# eth2 validator queue lengths

This is the full lifecyle of validator.

For more details about the lifecycle, please read: [A note on Ethereum 2.0 phase 0 validator lifecycle](https://notes.ethereum.org/@hww/lifecycle#4-Validator-lifecycle)

There are a number of delays here, as a validator goes from each stage of their lifecyle to the next. The potential largest delays are the activation queue (2.a) and the exit queue (3.a.1, 3.a.2).

Both queues are based on the churn limit.

### Churn limit

As per current eth2 parameters (can be changed via hard fork):
`churn_limit = max(4, V_active/65536)`
where V_active is the number of active validators at any given instant.

`churn_limit` helps specify the number of validator that can enter or exit in an epoch.

Here are some example values to make this intuitive.

| `V_active` (current active validators) | Current ETH staked | `churn_limit` (validators per epoch) | Projected churn_limit (validators per day) | Projected churn_limit (validators per month) | churn_limit/V_active (fraction of validators per month) |
| --------- | ----- | -- | ---- | ------- | - |
| 0 | 0 | 4 | 900 | 27,000 | infinite |
| 65,536 | 2.1M | 4 | 900 | 27,000 | 41.2% |
| 131,072 | 4.2M | 4 | 900 | 27,000 | 20.6% |
| 262,144 | 8.4M | 4 | 900 | 27,000 | 10.3% |
| 393,216 | 12.6M | 6 | 1350 | 40,500 | 10.3% |
| 524,288 | 16.8M | 8 | 1800 | 54,000 | 10.3% |
| 786,432 | 25.2M | 12 | 2700 | 81,000 | 10.3% |
| 1,048,576 | 33.6M | 16 | 3600 | 108,000 | 10.3% |

As we can see above, a large number of validators entering or exiting can take multiple months to occur. For anything above 262,144 validators, the churn is about 10.3% of active validators per month.

For instance, if the current active validators are 250k but you expect them to grow to about 500k, this growth is going to happen at a pace increasing from 25k validators per month to a pace of 50k validators per month. Assuming an average pace of 40k validators per month, this process will take over 6 months. Which means if all those validators entered the queue today, the queue would be 6 month long. In practice it's likely that entries will be spread out over time, this could still mean a queue length of 1-2 months at any given point in time.

### Entry considerations

**You are eligible for no rewards while sitting in the activation queue.**

How quickly the queue will grow as we near the Merge depends on how much yield is expected and how much ETH will be staked in anticipation of that yield. Check out [Yields from issued ETH](https://github.com/ethereum/consensus-specs/pull/971), however this excludes priority fees and MEV, which means total yield and ETH staked will likely be larger.

**If the queue is currently small but you expect congestion in the future, consider staking now.**

Conversely, if queue is currently large but you expect it to become smaller, you may consider staking later - although this still means the current queue has to clear.

A large number of exits could reduce the number of active validators which in turn will not allow the churn limit to grow.

### Exit considerations

**You are eligible for all the usual rewards and penalties while sitting in the exit queue.**

Exit queue is likely to grow if there's an adverse event that affects a larger number of validators and makes it impossible or undesirable for them to continue validating. This could be due to a natural disaster, oppresive regulations, an attack on the network, or any number of other factors. Some of these factors may be beyond your control - you are however still responsible for ensuring your validator is functional while you're in the queue waiting to be exited. Hence you are exposed to the same rewards and penalties. This is [by design](https://ethresear.ch/t/rate-limiting-entry-exits-not-withdrawals/4942/16?u=samueldashadrach).

**You may consider exiting if you anticipate any such adverse event to affect your own validators.**

Note that penalties will go up if the network enters a non-final state because of this. Rewards might also go up as others are offline, which is good for you if your validators can stay online.

As earlier, the exit queue can take multiple months to clear. Moreover, as the the active validators reduce, the churn rate will reduce which means slower exits for everyone who is still left.

A last reason the exit queue may grow rapidly is if yields are expected to go down. For instance, staking rewards may be modified in a hard fork. Transaction activity may move to new apps or rollups or simply reduce, reducing priority fees and MEV. Lending platform yields may go up, making staking less attractive.

**You may consider exiting if you expect yields to be unfavourable in the future.**