Slide
What’s New in Eth2 - 15 July 2022
2022-07-16
Ben Edgington (Eth2 at ConsenSys — all views expressed are my own)
Edition 97 at eth2.news
There is no ETH2 coin. Anyone offering you ETH2 coins is scamming you.
Apparently I need to say this
Top picks
The EF Research Team’s AMA number 8 is the read of the week.It’s happening!
So, we have finally uttered a potential Merge date in public!
Icymi, the week of September 19th[1] was suggested as a planning goal on yesterday’s consensus devs call.
Please treat this target with the appropriate caution. It’s not do-or-die, it’s a coordination point to help everyone plan. A lot of node operators have a lot of work to do between now and then, and targets help to focus minds. Tim Beiko’s cheese analogy likens the firmness of the mainnet date to the consistency of burrata.
Meanwhile, Goerli Merge plans are at the mature cheddar level of firmness. (This is such a great analogy.)
On the call we discussed doing the Goerli Bellatrix upgrade on the 8th of August and targeting the Goerli Merge for the 11th. The Goerli Merge TTD (terminal total difficulty) is expected to be chosen early next week, but my latest understanding is that Bellatrix may be brought forward to the 4th, and TTD to the 10th. Remember that the dates are targets: there is some variability in progress towards TTD on all the networks due to hash rate on PoW and some weird proof of authority process on Goerli, so it could move by a few days either way.
Testing the Merge
Sepolia
Since the last edition we’ve successfully Merged the Sepolia testnet onto Proof of Stake. TTD was hit and the first merged block was produced at 1400 UTC on July the 6th, just 24 seconds after the target time due to an incredible feat of hash rate tuning. Anthony Sassano hosted the Sepolia watch party alongside the EthStaker folk. It was much shorter than the Ropsten session.
Immediately after the Merge, network participation dropped to ~70%, which was a little concerning. The large number of missing validators turned out to be entirely due to user errors in configuring the nodes. Over the next few hours, as people fixed their stuff, participation returned to 95%, which is effectively 100%, since something like 5% of validator keys got lost at some point.
The node configuration issues were largely due to the way we decided to perform the testnet merges: first setting a dummy TTD, and later updating it to the real TTD. This was necessary to defend against people 30x-ing the hashrate to hit the TTD before the Bellatrix fork, as we saw happening on Ropsten. This will not be needed on mainnet. Indeed, the lesson that we need to make user experience as smooth as possible strongly informed our planning for the mainnet Merge in the consensus devs call yesterday.
Christine Kim covered the Sepolia Merge in her weekly Galaxy Digital research brief.
In short: it was a success! No client had any Merge related issue. But we need to be mindful of user experience in configuring clients.
Mainnet shadow forks
MSF8 went very well. Besu regressed to proposing empty blocks (but still kept proposing, which helps keep the beacon chain running), and Erigon didn’t get synced up in time. Otherwise all was smooth.
MSF9 had a few more issues. Cutting and pasting from my consensus call notes, as reported by Pari,
Lighthouse nodes were falling out of sync and not catching up until the next epoch. A fix was already in progress and is now deployed.
Besu: 4 of 5 nodes hit an invalid block - looks like a previous issue.
Nethermind is offline (deliberately). On MSF8 they ran into an issue they had seen on Ropsten. Therefore on MSF9 they ran lots of varied configurations to try to reproduce it, which they suceeded in doing.
Erigon nodes did not sync in time. Will use a snapshot in future tests.
Nimbus nodes: Pari had messed up the configuration.
So, a bit bumpy. Once the Nimbus configuration was fixed the chain successfully gained finality.
In view of the upcoming Goerli–Prater Merge the next shadow fork will be on Goerli next week, with mainnet shadow fork 10 to take place a week after that.
While you are testing the Merge, don’t forget that MEV-Boost also needs testing if you plan to use it. Flashbots has some info on this.
Prepping the Merge
A new section this week. Now that we know that it’s definitely happening, stakers and non-stakers alike need to start getting ready.
To that end, Superphiz is crowd-sourcing resources and media. This is a treasure trove for education around the Merge. GitPOAPs for all!
Over the next weeks Infura will be hosting a series of workshops on all aspects of the Merge. Follow ConsenSys on Crowdcast to receive updated information about when sessions will start and to view the recordings of past sessions. I hear that the first session was a huge success, with over 9,000 people attending or watching the recording. The next is on Tuesday the 19th on “Solo Stakers Forum: What You Need To Do To Get Ready”.
Speaking of Infura, in this video Rémy and Yorick from EthStaker discuss using Infura or any similar provider as a staker after the Merge.
Here’s a quick Merge setup guide for Lodestar.
The excellent beaconcha.in mobile app now has a Merge readiness checklist. I love that app - it just notified me that my (wife’s) node proposed a block while I was actually writing that sentence.
And I’ll leave this here as a permalink: the Launchpad’s own Merge readiness checklist.
Staking
The Lido apologists have been out in force on the podcasts in the last few weeks. As a counterpoint, I found the Unchained podcast with Ryan Berckmans and Alexandre Bergeron insightful. Ryan’s plea starting around 41m39s is outstanding. Here’s a couple of spicy quotes, but there’s much more to it - the whole podcast is very good.
I think this is going to be a long battle for the soul of Ethereum… Lido centralisation is probably the biggest issue ever to pop up… perhaps the biggest community issue since the DAO fork… This is a very, very potentially destructive issue over the long term.
and
You may be making Ethereum literally less valuable when you allocate into Lido, because the value of Ethereum is a function of its credible neutrality.
Elsewhere, you may have seen this stupid bit of self-promotion this week and been unduly worried by it. The “major attack vulnerability” turns out to be a well-known issue and to tweet it out like that was idiotic. The point is that it is possible to discover the IP addresses of Eth2 validators, and this could be used to DDoS block proposers, for example to steal their MEV. There are various ways to defend against this available now, but the correct long-term solution is single secret leader election. SSLE, however, is still in the research stage - I would love to see some more progress on this.
The Great Explainers
foobar gives us a nice broad overview of Ethereum Proof of Stake. There’s a solid section on misconceptions, and a quick glance into the future.
Ladislaus has produced a very helpful FAQ on MEV and mev-boost (for node runners). One way or another, all stakers are going to be impacted by MEV-Boost after the Merge. This will help you get to grips with the whole thing.
Zellic (a smart contract auditing firm) has produced a very good guide for Ethereum application developers, ETH 2 Proof-of-Stake: What Devs Need to Know. We’re changing the application layer (EVM) as little as we can through the Merge. Nonetheless, there are some implications for smart contract writers and application developers, so pay attention!
ethereumpools.info previously explained Ethereum’s post-Merge rewards, and now they’ve done penalties.
Jacek explores what you can do with light clients in a short Twitter thread.
I managed to pump out a couple of short book sections on committees and aggregator selection. Thinking of tackling Casper FFG next, which is a bit more substantial.
Media and stuff
The talks from the Layer 1 workshop at DevConnect in Amsterdam are finally available. We covered a ton of forward-looking stuff about life beyond the Merge. You may see some familiar faces
The Ethereum Cat Herders’ Know Your Client series continues with
Erigon & The Merge with Andrew Ashikhmin and Igor Mandrigin, and
Prysm & The Merge with Terence Tsao and Potuz.
The next installment of SSV network’s pre-Merge Twitter spaces is Developing on ETH, where next? with Anthony Sassano hosting Vitto Rivabella of Alchemy.
Tim Beiko appeared on the epicenter podcast. What do you think he was talking about?
Research
Oisín Kyne of Obol wants to propose some changes to the way that validators handle aggregation duties. This is to allow some more flexibility around implementing distributed validator technology.
Regular Calls
All Core Devs
All Core Devs call #142 took place on the 8th of July. It was the last Friday call! Moving to Thursdays from the 21st of July. This is the end of an era.
Agenda.
Video.
Tweet notes from Tim and Christine.
Consensus devs
Call #91 took place on the 14th of July.
Agenda
Video
My quick notes, and a Twitter note from Christine (whoa, what is this sorcery?).
Aside from Merge planning and shadow fork review, we spent some time discussing the proposal to delay the start of MEV-Boost to a little while after the Merge in order to simplify any debugging we need to do (fewer moving parts). In the end we agreed to simplify the proposal a little.
Merge Community Call
Merge Community Call #5 was today.
Agenda.
Video.
I haven’t had a chance to watch this yet, but Christine also covers it in the aforementioned Twitter wizardry.
Upcoming events
Tuesday, July 19th at 1600 UTC: Infura’s Solo Stakers Forum: What You Need To Do To Get Ready for the Merge.
Friday, July 29th at 1500 UTC: EthStaker validator Merge prep workshop. Please complete the Google form ahead of the workshop to help them prepare.
In other news
All Core Devs Update 012 from Tim Beiko. Always a great read.
The latest from Stereum. They have been busy.
Flashbots’ MEV-Boost Status Update - July 1-15, 2022. Don’t sleep on these updates if you are running a validator; there’s important stuff going on.
Michael Sproul is enhancing blockprint using a new tool called blockdreamer.
Hildobby’s Eth2 Staking deposits dashboard has been enhanced.
And finally…
Nearly didn’t get this finished at all this week as I got a bit distracted watching this guy on YouTube. This is epic. Love a bit of boogie-woogie!
Internet data
Edition 97 at eth2.news
There is no ETH2 coin. Anyone offering you ETH2 coins is scamming you.
Apparently I need to say this
Top picks
The EF Research Team’s AMA number 8 is the read of the week.It’s happening!
So, we have finally uttered a potential Merge date in public!
Icymi, the week of September 19th[1] was suggested as a planning goal on yesterday’s consensus devs call.
Please treat this target with the appropriate caution. It’s not do-or-die, it’s a coordination point to help everyone plan. A lot of node operators have a lot of work to do between now and then, and targets help to focus minds. Tim Beiko’s cheese analogy likens the firmness of the mainnet date to the consistency of burrata.
Meanwhile, Goerli Merge plans are at the mature cheddar level of firmness. (This is such a great analogy.)
On the call we discussed doing the Goerli Bellatrix upgrade on the 8th of August and targeting the Goerli Merge for the 11th. The Goerli Merge TTD (terminal total difficulty) is expected to be chosen early next week, but my latest understanding is that Bellatrix may be brought forward to the 4th, and TTD to the 10th. Remember that the dates are targets: there is some variability in progress towards TTD on all the networks due to hash rate on PoW and some weird proof of authority process on Goerli, so it could move by a few days either way.
Testing the Merge
Sepolia
Since the last edition we’ve successfully Merged the Sepolia testnet onto Proof of Stake. TTD was hit and the first merged block was produced at 1400 UTC on July the 6th, just 24 seconds after the target time due to an incredible feat of hash rate tuning. Anthony Sassano hosted the Sepolia watch party alongside the EthStaker folk. It was much shorter than the Ropsten session.
Immediately after the Merge, network participation dropped to ~70%, which was a little concerning. The large number of missing validators turned out to be entirely due to user errors in configuring the nodes. Over the next few hours, as people fixed their stuff, participation returned to 95%, which is effectively 100%, since something like 5% of validator keys got lost at some point.
The node configuration issues were largely due to the way we decided to perform the testnet merges: first setting a dummy TTD, and later updating it to the real TTD. This was necessary to defend against people 30x-ing the hashrate to hit the TTD before the Bellatrix fork, as we saw happening on Ropsten. This will not be needed on mainnet. Indeed, the lesson that we need to make user experience as smooth as possible strongly informed our planning for the mainnet Merge in the consensus devs call yesterday.
Christine Kim covered the Sepolia Merge in her weekly Galaxy Digital research brief.
In short: it was a success! No client had any Merge related issue. But we need to be mindful of user experience in configuring clients.
Mainnet shadow forks
MSF8 went very well. Besu regressed to proposing empty blocks (but still kept proposing, which helps keep the beacon chain running), and Erigon didn’t get synced up in time. Otherwise all was smooth.
MSF9 had a few more issues. Cutting and pasting from my consensus call notes, as reported by Pari,
Lighthouse nodes were falling out of sync and not catching up until the next epoch. A fix was already in progress and is now deployed.
Besu: 4 of 5 nodes hit an invalid block - looks like a previous issue.
Nethermind is offline (deliberately). On MSF8 they ran into an issue they had seen on Ropsten. Therefore on MSF9 they ran lots of varied configurations to try to reproduce it, which they suceeded in doing.
Erigon nodes did not sync in time. Will use a snapshot in future tests.
Nimbus nodes: Pari had messed up the configuration.
So, a bit bumpy. Once the Nimbus configuration was fixed the chain successfully gained finality.
In view of the upcoming Goerli–Prater Merge the next shadow fork will be on Goerli next week, with mainnet shadow fork 10 to take place a week after that.
While you are testing the Merge, don’t forget that MEV-Boost also needs testing if you plan to use it. Flashbots has some info on this.
Prepping the Merge
A new section this week. Now that we know that it’s definitely happening, stakers and non-stakers alike need to start getting ready.
To that end, Superphiz is crowd-sourcing resources and media. This is a treasure trove for education around the Merge. GitPOAPs for all!
Over the next weeks Infura will be hosting a series of workshops on all aspects of the Merge. Follow ConsenSys on Crowdcast to receive updated information about when sessions will start and to view the recordings of past sessions. I hear that the first session was a huge success, with over 9,000 people attending or watching the recording. The next is on Tuesday the 19th on “Solo Stakers Forum: What You Need To Do To Get Ready”.
Speaking of Infura, in this video Rémy and Yorick from EthStaker discuss using Infura or any similar provider as a staker after the Merge.
Here’s a quick Merge setup guide for Lodestar.
The excellent beaconcha.in mobile app now has a Merge readiness checklist. I love that app - it just notified me that my (wife’s) node proposed a block while I was actually writing that sentence.
And I’ll leave this here as a permalink: the Launchpad’s own Merge readiness checklist.
Staking
The Lido apologists have been out in force on the podcasts in the last few weeks. As a counterpoint, I found the Unchained podcast with Ryan Berckmans and Alexandre Bergeron insightful. Ryan’s plea starting around 41m39s is outstanding. Here’s a couple of spicy quotes, but there’s much more to it - the whole podcast is very good.
I think this is going to be a long battle for the soul of Ethereum… Lido centralisation is probably the biggest issue ever to pop up… perhaps the biggest community issue since the DAO fork… This is a very, very potentially destructive issue over the long term.
and
You may be making Ethereum literally less valuable when you allocate into Lido, because the value of Ethereum is a function of its credible neutrality.
Elsewhere, you may have seen this stupid bit of self-promotion this week and been unduly worried by it. The “major attack vulnerability” turns out to be a well-known issue and to tweet it out like that was idiotic. The point is that it is possible to discover the IP addresses of Eth2 validators, and this could be used to DDoS block proposers, for example to steal their MEV. There are various ways to defend against this available now, but the correct long-term solution is single secret leader election. SSLE, however, is still in the research stage - I would love to see some more progress on this.
The Great Explainers
foobar gives us a nice broad overview of Ethereum Proof of Stake. There’s a solid section on misconceptions, and a quick glance into the future.
Ladislaus has produced a very helpful FAQ on MEV and mev-boost (for node runners). One way or another, all stakers are going to be impacted by MEV-Boost after the Merge. This will help you get to grips with the whole thing.
Zellic (a smart contract auditing firm) has produced a very good guide for Ethereum application developers, ETH 2 Proof-of-Stake: What Devs Need to Know. We’re changing the application layer (EVM) as little as we can through the Merge. Nonetheless, there are some implications for smart contract writers and application developers, so pay attention!
ethereumpools.info previously explained Ethereum’s post-Merge rewards, and now they’ve done penalties.
Jacek explores what you can do with light clients in a short Twitter thread.
I managed to pump out a couple of short book sections on committees and aggregator selection. Thinking of tackling Casper FFG next, which is a bit more substantial.
Media and stuff
The talks from the Layer 1 workshop at DevConnect in Amsterdam are finally available. We covered a ton of forward-looking stuff about life beyond the Merge. You may see some familiar faces
The Ethereum Cat Herders’ Know Your Client series continues with
Erigon & The Merge with Andrew Ashikhmin and Igor Mandrigin, and
Prysm & The Merge with Terence Tsao and Potuz.
The next installment of SSV network’s pre-Merge Twitter spaces is Developing on ETH, where next? with Anthony Sassano hosting Vitto Rivabella of Alchemy.
Tim Beiko appeared on the epicenter podcast. What do you think he was talking about?
Research
Oisín Kyne of Obol wants to propose some changes to the way that validators handle aggregation duties. This is to allow some more flexibility around implementing distributed validator technology.
Regular Calls
All Core Devs
All Core Devs call #142 took place on the 8th of July. It was the last Friday call! Moving to Thursdays from the 21st of July. This is the end of an era.
Agenda.
Video.
Tweet notes from Tim and Christine.
Consensus devs
Call #91 took place on the 14th of July.
Agenda
Video
My quick notes, and a Twitter note from Christine (whoa, what is this sorcery?).
Aside from Merge planning and shadow fork review, we spent some time discussing the proposal to delay the start of MEV-Boost to a little while after the Merge in order to simplify any debugging we need to do (fewer moving parts). In the end we agreed to simplify the proposal a little.
Merge Community Call
Merge Community Call #5 was today.
Agenda.
Video.
I haven’t had a chance to watch this yet, but Christine also covers it in the aforementioned Twitter wizardry.
Upcoming events
Tuesday, July 19th at 1600 UTC: Infura’s Solo Stakers Forum: What You Need To Do To Get Ready for the Merge.
Friday, July 29th at 1500 UTC: EthStaker validator Merge prep workshop. Please complete the Google form ahead of the workshop to help them prepare.
In other news
All Core Devs Update 012 from Tim Beiko. Always a great read.
The latest from Stereum. They have been busy.
Flashbots’ MEV-Boost Status Update - July 1-15, 2022. Don’t sleep on these updates if you are running a validator; there’s important stuff going on.
Michael Sproul is enhancing blockprint using a new tool called blockdreamer.
Hildobby’s Eth2 Staking deposits dashboard has been enhanced.
And finally…
Nearly didn’t get this finished at all this week as I got a bit distracted watching this guy on YouTube. This is epic. Love a bit of boogie-woogie!
Internet data