/biz/ - Business & Finance

fud

>>57371319
Harming stakers when errors are made is a really bad choice. It just seems anti-open software.

>>57371319
You should add in the fact that centralized entities like infura, who provide critical user facing infrastructure for web3, use their own proprietary versions of everything for whatever reason and you have a real fucking dumpster fire waiting to spark up. Fortunately for Eth, it's the globohomo state sponsored useful idiot trap du jour so I'm sure they'll just keep painting over the rust and roll on down the road into their commie dictatorship bullshit faggot pedo fantasy world.

>>57371319
>- If geth is a super-majority, how can the minority clients even veto invalid blocks?
They effectively cannot if you assume all versions of geth are producing the same invalid blocks.
>- What is the slashing loss formula used to conclude that geth stakers will lose 90%
https://dankradfeist.de/ethereum/2022/03/24/run-the-majority-client-at-your-own-peril.html

My opinion is simply choose not to run geth and you will be protected from these potential brutal slashing events.

>>57371393
additional bailouts similar to what happened with the DAO are completely infeasible at this point

>>57371395
>infura
A massively overblown problem you could solve by simply not using metamask and learning about decentralized RPC providers

>>57371319
>This isn't being discussed much in forums yet
>>57371319
>It always seemed clear that the low staking returns in Ethereum weren't worth the risk
>>57371395
>You should add in the fact that centralized entities like infura

I can't believe you ppl are my counterparty in markets, it cant be this easy

>>57371484
It would also be great if every eth fag learned to run their own node so they could see what a slapped together layer cake of bullshit the protocol really is but I'm not holding my breath.

>>57371319
>Don't worry bros, We're gonna dank shart out of a wormhole into the Eigen layer.
Cant even make this shit up.

>>57371535
elaborate, say what you mean

New fud?

>>57371319
>It always seemed clear that the low staking returns in Ethereum weren't worth the risk
correct. linkies will also learn this the hard way.

>>57372436
uhmm sweaty the stakedotlink premium liquid staking service comprised of 15 top quality nodes will have an insurance pool to protect stakers from slashing events

>>57371319
t b h this board is too low iq at this point to discuss this seriously
half of the posters itt don't even understand your op
i expect a mongoloid to (You) my post with "fud" or "HEH good thing you're my counterparty"
but you're right, this is worrisome
on the bright side it IS worrisome
there is technically nothing preventing most node operators from switching to minority clients right this instant
surely coinbase, lido and all must hear the noise too
granted, this is not the moment to be complacent... if they don't switch fast, something will have to give
as for solo staking, while i agree with you the risk has never been worth the rewards, as a solo staker you can simply run a minority client and not be at risk
for a full breakdown of the slashing penalties you can read dankrad's blog (highlevel EF researcher)
https://dankradfeist.de/ethereum/2022/03/24/run-the-majority-client-at-your-own-peril.html
most relevant part is the quadratic inactivity leak
ngl, memes aside this is one of these moments that is making me rethink being allin eth. still cautiously optimistic a rational solution will be found, but this shit is a bigger problem than any of the so-called "black swans" of 2022-2023

>>57372536
You should never be all in on anything anon. I know stupid shit that hits mass usage can still grow in spite of itself (Windows) and there are way too many bagholders, myself included, for Eth to automagically go away and die. Thx for the post anon.

>>57372536
hi, thanks
do you perhaps have an idea as to which alternative to geth it is worth using now?

>>57371319
Load ze 2013 ETC fud

>>57371319
If you ever find yourself doubting what Eth is doing, keep in mind that if we could all agree on who s the No2, that project would had it 10x worse in all possible metrics. So it always kinda leads me back to Eth.

In other words, if anyone knows what they are doing it devs on Eth, everyone else is 10x worse... Hope this answers it for you.

>>57372536
HEH good thing you're my counterparty

>>57372799
>Uh oh guys watch out we got a real wolf of wall st over here.
Seriously bro fucking kys.

>>57371616
it's just stupid fud

>beta software has bug
>news at 10

>>57371450
Thanks for the info!
>My opinion is simply choose not to run geth and you will be protected from these potential brutal slashing events.
>>57372536
>as for solo staking, while i agree with you the risk has never been worth the rewards, as a solo staker you can simply run a minority client and not be at risk

"not be at risk"

If geth goes bad, then depending on how bad it's going to be (e.g. a bug may only be present in certain versions that not all stakers may be running) ETH could tank by something like 30-40% and may have weeks or months of uncertainty, and possibly even downtime and a rollback. With a minority client you still risk sitting in your validator exit queue while ETH bleeds down and eats all of your staking rewards and then some.

>>57372829
its legit not though, client diversity and geth supermajority situation could tank eth in one day

>>57372884
>validator exit queue
I never even thought about that aspect. You would have effectively barred the exit doors. Holy shit what a nightmare.

I'm too stupid to understand what's going on. Can someone get Vitalik on the line and ask him to debunk all of this please?

>>57372536
>HEH good thing you're my counterparty
This will be the most autistic thing I read all day

>>57374350
Yeah we've all heard of photoshop idiot, nobody is going to believe that's real loser!

>>57374350
Why? The only people who care about it are /pol/ keks who spend all day being judgemental on the internets.

>>57371319
>ticking time bomb

>>57375709
kek, no one tell him.

>>57372728
i don't stake, but iirc, nethermind and besu are the best options. in theory any minority client is fine, because the nature of penalties means you lose very little
i think erigon isn't maintained anymore, and reth is kinda new. but again, i'm not a staker

>>57372799
based

>>57372884
no argument there
"not be at risk" from this particular penalty scheme
overall risk, staking has never been worth the rewards

Ummm wtf eth bros? Can someone debunk this fud? I'm panicking and thinking of selling

>>57376240
Can't be debunked
ETH exploit and to 0 soon

>>57371393
> Harming stakers when errors are made is a really bad choice
you dont understand staking
you are trusting your asset with the nominator / validator to use your asset as a vote to participate in a system where if they act poorly or maliciously the system is designed to punish that actor

anyways
if its not proof of work
its a scam

>>57376240
I hope you're not baiting, but in short: this is not oversaturated FUD but definitely a very real concern.
If a supercatastrophic event does happen that involves all, or even a considerable part of validators running geth (that make up about 80% of all nodes at this moment) to start following a buggy chain,
the loss of ETH and the chain not being able to finalize for hours would pretty much be the worst thing that has ever happened to ethereum and it's unlikely that confidence in it's ecosystem will ever be the same again.

https://dankradfeist.de/ethereum/2022/03/24/run-the-majority-client-at-your-own-peril.html this is a very good read that goes into the technicalities of such an event occurring

But I highly doubt that we will let this happen. If you read the article I linked, you will see the writer point out that the Prysm consensus client had a 2/3 supermajority at the time of writing (May 2022)
We are now already sitting at a MUCH MUCH safer 40% share, which makes that catastrophic event I mentioned earlier impossible to happen from a consensus client bug.
The only thing that needs to happen now is a repeat of that, but for the execution client. Personally I've been running a geth node for years, but I will be switching to another exec client this week because of the recent nethermind bug.
The problem is that we need the institutions to do the same, as they are the ones running a majority of nodes right now. Are they gonna do it? Nobody knows.
This validator client "vulnerability" has been known for years though, and I am pretty much 99.9% sure that an ethereum-ending Geth bug like that should never see the light.
Not to mention that I'm quite optimistic about node operators switching clients as the issue is becoming more and more prevalent, as they did with the consensus client last year.