EIP-8080 Allowing the consolidation queue to be used for withdrawals (Let exits use the consolidation queue)

No call decision · Consensus & staking · EIP status Draft · created 2025-11-13

Allows validator withdrawals (exits) to use the consolidation queue's processing capacity when it is shorter. This opens up a workaround previously available only to validators with 2048 ETH or more, speeding up withdrawals for all validators.

Why it is needed

There is a loophole in the current consolidation mechanism. Validators with 2048 ETH or more can use the consolidation logic to route exits through the consolidation queue, and because the consolidation queue has consistently been shorter than the exit queue since the consolidation feature was introduced, this has become an attractive way to speed up exits. This method is unfair because it is only available to some validators. Meanwhile, the exit queue has remained over 40 days long for an extended period. A long queue degrades user experience, prevents operators from responding quickly to market or operational events, and also slows down how fast the network can change its stake composition after a problem occurs. Solo stakers who cannot issue liquid staking tokens or hold liquidity reserves are hit hardest. Meanwhile, when consolidation activity is low, the processing capacity allocated to the consolidation queue is simply wasted. This EIP proposes opening this up to everyone instead of removing the loophole, so the spare capacity can be used for processing exits.

What changes

Today

After EIP-8080

Who is affected

In depth

Background: Bypassing exits via the consolidation queue

The current consolidation mechanism design leaves a path by which validators with 2048 ETH or more can use a loophole in the consolidation logic to exit through the consolidation queue. Since the consolidation feature was introduced, the imbalance between the exit queue and the consolidation queue has consistently favored a shorter consolidation queue, making this loophole an attractive way to speed up exits. However, since only some validators can use it, it creates an unfair advantage. This EIP chooses to open this unintended feature to all exits rather than removing it. This makes more efficient use of the processing capacity allowed by the protocol's security parameter (the weak subjectivity period), enabling more exits to be processed without reducing security.

The modified function and routing condition

From the start of the epoch at which this EIP is activated, consensus layer clients MUST use the modified compute_exit_epoch_and_update_churn. The modification is a single branch added at the start of the function. earliest_consolidation_epoch is computed as max(state.earliest_consolidation_epoch, activation_exit_epoch), and if earliest_exit_epoch > earliest_consolidation_epoch, compute_consolidation_epoch_and_update_churn(state, 2 * exit_balance // 3) is called and its result returned. In other words, exits are routed to the consolidation queue only when the consolidation queue is shorter than the exit queue; otherwise the existing exit queue logic operates unchanged.

The meaning of the 2/3 conversion factor

The balance passed to the consolidation queue is 2 * exit_balance // 3 rather than the full exit_balance, because from a weak subjectivity standpoint, one unit of exit processing capacity corresponds to two thirds of a unit of consolidation processing capacity. Using this conversion factor keeps the impact on weak subjectivity the same whether the consolidation queue's capacity is used for consolidations or for exits. As a result, this maximizes the amount of work that can be processed while maintaining the same security guarantees. The Security Considerations section confirms this point: since only the allocation changes and total processing capacity stays the same, the weak subjectivity period is entirely unaffected.

Calculating maximum exit throughput

After this change, the maximum processable exit capacity becomes get_activation_exit_churn_limit(state) + 3 * get_consolidation_churn_limit(state) // 2. The 3 // 2 factor comes from one unit of exit capacity corresponding to two thirds of a unit of consolidation capacity. Under current parameters and a total stake of roughly 35.7 million ETH, this works out to 256 (the capped exit capacity) + (544 - 256) * 3/2 = 688 ETH per epoch, about 2.5 times the current maximum exit capacity of 256 ETH per epoch.

Why opening it up was chosen over removing it

The source gives two reasons for the decision to open this unintended feature up rather than remove it or leave it as is. The first is fairness: instead of only validators with 2048 ETH or more being able to exploit the loophole, all validators gain access to the same efficiency improvement. The second is efficiency: when consolidation activity is low, consolidation capacity is simply wasted, and letting exits use it improves staking liquidity without waste while preserving the same security guarantees. In short, on fairness alone either removal or opening up would work, but opening up is favored over removal because it delivers a large efficiency gain at no security cost.

Backward compatibility and current status

This EIP introduces a backward-incompatible change to the consensus layer and must be activated via a scheduled network upgrade. Test cases have not yet been written (TODO). Since the source's status is Draft, the details of the specification may still change. No record exists yet of a core developer call deciding to include it in a specific fork.

Original EIP · ← EIPs