AN INCENTIVE TO BRING DOWN ETHEREUM IS BORN
Network congestion is a big problem for any cryptocurrency, asset, or token. Unfortunately for Ethereum, and every asset or token based on this blockchain, these issues are quite apparent this week. It appears someone successfully identified how these problems come to be, and what could happen in the not-so-distant future if these things are not addressed. It only takes a relatively small amount of money for a group or individual to bring down the Ethereum network.
More specifically, the gas price used to send transaction on the Ethereum blockchain can be gamed quite easily. It is possible to write a smart contract which uses all gas in a block to create artificially long verification times. This issue has been apparent for some time now, but it was countered by introducing a lower gas limit, which was embraced by service providers and miners alike. In one way, this should have alleviated all of these concerns, but it has apparently only made things worse.
The recent surge of cryptocurrency ICOs creates an environment in which everyone wants to see transactions confirmed as soon as possible. With the gas amounts limited, only so many transactions can be confirmed in the next block. Users manually increase gas prices to skip ahead of the queue, which creates a situation that gets out of hand incredibly quickly. Miners prioritize profitable transaction fees over the lower ones, which effectively brought the Ethereum network to a snail’s pace earlier this week.
Unfortunately, it doesn’t appear these problems can be fixed all that easily. Miners are incentivized to create such situations during which gas prices spike. F2Pool has apparently experimented with this method of operation during the Status ICO. Changing the way ICOs are organized won’t alleviate these concerns by any means. The way things look now, anyone can exploit the smart contract building bricks to bring the Ethereum network down for an extended period of time.
A thorough Reddit post outlines how such a malicious smart contract would work exactly. It does not take all that much effort, even though the contract has to be funded with an amount of ether. This creates a situation in which entering the contract is profitable for everyone, as most people expect ICOs to sell out relatively quickly. Getting in as many transactions as early as possible is the number one objective. This is how most ICOs operate ever since the concept was brought to the masses, and little has changed.
If the malicious contract has a high enough balance in Ether, it is possible to run iterations on such a contract indefinitely. Doing so would slow down the Ethereum network and effectively bring to a halt. With US$1m of Ether, one could halt the network for 2 days. Increase that number to US$50m – an amount some nefarious entities may be willing to part with – and Ethereum could be reduced to an obscure project in a month or less. It is only a matter of time until all hell breaks loose, that much is evident.