["research"] timestamp dependence / transaction-ordering dependence

ArdanaLabs / audit

0 stars 0 forks source link

["research"] timestamp dependence / transaction-ordering dependence #20

Open quinn-dougherty opened 2 years ago

quinn-dougherty commented 2 years ago

Description

Can dishonest miners game the order in which transactions are processed?

Deliverable

yes/no
if yes we may want to have an audit section detailing what this attack would look like on Danaswap.

Notes

See book by Ma-Gorzny-Zulkoski-Bak-Mack for information on timestamp dependence attacks

morganthomas commented 2 years ago

Maybe relevant: https://eprint.iacr.org/2016/889.pdf

quinn-dougherty commented 2 years ago

discussed in https://github.com/ArdanaLabs/audit/issues/16#issuecomment-937825672

The Cardano mempool is designed to be "fair". Transactions are processed in a FIFO order regardless or how much in fees they pay (the ledger spec does support a fee market, but cardano-node doesn't take this into account)

source: https://github.com/input-output-hk/cardano-node/issues/3247

quinn-dougherty commented 2 years ago

This is currently listed in I.3. Insufficient literature because cardano-node commenter SebastienGllmt is the only source I've found, and subsequent commenters on that thread called into question some of their assumptions.