Skip to content

fix: add rollup config fields #7

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Jul 17, 2024

Conversation

yiweichi
Copy link
Member

Problem:
Field 'max_uncompressed_batch_bytes_size' is missing from docker/templates/rollup-config.json, Which will cause error when launching rollup-node.

Solution:
Add "max_uncompressed_batch_bytes_size": 634880 to docker/templates/rollup-config.json.

@yiweichi yiweichi requested a review from Thegaram July 17, 2024 08:53
@Thegaram Thegaram merged commit 01f2db0 into feat-deterministic-deployment Jul 17, 2024
2 checks passed
@Thegaram Thegaram deleted the fix-rollup-config branch July 17, 2024 09:11
dasdsadasca pushed a commit to dasdsadasca/scroll-contracts that referenced this pull request May 27, 2025
I've completed a meticulous verification of 8 of your specified vulnerability claims.

**Key Confirmed/Partially Confirmed Vulnerabilities:**

1.  **scroll-tech#7 Cross-Chain Message Replay via State Inconsistency (CONFIRMED - High/Critical):**
    *   **Issue:** In `L2ScrollMessenger._executeMessage` (and symmetrically in `L1ScrollMessenger.relayMessageWithProof`), the execution flag (`isL1MessageExecuted` / `isL2MessageExecuted`) is set *after* the external call.
    *   **Impact:** If the external call succeeds but the subsequent transaction fails (e.g., out-of-gas before setting the flag), and assuming the target contract's state changes from the successful call persist (as per your detailed exploit premise), the message can be replayed, leading to double execution (e.g., double minting/transfers).
    *   **Recommendation:** Set execution flags *before* the external call.

2.  **scroll-tech#8 Gas Price Oracle Manipulation (PARTIALLY CONFIRMED - High):**
    *   **Issue:** If a malicious L2 Sequencer can feed a significantly inflated L1 basefee to the `L1GasPriceOracle` contract on L2.
    *   **Impact:** Users on L2 would be overcharged for initiating L2->L1 messages, potentially making withdrawals/L2->L1 interactions economically unviable.
    *   **Mitigation:** Relies on trusted Sequencer operation and/or L2 node validation of L1 data.

3.  **scroll-tech#3 Enforced Batch Mode Bypass (PARTIALLY CONFIRMED - Medium/Low impact on direct censorship):**
    *   **Issue:** Sequencers/Provers can prevent the "finalization staleness" trigger for enforced mode by regularly finalizing batches, even if these batches exclude specific L2-originated transactions.
    *   **Impact:** If L1->L2 messages are not also stuck, enforced mode may not activate, allowing censorship of specific L2-native transactions. This is a limitation of the trigger's scope for L2 tx censorship resistance.
    *   **Mitigation:** Users facing L2 tx censorship would need to use L1->L2 messages.

**Refuted Critical/High Vulnerabilities (as per your specific claims):**
*   **scroll-tech#1 ScrollChain Batch Finalization Race Condition:** REFUTED. Critical state updates occur after proof verification.
*   **scroll-tech#2 L1ScrollMessenger Withdrawal Proof Bypass:** REFUTED. Comprehensive hashing and replay protection are effective.
*   **scroll-tech#4 Gateway Router Reentrancy Attack:** REFUTED. Layered defenses (context locks, nonReentrant guards) protect against claimed exploits.
*   **scroll-tech#5 Batch Bridge Hash Collision Attack:** REFUTED. Hashing mechanism is sound against collisions for fixed-size inputs.
*   **scroll-tech#6 Lido Bridge Rebasing Token Manipulation:** REFUTED. Bridge is designed for non-rebasing wstETH.

This information includes all prior documentation and detailed vulnerability analysis reports culminating in these verified findings.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants