# Stale sync gap inflates issuance or fee

Rate or reward multiplied by a stale or mis-anchored time gap, overstating the period

_Area: DeFi accounting and math bugs. Seen in 2 disclosed reports; latest 2024-05-10. Updated 2026-10-29._

## How it works
A rate, issuance, or reward formula multiplies a per-unit quantity by an elapsed-time or block delta. The bug is that the delta and the state it is measured against are not kept in step: the anchor (last-synced block, request time, or settlement interval) updates on a different schedule than the delta used in the formula. The formula then counts time or blocks that were never actually accrued, so the output overstates the real period. The broken assumption is that "delta since last update" equals "period actually owed."

## Where it shows up
- A state variable used in a rate formula that is only updated conditionally (threshold-gated, every N blocks, on callback).
- Elapsed time measured from a request or initiation timestamp rather than from when the result became available.
- Accrual formulas that use elapsed time for TWAP or interval gating but omit it from the increment itself.
- Reward or auction curves keyed to an external oracle callback.

## How to review
Identify every multiplier that represents a time or block gap. For each, confirm the anchor it subtracts from advances on the same schedule as the gap it measures. The invariant: over any window, the sum of per-step accruals must equal one accrual over the whole window (path-independence: N x dt == 1 x N*dt). For async oracles, the elapsed-time anchor must be fulfillment time, not request time. Test by comparing many short settlements against one long one.

## Variants seen
- Block-delta issuance over-counted because lastSyncedBlock updates only past a threshold while issuance uses the full delta [E1](https://bugbountylibrary.com/r/c4%3A2024-03-taiko%3AH-01.md).
- Auction reward timed from RNG request rather than result availability, counting oracle latency as auction time [E2](https://bugbountylibrary.com/r/sherlock%3A2024-05-pooltogether%3AM-12.md).
- Funding index incremented without elapsed-time scaling, making accrual depend on settlement frequency [E3](https://bugbountylibrary.com/r/c4%3A2025-08-gte-perps-and-launchpad%3AM-11.md).

## Impact and severity
Led to inflated base fees that could halt the chain [E1](https://bugbountylibrary.com/r/c4%3A2024-03-taiko%3AH-01.md), systematic reward overpayment [E2](https://bugbountylibrary.com/r/sherlock%3A2024-05-pooltogether%3AM-12.md), and distorted PnL with unfair liquidation risk [E3](https://bugbountylibrary.com/r/c4%3A2025-08-gte-perps-and-launchpad%3AM-11.md). Severity rises when the inflated quantity feeds a consensus-critical or liquidation path, and when the gap is large or recurring.

## Evidence
- E1: https://bugbountylibrary.com/r/c4%3A2024-03-taiko%3AH-01.md
- E2: https://bugbountylibrary.com/r/sherlock%3A2024-05-pooltogether%3AM-12.md
- E3: https://bugbountylibrary.com/r/c4%3A2025-08-gte-perps-and-launchpad%3AM-11.md