Medalla, the testnet that serves as a testing ground for validators of the upcoming Ethereum 2.0 upgrade, introduced a bug in the implementation of the Prysm client that caused the testnet to fork into 4 different networks.
At the time of this publication, the core developers of the testnet have not officially spoken out, but a report reveals that Medal has 4 major forks, which prevented the network from reaching consensus for the past 2 days. The forks, also known as hard fork, occurred after a bug in the Prysm client prevented most nodes from synchronizing. It is important to mention that most of the active validator nodes on the Medalla network use this client.
Source: TrustNode
For its part, Beacon Chain shows that the network is experiencing problems with consensus and block finalization, something that was also pointed out by the creator of BTConEthereum in your Twitter account.
Despite that Vitalik Buterin, co-founder of Ethereum, claimed that the ETH 2.0 upgrade would be ready by the end of this year, several in the community point towards a new delay in the long-awaited launch, although the developers have not said anything about it. Similarly, several users point out that Buterin and the rest of the network developers are designing a solution to solve the problem immediately.
It may interest you: Ethereum 2.0 developers announce the successful launch of Medalla, although the testnet presented some errors
Desynchronization and different points of view in the Medalla network
According to the source, the bug in the Prysm client caused “havoc with the synchronization” of nodes, resulting in nodes having different views of each other within the blockchain, resulting in some blocks being validated while several others were left lost or orphaned.
Likewise, one of Medalla's validators noted that the network “jump up and down everywhere”, arguing that many reorganizations occurred in Medalla and that because of this the nodes could not make a single decision in the network.
For his part, Raúl Jorda, developer of the Prysm client, points out that:
«There are a lot of different forks going on right now and some nodes are falling way behind, so you get all these requests for master locks to try and figure it out, but the master is currently showing up at eth2stats, which has consensus between Lighthouse and Prysm."
Despite the difficult situation, the developers of the Prysm and Lighthouse clients assure that in the last 12 hours the situation has been improving considerably, and that several nodes are reaching the longest main chain. It is expected that the termination error will be fixed soon and that the main client of the network will be active. Likewise, the Prysm developers ask the validators not to restart the nodes, but to let them run as much as possible, in order not to lose the current synchronization of the nodes.
So far, one of the most notable consequences of the recent forks in the Ethereum 2.0 testnet is the increase in node RAM requirements.
Source: TrustNode
Continue reading: Ethereum 2.0: What were the possible causes of Medalla's failures at the time of its launch?
