XRPL Validators Revoke Votes on XLS-30D, AMM Not Launching on Feb 14

The eagerly-anticipated XLS-30D amendment for the XRPL, set to introduce an Automated Market Maker (AMM), will not launch on Feb. 14, as more validators revoke their votes.

This is due to a critical bug discovered in the AMM code. RippleX, the development arm of Ripple, identified the issue during an extended testing, citing a potential obstacle for multiple AMM transactions executing in the same ledger.

The bug primarily affects an edge case where the trading fee is set below 0.01%. In this case, the bug occurs when there is a cluster of transactions on the AMM, causing some transactions to halt for one ledger, resulting in a three to five seconds delay, before execution.

1/ In extended integration testing to ensure the AMM performs as intended over time, we identified an issue that could prevent multiple AMM txs from executing in the same ledger. We have proposed and are reviewing a fix: https://t.co/MPMKuI8dDK

— RippleX (@RippleXDev) February 1, 2024

Notably, once the AMM is created, the transactions must wait for one ledger to close before execution. After this ledger closes, the transactions execute normally, except for the AMMvote transaction.

Notably, the first AMMvote transaction would execute correctly, but the AMMvote transactions that follow will continue to wait for one ledger to close within three to five seconds before execution.

RippleX has proposed a fix, and is currently reviewing it. The proposed fix addresses the issue, ensuring correct identification of default fields within AMM inner objects and reinforcing stability in both the core ledger and AMM code.

Despite the relatively minor impact of the bug, XRPL validators have taken a stand by revoking their votes on the XLS-30D amendment. Recall that the amendment reached the required validator consensus on Feb. 1, with implementation ETA at Feb. 14.

XRPL Validators Revoke Yes Votes

Vet, a prominent XRPL validator, expressed concern about the potential negative image implementing a feature with a bug might cast on the XRPL’s reputation.

I withdraw my vote for the XRP ledger Automated Market Maker.

The AMM edge case effect is very minor and will not cause an issues based on what we know right now, that didn’t change and my opinion about it hasn’t either.

On the other hand is not reflective of the image of the…

— Vet 🏴‍☠️ (@Vet_X0) February 7, 2024

The validator operator reflected on a similar situation with the XLS-20D amendment, emphasizing the importance of a flawless launch for such a critical feature and advocated for the fix to be implemented before the implementation of the AMM.

xSpectar, another validator, echoed this sentiment, emphasizing the critical nature of the feature and the need to address known issues before activation. The NFT project advocated for the community to uphold the highest standards for the XRPL.

After careful (re)consideration we’ve changed our vote on the AMM amendment to Nay

It’s a critical feature, there is a known issue, we need to fix first before activating.

Voting otherwise would be irresponsible and compromise our integrity and it would be a dangerous… https://t.co/K0WkCaKdGn

— xSPECTAR ™ (@xSPECTAR) February 7, 2024

The official xSpectar account stressed that it is the responsibility of validators to avoid setting a dangerous precedent by allowing amendments with known issues to pass.

Daniel, CTO at Eminence and operator of the Eminence validator, took a principled stand by withdrawing his vote on the proposed AMM amendment. He cited concerns about compromising the integrity of best practices and setting a dangerous precedent.

I have decided to withdraw my vote on the proposed AMM amendment. After careful consideration and discussion within our community, the proposed feature, as it stands, goes against the best practices we strive to uphold.

Specifically, my concern lies in releasing a feature with a…

— Daniel “No” Keller (@daniel_wwf) February 7, 2024

Daniel emphasized the importance of prioritizing stability and functionality, even if it meant delaying the feature. He suggested deprecating the current AMM and introducing AMMv2 with fixes for consideration.

XLS-30D Amendment Below 80% Consensus

XRPL Labs, the development team behind Xaman, have also rescinded their Yes vote. As of press time, the validator consensus has declined to 71.43%, below the required 80% consensus. As a result, the amendment will not pass.

Meanwhile, most validators, including Ripple, Bithomp, XRPScan, Alloy Networks, Bitrue and 20 others, have retained their support. XRPGoat, another dUNL validator, suggested that the AMM could be deployed since the bug is minor, advocating for a fix while the feature runs.

Source