Cardano (ADA) developer Adam Dean revealed a bug in Cardano node version 1.35.2 that caused the Cardano testnet to collapse in a Twitter thread on August 18th.
🧵 (1/n) It is important to point out today that #cardano #testnet It’s **catastrophically** broken due to a bug in Cardano Node v 1.35.2. This is Vasil Hardfork’s version previously said to be “tested and ready”. This bug was discovered…
— Adam Dean (@adamKDean) August 18, 2022
According to Dean, the bug was found on a node previously said to be “tested and ready”.
Dean continued that the bug created an incompatible fork and reduced chain density. The bug was discovered through on-chain investigations by Cardano operator supporter pools, ATADA Stakepool, and Pooltoll.
Upon investigation, we found versions 1.35.2 and 1.34.1 as problematic.
Dean highlighted how most of the operators upgraded “to 1.35.2 on the testnet to simulate the Vasil HFC event”, leading to version 1.35.3 being out of sync with the chain.
According to Dean, version 1.35.3 is currently being tested on “two ‘new’ testnets (that have passed all previous Cardano HFC events) that do not have a block history like the testnets, or a multinode Tested in version block simulation. chain.
Dean said Vasil’s hard fork “rush” unsettled him. He added that the Cardano network is about to hit “nuclear rock.”
He asks the Input Output engineering department and Charles Hoskinson to deploy the appropriate disaster recovery tools disclosed during the testnet launch.
Cardano disaster recovery schedule There is a mitigation process for prolonged network-wide partitions or prolonged global outages.
Hossinson answers
Charles Hoskinson reacted to the community reaction to news of a bug in testnet version 1.35.2.
It’s strange and disturbing to see these videos rushed in 1.35.3. The code in question has been thoroughly tested for months by everyone, including SPO. The code that caused the problem on the testnet has been removed.
— Charles Hoskinson (@IOHK_Charles) August 19, 2022
According to Hoskinson, the hype around the stake pool operator’s upgrade is strange, adding that “code that was problematic on the testnet has been removed.”
Hoskinson said it’s possible the community will decide to delay the start of the Vasil hard fork, but wonders if the decision is worth it for Dapp developers.
Meanwhile, some Cardano community members did not take Hoskinson’s comments properly, but Hoskinson urged them to upgrade their node to version 1.35.3.
Cardano’s Vasil upgrade has been delayed twice. Hoskinson revealed that testing uncovered three separate bugs in the development of three new software versions.
