01:12:04 @riekak:matrix.org: Monero is going up. 01:12:49 If I had more cash, I'd buy the dip. 01:47:29 cap 05:19:06 yo 05:19:29 gm 05:19:37 gm 05:23:51 where can I download the ban list at 05:37:14 <321bob321> @snifflz1:matrix.org: https://github.com/Boog900/monero-ban-list/blob/main/ban_list.txt 05:37:23 @snifflz1:matrix.org: https://github.com/Boog900/monero-ban-list 05:48:58 thanks. all good now 05:57:52 what does this ban list even do, is this useless if i have tx-proxy option? 06:04:40 not peering with them also helps other users 15:10:15 how? 16:14:08 @kiersten5821:matrix.org: The nodes in the ban list have proven to be untrustworthy. By using it your node will not communicate with them and as a result not share them with other nodes. It’s keeping the overall network safer to a degree. 18:19:42 if i have a synced daemon on one computer can i literally just copy the lmdb files over to a new computer to get the daemon there? 18:20:30 if you first gracefully stop the daemon before copying 18:20:41 should work 18:20:42 ok nice thanks 18:23:01 let's say i restore a wallet from a certain height, if i use a remote node, does my node need to download ALL block contents from that height until now so it can scan the entire output set for owned outputs? 18:23:29 my understanding is this is the case but idk if there were any improvements? 18:23:49 @lza_menace: Isn't it also advantageous to malicious node operators, since they know and will try to avoid all the "signature detection" 18:37:17 This is going on for a long time already, and they don't seem to be in a hurry to hide better. They also could improve the distribution of the nodes regarding IP numbers so it would be harder to filter them out with simple measures, but it seems they choose not to. Who knows, maybe the whole story is not that important for them. Or they are operating on a pretty tight budget. 18:38:19 We can dream up all kinds of scenarios where the NSA joins forces with the FSB to attack Monero with secret quantum computers, but reality often is a lot more boring :) 18:48:23 Hii!!!! 18:50:11 I'm looking for safe way to store my cryptocurrency. I had Ledger wallet but it's not safe as company can extract seed phrase from device easily. For right now I have TailsOS which I'm using to store and access my btc 18:51:18 With Monero I would like to host maybe local node. Of course I could just install Monero wallet on my machine, but I'm looking for safer way 18:53:49 I was thinking about Trezor but it's also relying on some stupid electron app 18:54:45 mlxdy: trezor is open hardware though 18:54:55 and open source too on the firmware and API side 18:55:49 actually wait 18:56:04 nvm i don't think they're open hardware 18:56:18 When OVK? 18:56:18 So I don't have to leave my hardware wallet connected and unlocked until it get one confirmation (Or having to unlock it again when it get confirmed) 18:56:26 I tought Trezor where open to some point 18:57:41 how easy is it to make your own hardware wallet 18:58:42 shove everything into a microcontroller (no external chips to reduce the chance of bus sniffing) 18:59:14 have protection against voltage glitching or any methods to trick the chip into unlocking protections and allowing access to internal data 18:59:21 some new trezor models have issues when being used with monero 18:59:35 not a monero issue as older moldels work fine 18:59:37 ravfx: soon hopefully 18:59:43 and hell, if it doesn't come with the hard fork 18:59:49 i'll make a carrot wallet myself anyway 18:59:51 *models 19:00:09 So how are you storing your monero? (if you can say of course) 19:00:25 Good question. > shove everything into a microcontroller (no external chips to reduce the chance of bus sniffing) 19:00:25 You would need some controller that is not vulnerable to getting raped by people who know how... 19:01:38 It's probably why they all use "secure enclave" things 19:01:38 Maybe make your wallet like some old arcade cartridge, store everything battery backuped SRAM. Make it so if the casing get tempered with, it cut the power and SRAM go byebye 19:02:26 depends on your level of paranoia, you can create a wallet while offline and later send to it 19:02:38 Use some Lithium 18600 cells or something with a charge controller so it charge when you plug it usb. But it's not used for anything but SRAM power 19:02:51 just because your paranoid doesn't mean that are not after you :) 19:03:13 write down your seed and store it somewhere safe 19:03:38 write it down again and put it somewhere else 19:03:49 hey, why not a third time 19:04:04 Just put it in plain sight so no one can find it 19:04:27 HW wallets are good if you want to transact and unsure of the security of the device you are using 19:04:48 I'm not sure about security of my Ledger device haha 19:04:51 ravfx:xmr.mx: you can get normal view keys out if you modify the cli to print them :P 19:05:03 with that you can make a view wallet 19:05:27 but yeah. you can't export key images so it's useless except for seeing the confirmations 19:06:06 Maybe I'll generate new seed offline in ledger live and I'll only download neccesary software like Monero etc. and then just I'll by using it with third party software 19:06:16 This Ledger Live application is what I hate the most 19:06:56 It have a lot of annoying popups and they try to sell their service for storing seed phrase online 19:07:28 storing seeds as a service lolol 19:08:23 yes xD 19:08:36 and in their app there's ads of it everywhere 19:08:51 they offering you free trial hahaha 19:09:11 first one is free, sounds like drugs 19:09:20 I'm dying from laugh when I think about this 19:10:08 And in terms of service there's information that they can share that seed phrase with goverment 19:10:16 0_o 19:10:37 at least they say so 19:11:20 It's sad that without KYC is impossible to buy crypto at normal prices 19:11:37 I can't pass KYC as I'm not 18 19:11:50 Of course in few months it will change 19:12:07 I remember that in the past we had local monero 19:13:32 DataHoarder: does not work with normal current view keys, you still need the hardware wallet to generate the key images 19:13:48 20:05:27 but yeah. you can't export key images so it's useless except for seeing the confirmations 19:13:52 you don't need the key images to view incoming :P 19:14:07 not the same but there is haveno with the network using haveno being called retoswap 19:14:10 which you may know of as you said "It's sad that without KYC is impossible to buy crypto at normal prices" 19:15:11 I very rarely use the ledger live thing. It's only to upgrade the firmware or monero app > It have a lot of annoying popups and they try to sell their service for storing seed phrase online 19:17:52 Ledger have to continue to make money once the crypto hype is gone 19:20:55 And goverment have to continiue to control their citizens 19:21:29 RavFX: have you tried to use Ledger as FIDO key? 19:22:07 mlxdy: Yeah, it just work 19:22:07 And I don't have to buy 3 to have proper backups 😂 19:22:39 The seed backup the FIDO key too 19:22:57 Okay so what if you'll lose your ledger? 19:23:28 mlxdy: I throw some monero dust at Amazon and get a new one? 19:23:28 Ledger Nano S are cheap anyway 19:23:49 Small, easy to hide... 19:24:13 I have no idea why people want to one with big screen, Stax or something like that.. At that point it's when it become a toy 19:26:38 I've seen that, is fucking joke. It's not even normal screen but like these in Kindles haha 19:27:13 I never lost a ledger, no idea why people do that too. 19:27:13 I did damage a ledger one time (don't use it as a finger fidget... Ledger meeting a computer screen at very high velocity can damage it 😂) 19:27:37 haha 19:28:15 mlxdy: Yeah, totally useless imo 19:28:30 I think it's so you can view your -99% NFT or something 19:29:44 ravfx: were you referring to the capcom suicide batteries? 19:31:20 Cindy: Yeah 19:31:21 But good lithium batteries are reliable now, if it's to power only sram it should last forever assuming you plug it frmm time to time 19:31:58 Maybe add extra light sensor circuit that also cut the juice to the sram, so if the person manage to open it without triggering the trap, then you have a backup trap 19:32:33 I wonder if one can get a chinese to make case that embdeeb micro metal wire all around inside the plastic 19:33:20 why not just sniff the bus while the device is running 19:34:41 Cindy: Because the private spend key won't leave the device. so you can't spy on it from USB. 19:34:41 And if you want to dig inside to reach it's actual memory bus then at that point you will have triggered the battery trap 19:34:54 capcom actually had motorola produce them a special kind of 68000 that would decrypt data after fetching them from the bus 19:35:19 Yeah, you can encrypt the ram and have your controller do all the encryption work too 19:35:43 and if you like have the usb data link encrypted too 19:35:54 yeah like that 19:36:16 nothing wrong with being extra careful 19:36:22 indeed 19:37:36 but are there any actual secure MCUs 19:37:49 like ones that aren't vulnerable to being tortured 19:37:58 Ah yeah, there many option 19:38:43 That hardest part might or might not be to implement the monero signing stuff. Have to do some math :) I should take a look at making my own signer, one day 19:39:12 Cindy: YEah, extra protection. 19:39:12 If someone manage to reach it without triggering any traps 19:39:33 ideally everything should be within the chip, like those TPMs 19:40:27 If you can find one that have sram that would be ideal 19:41:32 But yeah the secure mcu stuff are what's in Ledger and Trezors 19:42:22 20:22:57 Okay so what if you'll lose your ledger? 19:42:24 did you backup seed words? :P 19:42:36 you can restore (or even use offline tools to generate the paths) and recover all material 19:42:39 the problem with secure MCUs is voltage glitching 19:42:52 ive heard of the STM32's read out protection being bypassed by just fucking with the power source 19:43:03 > stm32 > secure 19:43:09 DataHoarder: Yeah, you can convert LEdger 24 words to Monero 25 words and just pluck it in Feather. Like in an emergency or something 19:43:27 they usually are external secure coprocessors 19:43:33 Cindy: yeah, initial Trezor key leaking bug 19:43:55 It's why you want the data stored in sram and have trap to erase it. 19:43:57 but it's becoming even better nowadays Cindy you can laser etch and do laser faults 19:45:09 ideally i'd want a fuse in the chip to get blown and permanently brick it if the voltage gets fucked with too much 19:45:26 i've seen that before i think 19:46:13 i'd rather want the hardware signer to die than to spill its secrets and live 19:46:44 Have a boot counter in the code, that save in sram too. 19:46:44 And have the usb power to be stabilitzed by the voltage regulator (the same that charge the battery) 19:47:09 they can unsolder it 19:47:12 Actually save the boot counter in the internet eeprom 19:47:19 internal eeprom** 19:47:23 :P 19:47:25 internet eeprom LOL 19:48:05 That way if they can manage to open it without traping the traps, and power glitch it (rebooting it in the same process), boot counter will be more than 1, if boot counter > 1, blow the fuses 19:48:29 It should never boot more than one time thanks to the nice lithium batteries 19:48:31 you can have this same issues on normal setups btw :P 19:48:33 so microcontroller will always be running? 19:48:44 ^ sram like some bank tokens 19:48:47 yes 19:49:22 sram because it's don't really use power when idle. 19:49:22 And there not vulnerable to cold boot attack 19:49:34 once it lost power, all the transistor instantly unlatch 19:52:40 microcontrollers probably have sleep modes too when not running. 19:52:40 Would have to check there power consumption when idle and size the battery appropriatly (maybe so it last 3 months would be ideal). 19:52:40 If only power the sram it would last years 19:54:40 also i thought SRAM was expensive 19:54:48 more expensive than DRAM 19:54:55 at least 19:55:58 It's not like you need a lot of ram 19:55:58 Don't you have a bucket of 486 cache chips somewhere or something? 19:55:58 That's 32K, it's wayyyy too much for the need 😂 19:57:29 Cindy: it's expensive, but cpu caches are SRAM 19:57:59 and yeah. you just need a couple of bytes 19:59:12 also since the chip is external, wouldn't they be able to just, i dunno 19:59:24 unsolder the read/write pin and solder it permanently to be read? 20:00:15 by hooking it to ground? 20:01:09 Cindy: That why the casing need to contain a trap, you know, so it lose power if you damage or open the case. 20:01:09 Plus the SRAM you encrypt it, the key for it reside in the mcu 20:02:12 And lets get real, most LE are arguably funny and will just crack it open and oops. 20:02:12 Same for thiefs... Lets harvest the gold or something 20:02:33 not lose power btw 20:02:42 you need to flush, as it has data remanence 20:02:58 DataHoarder: SRAM, not DRAM 20:03:02 SRAM 20:03:02 yes 20:03:06 it has data remanence 20:03:20 For how long, considering it does not have the tiny capacitors... 20:03:29 https://www.cl.cam.ac.uk/techreports/UCAM-CL-TR-536.html 20:03:31 even without power 20:03:34 DRAM needs the capacitors 20:03:38 <321bob321> EMP 20:04:03 this is from 2002 20:04:05 better attacks are done nowadays 20:05:42 so if you detect tamper cutting power is not enough, you should flush data first 20:06:59 and yeah, power bus issues :P https://cacm.acm.org/research-highlights/sram-has-no-chill-exploiting-power-domain-separation-to-steal-on-chip-secrets/ 20:08:01 also btw 20:08:14 don't microcontrollers have a internal power-up counter register? 20:08:41 Cindy: Some probably have yeah 20:09:12 Research shows that SRAM can partially retain data for a few milliseconds under extremely low temperatures (below 20:09:53 minutes :P 20:10:23 modern research: 20:10:25 > Data retention in SRAM is also significantly increased with lowering the operation temperature of the SoC. Data may be retained for around 1 ½ hours at 75°C, 3 days at 50°C, nearly two months at 20°C, and approximately 3 years at 0°C. 20:10:31 note this is retaining specific bits, not entire content 20:11:30 Your talking about a SOC, what about discreete SRAM chip? 20:11:38 similar numbers nowadays 20:12:08 see the first link 20:12:14 This article introduces Volt Boot, a method for executing physical memory disclosure attacks on on-chip SRAM memories by exploiting SoCs’ power domain separation 20:12:25 But if it's not in a SoC... 20:12:26 and the correlation to temperature 20:12:28 21:03:13 https://www.cl.cam.ac.uk/techreports/UCAM-CL-TR-536.html 20:12:31 this is old research that looks at chips 20:12:39 but it's become worse since then 20:12:47 On-chip memories, primarily SRAM, are integrated directly into the processor die, offering greater security against physical attacks (for example, probing attacks11,26) compared to off-chip memories. Research shows that SRAM can partially retain data for a few milliseconds under extremely low temperatures (below 20:12:47 – 20:12:47 110 20:12:47 ∘ 20:12:47 C[... more lines follow, see https://mrelay.p2pool.observer/e/je-i0OEKbmZvUFI5 ] 20:12:57 That specific article seam to be for attacking SoC sram 20:13:28 Chip are more godies inside, probably micro capacitors too, right? 20:13:31 see again the first link 20:13:33 which is attacking chips directly 20:13:35 no. 20:13:37 it's sram 20:13:41 it doesn't have micro capacitors 20:13:44 They say SoC 20:14:09 open the FIRST link 20:14:12 21:12:23 21:03:13 https://www.cl.cam.ac.uk/techreports/UCAM-CL-TR-536.html 20:14:30 Power domain separation. SoCs integrate numerous circuit blocks, each exhibiting unique analog characteristics. To meet stringent performance and power efficiency requirements, these blocks are divided into separate voltage domains. The power management unit (PMU) within the SoC dynamically manages the voltage levels for these domains at runtime, tailoring them to the workload of each domain. 20:14:30 In modern, complex SoCs, dozens of off-chip supply pins connect to various power domains, enabling precise control over analog circuit behavior. This setup mitigates challenges such as ground bounce, power-supply noise, and per-pin current limitations. We broadly categorize the power-supply domains of an SoC into three main areas, as illustrated in Figure 2: 20:14:42 I linked that second one cause it was interesting while searching for papers that came from this 20:15:12 ravfx:xmr.mx: you are still reading the wrong paper 20:15:14 again open the first link 20:15:16 21:14:03 21:12:23 21:03:13 https://www.cl.cam.ac.uk/techreports/UCAM-CL-TR-536.html 20:15:47 that has the PDF 20:16:15 > We built a special circuit board for testing static RAM chips. 20:16:17 they even list the models 20:16:46 it's 2003, and has become worse over time 20:18:14 is there a SoC that literally kills itself over voltage glitching 20:18:25 instead of handling it as a regular hardware fault 20:18:54 it's weird that MCU manufacturers haven't done this 20:19:09 yes, see rpi talk on recent ccc 20:19:40 https://www.youtube.com/watch?v=hbFRnPNQnoY 20:19:42 this had fault detectors 20:19:44 as far as cold attack on sram, you still need to reach the chip to freeze it. 20:19:44 At that point there is nothing truly secure 20:20:13 and they go over how some were also bypassed by doing locality attacks 20:20:15 well yeah, ravfx:xmr.mx, that is the point 20:20:17 you need tamper detectors to cover this :P 20:20:47 wait Cindy 20:20:49 wrong talk 20:20:51 THIS is the one https://www.youtube.com/watch?v=V5KvW4elzXU 20:20:53 > 39C3 - Of Boot Vectors and Double Glitches: Bypassing RP2350's Secure Boot 20:20:55 https://media.ccc.de/v/39c3-of-boot-vectors-and-double-glitches-bypassing-rp2350-s-secure-boot 20:21:44 YEah, the trap I mentionned about should cut the power when the casing is broken or open (micro wires embdeebed in the case, light sensor. 20:21:44 Now would have to have also a temperature sensor that work with a different power source, that would write "random" all over the sram assuming it's getting cold 20:23:22 Get a NEC D 20:23:32 NEC D4364C :p 20:23:34 https://mrelay.p2pool.observer/m/xmr.mx/epFasLlyHhklyvfAUKjcFzIb.png (clipboard.png) 20:24:27 Anyway, assuming the sram is encrypted, it's just extra protection for the mcu. Just an extra anti tamper thing. 20:24:39 more thing can break when you mess with it 20:24:41 then you read fuses from mcu to decrypt sram :P 20:24:56 you have all the time in the world 20:25:05 burning fuses takes relatively a lot of power, so you can't reliably burn these on the spot 20:27:21 So what about, instead of cuting the power to the sram, what about writing just 1 or 0's in it? 20:27:21 Cutting the power the moment the case is open 20:27:21 writting 1's all over the moment the case is open/broken 20:29:18 At an extra shenanigans that rely on some rf signal 20:29:18 signal gone -> erase (like in an event the device is relocated) 20:29:18 That way we can probably just use the mcu sram instead of having a dedicated chip. 20:30:14 Or actually leave it on a seperate chip for more fun and shenanigans 20:30:23 But normally people steal the stuff before cracking it apart 20:38:58 ideally if someone want to make a custom hardware wallet, one could make it so it does not look like an hardware wallet or something weird . 20:38:58 like some random idea 20:39:00 https://mrelay.p2pool.observer/m/xmr.mx/hIqlmTFYiTZibrOPJUYPEqcc.png (clipboard.png) 20:39:09 hidden in plein sight, on the fridge! 20:39:26 ravfx: cover the interior of the case with a magnetic strip 20:39:33 and then put magnetic sensors all over the board 20:39:35 There plenty if space in that thing 20:40:39