1

I'd like to know what would happen in terms of penalization of staked validators if there was some sort of major global netsplit that -- for the sake of this question -- resulted in exactly 50% of all beconchain nodes being separated from the other 50%.

As I understand things, if a beconchain node goes offline, it is peanalized by having ETH deducted from it's staked balance until it gets to under 16 ETH and then it is excluded from the network. As a blockchain is decentralized, I understand that decisions relating to this are handled collectively by all of the staked nodes.

I've heard people say that the whole practice of excluding offline nodes is to maintain the integrity of the network and handle things such as a nuclear attack. So even if 90% of nodes were suddenly destroyed, the remaining 10% could quickly exclude them and the network would heal.

But in this case, I'm not asking about 50% of the nodes being destroyed. I'm talking about what would happen if half of the nodes couldn't see the other half of the nodes due to a major netsplit. This could be a technical fault, a war or even a political desicion between major world powers. In this scenario, I presume that nodes on each half of the divide, would consider the nodes on the other half of the divide to be offline. As there would then effectively be two separate beaconchain mainnets, each one would then start the process of peanalizing the nodes they couldn't see, reducing them to less than 16 ETH and then eventually excluding them from the network.

If the netsplit was permanent, then I suppose this wouldn't matter. But, let's say that after a week (or a month, or a year) the internet was joined back together again. At this point, could the beaconchain become one again? Would nodes on each side of the split that had had 16 ETH of their stake eaten by the nodes on the other side (which would effectively be all nodes which existed prior to the split as far as I understand it) get their 16 ETH back? How could the stability of the network be maintained if every node had both 32 ETH and 16 ETH at the same time, depending on which part of the split you were looking at?

I've always understood that on Bitcoin and other PoW blockchains that when the two sides of the split were able to see each other again, the longer chain would 'win' and the shorter chain 'orphaned' but that any spends on the orphan chain that hadn't been double-spent on the winning chain would be integrated. But I don't understand how this would with a PoS blockchain, especially with 50/50 split followed by a reintegration.

If this is answered in any documentation or whitepaper then being pointed to the relevant information would be just as valuable as receiving an answer. With such limited knowledge of Ethereum to begin with, it's difficult to even know which terminology to use to find the correct answer. Even Googling something like 'ethereum netsplit beaconchain' return's the dreaded 'It looks like there aren't many great matches for your search' message.

0

There is an "inactivity leak" that drains the inactive validators (on each partition), eventually slashing them, so that the remaining active validators will again reach 2/3 majority so that finality resumes.

An important point is that the inactivity leak takes about 36 days.

The INACTIVITY_PENALTY_QUOTIENT equals INVERSE_SQRT_E_DROP_TIME*2 where INVERSE_SQRT_E_DROP_TIME := 2**13 epochs (about 36 days) is the time it takes the inactivity penalty to reduce the balance of non-participating validators to about 1/sqrt(e) ~= 60.6%.

There is no design currently in the Beacon Chain that in the event of a reintegration, gives back ETH to the slashed validators.

Explaining the liveness guarantee is a thread on ethresear.ch with some discussion. "Recovering" from a prolonged netsplit has not been prioritized over liveness. A reply from vbuterin:

I feel that 50/50 network partitions/splits are massively overrated as a threat. When has this historically happened, anywhere, and not been resolved soon? What would even be a coherent story by which two parts of the world stay coherent internally but communication between them is not possible? The incentives for maintaining global communication are massive, and there’s no reason why parts of the world that are still capable of maintaining communication internally would not be able to figure something out to talk to each other within a week or two. The thing that’s much more likely, whether in normal life, or due to government censorship, or due to a war, is nodes going offline, either because something happens to their operators or because their operators get cut off from the entire internet.

The inactivity leak is primarily there to deal with this “3/4 of the network goes offline at the same time” risk.

2
  • Thank you for the answer and the link to additional reading on the subject. – Joseph Dec 27 '20 at 12:19
  • Welcome to the Ethereum Stack Exchange and we're happy to have you! – eth Jan 29 at 2:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.