narf: monero is the best, which means it's fairly good but not great

right

I'd like to know how much time it'll take for the monero daemon to sync the blockchain, (from start), with a Internet connection of about 24.07 Mbit (download) and 65.80 Mbit (upload)

^ note: in theory the monerod says I am using a hdd when I am in theory using a ssd

You could get an Internet connection of about 24.07 Mbit (download) and 65.80 Mbit (upload), and measure it.

That said, usually internet speeds aren't the bottleneck.

Disk random access typically is. Spinning disks are atrocious, and syncing will take days with one of these.

Am I still able to send a transaction if my node is still 3 months behind and syncing?

Yes, in theory but (1) the daemon might reject it and (2) the fake outs will be old, which will leak out the fact the tx is either old or made with a partway synced daemon. Not a big problem I guess.

As a side effect, it'll leak that the real spend is at least 3 months old.

None of those a really a massive deal breaker, but on balance I could wait another 3 days for the node to finish syncing before making the transaction.

I really should look at setting up a graphana dashboard to monitor my node, does anyone have any templates?

Nevermind - found this: reddit.com/r/Monero/comments/nz7ysc/dashboards_with_monero_metrics

3 days for 3 months ? Wheee.

yeah, its basically taking a day to sync a month. Good ol rust drives

moneromooo: I am having a larger problem, the issue is that when I order my init system to kill the daemon that runs monero, the daemon is killed but the monero process is kept at uninterruptible process type, (which makes me not be able to run other monerod instances)

AFAIK, that means stuck in kernel space, so a kernel bug or stuck waiting on I/O on bad HW.

If you can get a stack trace, it might shed some light.

^ I'd have to get one of the processes, and I have 4 to pick, in which should I attach gdb?

^ what, they just disappeared

^ well, better for me (ig)

moneromooo: also, (completly unrelated), but if I started mining monero on some pool let's say pool x, I can use a subaddress for pool x and another address for pool y right?, except that apparently P2Pool doesn't allow me to run the daemon without specifing a main addr...

You can't mine to subaddresses. It's not supported.

oh, so I lost those coins completly?

well, they wheren't much so idc..

You mean you gave a pool a subaddress, and it did not complain, but just accept?

rbrunner: yeah, if you want I can say the pool name

Hmm, yes, of course a pool can pay you out to a subbaddress. I was talking about solo-mining.

Yes.

And p2pool is technically something like solo mining.

what's the difference between solo-mining and pool mining? (I asume pool mining uses the normal way of transfering money in monero?)

With pool mining you only submit "shares". With solo mining you submit a block - if you can find one, of course.

so if p2pool is like solo mining even if I mine normally I'd only get paid less? how can that be then?

Solo mining is, you solve a template your daemon made to your address. Pool mining is, you solve a template the pool's daemon made to its own address.

And you trust hte pool to send you some monero back to thank you.

so, a there will be a different challenge even for the same block?

^ because the template will be different because the daemon will make it different as the template needs to contain the address that the money will go to, (if valid obsly)

The template changes, but the target diff is the same

what does that mean?

The network difficulty determines how much work (on average) is required to find a block

And it's the same for everyone

The difference between solo and pool is that solo is all-or-nothing, while pool pays small amounts more regularly

diff by difficulty not diff by difference

And you submit "shares" to the pool to prove that you are doing work

Yes sorry, in the context of mining "diff" means mining difficulty

^ no problem, I should have alredy knew that :P

(And those shares are what get you paid)

P2pool is a little special, because there is no trusted middleman - but at the end of the day it's still pool mining

so everyone needs to "review" your blocks if they made a proper block?, wouldn't that take soo much time that the block would get invalid by then?

^ (like a valid block but outside of the time window)

just one p2pool node checks the block validity. then it submits it to mainnet if it's good

and the idea of sending the money to the other users, or is it sent to a special address?

P2pool blocks are constructed in such a way that every p2pool user gets their share of the block reward (according to their contributions)

so blocks need to be constructing a transtation with the block reward inside of it?

all blocks do that

^ but then how come that subaddrs aren't allowed?

just because. subaddrs were invented later, the encoding isn't supported for coinbase txns

Because the mining protocol dictates that rewards can only be sent to the primary address

thanks for the info

technically subaddresses are possible in coinbase txs

but they'll double the size of coinbase outputs

and there's no code written yet that can do it (neither in monerod nor in p2pool)

From the long and dark history of unmerged Monero PRs :)

right

looks like the changes weren't worth the bother

each subaddress requires a separate tx pubkey, this is why I didn't add support to p2pool - it add 32+1 bytes per output, almost double the size

also the wallet2 scanning is written in such a way that it will be O(N^2) if everyone in p2pool mines to subaddress

each of N outputs will scan all N pubkeys until it finds the right one

actually no, it's sorted by output index. But still, 2x transaction size is not worth it

sech1: there's a zmq related PR open, could you take a look?

ah, we discussed it with jberman before. I'll take a look

^ didn't even knew that by making a questionary about how pool mining works i'd indirectly make a PR be seen again back :P

hello