Aave has partially reopened WETH access after the KelpDAO rsETH bridge exploit forced one of DeFi’s sharpest emergency responses of 2026. The key detail is not just that Ethereum Core V3 WETH is unfrozen again. It is that Aave is reopening the least impaired venue while keeping broader cross-chain risk controls in place. That distinction matters for lenders, borrowers, and anyone tracking how major lending markets isolate contagion after a bridge failure.
Ethereum Core V3 WETH Reopens While Broader Restrictions Stay in Place
Aave said on April 21, 2026, that WETH reserves on its Ethereum Core V3 market had been unfrozen, allowing users to supply WETH there again. The update came a little more than 24 hours after the protocol froze WETH across Core, Prime, Arbitrum, Base, Mantle, and Linea at approximately 02:00 UTC on April 20, 2026, according to Aave’s incident report and reporting that cited Aave’s public statement. The same report shows the initial emergency response started earlier, at approximately 19:00 UTC on April 18, 2026, when Aave’s Protocol Guardian froze all rsETH and wrsETH reserves across Aave V3 deployments and set loan-to-value to zero. That sequence is the story: first isolate the compromised collateral, then ringfence WETH liquidity, then selectively reopen the market that appears least exposed.
The exploit itself was large enough to justify that caution. Multiple reports put the forged mint at 116,500 rsETH, worth roughly $290 million to $294 million at the time, after an attacker abused KelpDAO’s LayerZero-linked bridge path on April 18, 2026. Aave’s own incident report treats the event as external to Aave, stressing that its contracts, oracles, and liquidation mechanics functioned as designed. That distinction is important because it explains why governance focused on containment rather than protocol shutdown. The attacker then used the unbacked rsETH as collateral to borrow real WETH, with outside reporting putting the borrowed amount at roughly $190 million across Ethereum and Arbitrum, while other coverage framed the broader bad-debt risk closer to $236 million depending on recovery assumptions.
The Real Signal Is Risk Segmentation, Not a Full Return to Normal
Here is what competitors mostly missed: the partial unfreeze is less a vote of confidence in system-wide normalization than a sign that Aave is segmenting risk chain by chain. Aave’s incident report lays out two broad loss scenarios. In Scenario 1, where losses are socialized more broadly, estimated bad debt reaches $123.7 million. In Scenario 2, where losses are isolated to L2 rsETH, estimated bad debt rises to $230.1 million, all on L2s, while Ethereum Core remains unaffected. That is the core reason Ethereum Core V3 could reopen first. If Ethereum Core WETH is not directly impaired under the L2-isolated scenario, reopening supply there becomes a targeted liquidity measure rather than a blanket all-clear.
The report is unusually specific. It says Ethereum Core absorbs the largest absolute loss under one modeled path at $91.8 million, yet the shortfall there would still be only 1.54% because the reserve is deep. Mantle, by contrast, would face the highest proportional hit in that same scenario at 9.54%. Under the L2-isolated scenario, Mantle’s WETH shortfall reaches 71.45% and Arbitrum’s 26.67%, while Ethereum Core is unaffected. Those numbers explain why a partial unfreeze on mainnet is plausible even while other venues remain constrained. It is not inconsistency. It is triage.
Liquidity, Not Just Solvency, Forced Aave’s Hand
There is another layer here. Aave’s report says the WETH reserves on Ethereum, Arbitrum, Base, Linea, and Mantle were all at 100% utilization at the time of writing, with idle balances below $20 on every chain. That is a brutal liquidity picture. When utilization is maxed out, liquidators cannot easily receive underlying WETH from the pool and may have to accept aWETH instead, which slows liquidation throughput. In plain English: even if the protocol remains operational, clogged WETH liquidity can turn a collateral event into a market-function event. That is why the freeze on WETH mattered so much. It was about stopping stress from spreading into other reserves, including stablecoins, while Aave bought time.
Aave also adjusted rates before the WETH freeze. At approximately 14:30 UTC on April 19, 2026, the Risk Steward reduced Slope 2 on Arbitrum, Base, Mantle, and Linea to 1.50%, bringing the borrow rate at 100% utilization down from 8.5%–10.5% to 3.0% APR. Then, at approximately 05:00 UTC on April 20, it applied a similar adjustment on Core, setting Slope 1 to 2%, Slope 2 to 3%, and optimal utilization to 94%. That is not cosmetic. It shows Aave was trying to keep borrowing conditions from becoming disorderly while reserves were effectively jammed.
Why Ethereum Core Was the Logical First Market to Reopen
I think the cleanest interpretation is this: Aave reopened the venue with the strongest combination of reserve depth, safety backstop, and modeled resilience. The Ethereum Core market has an Umbrella WETH module sized at 23,507.63 WETH, or about $54.06 million in the incident report. The same report notes that the Umbrella Safety Module covers Ethereum Core reserves only and does not extend to L2 deployments. So if governance believes the most severe unresolved damage sits on L2s, Ethereum Core becomes the obvious candidate for a controlled reopening. That is a much narrower and more defensible move than unfreezing WETH everywhere at once.
The broader market context supports that reading. Aave’s report counts 6,077 WETH suppliers across five chains, with $4.8746 billion in WETH collateral and $2.8709 billion in uncorrelated debt. It also says Base and Arbitrum are the least buffered WETH markets, with first liquidations triggered by just 0.77% and 1.77% WETH price drops respectively, because looping positions sit around a health factor near 1.03. Mantle, by contrast, does not see first liquidation until a 22% WETH drop. Those are not abstract stress metrics. They are the map Aave is using to decide where liquidity can safely return first.
What This Means for Users and for DeFi Risk Models
For users, the immediate takeaway is simple: Ethereum Core V3 WETH supply is back, but that does not mean the KelpDAO fallout is resolved. Aave’s own report says no official decision by Kelp on loss allocation or recovery had been publicly confirmed at the time of writing. That leaves the final bad-debt outcome open. For DeFi more broadly, the episode is a reminder that bridge design can create systemic lending risk even when the lending protocol itself does not fail. Aave’s response shows the new playbook: freeze the compromised collateral, cap contagion, reprice liquidity, and reopen only the segments with the clearest insulation.
Frequently Asked Questions
What exactly did Aave unfreeze?
Aave said on April 21, 2026, that WETH reserves on the Ethereum Core V3 market had been unfrozen, meaning users could supply WETH there again. Reporting tied the move to Aave’s public statement, while the earlier protocol-wide WETH freeze had affected Core, Prime, Arbitrum, Base, Mantle, and Linea beginning around 02:00 UTC on April 20, 2026.
What caused the freeze in the first place?
The trigger was the KelpDAO rsETH bridge exploit on April 18, 2026. Reports say an attacker forged a cross-chain message and caused 116,500 rsETH, worth about $290 million to $294 million, to be minted or released without backing. The attacker then used that rsETH as collateral on Aave to borrow real WETH.
Did Aave itself get hacked?
Aave’s incident report says no. It describes the rsETH bridge exploit as an external event and states that Aave’s contracts, oracles, and liquidation mechanics functioned as designed. The problem was contagion from compromised collateral entering Aave, not a direct exploit of Aave’s core code.
Why was Ethereum Core V3 reopened before other markets?
Aave’s modeling suggests Ethereum Core may be less impaired than some L2 venues, especially under the scenario where losses are isolated to L2 rsETH. The report also notes that Ethereum Core has a dedicated Umbrella WETH module worth 23,507.63 WETH, about $54.06 million, while that coverage does not extend to L2 deployments.
Is the crisis over now?
Not yet. Aave’s report says no official Kelp decision on loss allocation or recovery had been publicly confirmed at the time of writing. Estimated bad debt still ranges widely depending on the scenario, from $123.7 million to $230.1 million. So the partial unfreeze is a stabilization step, not a final resolution.
What is the biggest lesson from this event?
The biggest lesson is that bridge risk can migrate into lending markets fast. Aave’s response shows that modern DeFi risk management is no longer just about collateral ratios. It is also about liquidity segmentation, cross-chain containment, and deciding which markets can reopen without importing unresolved losses from somewhere else.
Leave a comment