The rumors of Testnet Medalla’s death are grossly exaggeratedsays the editor of Prysmatic Labs, Raúl Jordanwho adds that despite the setback, The start plan for ETH 2.0 is still on track.
Medal – the last multi-client test network before the start of phase 0 of ETH 2.0 – crashed on August 14th when a bug disconnected most of the validators from the network.
The incident caused One of the six servers in the test network reports the time and date as a day in the futureto which the system responded by averaging the discrepancy – which resulted in the time held by each server being four hours incorrect.
Due, “Validators have mistakenly suggested blocks and certificates for future slots”According to the official Prysmatic incident report. The bug affected all Prysm users, which was the majority.
In a recent blog post, Jordan said the bug resulted in “a carnage with over 3000 clipping events … and all of our internal validators were cut off”. While the bug has been described as “the best thing that can happen to a testnet” because the mainnet incident would have been catastrophic with real money, competing networks are happy to endorse the bug.
The Bitcoin SV blog CoinGeek reported on a major catastrophe which shows that “Ethereum 2.0 is not ready to start … Considerable delays are to be expected in the start of ETH 2.0.” Any delay would be a huge problem for the Ethereum network, which is struggling with congestion and record fees.
Imagine a bug like this that brings the mainnet to a standstill. ETH 2.0 will need all the tests it can get. Maybe a couple of Ave Maria too. 18 months TM. https://t.co/6e1oQ7uVYu
– Eli Afram (@justicemate) 20th August 2020
Jordan downplayed the problems, calling it “a busy weekend”. This provided “the best learning experience to prevent this from happening again, especially on the mainnet”.
He stated that the problem was the result of Prysmatic uses approximate time cloud servers to inform users that their system time may be exhausted.“It was a good thing to dynamically adjust the time based on the responses from these servers, but not necessary at all and instead problematic.”
He said now they know that having a single point of failure on something so important and from now on is a security risk “They only depend on the system time.”
Jordan said claims that the test network was “dead” were false and said so As long as there are enough validators on board, the test network could be fully functional again. On August 19th he said:
“We still have hope. Participation has increased from 0-5% to 40%. The chain needs> 66% to get done.”
The Medalla test network is working again from today, although it’s not quite stable yet, from what they say.
Medal is saved!
INFO blockchain: Synchronized new block block = 0x68a46008 … epoch = 3428 finalizedEpoch = 3424 finalizedRoot = 0xf0e6b2fe … slot = 109700
– prestonvanloon.eth (@preston_vanloon) 19th August 2020
Jordan said Prysmatic Labs believed the incident “has no influence on the publication date” of ETH 2.0 and that the rollout of phase 0 can proceed as planned::
“We believe the planned launch target of 2-3 months after Medalla’s inception is still an ideal timeline. There will be a public list of requirements for the ETH2 launch and this incident will definitely add many new elements.” ready for customer hardship, security, and proper start. “