ZenHive mpp
Tempo fee sponsorship in mpp bounds each transaction but not aggregate exposure, allowing concurrent sponsor-wallet drain
Allocation of Resources Without Limits or Throttling in ZenHive mpp allows an unauthenticated remote client to drain the fee-payer wallet through concurrent sponsored payments, denying service to legitimate payers once it is empty. MPP.Methods.Tempo.FeePayerPolicy enforces its ceilings (max_gas, max_fee_per_gas, max_priority_fee_per_gas, the worst-case gas_limit * max_fee_per_gas <= max_total_fee budget cap, and a validity window) against one transaction at a time, and nothing accounts for exposure across concurrent requests. reserve_hash_atomic/2 is keyed on the transaction hash, so it prevents duplicate broadcast of the same signed transaction but not N distinct sponsored transactions carrying distinct expiring nonces. Committed sponsor exposure is therefore N times max_total_fee, bounded by nothing in the library, and the default 900 second validity window lets co-signed transactions stay broadcastable and uncounted for that entire period. This issue affects mpp: from 0.2.0 before 0.12.0.
Affected versions and mitigation
From the CVE record and the vendor's references.Affected
- >= 0.2.0, < 0.12.0
Fixed in 0.12.0
- >= d29d54e507918db00a5b65d90136b73166c017d7, < ddc46868fba57ccebb567c04709812b466123076
Fixed in ddc46868fba57ccebb567c04709812b466123076
Mitigation
Upgrade mpp to 0.12.0 or later.
Upgrade mpp to ddc46868fba57ccebb567c04709812b466123076 or later.
Workaround
Disable Tempo fee sponsorship by setting "fee_payer" => false (the default), so the client pays its own gas and the sponsor wallet is never committed. Where sponsorship must stay enabled, lowering max_total_fee and max_validity_window_seconds reduces the exposure each concurrent request can commit and how long it stays outstanding, without bounding the aggregate.
References
Timeline
Days relative to CVE publicationNo public PoC found yet; not in CISA KEV.
PoC repositories
Found on GitHub.No PoC repositories found yet.