Sepolia beacon chain #526
Comments
|
Did you discuss the option to have a small, permissioned validator set, i.e., something along the lines We should have one testnet that allows developers to expect a fairly stable application-test environment for dapp and contract developers (Sepolia?) whereas we have another one that can be used for consensus and execution client teams to test protocol mechanics (Goerli/Prater?). A small validator set would reduce the friction to maintain keys and validators long-term (IMO). |
|
We considered a small validator set for Sepolia but didn't want to miss the chance to have a new beaconchain with a large genesis set. The main reason is that in order to expand Prater to 2xmainnet we'd need to perform a non-insigificant number of deposits with an increasingly reducing supply of goerli ether. By making the Sepolia beaconchain much larger, we can stop trying to increase the validator set on Prater and reserve that goerli ether for other purposes. If we're fine continuing to increase the prater validator set size, then I agree that we can have a small network for sepolia that's a lot easier to run. |
|
Pro/Cons of various beaconchain sizes: Large beaconchain (2x mainnet):
Cons:
Small beaconchain
Cons:
|
|
As discussed on ACD, Sepolia will have a minimal validator set. I'd say 100 validator keys per entity taking part (Assuming that every client team wants to take part) => Validator set of 1212 (9 client teams + EF + Attestant/Vouch + 12 for EF security + Alex). Having a validator set of 100 per entity means that we can assign low number of keys (like 12) to entities such as EF security without them representing a significant portion of the network. If there are entities outside of client teams that would like to take part, please comment below. |
I'd be happy to participate on behalf of the Goerli community. |
|
On #528 we agreed that we should launch the sepolia beacon chain as soon as (reasonably) possible, and to confirm this with CL teams on the next CL call. |
|
Just wanted to leave a comment to say that I'm setting up a some infra to run a lot of validators on testnets (starting with prater) and would also be happy to also run on sepolia. |
We would be happy to participate too. |
|
I just read the CL Call#88 notes and noticed that the sepolia beaconchain "will be permissioned, with ~2k validators." (link) Aside from the 12 validators that the EF CL Security Research team is slated to get we were planning to run other validators to test various things like slashing. We have been mining sepolia ETH for the last few days to seed this other set of validators. If the validator set will be permissioned I would like to request that we get more than 12 validators (maybe ~50) or that there is some method provided for us to deposit our mined SEP in a permissioned manner. The plan is to have 12 honest, longer term validators that are running various sanitizers (MSAN, ASAN, UBSAN, TSAN, etc.) and to have others that will either be slashed on purpose or otherwise not reliable in some fashion. |
Please use Goerli/Prater for this. |
|
I think the idea is that they'd do this for each testnet, so they'd do this for goerli as well :) |
|
What for? |
|
I think Goerli will suffice for this but I am not sure we can mine Goerli as it is POA. |
|
|
Leaving a comment for discussion/approval (leave an emoji if you agree). Timeline:
Current validator set participants:
(Potentially some lido node operators will be onboarded later on to test their setups) We will be using a token gated deposit contract, we will have the ability to make new deposits later on if we want to onboard anyone. In case anyone is missing, please do leave a comment. |
|
I would like to run 100 validators on Sepolia. I'm currently running a validator on Kiln and I previously ran a validator on Kintsugi, now I would also like to run some validators on Sepolia. I have over 200K Sepolia ETH. |
|
@Emilijus436 Do you represent a staking pool or have a specific reason you'd like to be a validator on sepolia? We'd ideally like to keep the sepolia validator set closed to known entities that currently stake significantly on mainnet (and have done a proven good job). We want to keep it closed since we'd like to provide a great, stable testnet env for developers and it's a bit easier if the validator set is small as possible. |
|
@parithosh I started mining Sepolia on November 22, 2021 so that I can run a large amount of validators on Sepolia. For a long time, I was the only miner present on Sepolia. Getting 32 Goerli ETH is really hard right now. |
I can trade you Goerli Ether for Sepolia Ether 1:1. That might be the easiest for now? Send me a mail goerli at q9f dot cc or dm me on matrix @afr:tchncs.de (or join #goerli_chat:gitter.im). wdyt? See also eth-clients/goerli#106 |
|
@parithosh I'd take 100 validators to help with testnet stability and also client research. I believe I am the "Alex" in the top comment :) |
|
Simply Staking is part of the Ethereum Lido Node Operator set and we were wondering whether we could join the validator set to help in testing out the merge. Thanks in advance and have a nice day! |
|
@parithosh for anyone signing up to run validators on this network, I think that we should be clear that this is a long-term commitment rather than an opportunity to test the merge. As such, a confirmation that they are happy to run their validators in perpetuity would be a good idea. |
|
@mcdee Agreed, All of the people who are in the validator set reached out personally as well and I've clarified the requirements with them. The expectation is that the validators need to be run in perpetuity or handed over/exited when circumstances change. |
|
We are currently struggling to find our SepoliaETH to run our 100 Sepolia validators. Have a nice day! |
|
@nalepae You don't have to acquire any sepolia eth for the validators. I'll be creating the genesis state with your deposits baked in. Can you DM/Email me to discuss it? |
|
Sure! |
|
We would like to run 100 validators for sepolia.dev to use in funding our faucet |
|
We've unfortunately already generated the genesis config. But I'll be happy to redirect the fee_recipient address of the EF nodes to a faucet/group of faucets :) |
|
The genesis configs for sepolia can be found here: https://github.com/eth-clients/merge-testnets/tree/main/sepolia The validator keys have been sent to all parties that requested it, in case you haven't heard from me - please do ping me ASAP. |
|
@parithosh I am willing to be a sepolia genesis validator as I represent Obol core devs team. In Obol Network we run different validators with different testnets to develop and test our DV technology charon. We are willing to do the same with sepolia and we are ok to run validators for the long-term. |
|
I know it may be too late, but even though I was anticipating Sepolia, I only just noticed this discussion :-\ We've been mining Sepolia ETH for a few months already at SSV Network. @parithosh Is there a chance for a last day genesis inclusion? Regardless, we have a surplus of a few 1000's Sepolia ETH, if anyone needs 😉 |
|
@moshe-blox Unfortunately its too late to change the genesis state. But we can onboard post genesis and pre-merge. I'll create a guide and reach out to all the interested parties soon. Could you give me an idea of the scale of operations you run or what operator you represent on the SSV network? What's the best way to contact you? |
|
@parithosh Thanks, that's great to hear :-) At first, we're gonna run the Sepolia validators under Blox Staking to properly prepare it for The Merge. (since it already operates ~2k validators on mainnet) (Blox Staking runs a custom validator client with a non-custodial signing mechanism over Prysm & Lighthouse Beacon nodes.) Later, we might migrate them to SSV Network for internal testing, and then just migrate them back under Blox Staking forever (or, in coordination with you, exit / hand them over to someone else who needs.) Best way to contact me is via moshe@blox.io or via |
|
I'll be reaching out to the people who indicated interest in being validators pre-merge on sepolia. In the mean time, Since the sepolia beacon chain has successfully had its genesis we should close this issue. I suggest we move to this issue for long term requests regarding being a validator on sepolia: eth-clients/sepolia#12 I have made some initial documentation regarding the process to be a validator on sepolia, the requirements as well as how to use the tokens to join the permissioned validator set: https://notes.ethereum.org/zvkfSmYnT0-uxwwEegbCqg |
|
Hi @parithosh |
|
@aertoria You can run a CL without a validator and only follow the beacon chain. Any client will work. I personally run a Geth + Lodestar tuple on Sepolia. |
EDITED
from @parithosh:
As discussed on ACD, Sepolia will have a minimal validator set. I'd say 100 validator keys per entity taking part (Assuming that every client team wants to take part) => Validator set of 1212 (9 client teams + EF + Attestant/Vouch + 12 for EF security + Alex).
Having a validator set of 100 per entity means that we can assign low number of keys (like 12) to entities such as EF security without them representing a significant portion of the network.
If there are entities outside of client teams that would like to take part, please comment below.
The text was updated successfully, but these errors were encountered: