Maintenance-pool `deposit()` allows pool-id squatting via `TokenMismatch` griefing
I maintainer di solito rispondono entro 1 giorno
Valutazione
- Difficoltà
- 5/5
- Tempo stimato
- Più di una settimana
- Idoneità per principianti
- 20/100
- Tipo di issue
- Funzionalità
- Chiarezza
- Specificata chiaramente
- Stato di attività
- Ferma
- Stack tecnologico
- rust
- Ambito
- backend-api-design, security
Direzione di ricerca
Read contracts/maintenance-pool/src/lib.rs around deposit() and contracts/maintenance-pool/src/types.rs for the pool key, then inspect the pool ID generation assumption described in the issue. Review the open linked pull request before starting, since it may already be addressing this work. A complete fix must prevent token squatting, preserve permissionless recurring deposits and admin-authorized withdrawals, and include the described tests in contracts/maintenance-pool/src/test.rs; the issue specifies cargo test --workspace.
Scritto dal modello di indicizzazione a partire dal testo della issue.
Descrizione
Background / Context
contracts/maintenance-pool/src/lib.rs::deposit() creates a MaintenancePool on first deposit for a given pool_id, locking in whatever token the first depositor supplies (pool.deposit_count > 0 && pool.token != token → TokenMismatch). pool_id is described in types.rs as "an off-chain-assigned id for a repo/org," implying it's likely derived deterministically (e.g., hash of owner/repo) and therefore guessable ahead of time by anyone watching GitHub activity.
Problem Statement
An attacker can call deposit(pool_id, attacker, some_worthless_token, 1) for any repo/org they expect to receive real sponsor funding, permanently locking that pool to the worthless token. Every subsequent legitimate sponsor deposit in the intended asset (e.g., USDC on Stellar) reverts with TokenMismatch, with no recovery path in the contract (no admin override, no per-token sub-pools). This is a cheap, repeatable denial-of-service against the "recurring maintenance funding" product surface, distinct from and worth separating from the escrow-id-squatting issue because the recovery constraints differ (a pool is meant to be long-lived and repeatedly topped up, so there's no natural "just use a different id" workaround the way there might be for a one-off escrow).
Requirements
- Determine whether
pool_idgeneration is/should be unpredictable (out of scope to change the backend, but document the assumption) or whether the contract must be hardened regardless. - Design and implement a fix: candidates include admin-gated pool creation (a separate
create_pool(pool_id, token)admin call beforedepositis permitted), an allowlist of valid tokens set atinitialize, or a per-(pool_id, token)compound key so multiple tokens can coexist per pool without collision (requires rethinkingget_pool/balancesemantics). - Add tests in
contracts/maintenance-pool/src/test.rsreproducing the squat and validating the fix. - Ensure the fix doesn't regress the "sponsors can deposit repeatedly over time" recurring-funding use case central to this contract's purpose.
Acceptance Criteria
- Documented threat model for
pool_idpredictability - Implemented fix in
contracts/maintenance-pool/src/lib.rsandtypes.rsas needed - Tests proving squat attempts no longer block legitimate sponsors
- No breaking change to
withdraw's admin-authorized payout flow -
cargo test --workspacepasses
Technical Notes / Hints
- Code:
contracts/maintenance-pool/src/lib.rslines ~51-104,DataKey::Pool(u64)intypes.rs. - Consider interaction with
Deposithistory keyed by(pool_id, deposit_index)if you change how pools are created/keyed.
Difficulty Justification
The naive fix (reject deposits from non-admin on pool creation) conflicts with the contract's explicit design goal of being sponsor-initiated and permissionless for deposits. A correct solution must preserve permissionless recurring deposits while eliminating the squatting vector — this is a genuine access-control/design tradeoff requiring judgment, not a mechanical patch.
- Lingua principale
- Rust
- Stelle
- 1
- Fork
- 102
- Merge medio
- 4h 12m
- PR unite (30g)
- 30
Preparare l'ambiente
- Nessun Dockerfile né file Docker Compose
- Ha un modello di pull request
- Nessuna guida per i contributori
Come iniziare
- Leggi tutta la issue e poi la guida ai contributi del progetto.
- Commenta sulla issue per dire che te ne occupi tu — evita che due persone facciano lo stesso lavoro.
- Fai un fork del repository e lavora su un branch.
- Apri una pull request che faccia riferimento al numero della issue.
Altre issue di MergeFi/contracts
-
allocations map retains released issues — phantom ledger driftForse di nuovo libera @PINYOPATTANAWASANPORN l’ha presa 12 giorni fa e non c’è nessuna pull request aperta. Aperta
Difficoltà 2/5 1-3 ore Idoneità per principianti 25/100
MergeFi/contracts#608 · 3 commenti ·
I maintainer di solito rispondono entro 1 giorno
-
escrow::contribute and milestones::contribute accept new funds after deadline + GRACE_PERIOD, when anyone can already refund or cancelForse già presa @urielit1313-creator l’ha presa 29 giorni fa. Apertabug
Difficoltà 3/5 1-2 giorni Idoneità per principianti 25/100
MergeFi/contracts#568 · 3 commenti ·
I maintainer di solito rispondono entro 1 giorno
-
Makefile deploy targets deploy the unoptimised .wasm — `stellar contract optimize` writes *.optimized.wasmForse già presa @urielit1313-creator l’ha presa 29 giorni fa. Apertabug good first issue
Difficoltà 2/5 1-3 ore Idoneità per principianti 25/100
MergeFi/contracts#567 · 1 commento ·
I maintainer di solito rispondono entro 1 giorno
-
keep_alive never refreshes instance storage, so a contract kept alive only via keep_alive can still archive Admin/Oracle/TreasuryForse già presa @urielit1313-creator l’ha presa 29 giorni fa. Apertabug
Difficoltà 4/5 3-5 giorni Idoneità per principianti 25/100
MergeFi/contracts#566 · 3 commenti ·
I maintainer di solito rispondono entro 1 giorno
-
[Regression on closed #254] milestones::deallocate still has no test asserting admin authorizationForse già presa @urielit1313-creator l’ha presa 29 giorni fa. Apertasecurity testing
Difficoltà 2/5 1-3 ore Idoneità per principianti 25/100
MergeFi/contracts#565 · 1 commento ·
I maintainer di solito rispondono entro 1 giorno
Tutte le issue di MergeFi/contracts
Issue simili
-
Difficoltà 2/5 1-3 ore Idoneità per principianti 76/100
I maintainer di solito rispondono entro 1 giorno
-
design-data-spec
Difficoltà 2/5 1-3 ore Idoneità per principianti 78/100
adobe/spectrum-design-data#1744 ·
I maintainer di solito rispondono entro 1 giorno
-
Difficoltà 2/5 1-3 ore Idoneità per principianti 78/100
I maintainer di solito rispondono entro 1 giorno
-
USERID reads /etc/passwd directly instead of looking the user upForse già presa Una pull request collegata a questa issue è aperta o già unita. Apertadefect
Difficoltà 2/5 1-3 ore Idoneità per principianti 67/100
I maintainer di solito rispondono entro 1 giorno
-
Difficoltà 2/5 1-3 ore Idoneità per principianti 78/100
I maintainer di solito rispondono entro 1 giorno