Eth2 progress update (as of 5/8/2021)

Source | What's New in Eth2

Author | Ben Edgington

Source | What's New in Eth2

Author | Ben Edgingtonfirst level title。

As of the publication of this article, there are more than 135,000 active validator nodes on the Beacon Chain, and 4.45 million ETH pledged, worth $18 billion. 🤯 And the ratio of new pledged deposits has also increased recently.

The Beacon Chain has been running smoothly since a glitch two weeks ago. Then the Prysmatic Labs team conducted a detailed review of the mainnet accident. ECN translated the article:

"Ethereum 2.0 Mainnet Accident Review"

first level title

If users want to keep track of how well their validator nodes are performing their validating responsibilities, Paul Hauner has published a spreadsheet tool with instructions on how to use it. This is more comprehensive than the single "effectiveness" data provided by Beaconcha.in, and users can also see how their node's performance compares to the average.

Have readers heard of Secret Shared Validators (SSV, Secret Shared Validators)? If you haven't heard of it, it's time to learn it. Alon Muroch wrote the most recent SSV Phase 1 test.

first level title

Rayonism project

Rayonism is a style of Russian abstract art, and it is also an ongoing expansion project of Ethereum, focusing on the eth1->eth2 merged testnet and the development of sharding. The project is underway at EthGolobal's Scaling Ethereum hacker Panasonic, but will surely expand to other locations.

Lukasz Rozmej from the Nethermind team made a very useful video tutorial showing how to set up a merged testnet using Nethermind (Eth1/execution layer) and Teku (Eth2/consensus layer). There was big news: Steklo, the first post-merger developer testnet, went live for a few hours on April 30th. 🎉

Steklo means "glass" in Russian, and it is considered fragile. The same is true (only running for a few hours). To be fair, all teams went into development without preparing any test vectors, so this was a first, blind attempt. Decrypt posted a related article and wrote some previews about Steklo. The testnet showcased a total of 12 client combinations, pairing Nethermind, Besu, and Geth/Catalyst with Teku, Nimbus, Lighthouse, and Prysm. The client portfolio is shown in the figure below:

cr: tweet by @protolambda

Lukasz provided a summary on the Rayonism Discord channel:

➤ Lighthouse had consensus issues and was forked from the beginning (state root issue), but on this fork, Lighthouse worked well with engines on all execution layers

➤ Besu or Teku sometimes cannot reach consensus with other clients, but eventually they do. If it splits, it becomes two sets: (Teku-Geth, Teku-Nethermind) and (Teku-Besu, Nimbus-Geth, Nimbus-Besu, Nimbus-Nethermind).

➤ Nimbus has a communication problem with Nethermind at the transport layer, but this requires more investigation by the Nimbus and Nethermind teams. Of course, it works fine most of the time - sometimes a reboot is required

In short, there is a lot going on! However, a lot of things went in the right direction, and overall the testing process was very encouraging. After some fixes, Lighthouse can then reach consensus with Teku and Nimbus.

Last week we prepared some commonly used test vectors so that clients can debug individually before the next common test. The plan is to have a longer running merged developer testnet early next week.

first level title

Altair is a relatively minor beacon chain upgrade planned for mid-year.

According to a recent developer call, the client team is making fairly good progress implementing the Altair specification. We take this opportunity to draw up a program timeline (no commitment):

➤ Try to fork the current testnet by the end of June. I imagine Pyrmont will be the first, so it's best to migrate to Prater as soon as possible to avoid instability.

➤ Deploy to the beacon chain in late July/early August.

first level title

text

First, the transition process (also known as the docking process). We need to figure it out, write code and try it out in testnet/devtestnet. Second, the synchronization algorithm; the design work has already begun, and implementation and testing will follow. Technical optimizations in terms of block proposals, communication channel protocols, whether JSON-RPC or REST; JSON-RPC modifications for users (methods to add finalized blocks). Write specific EIPs related to execution layer modifications. Solve the BLOCKHASH randomness problem; conduct multiple rounds of testing. Blockchain tools, and access to proof that they work. The update of the infrastructure, the main question is what the block explorer will look like. We still have a lot of work to do, but we're optimistic that we'll deliver the merger before the end of the year. Caution sees delivery in the first quarter of 2022, but few expect the merger to be later than that.

Kotal was posted by my former colleague Mostafa Farghaly:

It supports nodes for Teku, Nimbus, Lighthouse and Prysm, IPFS and other Web3 stuff.herefirst level title

Paraphrase

here

Alex Stokes made a useful validator lifecycle diagram and shared the source code to generate the diagram.

In Viktor Bunin's eth2 update series 014, he offers some perspective on the issue of what it would look like if an accident like last week's happened on the beacon chain after the merge.

first level title

After the Beacon Chain incident, client diversity continues to be a hot topic. Phil.eth wrote a very good overview "Client diversity is good for the health of the beacon chain": Don't just focus on the clients you run. Superphiz is also very concerned about the issue of client diversity. He released a video "Ethereum Client Diversity & How to Solve the Problem". Just in time, I would like to remind readers that Superphiz's SOTs (State of the Stake) is a very good series.

text

@VitalikButerin: "Sounds reasonable."

Vitalik has a proposal to allow validators to change their signing keys: it provides a way for users to delegate to staking pools to participate in staking, while retaining management of their staked assets and being able to take responsibility for their own slashing events.

secondary title

• Agenda

• conference video

• implementers meeting

The 63rd meeting of client implementors was held on May 6th.

My shorthand, and Alex Stokes' twitter notes

secondary title

• Agenda

• conference video

• merge meeting

conference video

There are many detailed discussions and plans for Rayonism. Another useful piece of information: Dmitry Shmatko has been working on validator balance withdrawals and is about to work on the next phase of allowing partial withdrawals (staking the validator's staking rewards, but keeping the principal). There are plans to test the feature on the Rayonism network at some point.

ECN's translation work aims to deliver high-quality information and learning resources for the Chinese Ethereum community. The copyright of the article belongs to the original author, and the source of the original text and the ETH Chinese website must be indicated when reprinting. For long-term reprinting, please contact eth@ecn.co for authorization.