Solana staking reward records now lack visible commission data in certain conditions, creating a gap between the payouts stakers receive and the fees charged to generate them.
According to staking provider P2P.org, commission information disappeared from both voting and staking rewards beginning September 8. The provider reported that while commissions continued to be charged and reward amounts remained accurate, the metadata explaining those fees became unavailable.
The change stems from Agave pull request #12588, merged on May 20, which emptied the commission field in reward metadata when the custom-collector feature is active. Custom collectors allow validators to direct commission income to designated accounts. The feature was introduced under protocol specification SIMD-0232.
Why the data gap exists
An empty commission field does not indicate that no fee was charged. Rather, it means the metadata is unavailable in the reward record itself. Agave's calculation code still processes commission before suppressing it in the final reward record.
The PR's rationale notes that accumulated rewards combining income from multiple sources make a single commission rate inappropriate for that aggregate. The implementation also suppresses the rate for individual stake rewards, removing readily available evidence even for single-account examinations.
Historical rates complicate verification
Solana's reward system uses a timing delay under rule SIMD-0249. Commissions earned during epoch E are calculated at the start of epoch E+1 using rates captured at the start of epoch E−1. This separation means a validator's current commission rate may differ from the rate applied to a particular payout, making historical context essential for verification.
A July engineering discussion flagged that effective commission values exist in cached runtime state unavailable through standard account reads. An October RPC request sought optional response fields containing the effective inflation commission, its epoch, and the bank slot to address this gap.
Existing tools and proposed solutions
Several approaches can help reconstruct commission information. Jito's Validator History Program documents up to 512 epochs of commission history per validator, though this alone does not establish that observations match the delayed effective rate used for specific payouts. Staking protocol Marinade's snapshot parser extracts inflation commission from epoch-stakes snapshots with vintage metadata, providing a reconstruction method rather than a guarantee of complete archive availability.
Data provider Trillium uses a different method, measuring net staking returns without relying on displayed commission data, though this approach measures earnings rather than independently verifying the fee applied.
Proposed remedies include storing commission rates alongside their effective epochs and creating RPC views of effective rates already used by the runtime. Both remain proposals without deployed implementations.
P2P.org, which offers staking and reporting services, has a commercial interest in the issue. The company reports that reward amounts continue to reconcile correctly despite the visibility problem.
