--- tags: newineth2 description: The latest update on Ethereum 2.0 development image: https://benjaminion.xyz/f/favicon-96x96.png --- <style> a {text-decoration: underline;} a {color: #0000ee;} a:visited {color: #551a8b;} .ui-infobar {visibility: hidden; padding-top: 0;} .community-button {visibility: hidden;} .markdown-body {padding-top: 0;} </style> # What's New in Eth2 - 5&nbsp;November&nbsp;2021 ![My avatar](https://benjaminion.xyz/f/ms-icon-144x144.png =32x32) Ben Edgington (Eth2 at [ConsenSys](https://consensys.net/) — all views expressed are my own) Edition 81 at [eth2.news](https://eth2.news/) ## The Beacon Chain ### Altair success! Altair, the beacon chain's first upgrade passed by quite nicely last week. The build up has been very long, but in the event it was pretty straightforward. The [EthStaker livestream](https://youtu.be/Go4aacg-960?t=344) is available if you'd like to re-live the event. Song-a-day man, Jonathan Mann, [sang for for the occasion](https://twitter.com/songadaymann/status/1453036427558719496), and there was an Altair [POAP Art canvas](https://twitter.com/Logic_Beach/status/1453320728439365634). Ahead of the upgrade, it was quite unclear how many operators had updated their clients to be Altair compatible. As the fork epoch passed, attestation rates dropped from 99.8% to around 95%, which was at the better end of expectation. Over the next couple of hours, rates climbed back up to 98-99% as the remaining operators realised that they had dropped off the network and scrambled to update. There was one oddity. With 98% participation rates, we would expect to see 98% block production rates. That is, only 2% of blocks not being proposed. But in the hours following the upgrade, somewhere around 20% of blocks were missing, which is extremely strange, and affected the reward rates for the rest of the network. After some detective work, a very large staking operation was identified as the culprit ([reportedly](https://twitter.com/christine_dkim/status/1453720991272620056) a [Coinbase](https://www.reddit.com/r/ethstaker/comments/qh8dni/tons_of_miss_proposals_after_altair_is_the/hib4tcc/?context=3) platform). The exact cause is not public, but it's likely something to do with having updated their client software but not having updated their signing server. (Attestions did not change over the Altair fork, so the signer could still handle those, but the block structure did change.) In any case, within 12 hours or so, everything was back to normal and the beacon chain is once again chugging along very nicely. Many stakers have noticed either somewhat [lower](https://old.reddit.com/r/ethstaker/comments/qkoch5/income_lower_than_usual_during_last_few_days/) or somewhat [higher](https://old.reddit.com/r/ethstaker/comments/qjcqld/why_so_much_income_over_the_last_3_days_the_one/) rewards under Altair than previously. In fact, the long-run total reward is unchanged, but the apportionment has changed: - pre-Altair 97% of your reward was from daily attesting, with 3% from occasional block rewards; - post-Altair 84% is from daily attesting, with block proposals boosted to 12.5%, and the rest coming from the new sync committees. Being in a sync committee is very rare (on average just for one day in about 17 months), but for the lucky few it is a nice income boost. More info on this from [Michael Sproul](https://hackmd.io/@sproul/altair-rewards). And here's [Pintail's deep dive](https://pintail.xyz/posts/modelling-the-impact-of-altair/) again from a while back. Bottom line: most of the time your rewards will be 13% down, but occasionally much higher. For good general overviews of Altair, here's an [explainer from MyCrypto](https://blog.mycrypto.com/eth2-0-beacon-chain-altair-upgrade/), and Christine Kim's [research article](https://docsend.com/view/ak2q58qutvcx8s4f) digs a little deeper. ### Kintsugi is coming! With Altair behind us, all eyes are now on the Merge: the point at which we finally move Ethereum onto Proof of Stake. Since the success of the Amphora Merge devnet at the in-person [workshop](https://hackmd.io/@benjaminion/wnie2_211008#The-Merge-workshop) a month ago, we're now working towards a long-running testnet [called Kintsugi](https://twitter.com/TimBeiko/status/1455631393724067846). Kintsugi will be a step up in production readiness for the clients, and aims to implement a near-final Merge spec candidate. - Kintsugi [milestone tracker](https://notes.ethereum.org/@djrtwo/kintsugi-milestones) - Kintsugi [spec v1](https://hackmd.io/@n0ble/kintsugi-spec) (The latest Merge spec version will always be at http://spec.merge.wiki/. A v2 update is expected on Monday.) Here's a summary of the latest on The Merge from the [Other Side](https://hackmd.io/@timbeiko/acd/https%3A%2F%2Ftim.mirror.xyz%2FsR23jU02we6zXRgsF_oTUkttL83S3vyn05vJWnnp-Lc) (no, not Polkadot, I mean the Eth1 all core devs). ### Fork choice vulnerabilities From time to time theoretical attacks on the beacon chain's consensus mechanism are published, and I've been [fairly dismissive](https://hackmd.io/@benjaminion/wnie2_201018#Balancing-attack) about the feasibility of past attempts. However, a new set of attacks on the beacon chain's fork-choice rule [has been published](https://arxiv.org/abs/2110.10086) that are more concerning. The goal of the attacks is to orphan (exclude) blocks proposed by other validators, perhaps to steal their MEV after The Merge. Caspar, one of the authors, and an Ethereum Foundation researcher, gave a very accessible [short talk](https://vimeo.com/637529564) on the vulnerabilities at LisCon, and wrote a helpful [tweet thread](https://twitter.com/casparschwa/status/1454511836267692039). The interesting new issue is so-called "ex-ante" reorgs. In short, - _ex-post_ reorgs try to remove a previous block from the chain. This turns out [to be hard](https://www.paradigm.xyz/2021/07/ethereum-reorgs-after-the-merge/) under proof of stake. - _ex-ante_ reorgs try to prevent a future block or set of blocks being added to the chain. These are the new thing, and (in theory) require very little resource to pull off, just a good sense of timing. The good news, as Danny notes in [Finalized no. 31](https://blog.ethereum.org/2021/11/02/finalized-no-31/), is that we already have fixes for these vulnerabilities (and others). The first fix, and the one we've agreed to implement pre-Merge, is [proposer score boosting](https://github.com/ethereum/consensus-specs/pull/2353). This has been well studied, and would appear to be effective against all the easy attacks. A more recent idea, [proposer's view merge](https://ethresear.ch/t/change-fork-choice-rule-to-mitigate-balancing-and-reorging-attacks/11127), may be even more robust. ## Tooling Alex Stokes is [working on](https://twitter.com/ralexstokes/status/1454014021322121218) an implementation of the [ethereum consensus spec in Rust](https://github.com/ralexstokes/ethereum_consensus), primarily for R&D use cases. It's not supposed to be a competitor for Lighthouse. Franck Cassez and Joanne Fuller (both ConsenSys), with Aditya Asgaonkar (Ethereum Foundation), have published [Formal Verification of the Ethereum 2.0 Beacon Chain](https://cs.paperswithcode.com/paper/formal-verification-of-the-ethereum-2-0), a paper covering all aspects of their work on this important project. It's a good read! ## Staking After a slight pause, the countdown to the Rocket Pool launch [has resumed again](https://twitter.com/Rocket_Pool/status/1456421975396552707)! T-minus-zero is the 9th of November, 00:00 UTC (midnight on the 8th, I guess), and there will of course be an [EthStaker watch party](https://www.youtube.com/watch?v=CM1decziXkQ) starting on the 8th at 23:00 UTC :rocket: [Teku + Infura to 2.0: How to Handle Attestations](https://blog.infura.io/teku-infura-eth-2-how-to-handle-attestations/) has some important tips if you are staking while using a remote beacon node at Infura (or want to try this). There's also a nice little teaser in there for Teku's pioneering tree-storage archive format: it saves space while hugely boosting the speed of historic beacon chain state enquiries. I've just finished syncing up a new archive node with it and look forward to putting it through its paces next week. Meanwhile, here's a solid thread on [client diversity](https://twitter.com/technocrypto/status/1453128929418661892) from Jeff Coleman. ## Media and stuff The latest from the [treasure trove](https://www.youtube.com/channel/UCD9iiIwTRtLDYcEWONs2Q3A/search?query=PEEPanEIP) that is PEEPanEIP: - Paul Hauner of Lighthouse on [Merging Ethereum](https://www.youtube.com/watch?v=S1KFIVue248). - Coming soon, Sam Wilson on [Ethereum Execution Layer specs](https://twitter.com/poojaranjan19/status/1455533103204773902) 2021 [Ethereum Shanghai Meetup](https://twitter.com/EthereumCN/status/1455096930296422412): transcripts (in Chinese!) of talks with, - TimBeiko and Ping Chen on [EIP-1559 & The Merge](https://mp.weixin.qq.com/s/2PsM36E8f1h8leDHc8h7RA) - Hsiao-wei on [The Merge](https://mp.weixin.qq.com/s/SgkxXip6nOu1DLExH-POBw) - [Researchers on MEV](https://mp.weixin.qq.com/s/OC7vma_r0ZIKiqCTR4XxBg) ## Research Lots going on around MEV post-Merge. The Flashbots team has published [a proposed architecture](https://ethresear.ch/t/mev-boost-merge-ready-flashbots-architecture/11177?u=benjaminion) in which validators can acquire blocks from specialist third-party block builders (such as Flashbots). I know that they have been discussing this with client teams and the EF, among others. I am super happy that this issue is being looked at. The background concern is that we can assume large staking operations/exchanges will have the resources to seek out MEV and boost their own returns. This puts individual stakers, without access to those resources, at a disadvantage, and increases pressure towards centralisation of staking. The only realistic way to counter this is to make MEV opportunities available to all, and that is what Flashbots seeks to do. Good things about the proposed design are that it does not need changes at the consensus layer, so can be in-place with The Merge. It also nods towards being an open architecture that any block builders and intermediary relays could participate in. It is, however, quite complex, requiring several new components. And while it looks credibly neutral, I fear that there will be no effective market place for block builders or relays that will appear, and that Flashbots will end up with a de facto monopoly on block production. There are also technical concerns about the relay being the entity that delivers blocks to the p2p layer, which [Terence picked up on](https://ethresear.ch/t/mev-boost-merge-ready-flashbots-architecture/11177/9?u=benjaminion) as well. In an MEV world, all solutions incur a degree of of compromise, it seems. But, as I put it to Stephane (`thegostep`), if we avoid centralising staking but end up centralising block production, then I'm not sure we're winning. Personally, I would much rather see something closer to Flashbots' current model of simply delivering bundles or blocks directly to validators for them to propose, and kicking out those that abuse the trust. At least until we have a fully [trustless design](https://ethresear.ch/t/proposer-block-builder-separation-friendly-fee-market-designs/9725?u=benjaminion). Though scaling this up to a network as large as the beacon chain is a challenge, for sure. Even better would be for Flashbots to have a number of credible competitors so that we could have a proper market place for block builders - this would allay many concerns, I think. Also on [ethresear.ch](https://ethresear.ch/): - [The Safe Merge Proposal](https://ethresear.ch/t/the-safe-merge-proposal/11057?u=benjaminion), an alternative to relying on Terminal Total Difficulty on the PoW chain to trigger The Merge that allows repeated attempts at performing The Merge in case of failure. - [Beacon Chain Light Client Classification](https://ethresear.ch/t/beacon-chain-light-client-classification/11061?u=benjaminion), the beginnings of a taxonomy of light client types. ## Regular Calls ### Implementers Call #75 took place on the 4th of November. * [Agenda](https://github.com/ethereum/pm/issues/412) * [Video](https://youtu.be/9U_xj_zCMYg?t=69) * My [quick notes](https://hackmd.io/@benjaminion/B1wBrwbPF). For the next few weeks, the first part of these calls, and the first part of the Eth1 All Core Devs on the alternate Fridays, will be "Kintsugi office hours", with devs of both Eth1 and Eth2 clients having the chance to discuss progress and ask questions. Not much Kintsugi progress to report this time since we're just starting. But we agreed to start the first weekly devnet on November 18th, aiming to stand up the long-lived testnet at the start of December. In other business, we agreed to wind down the old Pyrmont testnet. Client teams will be turning off their validators soon. We also agreed that we would implement the "proposer score boosting" fix in clients before The Merge. This is to remove the threat from the fork choice vulnerabilities [discussed above](#Fork-choice-vulnerabilities). Teku already has this (behind a feature flag). ### Eth1 All Core Devs The ACD held meeting #125 on the 29th of October. - [Agenda](https://github.com/ethereum/pm/issues/401) - [Recording](https://www.youtube.com/watch?v=5cOWjMAuReI) - [Tweet notes](https://twitter.com/TimBeiko/status/1454188180609986564) from Tim Beiko. Not much Merge-related on this one. Nonetheless, it was an [excellent call](https://twitter.com/benjaminion_xyz/status/1454128461883314177), and I enjoyed it a lot. ### The Merge Community Call #1 Merge Community Call #1 [took place](https://github.com/ethereum/pm/issues/402) today. But there was no livestream as far as I could see :disappointed:, and client devs weren't invited. So I have no idea what was discussed. Stop press: [apparently](https://twitter.com/TimBeiko/status/1456668747704332291) the [recording](https://www.youtube.com/watch?v=_kfS9jAUY6g) is out and a transcript is coming. ## And finally... A toast :clinking_glasses: Farewell Proto! Yep, Protolambda is [joining Optimism](https://twitter.com/protolambda/status/1453899762613334019). I have frequently marvelled at the way that Ethereum draws in incredibly talented and driven people. I first became aware of Proto at the start of 2019, with his extraordinary, really quite bonkers, [beaconchain schematic](https://github.com/protolambda/beacon-schematic/blob/master/beacon_chain.svg). From there he went on to play an critical role in bringing the beacon chain to reality, often bridging large gaps between theory and engineering, spec and implementation. One of our key data structures, the fork-choice protoarray, is named for him. I could go on and on; his own [tweet thread](https://twitter.com/protolambda/status/1453899762613334019) barely scratches the surface of what he's achieved. It is hard to overstate Proto's influence on where we are today. But it is in the nature of super productive, super smart people that they run towards the hardest problems. That's what brought Proto into Eth2 in the first place. Now that Eth2 is on an excellent footing, scaling Ethereum via roll-ups presents some of the current hardest challenges. So this is not a surprise. And Proto is not going far! I don't normally write about people's comings and goings, but I did want to mark this one. Proto, we salute you :raised_hands: * * * [![[Twitter]](https://benjaminion.xyz/newineth2/img/twitter.svg =40x40)](https://twitter.com/benjaminion_xyz) Follow me on [Twitter](https://twitter.com/benjaminion_xyz) to hear when the next edition is out. [![[RSS]](https://benjaminion.xyz/newineth2/img/rss.svg =32x32)](https://benjaminion.xyz/newineth2/rss_feed.xml) We also have an [RSS feed](https://benjaminion.xyz/newineth2/rss_feed.xml). [Advertising](https://hackmd.io/@benjaminion/eth2_news/https%3A%2F%2Fhackmd.io%2F%40benjaminion%2Fadvertising) on this newsletter. [Give Feedback](https://docs.google.com/forms/d/e/1FAIpQLSfkESc4CmNfRGHHjWfNeF3ceLwrXDvynetda4sKfJFJ71Oabw/viewform).