Ronin Bridge incident, 2022-03-23

Ronin's bridge was secured by nine validators and needed five signatures to release funds. On paper that is a majority multisig; in practice four of the nine keys sat with one company, Sky Mavis, so compromising one organisation got the attacker 80% of the way. The fifth signature came from an operational shortcut: months earlier Axie DAO had allowed Sky Mavis to sign on its behalf during a traffic spike, and that permission was never revoked. Once inside Sky Mavis's systems, the attacker could use a gas-free RPC node to obtain Axie DAO's signature too (Halborn).

The second failure was detection. A $600M outflow went unnoticed for six days because no one was monitoring the bridge's balance against expectations. Validator diversity and withdrawal monitoring are now standard questions in bridge reviews, largely because of Ronin.

For users the lesson is that "9 validators" says little on its own. What matters is how many independent organisations hold the keys, whether any delegation exists, and whether large withdrawals are rate-limited or delayed.

Detection is the cheaper half of the fix: an alert on any withdrawal above a set share of the bridge's balance would have turned a six-day gap into minutes.

At the time

Users could not have detected the theft before the team did. The protective habit was not to keep idle balances in the bridge's wrapped assets on Ronin, and to bridge back to Ethereum only what was needed for play, limiting exposure to a validator set controlled mostly by one company.

Now

When a bridge is secured by a multisig or validator set, check how many distinct organisations hold keys and whether there is a withdrawal delay or rate limit. L2Beat's bridge pages list these details for many bridges; our reviews cite them in the risk card. Keep large amounts on the canonical chain rather than in bridged representations.