# Moonwell's cbETH market still shows negative liquidity six months after an oracle misconfiguration

Utilization above 100% means suppliers who were never liquidated cannot withdraw, and the protocol's latest recovery plan assigns nothing to closing the roughly $1.8 million gap.

- Published: 2026-08-26T05:49:32.350Z
- Canonical: https://polylog.news/crypto/2026-08-26/moonwell-s-cbeth-market-still-shows-negative-liquidity-six-m
- Publisher: Polylog (Crypto desk)
- Section: crypto
- Sources: [CryptoSlate](https://cryptoslate.com/moonwell-has-a-1-8m-liquidity-hole-and-its-latest-plan-allocates-0-to-fix-it/), [DailyCoin](https://dailycoin.com/1-8m-gone-in-minutes-moonwells-oracle-glitch-shakes-defi-lending), [Moonwell Governance Forum](https://forum.moonwell.fi/t/recovery-plan-cbeth-incident-and-moonwell-apollo-onboarding/2084)

Live contract data for the lending protocol Moonwell shows its market for Coinbase Wrapped Staked ETH (cbETH) on Base sitting at [100.43% utilization with negative available liquidity](https://cryptoslate.com/moonwell-has-a-1-8m-liquidity-hole-and-its-latest-plan-allocates-0-to-fix-it/). In a pooled lending market, utilization above 100% means borrowers have taken out more than the pool holds, so suppliers who did nothing wrong and were never liquidated cannot withdraw their collateral.

The cause was not an exploit in the usual sense. Between February 14 and 18, a single misconfiguration in the oracle pricing formula [valued cbETH at about $1.12 instead of roughly $2,200](https://dailycoin.com/1-8m-gone-in-minutes-moonwells-oracle-glitch-shakes-defi-lending), and the liquidation engine did exactly what it was written to do, seizing collateral whose value it read as having collapsed. Roughly 1,096 cbETH left the market, leaving a shortfall of about $1.8 million and roughly 181 affected borrowers on Base. The risk manager, Anthias Labs, cut supply and borrow caps to near zero to stop new exposure.

Moonwell's governance response [proposed about $2.68 million in compensation](https://forum.moonwell.fi/t/recovery-plan-cbeth-incident-and-moonwell-apollo-onboarding/2084), paid roughly $310,000 immediately from its Apollo treasury and committed the remaining balance to future protocol revenue. That plan addressed the borrowers who were liquidated at the wrong price. It did not fund the hole itself, so the suppliers unable to withdraw from a market at more than full utilization have no published path to get their funds back. The latest proposal allocates nothing to that gap.

Base carries $5.51 billion in total value locked and $12.57 billion in total value secured as a layer-2 network, up 14.4% over seven days, so the sums involved are small against the chain. The precedent is not small. A trusted off-contract input failed, the code executed correctly, and six months later a portion of user deposits remains frozen while the protocol repays from revenue it has not yet earned.

## What this means

Lending protocols price risk on the assumption that oracle inputs are correct, and an audit scoped to contract code cannot catch a wrong constant in a price feed configuration. Suppliers in the affected market bear the loss through frozen withdrawals rather than an explicit haircut, which delays recognition without removing it, and Moonwell's token holders bear the rest through revenue diverted to repayment. The two outcomes that matter are whether governance funds the gap from the treasury, which converts a user loss into a protocol loss, or leaves the market frozen until borrowers repay on their own.

## What to watch

- Whether a new Moonwell proposal explicitly funds the cbETH liquidity gap, which would show the protocol accepting the loss on its own balance sheet.
- Whether other lending markets publish oracle configuration reviews covering feed parameters and not just contract logic, since the failure sat in configuration rather than code.
- Whether cbETH borrowers repay and restore liquidity naturally, the only route out that costs the treasury nothing.