Thank you for leaving an insightful comment on this topic. I find this much more engaging than the typical drive by crypto sucks comment. Running a node comes with an interesting set of problems, there are additional problems when considering staking as well. Something to note is that when staking[0] it is suggested to use 32ETH. At today's market rate[1] that is nearly $50,000

[0] https://ethereum.org/en/staking/#how-to-stake-your-eth [1] https://finance.yahoo.com/quote/ETH-USD/history/

Clarification: it's not a suggestion but the minimum stake for a single independent validator on ETH mainnet.

It's also worth noting that staking pools exist that allow people to stake with less than 32 eth. Coinbase also allows you to stake.

The optimistic view of this is that you can run a node in a distributed payment and compute platform for a few thousands bucks. Consider the alternatives: the infrastructure running PayPal, the massive and archaic automated clearing house systems used by megabanks, whatever it is Western Union does to move money. You can participate in a distributed, open source alternative for what amounts to an affordable price for those in developed countries.

I am not claiming you should do this! What I'm attempting to illustrate is that this momentum towards freer and more open systems is what drives the technical optimism in cryptocurrencies and blockchains. There seems to be a Moore's law effect dropping the price of what is currently gatekept by governments and megacorporations. Hard work is once again creating exciting innovations that at least have the potential to improve the world.

Can you elaborate more on the distributed compute platform part? How will that work and prevent down-time issues and other mistakes that will cause slashing?

I think they're referring to being a single node in a distributed compute/payment platform.

> a few thousands bucks

I, too, consider 50000$ as just a few thousands of my millions in savings. /s

  Seems like an epic pain in the ass for a 4.1% yield 

I complain about gas prices as much as everyone else, but that seems pretty good. Does anyone know how much the typical miner was making with POW?

At the height of the last crypto boom, with elevated GPU prices, you could expect to make back your investment (with electric costs) in about 200 days. Assuming that means a 2.5% per-day return, that could be extrapolated to be ~60% return for the rest of the year, which figures out to about 27% return for the entire year. That's back of the napkin as it assumes you only run the machine for a year (warranty period of the equipment), crypto prices stay the same, and miner competition stays the same. It also doesn't account for facilities costs, salvage value of the equipment, or the (relatively limited) labor costs of setting up and maintaining the computers.

"Hive notes that the company’s ethereum mining operations have “historically generated 3 to 4 times more revenue per megawatt than bitcoin mining.”

https://news.bitcoin.com/publicly-listed-miner-hive-plans-to...

    epic pain in the ass for a 4.1% yield

Isn't the yield dynamic? If it is dynamic, then it will simply adjust to whatever price makes staking as painful/lucrative as any other investment. There is no free lunch.

Can anyone clarify this?

By which process does the market affect the yield on staking ETH?

The yield varies with a few factors, chief of which is how much total ETH is staked. More staked ETH means a lower yield. Here's a calculator: https://ethereumprice.org/eth-2-calculator/

The yield a staker receives from the protocol is composed of 2 parts *:

- Issuance

- Tips

The issuance is defined programmatically by the protocol and is dependent on the number of validators (stakers). There is a formula for it that basically scales with the square root of total amount of ETH staked. This is the part that gives ~4% nowadays. Given a number of stakers this part is guaranteed yield as it comes from new issuance. It's still subject to market dynamics in that if there are very few stakers, the yield is more interesting. If there are a lot of stakers the yield is lower.

The tips are completely market dependent and are a function of network activity and the price of ETH. When you pay a transaction you can include a tip for your transaction to be included faster. Low levels of activity lead to lower tips, high levels of activity result in higher tips. Tips are basically denominated in USD but paid in ETH, so high price of ETH leads to lower yield from tips, low price of ETH leads to higher price of ETH from tips. This part of the yield is obviously completely dynamic and subject to market forces.

With current (last 3 months) levels of activity the yield a validator can expect to experience is around ~5.5%. If network activity were to go back to Q4 2021 the yield would raise to ~25% at current prices.

*Technical note: There is a third part which is an effective yield resulting from burnt ETH and even a forth part that is outside of the protocol called MEV.

The current factors are strongly "market dependent":

- Staking base reward, depends on the total amount of ETH staked by all individuals - Priority fees and MEV fees from users

Currently, as blocks are still produced by PoW, user fees go to miners instead of stakers.

That's what I am afraid of the most. Do some devops mistake and get penalized. Are there any protection mechanisms to help you with that or are you totally on your own?

There are slashing protection features built into all clients, but the general advice is to don't do devops.

You only get slashed if you run malicious custom code, or if you run the same key on two machines at the same time.

The latter is a mistake that's very easy to make if you try to build auto-failover.

So don't build it. My validators operate with 'screen' and shell scripts. If I need to do anything (3-4 times a year), I do it manually.

I guess couple of questions out of curiosity:

- machine dies at 4am, what happens? do you manually restart process when you wake up? does this down time cause slashing?

- what mechanism prevents you from running 2 identical validators by accident?

An explicit goal is that running a validator at home for the enthusiast should be feasible. A couple of days total downtime per year for an individual validator should approximately only miss out on the rewards it would have generated by being available for the same period. Penalties increase non-linearly with downtime and number of validators unavailable at the same time (meaning assuming all else equal, being down for 100h at once is worse than being down 1h/d for 100 days, and a whale with 10 validators going down for 10h at the same time will lose more than an individual validator going down 10h at 10 different occasions. This is not contingent on those validators being attributable to the same entity)

If you have monitoring that alerts you for updates and when your node is unresponsive and can respond to downtime requiting intervention within a day or two you should still end up ETH-positive.

Thank you so much for this information

The rough breakeven time is 42% of the time. If your validator is online more than that you are making money, less than that you lose it. So the protocol is rather forgiving in uptime. That's why it's much better to not run failover of any kind. You can afford to be offline without problems but if your failover fails and you run two instances you get slashed.

There is also a penalty that kicks in if many validators fail at the same time resulting in over 1/3 of the network not being available. In that case penalties are also increased significantly. So if over 1/3 of the network were to be run in AWS those validators would risk heavy penalties if the region were to suffer some downtime.

>"There are slashing protection features built into all clients, but the general advice is to don't do devops."

I'm curious how the big crypto companies are managing these large fleets of servers without doing some form of Devops? Looking at the docs there's a stack that needs to be deployed and maintained:

https://ethereum.org/en/developers/docs/nodes-and-clients/ru...

They use special signing software (dirk or web3signer) that can have the keys split accross several physical machines. If they have separate access credentials for each system, it makes it difficult for anyone to have access to the complete keys. Then a central validator client can request signatures from each signer to complete attestations and block proposals.

This didn't really answer my question though. There's other software besides key management services that run on those nodes. How is that being managed if these orgs are not doing some form of Devops?

You only have to do the work of setting up the validators once, and then you get the 4.1% yield indefinitely, with only minor maintenance occasionally.

Not ìn the mood to read it, but why wound your stake get lost of you restore the wrong backup?

If your node operates on wrong or outdated data, it is very much possible to stake on invalid or wrong blocks, leading to the network acepting your stake, declaring you a bad validator, and witholding your stake (because that is axactly what the stake is for: don't stake on invalid data, or lose the stake)

Easier than shoveling coal

You are confusing "running a node" with "running a validator." You can run a node for free, without any staked ETH. If you want to participate in the protocol rewards, then yes you will have to do some work to receive them, and place your capital at risk of being slashed if you are not securing the network correctly. If you are a solo validator and make an honest mistake, the amount you will be slashed is probably very low.

If you want to receive slightly less rewards but do less work, you can delegate your stake to a staking pool or service and they will take a fee, see Lido and Rocket Pool.

Aside from protocol rewards, there is also rewards that may come in the form of priority fees or 'tips', and rewards that can come in the form of MEV. See MEV-Boost which validators will be incentivized to run, which can almost double their annual yield.[1]

[1] https://boost.flashbots.net/

Why would one want to run a node (and paying for its electricity and physical costs) if they won't get any reward for it?

Honest question, not being facetious, I'd like to understand the use case.

Same reason people will host and run services around IPFS, BitTorrent, Tor, Matrix, Mastodon, game rooms, and other servers even though they don't get a financial reward, and even though it may even cost them money or their time. They might just want to be part of the network, supporting it in their own way.

The protocol rewards are there to create a stronger incentive, so that it isn't just hobbyists and geeks paying out of pocket each month to secure a multi-billion dollar network.

Truly banking the unbanked.

Interestingly, this further incentivizes staking centralization, with coinbase and its ilk being the safe options.