<j​effro256:monero.social> Your pool probably requires TLS

<r​edhawk18:matrix.org> do you know of a pool that that I can just test with? It doesn't even have to be a real "pool" just something with the same schema for requests to know what I've wrote is correct.

<j​effro256:monero.social> I'd maybe look into github.com/jtgrassie/monero-pool

<r​edhawk18:matrix.org> I did... thats the exact pool I was trying to connect to...

<j​effro256:monero.social> Have you tried debugging the server side error messages ?

<r​edhawk18:matrix.org> I haven't gotten anything back to really debug, my unit test I wrote for this just fails with a eof which means the websocket was hung up on. Maybe the hosted version of the pool you linked has tls or something, I'm not really sure.

<r​edhawk18:matrix.org> Unless your saying I should built it myself to debug server messages?

<j​effro256:monero.social> ^ Yes

<j​effro256:monero.social> That will likely be more fruitful than only trying to debug client side

is the 64-64-24k implementation in external\supercop\crypto-sign\ed25519 also possible in C/C++? There are many files using qhasm

sech1 jberman?

<v​tnerd:monero.social> You could write c/c++, but you may want to ensure constant runtime to prevent private key leaks. For example, some of the assembly files touch every element in a table and use asm cmov instructions instead of branching to make the constant time guarantee. You could probably use the ref10 implementation as a guide on how to do this.

<v​tnerd:monero.social> Using simple indexes - which is the most straightforward forward to implement it - is not a direct port because of the constant time guarantee of the library. There is a valgrind plugin somewhere (I don't have link right now) that tries to validate whether a function is constant-time

hmm

I mean I would get something worse than the qhasm implementations but I wonder if I can still beat ref10

<r​edhawk18:matrix.org> I've been working on getting that pool compiled but the pool requires the monero project to be compiled as well and I'm having issues with that I keep having problems with boost

<r​edhawk18:matrix.org> ```bash

<r​edhawk18:matrix.org> -- Using linker security hardening flags: -pie -Wl,-z,relro -Wl,-z,now -Wl,-z,noexecstack

<r​edhawk18:matrix.org> CMake Error at /usr/share/cmake-3.29/Modules/FindPackageHandleStandardArgs.cmake:230 (message):

<r​edhawk18:matrix.org> Could NOT find Boost (missing: system filesystem thread date_time chrono

<r​edhawk18:matrix.org> regex serialization program_options) (found suitable version "1.81.0",

<r​edhawk18:matrix.org> minimum required is "1.58")

<r​edhawk18:matrix.org> Call Stack (most recent call first):

<r​edhawk18:matrix.org> /usr/share/cmake-3.29/Modules/FindPackageHandleStandardArgs.cmake:600 (_FPHSA_FAILURE_MESSAGE)

<r​edhawk18:matrix.org> /usr/share/cmake-3.29/Modules/FindBoost.cmake:2393 (find_package_handle_standard_args)

<r​edhawk18:matrix.org> CMakeLists.txt:1091 (find_package)

<r​edhawk18:matrix.org> Is it too new?

<j​effro256:monero.social> It looks like you might have multiple versions installed on your system

<j​effro256:monero.social> The "found suitable version" is 1.81

<j​effro256:monero.social> But your package manager shows 1.83

<j​effro256:monero.social> Can you install all `system filesystem thread date_time chrono regex serialization program_options` for Boost version 1.81?

<p​reland:monero.social> Oh no, I’ve seen this story before

<f​ede:xmr.mx> no, no, no

<f​ede:xmr.mx> pools don't support websocket

<f​ede:xmr.mx> stratum as described here: github.com/xmrig/xmrig-proxy/blob/master/doc/STRATUM.md is based on plain TCP

<f​ede:xmr.mx> and the TCP traffic can be optionally encrypted with SSK

<f​ede:xmr.mx> *SSL

<f​ede:xmr.mx> it's not WebSocket

<f​ede:xmr.mx> if you want to do XMR mining with a browser, you'll have to write and host your own WebSocket->TCP proxy

<r​edhawk18:matrix.org> This is what I saw that made me think I had to use web sockets, someone brought up that on browser they had to be web sockets and I thought I could use those. Do I just send the requests over a normal http client?

<f​ede:xmr.mx> if you aren't making an XMR browser miner, then just use TCP

Are there any working attacks on ED25519?

<r​edhawk18:matrix.org> Sorry I'm super new to lower level stuff so this is my first time learning all of this, so open a socket first and then somehow send a json through that?

.merge+ 9414

Added

<h​ardenedsteel:monero.social> is it possible to get block size from get_block_template? or am i looking to wrong place?

<j​effro256:monero.social> like the block weight ?

<h​ardenedsteel:monero.social> like the ``"expected_reward": 600000000000`` i would like to get or calculate the expected block size for the next block

<h​ardenedsteel:monero.social> it would be useful for block explorers

It's calculated internally, but it's not passed through to get_block_template RPC

Look for "cumulative_weight" in Blockchain::create_block_template

<h​ardenedsteel:monero.social> thanks, i will request feature for it, i dont know c++

<h​ardenedsteel:monero.social> monero-project/monero #9415

<r​edhawk18:matrix.org> When waiting to read input from a pool how big should my buffer size be for my response? since everything is over a tcp socket and jsons are sent over the wire

<r​edhawk18:matrix.org> When trying to reading responses from a pool how big should my buffer size be for my response? since everything is over a tcp socket and jsons are sent over the wire

XMRig uses 64 KB buffer, but you can probably get away with 1 KB buffer. Stratum responses are small.

<r​edhawk18:matrix.org> why 64k if I can only use 1k?