<plowsof11> "make a reddit post, wait for..." <- LOL. is it that easy to get free monero? make a sob story and wait for people to give u free monero?

<ureallyorigin[m]> "LOL. is it that easy to get free..." <- Great way to inspire usage and bring awareness. Certainly better than lectures and technical specs most won't understand.

<opsoyo[m]> "Great way to inspire usage and..." <- how does that bring awareness?

the donator is likely already aware of monero

anyone have a breached account to share the links of breached.vc/Thread-TSA-NoFly-List-Database-Leaked-Download

<BoBeR182> "anyone have a breached account..." <- are u thinking that ur on the no fly list?

I think it would be fun to search up some political type people

snowden and the like

but I don't have an email that's on the breached whitelist domain list

<ureallyorigin[m]> "LOL. is it that easy to get free..." <- Some absolute chad gave out 0.5 xmr on 4chan to 5 people

God bless /biz/

I'm curious on what you guys think but is buying monero from an atm a secure way to get it aside from P2P exchanges?

many ATM's require verification with ID, face, credit card or a combination of those. If you can pay with cash and without verification, ATM is a good option

ah ok thanks for the info

<ceetee[m]> "many ATM's require verification..." <- ATMs are physically in contact with you, no one knows what they can do

Wear full body clothing, gloves, and shoes that increase your height, also wear a mask, sunglass and a hat

Even after taking all those measures you still could be deanonymized if the agencies are after you

cockliuser[m]: are they really going _that_ far to identify ppl buying crypto at BATMS

Probably not but it depends on your threat model

Hello,

Can someone sugest me best place to exchange also buy & sell XMR Anonymously. (Without giving them passport details and stuff) ?

I want to buy XMR but i dont know the best place to go.

Since im new. :(

With fiat? Try localmonero.co

Yeah localmonero.co is just the only way to go if you want complete anonymity. Unfortunately, some countries just don't accept payments method like cash by mail etc...

Thank you for information. I will try :))

Can anyone send a post on Reddit about my CCS Proposal to discuss ? I don't like to be the one asking to be seen but my reddit account has 2 karma so i'm shadowban...

someoneelse495495: which ccs

Are there many successes sending cash by mail internationally?

<opsoyo[m]> "Are there many successes sending..." <- would highly doubt it

people send cash in the mail internationally to purchase services like mullvad and a significant number of them report not receiving time on their account meaning the mail was likely stolen

or lost

probably stolen though

<cockliuser[m]> "Probably not but it depends on..." <- well you are a cockliuser afterall, i imagine your threat model is higher than that guys

Who is responsible for taking getmonero.org down?

Maybe they're finally migrating out of Chinese shared hosting

Cloudflarefuck + ChinkCDN

someoneelse495495 if you make a thread and link it here, a kind reddit moderator can approve it or in #monero-community:monero.social

Stnby[m]: Also why did it start using cloudflare?

Siren[m]: Isn't Cloudflare based in USA?

Of course

has it always been cloudflare?

We have too smart of core

Siren[m]: DDoS i guess.

Monero has many adversaries.

Or perhaps it just started with CloudFlare

Fluffypony knows for sure

ask him

getmonero always used cloudflare

core are busy with it / no ddos

lol

guys

we're just upgrading the server and we were struggling with the IPMI being unresponsive

the CDN is still up, and we'll have it fixed in the next 12 hours

nothing to see here

BUT WE CANT DOWNLOAD MONERO

Very long downtime :D

lol

lol sech1

getmonero.org needs an i2p frontend by the way

Do you need help with migrating a static site without downtime?

Stnby[m]: feel free to contribute to this: github.com/monero-project/meta/blob/master/SERVER_SETUP_HARDENING.md

we're also not migrating anything

it was an OS upgrade that seems to have borked something

That was going to happen eventually, it clearly needs redundant setup, reading the MD right now.

Stnby[m]: why does it need a redundant setup? we have the CDN, binaries are on GitHub, a few hours of downtime isn't going to kill anyone

feel free to fund a redundant setup

fluffypony: problem is luigi always forgets to upload binaries to github

maybe someone with github access can add them

selsta: right but that's a fixable problem

Use CI/CD for that

we've had minimal downtime in the last 9 years, I don't think we need to stress about this

Binaries are not on github, they link to downloads.getmonero.org which is down

Siren[m]: can't, binaries need to be manually checked - that's the point of reproducible builds

otherwise it's an untrusted path

we need to get a movie to the oscars, hopefully this is fixed soon sir

rofl plowsof11

This does not require a mainframe, pretty sure we can fund it

oscars is overhyped

if someone asks for the binaries i have them linked here

Stnby[m]: super low physical tin requirements, but what it does require is a well-peered, low-latency 10gbps unmetered link at a DC, and that's the issue

<fluffypony> "Stnby: feel free to contribute..." <- Ditch Wazuh for Falco and you'll be able to monitor both the host and the docker containers

It's also lighter weight

10gbps? Do people download Monero that much?

selsta: yeah

sech1 I mean

You mentioned its behind a CDN, so what the server requires is certainly not a 10gbps unmetered link

even with the CDN we get a ton of cache misses

Stnby[m]: the CDN is mostly for punching through the GFW

Yes I know because we do not have a proper caching setup

Headers for cache do not exist basically

so submit a PR to the document

fluffypony: Manually checked for what?

I do not like current Docker setup

Siren[m]: to make sure that everyone agrees what the hashes of the binaries should be

so that nobody can backdoor the binaries

Siren[m]: en.wikipedia.org/wiki/Reproducible_builds

I know what reproducible builds are

so then you can see how going from an untrusted CI/CD pipeline into GitHub is a problem

Still doesn't make sense to me why you would do this by hand

fluffypony: Doesn't need to be a GitHub pipeline hell no

You have a gitlab instance ffs

right but it's still an untrusted build

someone could hack the CI/CD pipeline and backdoor the binaries

How?

the point of repro builds is for everyone to agree on the hashes BEFORE the builds are made publicly available

not after

s/everyone/a significant number of people who have a repro build environment

fluffypony: Not everything coming from ci has to be pushed straight to the download page

right but they'd still be available, even if the poisoned binaries are online for a couple of hours in a single location it's problematic

binaryFate already adds the release notes to GitHub, also attaching binaries shouldn't be an issue in the future

fluffypony: And how do you all have a reproducible build environment? do you use nix?

it's happened before, btw

Siren[m]: gitian

they were compromised for 35 minutes

Siren[m] github.com/monero-project/monero/tree/master/contrib/gitian

fluffypony: No they wouldn't be if it's multiple steps or pushes only on tag action etc

sech1: What were the investigation results?

fluffypony: Yeah and how did that happen?

* investigation results? (oopsie wrong reply)

Siren[m]: hard to tell; had the drives pulled and audited and the working theory was physical compromise at the DC we used back then

new DC is much better

actually cares about our opsec

You can get compromised at personal computer, Github, DC, CDN, Cloudflare all of these points 😉

fluffypony: Avoiding CI/CD won't protect you from that. Did you have Wazuh ruleset to watch over the binaries?

Siren[m]: this predates the hardened setup

the hardened setup was as a reaction to the incident

Docker needs to go out of this "hardened" setup

anyway, I deeply appreciate that everyone here is an expert

feel free to open a PR to the doc

Cloudflare will censor and potentially hurt users if US gov decides to sanction or does something stupid like that.

Can we come up with a setup for the website and website only that in our eyes is redundant, hardened and idk cheaper to run?

No Debian, no Docker "hardening" bs

Stnby[m]: you can do whatever you'd like, it's a free world

Siren[m]: we have fallbacks for a lot of the points of failure

Would any of you be interested in having something done from scratch and if someone is willing to review it we will attempt it

And "we" are who?

digilol.net

submit the PR and let that be the point of discussion and collaboration

Okay, thanks for the info, will attempt to throw something together.

<ofrnxmr[m]> "someoneelse495495: which ccs" <- repo.getmonero.org/monero-project/ccs-proposals/-/merge_requests/373

<Stnby[m]> "Okay, thanks for the info..." <- whats your pfp

i dont get it

Its a picture of me and my uncle

captaincrunchpre: en.wikipedia.org/wiki/Ottoman_military_band

your uncle is really classy

<Stnby[m]> "en.wikipedia.org/wiki..." <- are u a turk

captaincrunchpre: No

sounds like a turk flag behind

Stnby[m]: then why do u have a pfp with a turk flag

Same reason why you don't have any pfp

lol

someoneelse49549: I'm with an Ottoman empire flag, which does not exist anymore

wtf are you telling me you doesn't exist anymore?

* exist anymore? /s

Yeah I am quantum particle

lmao

<someoneelse49549> "repo.getmonero.org..." <- I'm going to copy paste the ccs and replace rust with c and no one can stop me

trollface.png

I'm trying to send some dust (just barely more than the fee) from the cli wallet and I get to 'Is this okay? (Y/Yes/N/No):' but when I type Y and hit enter, the enter key doesn't work, all I get is a ^M

is it because of the small amount?

Sweeping 0.000042260000 for a total fee of 0.000030640000. Is this okay? (Y/Yes/N/No): Y^M

anticarnist: what OS are you using? how did you install monero-wallet-cli?

selsta: pop_os. downloaded from getmonero.org. It worked yesterday, but that was with bigger amounts.

^M is carriage return, I don't see how this is related to the amount

did you do anything different compared to yesterday?

yes, doung sweep instead of tranfer

doing*

how many times did you try?

two

ctrl+m doesn't work either, pretty strange

what happens if you just enter a lowercase y and then enter?

same thing, enter just inputs a ^M

but enter works when entering transfer or sweep ?

the command itself i mean

what's the output of echo $TERM

<cockliuser[m]> "I'm going to copy paste the..." <- oh no a wide c maximalist appears /s

selsta: yes, works fine up until that prompt. echo output: st-256color

can you change it to one of these? github.com/monero-project/monero/bl…ends/patches/ncurses/fallback.c#L11

we don't have fallback for st-256color

try running stty sane or stty icrnl

okay, I tried to close the terminal and do it all again, but with lower y and enter instead of capital Y and that worked, lol

wo don't have any fallback values for st-256color, try `TERM= xterm-256color ./monero-wallet-cli` or maybe what Siren[m] said

capital Y should also work if everything is setup correctly

(without space)

maybe it was the fact that I closed and reopened the terminal?

what does it say when you do `echo $TERM` now?

same

Seems like your terminal got stuck in a cooked mode en.m.wikipedia.org/wiki/Terminal_mode

anticarnist: Could be that previously you accidentally disabled icrnl

I think it could either be issues with terminal emulator or monero or previous commands job

* or monero cli or previous

There were cases when monero cli didn't restore terminal settings after exiting

I have a feeling its monero cli related in here as well

Y/n selector should be in cooked mode

I tried again now, and now it worked with capital Y, so not sure what it was but closing and reopening the terminal solved it.

<selsta> "what does it say when you do `..." <- Same output?

What terminal emulator were you using?

st

Probably its fault sus terminal

Their philosophy is lines of code, not error handling

<someoneelse49549> "oh no a wide c maximalist..." <- Rust evangelist vs C maximalist battle

epic fight

I'd bet 20WOW that the C maxi wins :3

*Golang lover has joined the chat

QQ about the full blockchain sync, why does it go so much faster at the start than at the end?

checkpoints

they are added for each release

...and those have a cumulative impact on the validation?

yes

smathy[m]: Both checkpoints and block size

Thanks nioc

(newer monero blocks have more transactions than old blocks)

xfedex: ah cool, that was one of my guesses.

yes but mostly checkpoints

checkpoints are the main slowdown and the last release was months ago so a lot of blocks have to be synced without checkpoints

nioc: depends on your device

the blocks before ringCT are faster

Will the cumulative effect of the checkpoints eventually mean it becomes too slow?

if you have a hard disk, then disk speed will usually be the bottleneck

smathy[m]: No, not at all

xfedex[m]: true, I only use SSDs :)

What checkpoints do is saying "hey, blockchain is valid till block x, and its hash is y"

so your computer needs to crunch less numbers when syncing

Hmm, that seems like it ...right.

So why does that make it slower the further through the blockchain you get?

(happy to ask this in dev if that's better, seems like we quickly got away from the end user space here :)

no here is good

smathy[m]: Because checkpoints are added every once and while (usually during hardforks); they are hardcoded into source code.

Once you pass the last checkpoint, monerod must check manually that the blocks it receives are valid.

smathy[m]: are you using a HDD or SSD?

SSD

great :)

It's a MBP M1 Max.

HDDs can be painful for initial syncing

But if the checkpoints allow it to refer back to just the last checkpoint, rather than having to validate all the way back (right?) why does that explain why it takes like 3 minutes per 10% at the start of the blockchain vs hours for 90-100% at the end?

the beginning had different transaction type and fewer transactions

nioc: right, so I hear. Which was my other guess before asking, that this was a slowdown for something internal to lmdb once it gets a large value.

Right, ok. That was another guess, that the complexity/size/number/something of transactions and/or verification process has dramatically increased.

monerod ships with a list of known block hashes (checkpoints)

if a block is known it has to so less verifiable work

verification*

Right, so don't checkpoints improve the efficiency of the verification?

less work per block = faster sync

Answered :)

selsta: Will nodes reject a blockchain re-org that is deeper than the checkpoints?

all blocks that were mined after the last release are unknown

Rucknium[m]: Blockchain reorgs cannot be more than 10 blocks long

So yes, it will

I dont think that's true

Rucknium[m]: there are two different systems with the name checkpoint

So, unless as time has gone on, checkpoints have been more sparse (which [the code doesn't imply](github.com/monero-project/monero/bl…rc/checkpoints/checkpoints.cpp#L176)) then I don't see how that explains the sync slowing down.

Liek everything in Monero with the name ambiguity :P

one does what you describe, you can't rollback further, and the other speeds up sync with known block hashes

xfedex[m]: I also don't think that's true, there's no 10 block limit

selsta: Ah ok, I understand that, but that'd only be at the very end. Like even 40-50% is much slower than 10-20%.

smathy: Checkpoints have been utterly sparse since the last monerod version release :)

Blocks newer than 10 blocks are referred as "tips", blocks older than 10 blocks are permanent

selsta: Are you sure?

smathy[m]: that's due to different usage and different cryptography

Rucknium[m]: Infinitely sparse :)

2014-2015 barely had usage and simpler cryptography

Bitcoin Cash has a 10 block re-org depth limit right now. The perils of being a minority hashpower chain.

xfedex[m]: yes

wallets have a configurable reorg limit but not the daemon

(as far as I know)

I 51% attacked the Townforge testnet and re-orged about 24 hours of blocks. Townforge and Monero have about the same blockchain code.

So yeah, around the 44% mark it really slows down. This was just after a log ouput "Validating txpool for v4" which I'm guessing was a fork that involved a big change in verification/security/something.

v4 introduces RingCT

which uses inefficient cryptography

selsta: read this monero-project/research-lab #95

> Outputs in the PoW 'reorg zone' (the most recent ~1-10 blocks) can be completely removed from the chain by a reorg, because any output that can be reorged can be double-spent.

from this I deduce that reorgs cannot be longer than 10 blocks

once you get to v8 sync will speed up again due to advancements in cryptography

Ok cool.

xfedex @xfedex:matrix.org: selsta has read that lol

xfedex: I think the wording there is just unclear. It's the "re-org zone" since it's the chosen safety factor of the 10 block lock on spending tx outputs.

I would recommend asking UkoeHB in #monero-research-lab:monero.social to confirm the meaning of that statement and to revise if necessary.

Seeing as I have an SSD, during this v4-v8 stuff, this is all CPU-bound?

(fwiw, my htop suggests it's not really)

sync is not bound by anything in my system

random IO speed is important for sync

xfedex[m]: you'd have to ask UkoeHB to clarify, but I'm certain there's no 10 block re-org limit

Ok, nothing I can do about that, thanks

Checkpoints dont discard ALL data. Increased monero usage = more data to verify. Checkpoints decrease the workload, not eliminate it.

Syncing to ssd should take 8-24 hours on most systems

And yeah, it just wasn't making sense that checkpoints we're the cause of the slow down, but I wasn't understanding what y'all were talking about (the final stages, after the last checkpoint) nor was I making my actual question clear (ie that it slows down way before then in the 40% range (turned out to be v4))

Really appreciate all the help, very new to this and you've all been very helpful

Its like..... (full message at <libera.ems.host/_matrix/media/v3/do…202eac1c917e51838274770274fe638f6fb>)

Right, it's a checksum of all the blocks since the last checkpoint.