The State of Staking September 2022
The most significant upgrade in the history of Ethereum, the Merge, happened successfully on September 15th.
At the ecosystem level, it felt a little like Y2K – all the excitement as the millennium arrived after years of work and then… it was a second past midnight and all the lights were still on. With the Merge, block #15537394 carried transactions and a feeRecipient and was confirmed via Proof of Stake on the consensus layer (CL), and on we went.
However, at the micro level, there are some things that still need to be worked out.
There are more missed attestations. The whole exercise is more complex than it used to be, with more moving parts. Each CL client (like Teku or Lighthouse or Prysm) is connected on a one-to-one basis with an execution layer (EL) client (like Geth, Besu or Nethermind). Should the EL client come unstuck, falling behind or crashing, attestations also stop working correctly.
The good news is that the ecosystem is comfortable addressing these performance updates now that they are manifested on the actual public chain, rather than only on testnets.
Another issue that is occurring post-Merge is ‘empty blocks’. The EL client provides the CL client with a valid block but it has no transactions. The block is proposed and verified,and the validator earns its block reward. However, the validator misses out on any priority fees, and potentially loses a significant portion of its reward. We estimate this is happening around 1% of the time across the entire network at the moment.
Both CL and EL clients have fixes and improvements that will deal with these issues. However, given the size of the change, it is important to accept that there will be a period of greater instability than we’re used to with just the Beacon chain. Ethereum will go through a series of ‘firsts’ again for a while and we will have to adapt.
Before everyone starts yelling “wen withdrawal”, the Ethereum development community have taken the wise decision to take a break before coming back around the table to restart the work already done on withdrawals.
Codefi Staking
Aside from experiencing many of the issues mentioned above, Codefi Staking passed through the Merge relatively unscathed. Our distributed platform with client diversity across both layers meant that we weren’t wholly vulnerable to a single issue. Client feeRecipient addresses are mostly being picked up as expected with a few hiccups we’ve already resolved.
Our next focus is to improve our performance and resilience further in this new setup. A part of this will be adding MEV-Boost into our infrastructure. We were hoping to enable individual validators to point at different relays but we’re still waiting for that functionality to be released by Flashbots (you can imagine they’re busy right now!) and so, we’re resolved on applying something across our entire platform.
Other References
The bad news on this front is that Ben has stopped publishing his ‘What’s new in Eth2’. I daresay after two years and having finally delivered the Merge, he’s decided to put the Eth2 branding to rest. For myself, I’m sad to see it go. Fortunately, Anthony Sassano’s Daily Gwei remains updating us on all things Ethereum: https://thedailygwei.substack.com/.
