Ethereum’s Proof-of-Stake (PoS) blockchain, the Beacon Chain, suffered a rare seven block reorganization Wednesday, prompting fresh questions about the network’s transition away from Proof-of-Work (PoW) in what is colloquially called The Merge.
What’s the Beacon Chain?
The Beacon Chain is the next home for the Ethereum network. Launched in December 2020, the Beacon Chain uses deposits of Ether (ETH) as a ‘stake’ on the network's decision making, such as voting on who gets to build a block. The blockchain has no value as of now, but will have assets on top of it once The Merge is completed.
What’s a blockchain reorg?
A reorg is when two blocks are built by miners (or validators) simultaneously, but propagate around the network of nodes at different speeds. In a reorg, the network starts building the chain on one of the blocks, only to find out moments later that the other block is better to build on top of. The blocks built on the wrong block revert, and start rebuilding on the correct block.
Why do blockchains do this?
Miners and validators need a rule book to make decisions on which block to add to the blockchain. These rules are known as ‘fork choice rules’ and can vary by network.
Bitcoin uses ‘Nakamoto Consensus,’ which chooses the block with the highest total difficulty. In other words, if the network has to choose between Block A with a difficulty of 100 and Block B with a difficulty of 110, the network will choose the latter. Difficulty can be thought of as the ‘work’ required by the Bitcoin network to mine.
All PoW blockchains (and most PoS blockchains) can suffer reorgs. As a general rule, a blockchain's proneness to reorgs is correlated to the clarity of its fork choice rules and the latency of block propagation. Bitcoin’s clear fork choice rule (highest difficulty) and long block propagation time (10 minutes) make reorgs a highly unlikely event.
Ethereum’s fork choice
PoS networks necessarily have a different fork choice rule than Bitcoin. Why? Because there is no work to use as a selection criterion. The blockchain in question, the Beacon Chain, builds on a block created by a randomly selected validator. That block is then ‘attested’ by other randomly selected validators. You can read more about it in this article.
On Wednesday, what likely occurred on the Beacon Chain is some validators having slightly different fork choice rules than other validators. Some validators expected one block and other validators expected another block because the fork choice rules were not tight enough. In other words, an edge case found its way into the wild.
Blockchain reorgs are bad for user experience. There can be devastating consequences for networks like Ethereum that host many assets and applications. For example, imagine not being able to top off a risky lending position because the chain is unstable. Being liquidated because the baselayer can’t fulfil its purpose doesn’t make a case for adoption.
Luckily there's little concern in this case. The Beacon Chain's fork choice rules can be tweaked fairly trivially and there are no assets live on-chain. In other words, this reorg is merely another learning moment for blockchain engineers.