Ethereum: Bellatrix Hard Fork breakdown – Ready for The Merge

August 7, 2022

What is it?

Bellatrix is a hard fork (Bellatrix) of the Beacon Chain – a proof-of-stake network running in parallel to Mainnet (i.e., proof-of-work). No, this is not “The Merge”, but rather prepares the Beacon Chain for the merge.

Currently, the Beacon Chain is largely coming to consensus about its own state (i.e., blocks typically include information about attestations, deposits, validator slashings and exits). In other words, Mainnet (i.e., Proof-of-Work) is doing the heavy lifting for the network and coming to consensus about the canonical chain on which transactions are settling. None of this changes after Bellatrix – but the Beacon Chain will be ready to take over after the terminal block is reached on Mainnet (the beginning of The Merge).

(source)

What has to change to get ready for The Merge?

Generally, Bellatrix accomplishes a few vital tasks before The Merge. The Beacon Chain will need to recognize when it has been ‘passed the baton’ from Mainnet (i.e., The Merge has occurred) and what to do once it has

For a deeper dive into the specific changes, a review of the Bellatrix specs is recommended:

Of particular interest are likely the Beacon Chain changes and Honest validator guide changes sections. These two comprise functions/methods, which allow validators to propose blocks containing transactions (see ‘Block proposal’ in the Honest validator guide changes, for instance), among other things. Validators taking over the processing of transactions and determining the canonical chain are, arguably, the key features of The Merge. Not only that, it means that tips and MEV fees will accrue to validators post-merge.

What should stakers know?

In addition to the changes above, penalties (for more on slashing see here) have been increased to their final post-merge values.

(Source1 Source2)

Big picture, all of these values represent more punitive penalties.

Broadly, the idea is that for the initial phases of Beacon Chain validators would be given a grace period and not face full penalties as early operators. But now that The Merge is upon us, it’s time to bring those values in line with their post-merge levels.

INACTIVITY_PENALTY_QUOTIENT – as the name suggests this value feeds into penalty values as a divisor – so a smaller number results in a larger penalty. More specifically, it now takes about 21 days to lose half your balance, assuming your validator is offline during a period of inactivity leak; before Bellatrix it took approximately 37 days.

MIN_SLASHING_PENALTY_QUOTIENT – a validator’s effective balance is divided by this number to arrive at the initial slashing penalty (for more on slashing see here – bear in mind this reference is based on Altair). Assuming an effective balance of 32 ETH – the amount required to run a validator – the penalty has moved from 0.5 ETH to 1 ETH with the Bellatrix hard fork.

PROPORTIONAL_SLASHING_MULTIPLIER – if a validator commits a slashable offense the protocol adds a penalty that is related to the proportion of other validators that have engaged in a slashable offense; the idea being that if the network is under attack, it is likely that a higher proportion of validators have committed a slashable offense than if the offense happened due to random chance. (During Altair the penalty was two times the proportion of validators engaging in a slashable offense.)

Following Bellatrix, the number is now three; that is, three times the proportion of validators committing a slashable offense. For instance, if a validator commits a slashable offense and it is found that 5% of validators have committed a slashable offense at the same time then the validator will face a penalty of 3 * 5% = 15% of their ETH balance (prior to Bellatrix, i.e., Altair, this number would have been 2 * 5% = 10%).

How did it go?

(source)

The Bellatrix hard fork was successful and went relatively smoothly. That said, network participation did drop by about 5% (note the scale – it’s not as significant a drop as it appears in chart):

(beaconchai.in

The most likely culprit for the drop in participation is failure of some to update their nodes. That said, it looks like we are already starting to see a bounce back in participation.

What can we expect to come next?

Next up is Paris – the upgrade to Mainnet signaling The Merge, likely to take place around September 13-15th. Following that will be the Shanghai upgrade on the execution layer and Capella on the consensus layer – these upgrades will allow for withdrawals, among other things.

(source)

SHARE POST

Meet with us

Bring the Complete Staking Solution to Your Organization

Figment respects your privacy. By submitting this form, you are acknowledging that you have read and agree to our Privacy Policy, which details how we collect and use your information.