[ARC] Aave V3 Fantom - Freeze reserves

[ARC] Aave V3 Fantom - Freeze reserves

aip: TBA

title: Aave V3 Fantom - Freeze reserves

status: Proposal

author: Marc Zeller

created: 02/08/2022


Simple Summary

This ARC proposes to freeze the Aave V3 Fantom market, allowing users to repay their debts and withdraw but blocking deposits and further borrowing on this market.
After the Harmony bridge event and the recent Nomad bridge exploit, the aave community should consider the risk/benefits of keeping an active Aave V3 market on fantom as this network is dependent of anyswap (multichain) bridge.

The Fantom Aave V3 market did not gain noticeable traction with a current market size of 9M$ and 2.4M$ of open borrowing position, this led to a market generating on average 300$ of fees daily for the protocol (~30$ of daily fees for the Aave treasury)

This market is not currently supported by the Aave Safety module and unless the Aave governance decides otherwise by specific vote, a failure of the main Bridge to Fantom will lead to users with assets worth 0.

This proposal calls for the Aave Guardian to freeze all reserves on the Aave V3 deployment on Fantom. This action would protect users by disabling the ability to deposit or borrow assets in the Aave V3 Market on Famtom, while still allowing repayment of debt, liquidations, withdrawals, and changes to the interest rates. Freezing the reserves will not interfere with any direct transfers to any of the aToken contracts.

Motivation

The unbalanced risk/reward of keeping of Aave V3 market on Fantom is the motivation of this ARC,

The risk of exposing users to potentially losing millions of $ due to causes exterior to intrinsic Aave security is considered not worth the 30$ of daily fees accrued by the Aave treasury.

Anyswap is a team that proved its seriousness and its safety and this proposal is not meant to dismiss their hard work, it is simply considered that the potential risk even if considered small outweighs the benefits of keeping the Aave V3 market on fantom active.

In general, the “EVM-verse” of alternative L1 seems to be currently decided in favor of Avalanche & Polygon.
L2s (Optimism/Arbitrum) currently slowly gaining traction allows Aave to allocate focus on networks presenting fewer vectors of risks in terms of bridges.

Specification

If passed, this AIP will call upon the Aave Guardian to call the setReserveFreeze() function with the appropriate parameters on each asset in the Aave V3 Market on Fantom. In addition, passing this AIP will also authorize the necessary changes to the IPFS Aave UI to support this change (disabling supply and borrow) as well as implementing additional warnings and alerts to educate users attempting to interact with the Aave V3 Market on Fantom.

After a delay of 5 days after this thread publication, a Snapshot vote will be published to freeze the FTM V3 aave market allowing the community to vote on this topic.

15 Likes

Definitely supporting this proposal. The Fantom market is right now either not prepared for Aave or will never be in the future. But leaving open the risk regarding all those bridge hacks isn’t an option.

In general i think Aave v2 on Avalanche should be frozen too. Why?
V3 is already out there, there is no need for v2 to be “running” anymore.
Also v2 AMM should be frozen. The market is simply not being used. The only assets being borrowed are stablecoins, which are also available in the other markets and ETH.

Focusing on the main networks and reducing every kind of risk and confusion should be the goal.
In general, after listening to Hasu’s talk about DeFi governance, i think we need way more professional work in here. Someone who really takes care of the protocol and wants it to succeed. But this will only be possible if there is some kind of incentive.

4 Likes

@EzR3aL The motivation of this proposal is to reduce exposure to a multichain bridge. I don’t think we have to worry about freezing other markets just because they aren’t being used enough.

on additional “freeze proposals” here are some thoughts.

  1. on Aave V2 market on avalanche, I agree it’s the past of Aave and we should focus on V3 but
  • lack of LM incentives
  • this market doesn’t present noticeable additional bridge risk compared to V3
  • BGD labs working on Proof-Of-Reserve implementation to limit this kind of risks for Avalanche
  • recent governance vote (I voted against) to expand SM support to this market
    makes it a weird candidate for a freeze IMHO.
  1. V2 AMM is a market that I would consider obsolete, that being said, instead of freezing, a better strategy will be to wait for V3 on mainnet and expand via isolation mode & emode the V3 main market collateral to the ones in the AMM market then simply let the market decide that this V2 implementation is obsolete.
    in every case, being a mainnet market, the bridge risk associated with it is zero.
3 Likes

Speaking of risks, isn’t Abritrum considered a beta product and potentially vulnerable to a multisig exploit such as what happened to Harmony Bridge (not a “bridge” exploit but a multisig exploit on a bridge)?

From https://developer.offchainlabs.com/docs/mainnet#some-words-of-caution

“That said, given that Arbitrum is still a new, cutting edge, and complex system, we currently maintain various levels of control over the chains while they’re still in this early Beta phase; this includes contract upgradeability, the ability to pause the system, validator whitelisting, and token bridge whitelisting”

Trying to separate prudent protocol risk from L2 vs side-chain preference. If Fantom is culled based on a “potential risk/reward” of an attack that hasn’t happened, shouldn’t Arbritrum also?

2 Likes

I think the main issue of risk here is that activity on fantom market is quite less than expected. I am not sure why no interest has been taken by fantom foundation to prop up interest and incentivize depositors to fantom market. Anyway, I agree with the freeze but would allow fantom foundation to review the proposal (if they are able to) and take a considered decision.

Can someone help me understand the risk/reward argument? If we define the value of risk/reward to be (risk of bridge being hacked * total locked value) / total reward: Yes the reward on Fantom is low compared to other chains, but the total locked value is proportionally low as well. And assuming the risk of bridge being hacked is similar on all chains, shouldn’t the risk/reward value be roughly the same on all chains?

I hear what you are saying here…but also think they are quite different scenarios in reality.

I know, i just wanted to kickstart this conversation somehow too. But maybe an extra thread for this would be better.

Поддерживаю предложение о заморозке.
Наша обязанность минимизировать риски.
Мы сможем снова начать работу, когда увидим соответствующий прогресс в безопасности переноса токенов.

If a primary motivation for this is a bridge was hacked, why isn’t the proposal to freeze all bridged assets on every chain or to remove any asset that relies on two separate chains to function? I don’t understand why Fantom is singled out as if it’s the only iteration of Aave that’s vulnerable to this.

I also don’t understand the line of thought that says the more funds that are at risk from a bridging hacks, the less we care about it. Approaching $1B in bridged assets on blockchain x is clearly not an issue but less than $10M in bridged assets on Fantom is just too large of a threat to let stand.

To me this should be broken down into two separate proposals:

  1. If bridged assets and multi-chain assets truly are a threat then they should be removed. This should include all chains and all bridged and wrapped assets imo. I don’t understand the logic that says WBTC on Avalanche or Fantom is a threat but WBTC on Ethereum is obviously of no risk whatsoever or that bridged USDC on Fantom is a clear and present danger but bridged USDC on Polygon is obviously very safe and relatively risk-less.

  2. A proposal that, “any instance of Aave that doesn’t have a TVL of x by time y is depreciated.”

Threats only count if they’re on Fantom doesn’t fly imo.

2 Likes

the snapshot vote for this ARC has been published and vote will open tomorrow.

I think all sides had room to discuss this proposal it’s now up to the Aave token holders to decide by expressing their votes.

The snapshot vote is now closed with a majority of YAE vote.

Fantom V3 Aave Market reserve will be frozen by the community guardian shortly.

Users will be able to withdraw and repay their position but deposits, and new borrows won’t be available anymore.

The argument of bridged assest (in general) are risky, therefore, we should freeze a network does not make sense.
This argument can be generalized to most assets on any network. I am not aware that multichain has any inherent security risk.