01:12:20 <Mochi101> moneromooo, you were the one who did the work for pruning?
06:26:52 <moneromooo> Mochi101: yes
06:28:34 <Mochi101> moneromooo, any comments on maybe a type of blockchain pruning that would make it so some nodes would only keep the last 100k or so blocks of the chain?
06:29:36 <Mochi101> Or say, 300k for the last year of txs
06:30:31 <moneromooo> Blocks can be dumped in theory. They're not the heaviest thing though. But yes, you can dump more if you accept drawbacks.
06:30:55 <Mochi101> What's the heaviest thing?
06:31:05 <moneromooo> Still tx data AFAIK.
06:31:33 <moneromooo> Output amounts also heavy IIRC. 
06:31:58 <Mochi101> Sounds like I misunderstand how the data is stored.
06:33:45 <Mochi101> I thought by dumping blocks you'd be dumping all the tx data that's inside these blocks as well.
06:34:02 <moneromooo> https://paste.debian.net/hidden/534459e5/
06:34:21 <moneromooo> Well, if you dump tx data you save a lot, yes.
06:34:37 <moneromooo> But you becone mostly useless to other peers.
06:34:59 <Mochi101> That's understood.
06:34:59 <moneromooo> It's supposed to be a P2P network, not a baby2mother system.
06:35:43 <Mochi101> You think that a large majority would choose to run this "micro" node?
06:36:22 <moneromooo> Yes. Some would be people who'd not run a node otherwise though, so it'd be a win for those people.
06:36:38 <moneromooo> Like pools. Easier, so people mine in a pool.
06:36:54 <Mochi101> and when you say "mostly" useless... what's the part that doesn't fall into the mostly section?
06:36:58 <moneromooo> Actually, I take back the "yes". Still easier to use a stranger's node.
06:37:16 <moneromooo> I don't know, I'd have to think about it first.
06:37:48 <Mochi101> Well, I'm thinking more for stores, or services that mostly just need to verify txs.
06:38:18 <moneromooo> Pretty much everyone just needs to verify txes.
06:38:32 <Mochi101> A VPS that is capable of running a pruned node costs about $20 per month. I spend $10 on my cellular per month.
06:39:23 <Mochi101> yes, that's true hey. :)
06:39:34 <Mochi101> I guess that's what the whole network is about.
06:40:34 <moneromooo> I had made that file I pasted because I had started thinking about exactly that in fact. Every once in a while, I start warming up to the idea again, then drop it :)
06:41:58 <Mochi101> I know there are a lot of... drawbacks to it.
06:42:38 <Mochi101> But there are a lot of people out there that run full nodes and even that mine at a loss.
06:45:59 <Mochi101> It was just a thought.
06:49:06 <Mochi101> Thanks for entertaining me through the thought :)
06:56:04 <Mochi101> Probably better just the way it is moneromooo.
06:56:59 <Mochi101> moneromooo, what do you think of archiving the whole /bitmonero directory and making it available on getmonero.org with a verification hash?
06:58:00 <Mochi101> I'd appreciate a comment on it (good or bad) here: https://github.com/monero-project/monero-site/issues/2027
06:59:09 <Mochi101> This would allow people to download the archive, verify it and drop it in. Saving massive amounts of sync time for those who need/want it.
06:59:36 <Mochi101> Could make a full and pruned archive available too.
06:59:54 <Mochi101> Do different nodes prune differently?
07:00:06 <moneromooo> Same thing again. Giving people an easy way to do the wrong thing will cause more people to do the wrong thing.
07:00:22 <moneromooo> Same as pools. Same as a smaller pruned DB tha'ts useless to others.
07:00:44 <moneromooo> It sounds like a good idea until you realize it's good for the person and bad for the network.
07:01:01 <Mochi101> I know that it's bad for the network.
07:01:05 <Mochi101> I really do.
07:01:55 <Mochi101> Ever new node should start from block 0 and verify the whole chain.
07:01:57 <moneromooo> If you're going to have a ready made DB for download, why not remove tx checks in monerod, and just have the core team sign txes ? You get the same amount of trust needed.
07:02:11 <moneromooo> And a lot faster monerod.
07:02:33 <moneromooo> If you don't verify the chain, then why use a blockchain in the first place ?
07:03:20 <sech1> those are some deep philosophical questions
07:03:25 <Mochi101> yes
07:03:41 <moneromooo> This kind of trust is what a lot of people already give anyway when they download binaries from getmonero.org (as opposed to via monero-update).
07:03:58 <Mochi101> Really, anybody could just archive /bitmonero and make it available as a torrent anyways.
07:04:01 <sech1> but they verify hashes, right? right?
07:04:04 <Mochi101> Doesn't have to come from core.
07:04:23 <moneromooo> Yes, but the hashes are made by the same party that'd hash a ready made data.mdb.
07:04:39 <moneromooo> So the amount of trust would not change technically.
07:05:09 <moneromooo> monero-update uses the deterministic build and chceks sigs, but the GUI ended up using the core team signed binaries. Shrug.
07:06:01 <moneromooo> I dunno, maybe it's worth it for now.
07:06:55 <moneromooo> Initial sync is a fairly large barrier and since most people already trust the core team... But if/when we switch to better (ie, checking gitian sigs), then that'd have to go I guess.
07:07:23 <moneromooo> Or we'd end up with, again, making it easier to do the wrong thing.
07:07:40 <Mochi101> yeah
07:07:44 <moneromooo> Actually... 
07:08:34 <moneromooo> LMDB syncing can be bit exact if you're lucky. Maybe a deterministic sync mode could help here. You'd get several people to sync to block N (with N being current height minus... 5k ?), then sign it.
07:08:51 <moneromooo> Then you get what you want, with somewhat better trust.
07:09:04 <Mochi101> ooooh
07:09:16 <Mochi101> oh that's a good idea
07:09:21 <moneromooo> I know that when I tested by pruning changes a few days ago, the pruned DB was bit exact if the same seed was picked.
07:09:47 <Mochi101> So different nodes prune differently?
07:09:52 <moneromooo> I'm not sure how often it'd end up bit exact when syncing from zero without reorgs.
07:10:25 <moneromooo> The same way, but they randomly pick which eighth of the prunable data to keep.
07:10:44 <Mochi101> I see
07:11:27 <Mochi101> deterministic sync node sounds really interesting
07:11:55 <moneromooo> I am speculating it can be done easily enough. I do not know for a fact.
07:12:06 <Mochi101> yeah
07:12:11 <moneromooo> You'd need to add a "max-height" param to bitmonerod.
07:12:29 <moneromooo> And I think you'd need to enforce 20 block spans.
07:12:57 <moneromooo> Testing that is a bitch though. 10 hours per test :)
07:13:08 <Mochi101> 38 for me
07:13:10 <Mochi101> ;)
07:13:58 <Mochi101> Almost done my 3rd sync here.. this one with the patch and --db-sync-mode=safe will be the longes.
07:14:21 <Mochi101> I'm guessing 40-42 hours
07:14:26 <moneromooo> I guess a first step is asking hyc whether it's obviously not going to work because too much change of variance in db writes.
07:14:30 <moneromooo> hyc: ^ ? :)
07:15:31 <moneromooo> hyc: in a nutshell, if you ensure several monerod sync from scratch to the same height, assuming good connectivity, is it likely the end DBs will be bit identical ?
07:15:34 <Mochi101> but we'd be testing it on testnet or stagenet right.
07:15:52 <Mochi101> So no 10 hour syncs until later on in testing.
07:15:55 <moneromooo> Now that's an idea I wish I had had.
08:18:32 <moneromooo> Who here manages the matrix relay ?
11:13:01 <nioc> this room has no matrix relay, the matrix monero room is separate 
11:13:40 <moneromooo> It needs one, as people come to -dev with "help my find my monero" questions and we can't point them to #monero if it's not relayed :/
11:13:45 <nioc> I believe due to size limitations ir needs something special, maybe from libera
11:14:25 <nioc> much of that help happens now in -community 
11:14:41 <moneromooo> That is relayed to matrix ?
11:14:45 <nioc> yes
11:14:53 <moneromooo> Nice! Thanks.
11:14:58 <nioc> meow
11:23:13 <merope> For some reason, the bridge for this room never got fixed after the move to libera 
11:23:31 <merope> So #monero is two separate rooms on irc and matrix
11:38:25 <moneromooo> I dunno, people who come in dev typically find #monero-dev but not #monero, so it looks like there may not even be a matrix-only #monero channel.
12:11:50 <merope> there is (I'm in it)
12:12:16 <merope> I guess it's simply due to the association "wallet issues => ask the developers directly"
14:43:23 <Mochi101> hyc, moneromooo, 42 hours 48 mins with the --db-sync-mode=safe flag
14:44:30 <moneromooo> vs 38 ? That's not bad at all, I thought safe would be much slower.
14:44:38 <Mochi101> yes
14:45:18 <moneromooo> We need settings for safer, safest and windows-proof.
14:48:53 <Mochi101> :D
14:57:58 <Mochi101> .usd
14:57:58 <Wallet> Mochi101: DURR.. 1 USD = 1 USD genious! [sic]
16:30:23 <hyc> yeah, 10% diff, that's pretty reasonable. we could just make that the windows default.
16:33:22 <RubyQuby> does anyone have any hashrates for XMR using a GTX3090Ti?
16:37:08 <hyc> nobody mines on GPUs
16:37:40 <hyc> if you want to use that GPU you should be mining some other coin than XMR
16:39:01 <RubyQuby> I was given one for free, and been wondering if mining is still a thing
16:39:09 <RubyQuby> what coin?
16:43:48 <moneromooo> Mining on monero is definitely a thing, but on CPU.
16:46:26 <RubyQuby> moneromooo - oh ok, thanks
16:50:32 <tevador> Here are some older nvidia GPU benchmarks: https://github.com/SChernykh/RandomX_CUDA#randomx-cuda-implementation
16:51:24 <tevador> High-end GPUs perform similar to a laptop CPU, so mining is not really worth it.
17:21:20 <version2> wtf up with monero today?
17:21:48 <version2> like earlier it needed 50 confirmations until balance finally upadted after 2-3 hrs and then on this site
17:22:22 <version2> guarda.co wallet i exchanged some other coins to monero and now (even though the exchange transaction is showing up everywhere) both wallets the monero and the other crypto is empty now wtf is this
17:22:59 <version2> i need these coins now, i actually just clicked on exchange accidentally, i would have used another exchange service oh man
17:26:07 <Mochi101> version2, I think I read somewhere that guarda was having some problems with the network upgrade.
17:29:46 <version2> since when?
17:29:48 <version2> when did u read that?
17:29:50 <version2> and when was the upgrade?
17:29:56 <version2> @ Mochi101 
17:30:05 <version2> does that mean my coins are lost now for uknown time?
17:30:11 <Mochi101> no
17:30:29 <Mochi101> Ask guarda for support.
17:30:32 <version2> i also read the announcement "There can be problems with xrmr blah blah"
17:30:34 <version2> no what?
17:30:36 <version2> when did u read it?
17:30:41 <Mochi101> Ask guarda for support.
17:30:44 <version2> since when does these problems occur?
17:30:53 <version2> you think i didnt do that already?
17:31:06 <Mochi101> So why do you come here and think I owe you support?
17:31:21 <version2> what please
17:31:29 <version2> no one owes anyone anything
17:31:38 <Mochi101> Ask nicely and you may get better results.
17:31:39 <version2> we're all just talking its irc
17:32:01 <version2> but you could answer when did u read it at least so that i know since when this problem is occuring already
17:32:19 <version2> or when this network upgrade was? have no idea i was completely offline for 2 weeks
17:32:55 <version2> "ask nicely"
17:32:55 <version2> i am
17:32:59 <Mochi101> https://www.getmonero.org/2022/04/20/network-upgrade-july-2022.html
17:33:01 <version2> im asking normally
17:33:29 <version2> oh so since 6 days
17:33:40 <version2> so the coins are lost for unknown time
17:33:48 <version2> that pretty sucks
17:34:00 <Mochi101> or you can get the cli and restore your wallet there
17:34:34 <version2> already searched for seed code or something didnt find anything
17:34:50 <rbrunner> https://old.reddit.com/r/Monero/comments/wqkqgd/xmr_guarda_wallet/
17:35:26 <version2> or no, found it
17:36:10 <version2> but i am not sure if this will solve that problem as its an inside-the-system coin exchange
17:36:21 <Mochi101> :)
17:36:37 <version2> pretty shit, it was just an "accidental" click, usually i would change it on fixedfloat
17:37:51 <Mochi101> version2, it probably didn't follow the new fork of Monero and went off on its own fork... which would mean that your real XMR is untouched.
17:38:52 <Mochi101> and that restoring your wallet with the latest CLI will give you access to your real XMR.
17:39:28 <version2> did just do it
17:39:32 <version2> balance 0 XMR
17:39:41 <version2> nothing
17:39:47 <Mochi101> You restored your wallet in 2 minutes?
17:39:52 <version2> yea
17:40:07 <Mochi101> You're doing something wrong.
17:40:21 <version2> only copy and paste mnemonic key
17:40:31 <Mochi101> into the CLI?
17:40:36 <version2> or what do you mean else with restoring CLI
17:40:53 <version2> no gui wallet
17:41:12 <Mochi101> and your GUI wallet is synced?
17:41:14 <version2> oh wait, now its showing up
17:41:15 <version2> suddenly
17:41:18 <Mochi101> :o
17:41:20 <Mochi101> omg
17:41:22 <version2> the same as before
17:41:29 <version2> first 0 and then some minutes later it just shows up
17:41:38 <Mochi101> syncing
17:41:43 <version2> i see
17:42:08 <version2> alright
17:42:24 <Mochi101> Lesson to be learned here...
17:43:17 <Mochi101> Use a real wallet or one with a dedicated team that makes sure it keeps its software up to date and follows the crypto community that it claims support.
17:43:45 <Mochi101> eg: not Guarda
17:45:35 <version2> yea i usually do
17:45:53 <version2> i just used this guarda thing today for another coin
17:46:27 <version2> for monero i usually use mymonero but just found the feather wallet is also pretty nice
17:49:38 <Mochi101> So we can agree now that the problem was user error and not Monero?
17:50:20 <version2> also mymoner seems to have some issues too with the update
17:50:36 <Mochi101> yes
17:50:37 <version2> yea i never claimed that that its monero's problem
17:51:03 <Mochi101> <version2> wtf up with monero today?
17:51:59 <rbrunner> Yeah, I wonder about MyMonero's problems. Really doesn't look their style.
17:53:41 <version2> i meant with that whats up as many wallets seemed to have problems with it today first mymonero then this guardia thing - but as it turned out restoring the wallet in another software just solved it all
17:55:32 <version2> yea, i actually liked mymonero too
17:55:49 <version2> and when you wanna write mymonero support it says they will be back on monday ^_^
19:09:36 <selsta> MyMonero did some work to prepare for the hardfork but seems that they didn't finish in time.
20:50:30 <garth> Maybe I need to restart monerod after enabling large pages?
20:51:00 <garth> Or is restarting p2pool enough?
22:56:01 <Torr> Just finished Kristin Dawkins's Gene Wars, interesting read.