Paradex Bitcoin Price Error to $0 Triggers Liquidations and Chain Rollback

Paradex Bitcoin Price Error to $0 Triggers Liquidations and Chain Rollback

N
News Editor 01
2026-07-23 04:55:14
Paradex briefly priced Bitcoin at $0 during maintenance, triggering liquidations on leveraged positions. The platform paused for about six to seven hours and rolled back to block 1,604,710 to reverse affected transactions.
ParadexBitcoinPerpetual DEXStarknetChain Rollback

Paradex, a Starknet-based perpetual DEX, suffered a major failure on January 19 after a database migration mistakenly priced Bitcoin at $0. Within minutes, the error triggered automated liquidations across many leveraged positions. The platform then halted services for roughly six to seven hours.

To address the damage, Paradex’s engineering team rolled the chain back to block 1,604,710 and reversed the affected transactions. It also removed pending trades submitted by users that had not been executed because of the incident. The immediate issue was contained, but the rollback became a separate point of debate.

Maintenance Error Broke the Link Between Internal Pricing and On-Chain Data

According to the source material, the glitch happened during routine maintenance. Paradex’s internal pricing system became de-synced from on-chain data, causing Bitcoin, and possibly other assets, to briefly display a value of zero. In a perpetuals venue with automatic liquidation logic, that kind of pricing fault can hit leveraged positions almost instantly.

The problem was not described as a matching engine failure. It came from off-chain dependency. That distinction matters. DeFi trading systems may execute on-chain, but databases, maintenance scripts, and data pipelines still sit in the critical path, and a fault in any of them can produce chain-level consequences.

Rollback Response Raises Questions About Finality

Reaction in the community was mixed. Some viewed the recovery as fast and practical because it reversed losses tied to an obvious system error. Others focused on the rollback itself, arguing that such intervention cuts against the expectation of immutable blockchain records. The incident reopened a familiar question for application-specific chains and exchange infrastructure: how much discretion should operators have once trades have touched the chain.

The source also notes that STRK fell 3% to 4%, while the broader crypto market was down about 2%. It adds that the wider market weakness was driven mainly by news related to Trump tariffs rather than the Paradex incident. That leaves the market impact relatively contained, even as the platform-specific implications remain serious.

Past DeFi Incidents Show the Same Pricing Risk From Different Angles

The material points to several earlier cases where bad pricing data or oracle issues led to liquidations and losses. It cites the 2022 Mango Markets exploit, where attackers withdrew about $117 million through oracle price manipulation. It also mentions the bZx flash loan attacks dating from 2020, which involved oracle-related weaknesses and caused liquidations worth millions, along with a Cosmos-related flash crash case in which token prices moved close to zero.

Paradex’s failure was described as an internal systems problem rather than an oracle attack, but the weakness exposed is similar. High leverage, automated liquidation engines, and reliance on pricing inputs create a tightly coupled system. Once one component breaks, the rest can react exactly as designed and still produce a destructive outcome.

This article was originally published by Bit.Fan. For more cryptocurrency news and market insights, visit www.bit.fan.
200

Disclaimer:

The market information, project data, and third-party content displayed on this platform are for industry information sharing only and do not constitute any form of investment advice or return commitment.

Cryptocurrency trading carries high risks. Users should fully assess their risk tolerance and make independent decisions. All profits, losses, and legal responsibilities are borne by the users themselves.