsal
14:33:48 <d4ndo[m]> Compare that to Cryptokitties and you will believe.
14:35:46 <d4ndo[m]> I think they do store the history of every move you make on Litecoin. But not sure.
14:36:17 <Inge-> sounds like a GREAT idea
15:39:56 <xmrmatterbridge> <mxudi> hello, is this the right monero channel to ask questions?
15:40:15 <moneromooo> Yes.
15:43:44 <xmrmatterbridge> <mxudi> I have the following problem: I need my payment details from an earlier payment. but I only have my wallet address at hand, and the date and approximate amount of the payment. can I have a chance to find the payment details on a ledger somewhere?
15:46:01 * moneromooo does not know much about ledger and how to restore a ledger wallet, passes and wits for someone who knows
15:49:48 <endor00[m]> I don't think he's talking about the Ledger wallet
15:51:26 <Mochi101> mxudi, no
15:51:46 <xmrmatterbridge> <mxudi> I am just asking about whether any record (public or other) exists, where I can find the referred payment
15:51:58 <Mochi101> Monero is private, those details are not available to the public.
15:52:29 <Mochi101> If you have access to your wallet, you can get some details of that transaction though.
15:52:46 <endor00[m]> Unless you have that wallet's private viewkey or that transaction's private key, you won't see anything
15:53:45 <endor00[m]> And if you're trying to uncover somebody else's tx details out of those info, you're out of luck ;)
15:53:58 <Mochi101> Praise Monero!
15:58:14 <xmrmatterbridge> <mxudi> the keys are gone. I dont have them anymore
15:58:57 <endor00[m]> Did you write down your seed?
15:59:13 <xmrmatterbridge> <mxudi> I am trying to uncover those of my own
15:59:45 <endor00[m]> Rip in pepperoni
16:01:03 <endor00[m]> Today is the day you've learned that you must always write down your seed and store it somewhere safe, because you never know when you'll need it
16:04:08 <xmrmatterbridge> <mxudi> today I feel very mnemonic
16:06:22 <moneromooo> That's what she seed.
16:06:35 <xmrmatterbridge> <mxudi> does the big bulky blockchain not provide any access with the details that I have mentioned above?
16:06:57 <moneromooo> It only does if you have the keys.
16:07:19 <xmrmatterbridge> <mxudi> (like my wallet address)
16:07:23 <moneromooo> I mean, you can put your money in a small purse or a big vault. Is the big vault easier to get stuff from ?
16:07:42 <kinghat[m]> endor00: from supportxmr back in the day?
16:08:36 <endor00[m]> Yeah man
16:09:35 <kinghat[m]> why did you add the 00?
16:11:18 <xmrmatterbridge> <mxudi> I see
16:11:55 <xmrmatterbridge> <mxudi> thank you, I am at least convinced there is no need to search for it
17:05:14 <endor00[m]> <kinghat[m] "why did you add the 00?"> The name without the 00 was already taken iirc
17:06:35 <kinghat[m]> are you done with school and shit now? engineering, correct?
17:16:48 <endor00[m]> Still working on it
17:18:18 <d4ndo[m]> There are only 10 types of people in the world: those who understand binary, and those who don't.
18:28:38 <upsdnz> QUESTION: what is the correct way to check if my blockchain data is pruned ?.
18:29:37 <moneromooo> check_blockchain_pruning
18:30:05 <moneromooo> It will also check some consistency so is not instant.
18:30:29 <upsdnz> k
22:28:55 <hondza> Hello. Can the change address for the transfer command be specified, or is it always the subaddress with index=0 in the current account?
22:31:28 <moneromooo> Always automatic.
22:44:46 <hondza> Doesn't that leak anything, though? Or is the change address indistinguishable from a second output (which I guess is a stealth address)?
22:45:53 <moneromooo> Well, the recipient can be... 95% certain I guess... that the other ouput is change.
22:46:08 <moneromooo> Define leak, and to whom.
22:47:34 <ndorf> it is a stealth address, and only you (the recipient of the change output) can decrypt the amount
22:54:49 <hondza> So the recipient of the first output doesn't learn anything? For instance if I repeated the exact same transfer command, could the recipient infer that it is from me again? Or will those stealth addresses be different?
22:55:46 <moneromooo> Oh I see your concern. Monero is a private cryptocurrency. Addresses do not appear on the chain. All the recipient sees are one time output public keys.
22:56:18 <moneromooo> So if you send several times to the same address, it doesn't show up. They're different outputs, each which its own public key.
23:09:14 <hondza> Thanks. One more question: the organization of wallet -> accounts -> addresses is mainly so that they could be derived deterministically from the seed, correct?  I mean if I somehow knew the private key of a subaddress, this key alone is enough to operate with the unspent outputs on that address, yes?
23:25:56 <xmrmatterbridge> <xmrhaelan> Is there a tutorial for noobs who want to make a pull request for getmonero website?
23:39:39 <moneromooo> Yes, accounts/addresses can be restored deterministially from the seed.
23:39:53 <moneromooo> I'm not sure about the second question.