Marathon Digital Holdings, a publicly listed Bitcoin mining firm, confirmed that it had mined an invalid block on the Bitcoin network.
In a statement posted to X on Wednesday, the firm said the invalid block was produced as a result of an error with one of its experiments, clarifying that it was not meant to alter Bitcoin Core in any way.
We can confirm that Marathon did mine an invalid block. We utilize a small portion of our hash rate to experiment with our development pool and research potential methods to optimize our operations. The error was the result of an unanticipated bug that came from one of our…
— MARA (@MarathonDH) September 27, 2023
“We utilize a small portion of our hash rate to experiment with our development pool and research potential methods to optimize our operations,” said the firm.
According to the statement, the firm noticed the invalid block at around the same time as the rest of the world, and immediately corrected the error.
“This incident, while unintended, underscores the robust security of the Bitcoin network, which rejected and rectified the anomaly,” said Marathon.
The invalid block was first noticed on the mainnet at around 5:51 a.m. on Wednesday by users, with one X account, “0xB10C”, pointing to the use of custom code.
It seems like MARAPool had a transaction ordering issue.
66dfefcdc3eeec2745c11f511f6068d62f34c34c767ba0feed47f63f01ccc2d8 is the 6th tx in the invalid block.
It spends from 7d18f0eefce0497b5d0c9b61fdf816b7744587c7e5e57acc53de71d1dae59725, which is the 1454th tx in the block. https://t.co/P3txhKqVS0
— b10c (@0xB10C) September 27, 2023
BitMEX Research found that the block was invalidated because of a transaction ordering problem. The Bitcoin blockchain ensures that transactions are ordered chronologically, which makes it resistant to double spending.
Market participants criticized Marathon for not using the testnet for experiments, with some users taking aim at the use of investors’ hashrate for the cause.
However, one X user opined that Marathon’s optimizing hashrate would only come via experimentation.
“It’s like getting mad at Google for messing up a project while exploring new AI advances. Coming from the SaaS world, it seems easy to test non production on a small scale, but I don’t know how you’d replicate a dev Bitcoin environment to test hash-rate-function at scale,” he said.
