feat(prover, coordinator): update vk handling logic (#931 )

Co-authored-by: Steven Gu <asongala@163.com> Co-authored-by: HAOYUatHZ <HAOYUatHZ@users.noreply.github.com>
feat: libzkp v0.8.1 (#894 )
2026-01-12 07:28:08 -05:00 · 2023-09-07 10:59:47 +08:00 · 2023-09-07 10:34:35 +08:00 · 2023-09-06 18:22:25 +08:00 · 2023-09-06 16:08:41 +08:00 · 2023-09-06 15:28:31 +08:00
56 changed files with 1431 additions and 664 deletions
--- a/.github/workflows/contracts.yaml
+++ b/.github/workflows/contracts.yaml
--- a/.github/workflows/docker.yaml
+++ b/.github/workflows/docker.yaml
--- a/.github/workflows/integration.yaml
+++ b/.github/workflows/integration.yaml
--- a/.gitignore
+++ b/.gitignore
@@ -1,9 +1,22 @@
-.idea
+# Asset files
 assets/params*
 assets/seed
-coverage.txt
+
+# Built binaries
 build/bin
+
+coverage.txt
 *.integration.txt

+# Visual Studio Code
+.vscode
+
+# IntelliJ
+.idea
+
+# MacOS
+.DS_Store
+
 # misc
 sftp-config.json
+*~
--- a/CODE_OF_CONDUCT.md
+++ b/CODE_OF_CONDUCT.md
@@ -0,0 +1,128 @@
+# Contributor Covenant Code of Conduct
+
+## Our Pledge
+
+We as members, contributors, and leaders pledge to make participation in our
+community a harassment-free experience for everyone, regardless of age, body
+size, visible or invisible disability, ethnicity, sex characteristics, gender
+identity and expression, level of experience, education, socio-economic status,
+nationality, personal appearance, race, religion, or sexual identity
+and orientation.
+
+We pledge to act and interact in ways that contribute to an open, welcoming,
+diverse, inclusive, and healthy community.
+
+## Our Standards
+
+Examples of behavior that contributes to a positive environment for our
+community include:
+
+* Demonstrating empathy and kindness toward other people
+* Being respectful of differing opinions, viewpoints, and experiences
+* Giving and gracefully accepting constructive feedback
+* Accepting responsibility and apologizing to those affected by our mistakes,
+  and learning from the experience
+* Focusing on what is best not just for us as individuals, but for the
+  overall community
+
+Examples of unacceptable behavior include:
+
+* The use of sexualized language or imagery, and sexual attention or
+  advances of any kind
+* Trolling, insulting or derogatory comments, and personal or political attacks
+* Public or private harassment
+* Publishing others' private information, such as a physical or email
+  address, without their explicit permission
+* Other conduct which could reasonably be considered inappropriate in a
+  professional setting
+
+## Enforcement Responsibilities
+
+Community leaders are responsible for clarifying and enforcing our standards of
+acceptable behavior and will take appropriate and fair corrective action in
+response to any behavior that they deem inappropriate, threatening, offensive,
+or harmful.
+
+Community leaders have the right and responsibility to remove, edit, or reject
+comments, commits, code, wiki edits, issues, and other contributions that are
+not aligned to this Code of Conduct, and will communicate reasons for moderation
+decisions when appropriate.
+
+## Scope
+
+This Code of Conduct applies within all community spaces, and also applies when
+an individual is officially representing the community in public spaces.
+Examples of representing our community include using an official e-mail address,
+posting via an official social media account, or acting as an appointed
+representative at an online or offline event.
+
+## Enforcement
+
+Instances of abusive, harassing, or otherwise unacceptable behavior may be
+reported to the community leaders responsible for enforcement at
+contributor@scroll.io.
+All complaints will be reviewed and investigated promptly and fairly.
+
+All community leaders are obligated to respect the privacy and security of the
+reporter of any incident.
+
+## Enforcement Guidelines
+
+Community leaders will follow these Community Impact Guidelines in determining
+the consequences for any action they deem in violation of this Code of Conduct:
+
+### 1. Correction
+
+**Community Impact**: Use of inappropriate language or other behavior deemed
+unprofessional or unwelcome in the community.
+
+**Consequence**: A private, written warning from community leaders, providing
+clarity around the nature of the violation and an explanation of why the
+behavior was inappropriate. A public apology may be requested.
+
+### 2. Warning
+
+**Community Impact**: A violation through a single incident or series
+of actions.
+
+**Consequence**: A warning with consequences for continued behavior. No
+interaction with the people involved, including unsolicited interaction with
+those enforcing the Code of Conduct, for a specified period of time. This
+includes avoiding interactions in community spaces as well as external channels
+like social media. Violating these terms may lead to a temporary or
+permanent ban.
+
+### 3. Temporary Ban
+
+**Community Impact**: A serious violation of community standards, including
+sustained inappropriate behavior.
+
+**Consequence**: A temporary ban from any sort of interaction or public
+communication with the community for a specified period of time. No public or
+private interaction with the people involved, including unsolicited interaction
+with those enforcing the Code of Conduct, is allowed during this period.
+Violating these terms may lead to a permanent ban.
+
+### 4. Permanent Ban
+
+**Community Impact**: Demonstrating a pattern of violation of community
+standards, including sustained inappropriate behavior,  harassment of an
+individual, or aggression toward or disparagement of classes of individuals.
+
+**Consequence**: A permanent ban from any sort of public interaction within
+the community.
+
+## Attribution
+
+This Code of Conduct is adapted from the [Contributor Covenant][homepage],
+version 2.0, available at
+https://www.contributor-covenant.org/version/2/0/code_of_conduct.html.
+
+Community Impact Guidelines were inspired by [Mozilla's code of conduct
+enforcement ladder](https://github.com/mozilla/diversity).
+
+[homepage]: https://www.contributor-covenant.org
+
+For answers to common questions about this code of conduct, see the FAQ at
+https://www.contributor-covenant.org/faq. Translations are available at
+https://www.contributor-covenant.org/translations.
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@@ -0,0 +1,42 @@
+## Contributing
+
+[fork]: /fork
+[pr]: /compare
+[style]: https://standardjs.com/
+[code-of-conduct]: CODE_OF_CONDUCT.md
+
+Hi there! We're thrilled that you'd like to contribute to this project. Your help is essential for keeping it great.
+
+Please note that this project is released with a [Contributor Code of Conduct][code-of-conduct]. By participating in this project you agree to abide by its terms.
+
+## Contribute to Scroll
+
+Did you know there are many ways of contributing to Scroll? If you are looking to contribute to by adding Scroll to existing Dev Tools or by doing integrations please go to the [Contribute to Scroll](https://github.com/scroll-tech/contribute-to-scroll) repo instead. If you are looking to contribute to Scroll's Halo2 zkEVM circuits please refer to the [zkEVM circuits](https://github.com/scroll-tech/zkevm-circuits) repo. This repository covers the Scroll infrastructure and smart contracts, if you want to contribute to these areas continue reading this document.
+
+## Issues and PRs
+
+If you have suggestions for how this project could be improved, or want to report a bug, open an issue! We'd love all and any contributions. If you have questions, too, we'd love to hear them.
+
+We'd also love PRs. If you're thinking of a large PR, we advise opening up an issue first to talk about it, though! Look at the links below if you're not sure how to open a PR.
+
+## Submitting a pull request
+
+1. [Fork][fork] and clone the repository.
+1. Create a new branch: `git checkout -b my-branch-name`.
+1. Make your change, add tests, and make sure the tests still pass.
+1. Push to your fork and [submit a pull request][pr].
+1. Pat yourself on the back and wait for your pull request to be reviewed and merged.
+
+Here are a few things you can do that will increase the likelihood of your pull request being accepted:
+
+- Write and update tests.
+- Keep your changes as focused as possible. If there are multiple changes you would like to make that are not dependent upon each other, consider submitting them as separate pull requests.
+- Write a [good commit message](http://tbaggery.com/2008/04/19/a-note-about-git-commit-messages.html).
+
+Work in Progress pull requests are also welcome to get feedback early on, or if there is something that blocked you.
+
+## Resources
+
+- [How to Contribute to Open Source](https://opensource.guide/how-to-contribute/)
+- [Using Pull Requests](https://help.github.com/articles/about-pull-requests/)
+- [GitHub Help](https://help.github.com)
--- a/README.md
+++ b/README.md
@@ -1,7 +1,34 @@
 # Scroll Monorepo

+[![rollup](https://github.com/scroll-tech/scroll/actions/workflows/rollup.yml/badge.svg)](https://github.com/scroll-tech/scroll/actions/workflows/rollup.yml)
+[![contracts](https://github.com/scroll-tech/scroll/actions/workflows/contracts.yml/badge.svg)](https://github.com/scroll-tech/scroll/actions/workflows/contracts.yml)
+[![bridge-history](https://github.com/scroll-tech/scroll/actions/workflows/bridge_history_api.yml/badge.svg)](https://github.com/scroll-tech/scroll/actions/workflows/bridge_history_api.yml)
+[![coordinator](https://github.com/scroll-tech/scroll/actions/workflows/coordinator.yml/badge.svg)](https://github.com/scroll-tech/scroll/actions/workflows/coordinator.yml)
+[![prover](https://github.com/scroll-tech/scroll/actions/workflows/prover.yml/badge.svg)](https://github.com/scroll-tech/scroll/actions/workflows/prover.yml)
+[![integration](https://github.com/scroll-tech/scroll/actions/workflows/integration.yml/badge.svg)](https://github.com/scroll-tech/scroll/actions/workflows/integration.yml)
 [![codecov](https://codecov.io/gh/scroll-tech/scroll/branch/develop/graph/badge.svg?token=VJVHNQWGGW)](https://codecov.io/gh/scroll-tech/scroll)

+<a href="https://scroll.io">Scroll</a> is a zkRollup Layer 2 dedicated to enhance Ethereum scalability through a bytecode-equivalent [zkEVM](https://github.com/scroll-tech/zkevm-circuits) circuit. This monorepo encompasses essential infrastructure components of the Scroll protocol. It contains the L1 and L2 contracts, the rollup node, the prover client, and the prover coordinator.
+
+## Directory Structure
+
+<pre>
+├── <a href="./bridge-history-api/">bridge-history-api</a>: Bridge history service that collects deposit and withdraw events from both L1 and L2 chain and generates withdrawal proofs
+├── <a href="./common/">common</a>: Common libraries and types
+├── <a href="./coordinator/">coordinator</a>: Prover coordinator service that dispatches proving tasks to provers
+├── <a href="./database">database</a>: Database client and schema definition
+├── <a href="./src">l2geth</a>: Scroll execution node
+├── <a href="./prover">prover</a>: Prover client that runs proof generation for zkEVM circuit and aggregation circuit
+├── <a href="./prover-stats-api">prover-stats-api</a>: Collect and show prover statistics
+├── <a href="./rollup">rollup</a>: Rollup-related services
+├── <a href="./rpc-gateway">rpc-gateway</a>: RPC gateway external repo
+└── <a href="./tests">tests</a>: Integration tests
+</pre>
+
+## Contributing
+
+We welcome community contributions to this repository. Before you submit any issues or PRs, please read the [Code of Conduct](CODE_OF_CONDUCT.md) and the [Contribution Guideline](CONTRIBUTING.md).
+
 ## Prerequisites
 + Go 1.19
 + Rust (for version, see [rust-toolchain](./common/libzkp/impl/rust-toolchain))
@@ -63,6 +90,10 @@ go test -v -race -covermode=atomic scroll-tech/common/...

 ## Testing Contracts

-You can find the unit tests in [`<REPO_DIR>/contracts/src/test/`](/contracts/src/test/), and integration tests in [`<REPO_DIR>/contracts/integration-test/`](/contracts/integration-test/).
+You can find the unit tests in [`contracts/src/test/`](/contracts/src/test/), and integration tests in [`contracts/integration-test/`](/contracts/integration-test/).

-For more details on contracts, see [`/contracts`](/contracts).
+See [`contracts`](/contracts) for more details on the contracts.
+
+## License
+
+Scroll Monorepo is licensed under the [MIT](./LICENSE) license.
--- a/common/libzkp/impl/Cargo.lock
+++ b/common/libzkp/impl/Cargo.lock
@@ -32,7 +32,7 @@ dependencies = [
 [[package]]
 name = "aggregator"
 version = "0.1.0"
-source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.7.5#aa9a9aff698a5b253d1f3c29ea3d3006364777bf"
+source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.8.1#fd703cc269a07f28b1febc7ce021792e2733564b"
 dependencies = [
 "ark-std",
 "env_logger 0.10.0",
@@ -380,7 +380,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "4114279215a005bc675e386011e594e1d9b800918cea18fcadadcce864a2046b"
 dependencies = [
 "borsh-derive",
- "hashbrown 0.12.3",
+ "hashbrown 0.13.2",
 ]

 [[package]]
@@ -433,7 +433,7 @@ checksum = "a3e2c3daef883ecc1b5d58c15adae93470a91d425f3532ba1695849656af3fc1"
 [[package]]
 name = "bus-mapping"
 version = "0.1.0"
-source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.7.5#aa9a9aff698a5b253d1f3c29ea3d3006364777bf"
+source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.8.1#fd703cc269a07f28b1febc7ce021792e2733564b"
 dependencies = [
 "eth-types",
 "ethers-core",
@@ -447,6 +447,7 @@ dependencies = [
 "lazy_static",
 "log",
 "mock",
+ "mpt-zktrie",
 "once_cell",
 "poseidon-circuit",
 "rand",
@@ -1049,7 +1050,7 @@ dependencies = [
 [[package]]
 name = "eth-types"
 version = "0.1.0"
-source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.7.5#aa9a9aff698a5b253d1f3c29ea3d3006364777bf"
+source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.8.1#fd703cc269a07f28b1febc7ce021792e2733564b"
 dependencies = [
 "ethers-core",
 "ethers-signers",
@@ -1226,7 +1227,7 @@ dependencies = [
 [[package]]
 name = "external-tracer"
 version = "0.1.0"
-source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.7.5#aa9a9aff698a5b253d1f3c29ea3d3006364777bf"
+source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.8.1#fd703cc269a07f28b1febc7ce021792e2733564b"
 dependencies = [
 "eth-types",
 "geth-utils",
@@ -1439,7 +1440,7 @@ dependencies = [
 [[package]]
 name = "gadgets"
 version = "0.1.0"
-source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.7.5#aa9a9aff698a5b253d1f3c29ea3d3006364777bf"
+source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.8.1#fd703cc269a07f28b1febc7ce021792e2733564b"
 dependencies = [
 "digest 0.7.6",
 "eth-types",
@@ -1479,7 +1480,7 @@ dependencies = [
 [[package]]
 name = "geth-utils"
 version = "0.1.0"
-source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.7.5#aa9a9aff698a5b253d1f3c29ea3d3006364777bf"
+source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.8.1#fd703cc269a07f28b1febc7ce021792e2733564b"
 dependencies = [
 "env_logger 0.9.3",
 "gobuild 0.1.0-alpha.2 (git+https://github.com/scroll-tech/gobuild.git)",
@@ -1583,7 +1584,7 @@ dependencies = [
 [[package]]
 name = "halo2-base"
 version = "0.2.2"
-source = "git+https://github.com/scroll-tech/halo2-lib?tag=v0.1.0#2c225864227e74b207d9f4b9e08c4d5f1afc69a1"
+source = "git+https://github.com/scroll-tech/halo2-lib?tag=v0.1.3#33b3b4d240ba8b7e0fbdca241c677ba84ab16db5"
 dependencies = [
 "ff",
 "halo2_proofs",
@@ -1598,7 +1599,7 @@ dependencies = [
 [[package]]
 name = "halo2-ecc"
 version = "0.2.2"
-source = "git+https://github.com/scroll-tech/halo2-lib?tag=v0.1.0#2c225864227e74b207d9f4b9e08c4d5f1afc69a1"
+source = "git+https://github.com/scroll-tech/halo2-lib?tag=v0.1.3#33b3b4d240ba8b7e0fbdca241c677ba84ab16db5"
 dependencies = [
 "ff",
 "group",
@@ -1633,7 +1634,7 @@ dependencies = [
 [[package]]
 name = "halo2-mpt-circuits"
 version = "0.1.0"
-source = "git+https://github.com/scroll-tech/mpt-circuit.git?tag=v0.5.1#2163a9c436ed85363c954ecf7e6e1044a1b991dc"
+source = "git+https://github.com/scroll-tech/mpt-circuit.git?tag=v0.6.2#cafcdeb2c7fd6602d0ddac183c1fb5396a135f9e"
 dependencies = [
 "ethers-core",
 "halo2_proofs",
@@ -1655,7 +1656,7 @@ dependencies = [
 [[package]]
 name = "halo2_proofs"
 version = "0.2.0"
-source = "git+https://github.com/scroll-tech/halo2.git?branch=develop#19de67c07a9b9b567580466763f93ebfbc3bb799"
+source = "git+https://github.com/scroll-tech/halo2.git?branch=develop#aa86c107aeb62282d81ebce5c4930ec0c0aa540b"
 dependencies = [
 "ark-std",
 "blake2b_simd",
@@ -1769,9 +1770,9 @@ checksum = "7f24254aa9a54b5c858eaee2f5bccdb46aaf0e486a595ed5fd8f86ba55232a70"

 [[package]]
 name = "hex-literal"
-version = "0.4.1"
+version = "0.3.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6fe2267d4ed49bc07b63801559be28c718ea06c4738b7a03c94df7386d2cde46"
+checksum = "7ebdb29d2ea9ed0083cd8cece49bbd968021bd99b0849edb4a9a7ee0fdf6a4e0"

 [[package]]
 name = "hmac"
@@ -2077,7 +2078,7 @@ dependencies = [
 [[package]]
 name = "keccak256"
 version = "0.1.0"
-source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.7.5#aa9a9aff698a5b253d1f3c29ea3d3006364777bf"
+source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.8.1#fd703cc269a07f28b1febc7ce021792e2733564b"
 dependencies = [
 "env_logger 0.9.3",
 "eth-types",
@@ -2261,10 +2262,27 @@ dependencies = [
 "windows-sys 0.48.0",
 ]

+[[package]]
+name = "misc-precompiled-circuit"
+version = "0.1.0"
+source = "git+https://github.com/scroll-tech/misc-precompiled-circuit.git?tag=v0.1.0#f647341f9951f5c2399035728d4f6765564e2e02"
+dependencies = [
+ "halo2-gate-generator",
+ "halo2_proofs",
+ "lazy_static",
+ "num-bigint",
+ "rand",
+ "serde",
+ "serde_json",
+ "strum",
+ "strum_macros",
+ "subtle",
+]
+
 [[package]]
 name = "mock"
 version = "0.1.0"
-source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.7.5#aa9a9aff698a5b253d1f3c29ea3d3006364777bf"
+source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.8.1#fd703cc269a07f28b1febc7ce021792e2733564b"
 dependencies = [
 "eth-types",
 "ethers-core",
@@ -2272,6 +2290,7 @@ dependencies = [
 "external-tracer",
 "itertools",
 "lazy_static",
+ "log",
 "rand",
 "rand_chacha",
 ]
@@ -2279,9 +2298,8 @@ dependencies = [
 [[package]]
 name = "mpt-zktrie"
 version = "0.1.0"
-source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.7.5#aa9a9aff698a5b253d1f3c29ea3d3006364777bf"
+source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.8.1#fd703cc269a07f28b1febc7ce021792e2733564b"
 dependencies = [
- "bus-mapping",
 "eth-types",
 "halo2-mpt-circuits",
 "halo2_proofs",
@@ -2637,7 +2655,7 @@ dependencies = [
 [[package]]
 name = "poseidon-circuit"
 version = "0.1.0"
-source = "git+https://github.com/scroll-tech/poseidon-circuit.git?branch=scroll-dev-0723#1652d54bf7ca9d8f286b53fe077d9efefdcf6d5f"
+source = "git+https://github.com/scroll-tech/poseidon-circuit.git?branch=scroll-dev-0901#69524f42bdc55c581088c2fe64c2ab9a2921146b"
 dependencies = [
 "bitvec 1.0.1",
 "halo2_proofs",
@@ -2754,11 +2772,12 @@ dependencies = [

 [[package]]
 name = "prover"
-version = "0.7.5"
-source = "git+https://github.com/scroll-tech/scroll-prover?tag=v0.7.5#9699d40940aed2f14d8e1958167d714bca2c9984"
+version = "0.8.1"
+source = "git+https://github.com/scroll-tech/scroll-prover?tag=v0.8.1#5b94df914877aa7f20f7c37fabf80d73f1c8cf2c"
 dependencies = [
 "aggregator",
 "anyhow",
+ "base64 0.13.1",
 "blake2",
 "bus-mapping",
 "chrono",
@@ -2789,7 +2808,6 @@ dependencies = [
 "snark-verifier-sdk",
 "strum",
 "strum_macros",
- "types",
 "zkevm-circuits",
 ]

@@ -3019,8 +3037,7 @@ dependencies = [
 [[package]]
 name = "revm-precompile"
 version = "2.0.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "66837781605c6dcb7f07ad87604eeab3119dae9149d69d8839073dd6f188673d"
+source = "git+https://github.com/scroll-tech/revm?branch=scroll-fix#aebf2e591e622e6bcce2c5d4bf3336935a68cf11"
 dependencies = [
 "k256",
 "num",
@@ -3035,12 +3052,10 @@ dependencies = [

 [[package]]
 name = "revm-primitives"
-version = "1.1.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "304d998f466ffef72d76c7f20b05bf08a96801736a6fb1fdef47d49a292618df"
+version = "1.0.0"
+source = "git+https://github.com/scroll-tech/revm?branch=scroll-fix#aebf2e591e622e6bcce2c5d4bf3336935a68cf11"
 dependencies = [
 "auto_impl",
- "bitvec 1.0.1",
 "bytes",
 "derive_more",
 "enumn",
@@ -3048,7 +3063,6 @@ dependencies = [
 "hashbrown 0.13.2",
 "hex",
 "hex-literal",
- "primitive-types 0.12.1",
 "rlp",
 "ruint",
 "sha3 0.10.8",
@@ -3156,23 +3170,6 @@ dependencies = [
 "syn 1.0.109",
 ]

-[[package]]
-name = "rmd160-circuits"
-version = "0.1.0"
-source = "git+https://github.com/scroll-tech/misc-precompiled-circuit.git?branch=integration#31c41ca4365dcf2b6ed4f2cdcd3dc8d2e8f080df"
-dependencies = [
- "halo2-gate-generator",
- "halo2_proofs",
- "lazy_static",
- "num-bigint",
- "rand",
- "serde",
- "serde_json",
- "strum",
- "strum_macros",
- "subtle",
-]
-
 [[package]]
 name = "ruint"
 version = "1.9.0"
@@ -3454,17 +3451,6 @@ dependencies = [
 "serde",
 ]

-[[package]]
-name = "serde_repr"
-version = "0.1.16"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8725e1dfadb3a50f7e5ce0b1a540466f6ed3fe7a0fca2ac2b8b831d31316bd00"
-dependencies = [
- "proc-macro2",
- "quote",
- "syn 2.0.27",
-]
-
 [[package]]
 name = "serde_stacker"
 version = "0.1.10"
@@ -3624,7 +3610,7 @@ checksum = "62bb4feee49fdd9f707ef802e22365a35de4b7b299de4763d44bfea899442ff9"
 [[package]]
 name = "snark-verifier"
 version = "0.1.0"
-source = "git+https://github.com/scroll-tech/snark-verifier?tag=v0.1.2#4466059ce9a6dfaf26455e4ffb61d72af775cf52"
+source = "git+https://github.com/scroll-tech/snark-verifier?tag=v0.1.4#c60146b786ab14ba60636eef0a7e427c3eb0baee"
 dependencies = [
 "bytes",
 "ethereum-types 0.14.1",
@@ -3648,7 +3634,7 @@ dependencies = [
 [[package]]
 name = "snark-verifier-sdk"
 version = "0.0.1"
-source = "git+https://github.com/scroll-tech/snark-verifier?tag=v0.1.2#4466059ce9a6dfaf26455e4ffb61d72af775cf52"
+source = "git+https://github.com/scroll-tech/snark-verifier?tag=v0.1.4#c60146b786ab14ba60636eef0a7e427c3eb0baee"
 dependencies = [
 "bincode",
 "env_logger 0.10.0",
@@ -4037,21 +4023,6 @@ version = "1.16.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "497961ef93d974e23eb6f433eb5fe1b7930b659f06d12dec6fc44a8f554c0bba"

-[[package]]
-name = "types"
-version = "0.7.5"
-source = "git+https://github.com/scroll-tech/scroll-prover?tag=v0.7.5#9699d40940aed2f14d8e1958167d714bca2c9984"
-dependencies = [
- "base64 0.13.1",
- "blake2",
- "eth-types",
- "ethers-core",
- "serde",
- "serde_derive",
- "serde_json",
- "serde_repr",
-]
-
 [[package]]
 name = "uint"
 version = "0.9.5"
@@ -4491,7 +4462,7 @@ checksum = "2a0956f1ba7c7909bfb66c2e9e4124ab6f6482560f6628b5aaeba39207c9aad9"
 [[package]]
 name = "zkevm-circuits"
 version = "0.1.0"
-source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.7.5#aa9a9aff698a5b253d1f3c29ea3d3006364777bf"
+source = "git+https://github.com/scroll-tech/zkevm-circuits.git?tag=v0.8.1#fd703cc269a07f28b1febc7ce021792e2733564b"
 dependencies = [
 "array-init",
 "bus-mapping",
@@ -4511,6 +4482,7 @@ dependencies = [
 "libsecp256k1",
 "log",
 "maingate",
+ "misc-precompiled-circuit",
 "mock",
 "mpt-zktrie",
 "num",
@@ -4521,7 +4493,6 @@ dependencies = [
 "rand_chacha",
 "rand_xorshift",
 "rayon",
- "rmd160-circuits",
 "serde",
 "serde_json",
 "sha3 0.10.8",
@@ -4546,7 +4517,6 @@ dependencies = [
 "serde",
 "serde_derive",
 "serde_json",
- "types",
 ]

 [[package]]
--- a/common/libzkp/impl/Cargo.toml
+++ b/common/libzkp/impl/Cargo.toml
@@ -20,8 +20,7 @@ maingate = { git = "https://github.com/scroll-tech/halo2wrong", branch = "halo2-
 halo2curves = { git = "https://github.com/scroll-tech/halo2curves.git", branch = "0.3.1-derive-serde" }

 [dependencies]
-prover = { git = "https://github.com/scroll-tech/scroll-prover", tag = "v0.7.5" }
-types = { git = "https://github.com/scroll-tech/scroll-prover", tag = "v0.7.5" }
+prover = { git = "https://github.com/scroll-tech/scroll-prover", tag = "v0.8.1" }
 halo2_proofs = { git = "https://github.com/scroll-tech/halo2.git", branch = "develop" }

 base64 = "0.13.0"
--- a/common/libzkp/impl/Makefile
+++ b/common/libzkp/impl/Makefile
@@ -0,0 +1,8 @@
+.PHONY: help fmt clippy test test-ci test-all
+
+fmt:
+	@cargo fmt --all -- --check
+
+clippy:
+	@cargo check --all-features
+	@cargo clippy --release -- -D warnings
--- a/common/libzkp/impl/src/batch.rs
+++ b/common/libzkp/impl/src/batch.rs
@@ -1,15 +1,18 @@
 use crate::{
    types::{CheckChunkProofsResponse, ProofResult},
-    utils::{c_char_to_str, c_char_to_vec, string_to_c_char, vec_to_c_char, OUTPUT_DIR},
+    utils::{
+        c_char_to_str, c_char_to_vec, file_exists, string_to_c_char, vec_to_c_char, OUTPUT_DIR,
+    },
 };
 use libc::c_char;
 use prover::{
    aggregator::{Prover, Verifier},
+    consts::AGG_VK_FILENAME,
+    types::eth::BlockTrace,
    utils::{chunk_trace_to_witness_block, init_env_and_log},
    BatchProof, ChunkHash, ChunkProof,
 };
 use std::{cell::OnceCell, env, panic, ptr::null};
-use types::eth::BlockTrace;

 static mut PROVER: OnceCell<Prover> = OnceCell::new();
 static mut VERIFIER: OnceCell<Verifier> = OnceCell::new();
@@ -24,6 +27,12 @@ pub unsafe extern "C" fn init_batch_prover(params_dir: *const c_char, assets_dir

    // TODO: add a settings in scroll-prover.
    env::set_var("SCROLL_PROVER_ASSETS_DIR", assets_dir);
+
+    // VK file must exist, it is optional and logged as a warning in prover.
+    if !file_exists(assets_dir, &AGG_VK_FILENAME) {
+        panic!("{} must exist in folder {}", *AGG_VK_FILENAME, assets_dir);
+    }
+
    let prover = Prover::from_dirs(params_dir, assets_dir);

    PROVER.set(prover).unwrap();
--- a/common/libzkp/impl/src/chunk.rs
+++ b/common/libzkp/impl/src/chunk.rs
@@ -1,15 +1,18 @@
 use crate::{
    types::ProofResult,
-    utils::{c_char_to_str, c_char_to_vec, string_to_c_char, vec_to_c_char, OUTPUT_DIR},
+    utils::{
+        c_char_to_str, c_char_to_vec, file_exists, string_to_c_char, vec_to_c_char, OUTPUT_DIR,
+    },
 };
 use libc::c_char;
 use prover::{
+    consts::CHUNK_VK_FILENAME,
+    types::eth::BlockTrace,
    utils::init_env_and_log,
    zkevm::{Prover, Verifier},
    ChunkProof,
 };
 use std::{cell::OnceCell, env, panic, ptr::null};
-use types::eth::BlockTrace;

 static mut PROVER: OnceCell<Prover> = OnceCell::new();
 static mut VERIFIER: OnceCell<Verifier> = OnceCell::new();
@@ -24,7 +27,13 @@ pub unsafe extern "C" fn init_chunk_prover(params_dir: *const c_char, assets_dir

    // TODO: add a settings in scroll-prover.
    env::set_var("SCROLL_PROVER_ASSETS_DIR", assets_dir);
-    let prover = Prover::from_params_dir(params_dir);
+
+    // VK file must exist, it is optional and logged as a warning in prover.
+    if !file_exists(assets_dir, &CHUNK_VK_FILENAME) {
+        panic!("{} must exist in folder {}", *CHUNK_VK_FILENAME, assets_dir);
+    }
+
+    let prover = Prover::from_dirs(params_dir, assets_dir);

    PROVER.set(prover).unwrap();
 }
--- a/common/libzkp/impl/src/utils.rs
+++ b/common/libzkp/impl/src/utils.rs
@@ -3,6 +3,7 @@ use std::{
    env,
    ffi::{CStr, CString},
    os::raw::c_char,
+    path::PathBuf,
 };

 // Only used for debugging.
@@ -26,3 +27,10 @@ pub(crate) fn string_to_c_char(string: String) -> *const c_char {
 pub(crate) fn vec_to_c_char(bytes: Vec<u8>) -> *const c_char {
    CString::new(bytes).unwrap().into_raw()
 }
+
+pub(crate) fn file_exists(dir: &str, filename: &str) -> bool {
+    let mut path = PathBuf::from(dir);
+    path.push(filename);
+
+    path.exists()
+}
--- a/common/types/db.go
+++ b/common/types/db.go
@@ -103,6 +103,12 @@ const (
 	ProverTaskFailureTypeUndefined ProverTaskFailureType = iota
 	// ProverTaskFailureTypeTimeout prover task failure of timeout
 	ProverTaskFailureTypeTimeout
+	// ProverTaskFailureTypeSubmitStatusNotOk prover task failure of validated failed by coordinator
+	ProverTaskFailureTypeSubmitStatusNotOk
+	// ProverTaskFailureTypeVerifiedFailed prover task failure of verified failed by coordinator
+	ProverTaskFailureTypeVerifiedFailed
+	// ProverTaskFailureTypeServerError collect occur error
+	ProverTaskFailureTypeServerError
 )

 func (r ProverTaskFailureType) String() string {
@@ -111,8 +117,14 @@ func (r ProverTaskFailureType) String() string {
 		return "prover task failure undefined"
 	case ProverTaskFailureTypeTimeout:
 		return "prover task failure timeout"
+	case ProverTaskFailureTypeSubmitStatusNotOk:
+		return "prover task failure validated submit proof status not ok"
+	case ProverTaskFailureTypeVerifiedFailed:
+		return "prover task failure verified failed"
+	case ProverTaskFailureTypeServerError:
+		return "prover task failure server exception"
 	default:
-		return "illegal prover task failure type"
+		return fmt.Sprintf("illegal prover task failure type (%d)", int32(r))
 	}
 }

--- a/common/version/prover_version.go
+++ b/common/version/prover_version.go
@@ -1,7 +1,6 @@
 package version

 import (
-	"strconv"
 	"strings"
 )

@@ -20,36 +19,3 @@ func CheckScrollProverVersion(proverVersion string) bool {
 	// compare the `scroll_prover` version
 	return remote[2] == local[2]
 }
-
-// CheckScrollProverVersionTag check the "scroll-prover" version's tag, if it's too old, return false
-func CheckScrollProverVersionTag(proverVersion string) bool {
-	// note the the version is in fact in the format of "tag-commit-scroll_prover-halo2",
-	// so split-by-'-' length should be 4
-	remote := strings.Split(proverVersion, "-")
-	if len(remote) != 4 {
-		return false
-	}
-	remoteTagNums := strings.Split(strings.TrimPrefix(remote[0], "v"), ".")
-	if len(remoteTagNums) != 3 {
-		return false
-	}
-	remoteTagMajor, err := strconv.Atoi(remoteTagNums[0])
-	if err != nil {
-		return false
-	}
-	remoteTagMinor, err := strconv.Atoi(remoteTagNums[1])
-	if err != nil {
-		return false
-	}
-	remoteTagPatch, err := strconv.Atoi(remoteTagNums[2])
-	if err != nil {
-		return false
-	}
-	if remoteTagMajor < 4 {
-		return false
-	}
-	if remoteTagMinor == 1 && remoteTagPatch < 98 {
-		return false
-	}
-	return true
-}
--- a/common/version/version.go
+++ b/common/version/version.go
@@ -5,7 +5,7 @@ import (
 	"runtime/debug"
 )

-var tag = "v4.2.17"
+var tag = "v4.3.1"

 var commit = func() string {
 	if info, ok := debug.ReadBuildInfo(); ok {
--- a/contracts/integration-test/GasSwap.spec.ts
+++ b/contracts/integration-test/GasSwap.spec.ts
@@ -1,7 +1,7 @@
 /* eslint-disable node/no-unpublished-import */
 /* eslint-disable node/no-missing-import */
 import { ethers } from "hardhat";
-import { GasSwap, MinimalForwarder, MockERC20, MockGasSwapTarget } from "../typechain";
+import { GasSwap, ERC2771Forwarder, MockERC20, MockGasSwapTarget } from "../typechain";
 import { SignerWithAddress } from "@nomiclabs/hardhat-ethers/signers";
 import { expect } from "chai";
 import { BigNumber, constants } from "ethers";
@@ -11,7 +11,7 @@ describe("GasSwap.spec", async () => {
  let deployer: SignerWithAddress;
  let signer: SignerWithAddress;

-  let forwarder: MinimalForwarder;
+  let forwarder: ERC2771Forwarder;
  let swap: GasSwap;
  let target: MockGasSwapTarget;
  let token: MockERC20;
@@ -19,8 +19,8 @@ describe("GasSwap.spec", async () => {
  beforeEach(async () => {
    [deployer, signer] = await ethers.getSigners();

-    const MinimalForwarder = await ethers.getContractFactory("MinimalForwarder", deployer);
-    forwarder = await MinimalForwarder.deploy();
+    const ERC2771Forwarder = await ethers.getContractFactory("ERC2771Forwarder", deployer);
+    forwarder = await ERC2771Forwarder.deploy("ERC2771Forwarder");
    await forwarder.deployed();

    const GasSwap = await ethers.getContractFactory("GasSwap", deployer);
@@ -253,12 +253,13 @@ describe("GasSwap.spec", async () => {
          await swap.updateFeeRatio(ethers.utils.parseEther(feeRatio).div(100));
          const fee = amountOut.mul(feeRatio).div(100);

-          const req = {
+          const reqWithoutSignature = {
            from: signer.address,
            to: swap.address,
            value: constants.Zero,
            gas: 1000000,
-            nonce: 0,
+            nonce: await forwarder.nonces(signer.address),
+            deadline: 2000000000,
            data: swap.interface.encodeFunctionData("swap", [
              {
                token: token.address,
@@ -278,8 +279,8 @@ describe("GasSwap.spec", async () => {

          const signature = await signer._signTypedData(
            {
-              name: "MinimalForwarder",
-              version: "0.0.1",
+              name: "ERC2771Forwarder",
+              version: "1",
              chainId: (await ethers.provider.getNetwork()).chainId,
              verifyingContract: forwarder.address,
            },
@@ -305,17 +306,29 @@ describe("GasSwap.spec", async () => {
                  name: "nonce",
                  type: "uint256",
                },
+                {
+                  name: "deadline",
+                  type: "uint48",
+                },
                {
                  name: "data",
                  type: "bytes",
                },
              ],
            },
-            req
+            reqWithoutSignature
          );

          const balanceBefore = await signer.getBalance();
-          await forwarder.execute(req, signature);
+          await forwarder.execute({
+            from: reqWithoutSignature.from,
+            to: reqWithoutSignature.to,
+            value: reqWithoutSignature.value,
+            gas: reqWithoutSignature.gas,
+            deadline: reqWithoutSignature.deadline,
+            data: reqWithoutSignature.data,
+            signature,
+          });
          const balanceAfter = await signer.getBalance();
          expect(balanceAfter.sub(balanceBefore)).to.eq(amountOut.sub(fee));
          expect(await token.balanceOf(signer.address)).to.eq(amountIn.mul(refundRatio).div(100));
--- a/contracts/scripts/foundry/InitializeL1BridgeContracts.s.sol
+++ b/contracts/scripts/foundry/InitializeL1BridgeContracts.s.sol
@@ -22,7 +22,7 @@ contract InitializeL1BridgeContracts is Script {
    uint256 L1_DEPLOYER_PRIVATE_KEY = vm.envUint("L1_DEPLOYER_PRIVATE_KEY");

    uint256 CHAIN_ID_L2 = vm.envUint("CHAIN_ID_L2");
-    uint256 MAX_L2_TX_IN_CHUNK = vm.envUint("MAX_L2_TX_IN_CHUNK");
+    uint256 MAX_TX_IN_CHUNK = vm.envUint("MAX_TX_IN_CHUNK");
    uint256 MAX_L1_MESSAGE_GAS_LIMIT = vm.envUint("MAX_L1_MESSAGE_GAS_LIMIT");
    address L1_COMMIT_SENDER_ADDRESS = vm.envAddress("L1_COMMIT_SENDER_ADDRESS");
    address L1_FINALIZE_SENDER_ADDRESS = vm.envAddress("L1_FINALIZE_SENDER_ADDRESS");
@@ -67,7 +67,7 @@ contract InitializeL1BridgeContracts is Script {
        ScrollChain(L1_SCROLL_CHAIN_PROXY_ADDR).initialize(
            L1_MESSAGE_QUEUE_PROXY_ADDR,
            L1_MULTIPLE_VERSION_ROLLUP_VERIFIER_ADDR,
-            MAX_L2_TX_IN_CHUNK
+            MAX_TX_IN_CHUNK
        );
        ScrollChain(L1_SCROLL_CHAIN_PROXY_ADDR).addSequencer(L1_COMMIT_SENDER_ADDRESS);
        ScrollChain(L1_SCROLL_CHAIN_PROXY_ADDR).addProver(L1_FINALIZE_SENDER_ADDRESS);
--- a/contracts/src/L1/gateways/L1ERC20Gateway.sol
+++ b/contracts/src/L1/gateways/L1ERC20Gateway.sol
@@ -9,8 +9,6 @@ import {IL1ERC20Gateway} from "./IL1ERC20Gateway.sol";
 import {IL1GatewayRouter} from "./IL1GatewayRouter.sol";

 import {IL2ERC20Gateway} from "../../L2/gateways/IL2ERC20Gateway.sol";
-import {IScrollMessenger} from "../../libraries/IScrollMessenger.sol";
-import {ScrollConstants} from "../../libraries/constants/ScrollConstants.sol";
 import {ScrollGatewayBase} from "../../libraries/gateway/ScrollGatewayBase.sol";
 import {IMessageDropCallback} from "../../libraries/callbacks/IMessageDropCallback.sol";

--- a/contracts/src/L1/gateways/usdc/L1USDCGateway.sol
+++ b/contracts/src/L1/gateways/usdc/L1USDCGateway.sol
@@ -2,9 +2,6 @@

 pragma solidity =0.8.16;

-import {OwnableUpgradeable} from "@openzeppelin/contracts-upgradeable/access/OwnableUpgradeable.sol";
-import {IERC20Upgradeable} from "@openzeppelin/contracts-upgradeable/token/ERC20/IERC20Upgradeable.sol";
-
 import {IFiatToken} from "../../../interfaces/IFiatToken.sol";
 import {IUSDCBurnableSourceBridge} from "../../../interfaces/IUSDCBurnableSourceBridge.sol";
 import {IL2ERC20Gateway} from "../../../L2/gateways/IL2ERC20Gateway.sol";
--- a/contracts/src/L2/gateways/usdc/L2USDCGateway.sol
+++ b/contracts/src/L2/gateways/usdc/L2USDCGateway.sol
@@ -12,7 +12,7 @@ import {IL1ERC20Gateway} from "../../../L1/gateways/IL1ERC20Gateway.sol";
 import {IL2ScrollMessenger} from "../../IL2ScrollMessenger.sol";
 import {IL2ERC20Gateway} from "../IL2ERC20Gateway.sol";

-import {ScrollGatewayBase, IScrollGateway} from "../../../libraries/gateway/ScrollGatewayBase.sol";
+import {ScrollGatewayBase} from "../../../libraries/gateway/ScrollGatewayBase.sol";
 import {L2ERC20Gateway} from "../L2ERC20Gateway.sol";

 /// @title L2USDCGateway
@@ -169,7 +169,7 @@ contract L2USDCGateway is L2ERC20Gateway, IUSDCDestinationBridge {
            (_l1USDC, _token, _from, _to, _amount, _data)
        );

-        // 4. Send message to L1ScrollMessenger.
+        // 4. Send message to L2ScrollMessenger.
        IL2ScrollMessenger(messenger).sendMessage{value: msg.value}(counterpart, 0, _message, _gasLimit);

        emit WithdrawERC20(_l1USDC, _token, _from, _to, _amount, _data);
--- a/contracts/src/gas-swap/GasSwap.sol
+++ b/contracts/src/gas-swap/GasSwap.sol
@@ -104,6 +104,9 @@ contract GasSwap is ERC2771Context, ReentrancyGuard, OwnableBase {
            _permit.s
        );

+        // record token balance in this contract
+        uint256 _balance = IERC20(_permit.token).balanceOf(address(this));
+
        // transfer token
        IERC20(_permit.token).safeTransferFrom(_sender, address(this), _permit.value);

@@ -128,7 +131,7 @@ contract GasSwap is ERC2771Context, ReentrancyGuard, OwnableBase {
        require(_success, "transfer ETH failed");

        // refund rest token
-        uint256 _dust = IERC20(_permit.token).balanceOf(address(this));
+        uint256 _dust = IERC20(_permit.token).balanceOf(address(this)) - _balance;
        if (_dust > 0) {
            IERC20(_permit.token).safeTransfer(_sender, _dust);
        }
--- a/contracts/src/libraries/FeeVault.sol
+++ b/contracts/src/libraries/FeeVault.sol
@@ -103,29 +103,34 @@ abstract contract FeeVault is OwnableBase {
    receive() external payable {}

    /// @notice Triggers a withdrawal of funds to the L1 fee wallet.
-    function withdraw() external {
-        uint256 value = address(this).balance;
-
+    /// @param _value The amount of ETH to withdraw.
+    function withdraw(uint256 _value) public {
        require(
-            value >= minWithdrawAmount,
+            _value >= minWithdrawAmount,
            "FeeVault: withdrawal amount must be greater than minimum withdrawal amount"
        );

        unchecked {
-            totalProcessed += value;
+            totalProcessed += _value;
        }

-        emit Withdrawal(value, recipient, msg.sender);
+        emit Withdrawal(_value, recipient, msg.sender);

        // no fee provided
-        IL2ScrollMessenger(messenger).sendMessage{value: value}(
+        IL2ScrollMessenger(messenger).sendMessage{value: _value}(
            recipient,
-            value,
+            _value,
            bytes(""), // no message (simple eth transfer)
            0 // _gasLimit can be zero for fee vault.
        );
    }

+    /// @notice Triggers a withdrawal of all available funds to the L1 fee wallet.
+    function withdraw() external {
+        uint256 value = address(this).balance;
+        withdraw(value);
+    }
+
    /************************
     * Restricted Functions *
     ************************/
--- a/contracts/src/misc/ERC2771Forwarder.sol
+++ b/contracts/src/misc/ERC2771Forwarder.sol
@@ -0,0 +1,391 @@
+// SPDX-License-Identifier: MIT
+
+// @note This file is directly copied from OpenZeppelin's master branch:
+// https://github.com/OpenZeppelin/openzeppelin-contracts/blob/master/contracts/metatx/ERC2771Forwarder.sol
+// Modifications are made to make it compatible with solidity 0.8.16.
+
+pragma solidity =0.8.16;
+
+import {ERC2771Context} from "@openzeppelin/contracts/metatx/ERC2771Context.sol";
+import {ECDSA} from "@openzeppelin/contracts/utils/cryptography/ECDSA.sol";
+import {EIP712} from "@openzeppelin/contracts/utils/cryptography/EIP712.sol";
+import {Nonces} from "./Nonces.sol";
+import {Address} from "@openzeppelin/contracts/utils/Address.sol";
+
+/**
+ * @dev A forwarder compatible with ERC2771 contracts. See {ERC2771Context}.
+ *
+ * This forwarder operates on forward requests that include:
+ *
+ * * `from`: An address to operate on behalf of. It is required to be equal to the request signer.
+ * * `to`: The address that should be called.
+ * * `value`: The amount of native token to attach with the requested call.
+ * * `gas`: The amount of gas limit that will be forwarded with the requested call.
+ * * `nonce`: A unique transaction ordering identifier to avoid replayability and request invalidation.
+ * * `deadline`: A timestamp after which the request is not executable anymore.
+ * * `data`: Encoded `msg.data` to send with the requested call.
+ *
+ * Relayers are able to submit batches if they are processing a high volume of requests. With high
+ * throughput, relayers may run into limitations of the chain such as limits on the number of
+ * transactions in the mempool. In these cases the recommendation is to distribute the load among
+ * multiple accounts.
+ *
+ * NOTE: Batching requests includes an optional refund for unused `msg.value` that is achieved by
+ * performing a call with empty calldata. While this is within the bounds of ERC-2771 compliance,
+ * if the refund receiver happens to consider the forwarder a trusted forwarder, it MUST properly
+ * handle `msg.data.length == 0`. `ERC2771Context` in OpenZeppelin Contracts versions prior to 4.9.3
+ * do not handle this properly.
+ *
+ * ==== Security Considerations
+ *
+ * If a relayer submits a forward request, it should be willing to pay up to 100% of the gas amount
+ * specified in the request. This contract does not implement any kind of retribution for this gas,
+ * and it is assumed that there is an out of band incentive for relayers to pay for execution on
+ * behalf of signers. Often, the relayer is operated by a project that will consider it a user
+ * acquisition cost.
+ *
+ * By offering to pay for gas, relayers are at risk of having that gas used by an attacker toward
+ * some other purpose that is not aligned with the expected out of band incentives. If you operate a
+ * relayer, consider whitelisting target contracts and function selectors. When relaying ERC-721 or
+ * ERC-1155 transfers specifically, consider rejecting the use of the `data` field, since it can be
+ * used to execute arbitrary code.
+ */
+contract ERC2771Forwarder is EIP712, Nonces {
+    using ECDSA for bytes32;
+
+    struct ForwardRequestData {
+        address from;
+        address to;
+        uint256 value;
+        uint256 gas;
+        uint48 deadline;
+        bytes data;
+        bytes signature;
+    }
+
+    bytes32 internal constant _FORWARD_REQUEST_TYPEHASH =
+        keccak256(
+            "ForwardRequest(address from,address to,uint256 value,uint256 gas,uint256 nonce,uint48 deadline,bytes data)"
+        );
+
+    /**
+     * @dev Emitted when a `ForwardRequest` is executed.
+     *
+     * NOTE: An unsuccessful forward request could be due to an invalid signature, an expired deadline,
+     * or simply a revert in the requested call. The contract guarantees that the relayer is not able to force
+     * the requested call to run out of gas.
+     */
+    event ExecutedForwardRequest(address indexed signer, uint256 nonce, bool success);
+
+    /**
+     * @dev The request `from` doesn't match with the recovered `signer`.
+     */
+    error ERC2771ForwarderInvalidSigner(address signer, address from);
+
+    /**
+     * @dev The `requestedValue` doesn't match with the available `msgValue`.
+     */
+    error ERC2771ForwarderMismatchedValue(uint256 requestedValue, uint256 msgValue);
+
+    /**
+     * @dev The request `deadline` has expired.
+     */
+    error ERC2771ForwarderExpiredRequest(uint48 deadline);
+
+    /**
+     * @dev The request target doesn't trust the `forwarder`.
+     */
+    error ERC2771UntrustfulTarget(address target, address forwarder);
+
+    /**
+     * @dev A call to an address target failed. The target may have reverted.
+     */
+    error FailedInnerCall();
+
+    /**
+     * @dev See {EIP712-constructor}.
+     */
+    constructor(string memory name) EIP712(name, "1") {}
+
+    /**
+     * @dev Returns `true` if a request is valid for a provided `signature` at the current block timestamp.
+     *
+     * A transaction is considered valid when the target trusts this forwarder, the request hasn't expired
+     * (deadline is not met), and the signer matches the `from` parameter of the signed request.
+     *
+     * NOTE: A request may return false here but it won't cause {executeBatch} to revert if a refund
+     * receiver is provided.
+     */
+    function verify(ForwardRequestData calldata request) public view virtual returns (bool) {
+        (bool isTrustedForwarder, bool active, bool signerMatch, ) = _validate(request);
+        return isTrustedForwarder && active && signerMatch;
+    }
+
+    /**
+     * @dev Executes a `request` on behalf of `signature`'s signer using the ERC-2771 protocol. The gas
+     * provided to the requested call may not be exactly the amount requested, but the call will not run
+     * out of gas. Will revert if the request is invalid or the call reverts, in this case the nonce is not consumed.
+     *
+     * Requirements:
+     *
+     * - The request value should be equal to the provided `msg.value`.
+     * - The request should be valid according to {verify}.
+     */
+    function execute(ForwardRequestData calldata request) public payable virtual {
+        // We make sure that msg.value and request.value match exactly.
+        // If the request is invalid or the call reverts, this whole function
+        // will revert, ensuring value isn't stuck.
+        if (msg.value != request.value) {
+            revert ERC2771ForwarderMismatchedValue(request.value, msg.value);
+        }
+
+        if (!_execute(request, true)) {
+            revert FailedInnerCall();
+        }
+    }
+
+    /**
+     * @dev Batch version of {execute} with optional refunding and atomic execution.
+     *
+     * In case a batch contains at least one invalid request (see {verify}), the
+     * request will be skipped and the `refundReceiver` parameter will receive back the
+     * unused requested value at the end of the execution. This is done to prevent reverting
+     * the entire batch when a request is invalid or has already been submitted.
+     *
+     * If the `refundReceiver` is the `address(0)`, this function will revert when at least
+     * one of the requests was not valid instead of skipping it. This could be useful if
+     * a batch is required to get executed atomically (at least at the top-level). For example,
+     * refunding (and thus atomicity) can be opt-out if the relayer is using a service that avoids
+     * including reverted transactions.
+     *
+     * Requirements:
+     *
+     * - The sum of the requests' values should be equal to the provided `msg.value`.
+     * - All of the requests should be valid (see {verify}) when `refundReceiver` is the zero address.
+     *
+     * NOTE: Setting a zero `refundReceiver` guarantees an all-or-nothing requests execution only for
+     * the first-level forwarded calls. In case a forwarded request calls to a contract with another
+     * subcall, the second-level call may revert without the top-level call reverting.
+     */
+    function executeBatch(ForwardRequestData[] calldata requests, address payable refundReceiver)
+        public
+        payable
+        virtual
+    {
+        bool atomic = refundReceiver == address(0);
+
+        uint256 requestsValue;
+        uint256 refundValue;
+
+        for (uint256 i; i < requests.length; ++i) {
+            requestsValue += requests[i].value;
+            bool success = _execute(requests[i], atomic);
+            if (!success) {
+                refundValue += requests[i].value;
+            }
+        }
+
+        // The batch should revert if there's a mismatched msg.value provided
+        // to avoid request value tampering
+        if (requestsValue != msg.value) {
+            revert ERC2771ForwarderMismatchedValue(requestsValue, msg.value);
+        }
+
+        // Some requests with value were invalid (possibly due to frontrunning).
+        // To avoid leaving ETH in the contract this value is refunded.
+        if (refundValue != 0) {
+            // We know refundReceiver != address(0) && requestsValue == msg.value
+            // meaning we can ensure refundValue is not taken from the original contract's balance
+            // and refundReceiver is a known account.
+            Address.sendValue(refundReceiver, refundValue);
+        }
+    }
+
+    /**
+     * @dev Validates if the provided request can be executed at current block timestamp with
+     * the given `request.signature` on behalf of `request.signer`.
+     */
+    function _validate(ForwardRequestData calldata request)
+        internal
+        view
+        virtual
+        returns (
+            bool isTrustedForwarder,
+            bool active,
+            bool signerMatch,
+            address signer
+        )
+    {
+        (bool isValid, address recovered) = _recoverForwardRequestSigner(request);
+
+        return (
+            _isTrustedByTarget(request.to),
+            request.deadline >= block.timestamp,
+            isValid && recovered == request.from,
+            recovered
+        );
+    }
+
+    /**
+     * @dev Returns a tuple with the recovered the signer of an EIP712 forward request message hash
+     * and a boolean indicating if the signature is valid.
+     *
+     * NOTE: The signature is considered valid if {ECDSA-tryRecover} indicates no recover error for it.
+     */
+    function _recoverForwardRequestSigner(ForwardRequestData calldata request)
+        internal
+        view
+        virtual
+        returns (bool, address)
+    {
+        (address recovered, ECDSA.RecoverError err) = _hashTypedDataV4(
+            keccak256(
+                abi.encode(
+                    _FORWARD_REQUEST_TYPEHASH,
+                    request.from,
+                    request.to,
+                    request.value,
+                    request.gas,
+                    nonces(request.from),
+                    request.deadline,
+                    keccak256(request.data)
+                )
+            )
+        ).tryRecover(request.signature);
+
+        return (err == ECDSA.RecoverError.NoError, recovered);
+    }
+
+    /**
+     * @dev Validates and executes a signed request returning the request call `success` value.
+     *
+     * Internal function without msg.value validation.
+     *
+     * Requirements:
+     *
+     * - The caller must have provided enough gas to forward with the call.
+     * - The request must be valid (see {verify}) if the `requireValidRequest` is true.
+     *
+     * Emits an {ExecutedForwardRequest} event.
+     *
+     * IMPORTANT: Using this function doesn't check that all the `msg.value` was sent, potentially
+     * leaving value stuck in the contract.
+     */
+    function _execute(ForwardRequestData calldata request, bool requireValidRequest)
+        internal
+        virtual
+        returns (bool success)
+    {
+        (bool isTrustedForwarder, bool active, bool signerMatch, address signer) = _validate(request);
+
+        // Need to explicitly specify if a revert is required since non-reverting is default for
+        // batches and reversion is opt-in since it could be useful in some scenarios
+        if (requireValidRequest) {
+            if (!isTrustedForwarder) {
+                revert ERC2771UntrustfulTarget(request.to, address(this));
+            }
+
+            if (!active) {
+                revert ERC2771ForwarderExpiredRequest(request.deadline);
+            }
+
+            if (!signerMatch) {
+                revert ERC2771ForwarderInvalidSigner(signer, request.from);
+            }
+        }
+
+        // Ignore an invalid request because requireValidRequest = false
+        if (isTrustedForwarder && signerMatch && active) {
+            // Nonce should be used before the call to prevent reusing by reentrancy
+            uint256 currentNonce = _useNonce(signer);
+
+            uint256 reqGas = request.gas;
+            address to = request.to;
+            uint256 value = request.value;
+            bytes memory data = abi.encodePacked(request.data, request.from);
+
+            uint256 gasLeft;
+
+            assembly {
+                success := call(reqGas, to, value, add(data, 0x20), mload(data), 0, 0)
+                gasLeft := gas()
+            }
+
+            _checkForwardedGas(gasLeft, request);
+
+            emit ExecutedForwardRequest(signer, currentNonce, success);
+        }
+    }
+
+    /**
+     * @dev Returns whether the target trusts this forwarder.
+     *
+     * This function performs a static call to the target contract calling the
+     * {ERC2771Context-isTrustedForwarder} function.
+     */
+    function _isTrustedByTarget(address target) private view returns (bool) {
+        bytes memory encodedParams = abi.encodeCall(ERC2771Context.isTrustedForwarder, (address(this)));
+
+        bool success;
+        uint256 returnSize;
+        uint256 returnValue;
+        /// @solidity memory-safe-assembly
+        assembly {
+            // Perform the staticcal and save the result in the scratch space.
+            // | Location  | Content  | Content (Hex)                                                      |
+            // |-----------|----------|--------------------------------------------------------------------|
+            // |           |          |                                                           result ↓ |
+            // | 0x00:0x1F | selector | 0x0000000000000000000000000000000000000000000000000000000000000001 |
+            success := staticcall(gas(), target, add(encodedParams, 0x20), mload(encodedParams), 0, 0x20)
+            returnSize := returndatasize()
+            returnValue := mload(0)
+        }
+
+        return success && returnSize >= 0x20 && returnValue > 0;
+    }
+
+    /**
+     * @dev Checks if the requested gas was correctly forwarded to the callee.
+     *
+     * As a consequence of https://eips.ethereum.org/EIPS/eip-150[EIP-150]:
+     * - At most `gasleft() - floor(gasleft() / 64)` is forwarded to the callee.
+     * - At least `floor(gasleft() / 64)` is kept in the caller.
+     *
+     * It reverts consuming all the available gas if the forwarded gas is not the requested gas.
+     *
+     * IMPORTANT: The `gasLeft` parameter should be measured exactly at the end of the forwarded call.
+     * Any gas consumed in between will make room for bypassing this check.
+     */
+    function _checkForwardedGas(uint256 gasLeft, ForwardRequestData calldata request) private pure {
+        // To avoid insufficient gas griefing attacks, as referenced in https://ronan.eth.limo/blog/ethereum-gas-dangers/
+        //
+        // A malicious relayer can attempt to shrink the gas forwarded so that the underlying call reverts out-of-gas
+        // but the forwarding itself still succeeds. In order to make sure that the subcall received sufficient gas,
+        // we will inspect gasleft() after the forwarding.
+        //
+        // Let X be the gas available before the subcall, such that the subcall gets at most X * 63 / 64.
+        // We can't know X after CALL dynamic costs, but we want it to be such that X * 63 / 64 >= req.gas.
+        // Let Y be the gas used in the subcall. gasleft() measured immediately after the subcall will be gasleft() = X - Y.
+        // If the subcall ran out of gas, then Y = X * 63 / 64 and gasleft() = X - Y = X / 64.
+        // Under this assumption req.gas / 63 > gasleft() is true is true if and only if
+        // req.gas / 63 > X / 64, or equivalently req.gas > X * 63 / 64.
+        // This means that if the subcall runs out of gas we are able to detect that insufficient gas was passed.
+        //
+        // We will now also see that req.gas / 63 > gasleft() implies that req.gas >= X * 63 / 64.
+        // The contract guarantees Y <= req.gas, thus gasleft() = X - Y >= X - req.gas.
+        // -    req.gas / 63 > gasleft()
+        // -    req.gas / 63 >= X - req.gas
+        // -    req.gas >= X * 63 / 64
+        // In other words if req.gas < X * 63 / 64 then req.gas / 63 <= gasleft(), thus if the relayer behaves honestly
+        // the forwarding does not revert.
+        if (gasLeft < request.gas / 63) {
+            // We explicitly trigger invalid opcode to consume all gas and bubble-up the effects, since
+            // neither revert or assert consume all gas since Solidity 0.8.20
+            // https://docs.soliditylang.org/en/v0.8.20/control-structures.html#panic-via-assert-and-error-via-require
+            /// @solidity memory-safe-assembly
+            assembly {
+                invalid()
+            }
+        }
+    }
+}
--- a/contracts/src/misc/Nonces.sol
+++ b/contracts/src/misc/Nonces.sol
@@ -0,0 +1,51 @@
+// SPDX-License-Identifier: MIT
+
+// @note This file is directly copied from OpenZeppelin's master branch:
+// https://github.com/OpenZeppelin/openzeppelin-contracts/blob/master/contracts/utils/Nonces.sol
+// Modifications are made to make it compatible with solidity 0.8.16.
+
+pragma solidity ^0.8.16;
+
+/**
+ * @dev Provides tracking nonces for addresses. Nonces will only increment.
+ */
+abstract contract Nonces {
+    /**
+     * @dev The nonce used for an `account` is not the expected current nonce.
+     */
+    error InvalidAccountNonce(address account, uint256 currentNonce);
+
+    mapping(address => uint256) private _nonces;
+
+    /**
+     * @dev Returns an the next unused nonce for an address.
+     */
+    function nonces(address owner) public view virtual returns (uint256) {
+        return _nonces[owner];
+    }
+
+    /**
+     * @dev Consumes a nonce.
+     *
+     * Returns the current value and increments nonce.
+     */
+    function _useNonce(address owner) internal virtual returns (uint256) {
+        // For each account, the nonce has an initial value of 0, can only be incremented by one, and cannot be
+        // decremented or reset. This guarantees that the nonce never overflows.
+        unchecked {
+            // It is important to do x++ and not ++x here.
+            return _nonces[owner]++;
+        }
+    }
+
+    /**
+     * @dev Same as {_useNonce} but checking that `nonce` is the next valid for `owner`.
+     */
+    function _useCheckedNonce(address owner, uint256 nonce) internal virtual returns (uint256) {
+        uint256 current = _useNonce(owner);
+        if (nonce != current) {
+            revert InvalidAccountNonce(owner, current);
+        }
+        return current;
+    }
+}
--- a/contracts/src/test/L2TxFeeVault.t.sol
+++ b/contracts/src/test/L2TxFeeVault.t.sol
@@ -23,6 +23,20 @@ contract L2TxFeeVaultTest is DSTestPlus {
        vault.withdraw();
    }

+    function testCantWithdrawAmountBelowMinimum(uint256 amount) public {
+        amount = bound(amount, 0 ether, 10 ether - 1);
+        hevm.deal(address(vault), 100 ether);
+        hevm.expectRevert("FeeVault: withdrawal amount must be greater than minimum withdrawal amount");
+        vault.withdraw(amount);
+    }
+
+    function testCantWithdrawMoreThanBalance(uint256 amount) public {
+        hevm.assume(amount >= 10 ether);
+        hevm.deal(address(vault), amount - 1);
+        hevm.expectRevert(new bytes(0));
+        vault.withdraw(amount);
+    }
+
    function testWithdrawOnce() public {
        hevm.deal(address(vault), 11 ether);
        vault.withdraw();
@@ -30,6 +44,17 @@ contract L2TxFeeVaultTest is DSTestPlus {
        assertEq(vault.totalProcessed(), 11 ether);
    }

+    function testWithdrawAmountOnce(uint256 amount) public {
+        amount = bound(amount, 10 ether, 100 ether);
+
+        hevm.deal(address(vault), 100 ether);
+        vault.withdraw(amount);
+
+        assertEq(address(messenger).balance, amount);
+        assertEq(vault.totalProcessed(), amount);
+        assertEq(address(vault).balance, 100 ether - amount);
+    }
+
    function testWithdrawTwice() public {
        hevm.deal(address(vault), 11 ether);
        vault.withdraw();
@@ -41,4 +66,21 @@ contract L2TxFeeVaultTest is DSTestPlus {
        assertEq(address(messenger).balance, 33 ether);
        assertEq(vault.totalProcessed(), 33 ether);
    }
+
+    function testWithdrawAmountTwice(uint256 amount1, uint256 amount2) public {
+        amount1 = bound(amount1, 10 ether, 100 ether);
+        amount2 = bound(amount2, 10 ether, 100 ether);
+
+        hevm.deal(address(vault), 200 ether);
+
+        vault.withdraw(amount1);
+        assertEq(address(messenger).balance, amount1);
+        assertEq(vault.totalProcessed(), amount1);
+
+        vault.withdraw(amount2);
+        assertEq(address(messenger).balance, amount1 + amount2);
+        assertEq(vault.totalProcessed(), amount1 + amount2);
+
+        assertEq(address(vault).balance, 200 ether - amount1 - amount2);
+    }
 }
--- a/coordinator/internal/controller/cron/collect_proof.go
+++ b/coordinator/internal/controller/cron/collect_proof.go
@@ -159,32 +159,39 @@ func (c *Collector) check(assignedProverTasks []orm.ProverTask, timeout promethe
 		}

 		timeout.Inc()
+
 		log.Warn("proof task have reach the timeout", "task id", assignedProverTask.TaskID,
 			"prover public key", assignedProverTask.ProverPublicKey, "prover name", assignedProverTask.ProverName, "task type", assignedProverTask.TaskType)
+
 		err := c.db.Transaction(func(tx *gorm.DB) error {
-			// update prover task proving status as ProverProofInvalid
-			if err := c.proverTaskOrm.UpdateProverTaskProvingStatus(c.ctx, assignedProverTask.UUID, types.ProverProofInvalid, tx); err != nil {
+			if err := c.proverTaskOrm.UpdateProverTaskProvingStatusAndFailureType(c.ctx, assignedProverTask.UUID, types.ProverProofInvalid, types.ProverTaskFailureTypeTimeout, tx); err != nil {
 				log.Error("update prover task proving status failure", "uuid", assignedProverTask.UUID, "hash", assignedProverTask.TaskID, "pubKey", assignedProverTask.ProverPublicKey, "err", err)
 				return err
 			}

-			// update prover task failure type
-			if err := c.proverTaskOrm.UpdateProverTaskFailureType(c.ctx, assignedProverTask.UUID, types.ProverTaskFailureTypeTimeout, tx); err != nil {
-				log.Error("update prover task failure type failure", "uuid", assignedProverTask.UUID, "hash", assignedProverTask.TaskID, "pubKey", assignedProverTask.ProverPublicKey, "err", err)
-				return err
+			switch message.ProofType(assignedProverTask.TaskType) {
+			case message.ProofTypeChunk:
+				if err := c.chunkOrm.DecreaseActiveAttemptsByHash(c.ctx, assignedProverTask.TaskID, tx); err != nil {
+					log.Error("decrease chunk active attempts failure", "uuid", assignedProverTask.UUID, "hash", assignedProverTask.TaskID, "pubKey", assignedProverTask.ProverPublicKey, "err", err)
+					return err
+				}
+
+				if err := c.chunkOrm.UpdateProvingStatusFailed(c.ctx, assignedProverTask.TaskID, c.cfg.ProverManager.SessionAttempts, tx); err != nil {
+					log.Error("update proving status failed failure", "uuid", assignedProverTask.UUID, "hash", assignedProverTask.TaskID, "pubKey", assignedProverTask.ProverPublicKey, "err", err)
+					return err
+				}
+			case message.ProofTypeBatch:
+				if err := c.batchOrm.DecreaseActiveAttemptsByHash(c.ctx, assignedProverTask.TaskID, tx); err != nil {
+					log.Error("decrease batch active attempts failure", "uuid", assignedProverTask.UUID, "hash", assignedProverTask.TaskID, "pubKey", assignedProverTask.ProverPublicKey, "err", err)
+					return err
+				}
+
+				if err := c.batchOrm.UpdateProvingStatusFailed(c.ctx, assignedProverTask.TaskID, c.cfg.ProverManager.SessionAttempts, tx); err != nil {
+					log.Error("update proving status failed failure", "uuid", assignedProverTask.UUID, "hash", assignedProverTask.TaskID, "pubKey", assignedProverTask.ProverPublicKey, "err", err)
+					return err
+				}
 			}

-			// update the task to unassigned, let collector restart it
-			if message.ProofType(assignedProverTask.TaskType) == message.ProofTypeChunk {
-				if err := c.chunkOrm.UpdateProvingStatus(c.ctx, assignedProverTask.TaskID, types.ProvingTaskUnassigned, tx); err != nil {
-					log.Error("update chunk proving status to unassigned to restart it failure", "hash", assignedProverTask.TaskID, "err", err)
-				}
-			}
-			if message.ProofType(assignedProverTask.TaskType) == message.ProofTypeBatch {
-				if err := c.batchOrm.UpdateProvingStatus(c.ctx, assignedProverTask.TaskID, types.ProvingTaskUnassigned, tx); err != nil {
-					log.Error("update batch proving status to unassigned to restart it failure", "hash", assignedProverTask.TaskID, "err", err)
-				}
-			}
 			return nil
 		})
 		if err != nil {
--- a/coordinator/internal/logic/provertask/batch_prover_task.go
+++ b/coordinator/internal/logic/provertask/batch_prover_task.go
@@ -15,7 +15,6 @@ import (
 	"scroll-tech/common/types"
 	"scroll-tech/common/types/message"
 	"scroll-tech/common/utils"
-	"scroll-tech/common/version"

 	"scroll-tech/coordinator/internal/config"
 	"scroll-tech/coordinator/internal/orm"
@@ -25,7 +24,6 @@ import (
 // BatchProverTask is prover task implement for batch proof
 type BatchProverTask struct {
 	BaseProverTask
-	vk string

 	batchAttemptsExceedTotal prometheus.Counter
 	batchTaskGetTaskTotal    prometheus.Counter
@@ -35,13 +33,13 @@ type BatchProverTask struct {
 func NewBatchProverTask(cfg *config.Config, db *gorm.DB, vk string, reg prometheus.Registerer) *BatchProverTask {
 	bp := &BatchProverTask{
 		BaseProverTask: BaseProverTask{
+			vk:            vk,
 			db:            db,
 			cfg:           cfg,
 			chunkOrm:      orm.NewChunk(db),
 			batchOrm:      orm.NewBatch(db),
 			proverTaskOrm: orm.NewProverTask(db),
 		},
-		vk: vk,
 		batchAttemptsExceedTotal: promauto.With(reg).NewCounter(prometheus.CounterOpts{
 			Name: "coordinator_batch_attempts_exceed_total",
 			Help: "Total number of batch attempts exceed.",
@@ -56,71 +54,31 @@ func NewBatchProverTask(cfg *config.Config, db *gorm.DB, vk string, reg promethe

 // Assign load and assign batch tasks
 func (bp *BatchProverTask) Assign(ctx *gin.Context, getTaskParameter *coordinatorType.GetTaskParameter) (*coordinatorType.GetTaskSchema, error) {
-	publicKey, publicKeyExist := ctx.Get(coordinatorType.PublicKey)
-	if !publicKeyExist {
-		return nil, fmt.Errorf("get public key from context failed")
+	taskCtx, err := bp.checkParameter(ctx, getTaskParameter)
+	if err != nil || taskCtx == nil {
+		return nil, fmt.Errorf("check prover task parameter failed, error:%w", err)
 	}

-	proverName, proverNameExist := ctx.Get(coordinatorType.ProverName)
-	if !proverNameExist {
-		return nil, fmt.Errorf("get prover name from context failed")
-	}
-
-	proverVersion, proverVersionExist := ctx.Get(coordinatorType.ProverVersion)
-	if !proverVersionExist {
-		return nil, fmt.Errorf("get prover version from context failed")
-	}
-	if getTaskParameter.VK == "" { // allow vk being empty, because for the first time the prover may not know its vk
-		if !version.CheckScrollProverVersionTag(proverVersion.(string)) { // but reject too-old provers
-			return nil, fmt.Errorf("incompatible prover version. please upgrade your prover, expect version: %s, actual version: %s", version.Version, proverVersion.(string))
-		}
-	} else if getTaskParameter.VK != bp.vk { // non-empty vk but different
-		if version.CheckScrollProverVersion(proverVersion.(string)) { // same prover version but different vks
-			return nil, fmt.Errorf("incompatible vk. please check your params files or config files")
-		}
-		// different prover versions and different vks
-		return nil, fmt.Errorf("incompatible prover version. please upgrade your prover, expect version: %s, actual version: %s", version.Version, proverVersion.(string))
-	}
-
-	isAssigned, err := bp.proverTaskOrm.IsProverAssigned(ctx, publicKey.(string))
+	maxActiveAttempts := bp.cfg.ProverManager.ProversPerSession
+	maxTotalAttempts := bp.cfg.ProverManager.SessionAttempts
+	batchTask, err := bp.batchOrm.UpdateBatchAttemptsReturning(ctx, maxActiveAttempts, maxTotalAttempts)
 	if err != nil {
-		return nil, fmt.Errorf("failed to check if prover is assigned a task: %w", err)
-	}
-
-	if isAssigned {
-		return nil, fmt.Errorf("prover with publicKey %s is already assigned a task", publicKey)
-	}
-
-	batchTasks, err := bp.batchOrm.UpdateUnassignedBatchReturning(ctx, 1)
-	if err != nil {
-		return nil, fmt.Errorf("failed to get unassigned batch proving tasks, error:%w", err)
-	}
-
-	if len(batchTasks) == 0 {
-		return nil, nil
-	}
-
-	if len(batchTasks) != 1 {
-		log.Error("get unassigned batch proving task len not 1", "length", len(batchTasks), "batch tasks", batchTasks)
+		log.Error("failed to get unassigned batch proving tasks", "err", err)
 		return nil, ErrCoordinatorInternalFailure
 	}

-	batchTask := batchTasks[0]
-	log.Info("start batch proof generation session", "id", batchTask.Hash, "public key", publicKey, "prover name", proverName)
-
-	if !bp.checkAttemptsExceeded(batchTask.Hash, message.ProofTypeBatch) {
-		bp.batchAttemptsExceedTotal.Inc()
-		// TODO: retry fetching unassigned batch proving task
-		log.Error("batch task proving attempts reach the maximum", "hash", batchTask.Hash)
+	if batchTask == nil {
 		return nil, nil
 	}

+	log.Info("start batch proof generation session", "id", batchTask.Hash, "public key", taskCtx.PublicKey, "prover name", taskCtx.ProverName)
+
 	proverTask := orm.ProverTask{
 		TaskID:          batchTask.Hash,
-		ProverPublicKey: publicKey.(string),
+		ProverPublicKey: taskCtx.PublicKey,
 		TaskType:        int16(message.ProofTypeBatch),
-		ProverName:      proverName.(string),
-		ProverVersion:   proverVersion.(string),
+		ProverName:      taskCtx.ProverName,
+		ProverVersion:   taskCtx.ProverVersion,
 		ProvingStatus:   int16(types.ProverAssigned),
 		FailureType:     int16(types.ProverTaskFailureTypeUndefined),
 		// here why need use UTC time. see scroll/common/databased/db.go
@@ -129,14 +87,14 @@ func (bp *BatchProverTask) Assign(ctx *gin.Context, getTaskParameter *coordinato

 	// Store session info.
 	if err = bp.proverTaskOrm.InsertProverTask(ctx, &proverTask); err != nil {
-		bp.recoverProvingStatus(ctx, batchTask)
-		log.Error("insert batch prover task info fail", "taskID", batchTask.Hash, "publicKey", publicKey, "err", err)
+		bp.recoverActiveAttempts(ctx, batchTask)
+		log.Error("insert batch prover task info fail", "taskID", batchTask.Hash, "publicKey", taskCtx.PublicKey, "err", err)
 		return nil, ErrCoordinatorInternalFailure
 	}

 	taskMsg, err := bp.formatProverTask(ctx, &proverTask)
 	if err != nil {
-		bp.recoverProvingStatus(ctx, batchTask)
+		bp.recoverActiveAttempts(ctx, batchTask)
 		log.Error("format prover task failure", "hash", batchTask.Hash, "err", err)
 		return nil, ErrCoordinatorInternalFailure
 	}
@@ -193,10 +151,8 @@ func (bp *BatchProverTask) formatProverTask(ctx context.Context, task *orm.Prove
 	return taskMsg, nil
 }

-// recoverProvingStatus if not return the batch task to prover success,
-// need recover the proving status to unassigned
-func (bp *BatchProverTask) recoverProvingStatus(ctx *gin.Context, batchTask *orm.Batch) {
-	if err := bp.batchOrm.UpdateProvingStatus(ctx, batchTask.Hash, types.ProvingTaskUnassigned); err != nil {
-		log.Warn("failed to recover batch proving status", "hash:", batchTask.Hash, "error", err)
+func (bp *BatchProverTask) recoverActiveAttempts(ctx *gin.Context, batchTask *orm.Batch) {
+	if err := bp.chunkOrm.DecreaseActiveAttemptsByHash(ctx, batchTask.Hash); err != nil {
+		log.Error("failed to recover batch active attempts", "hash", batchTask.Hash, "error", err)
 	}
 }
--- a/coordinator/internal/logic/provertask/chunk_prover_task.go
+++ b/coordinator/internal/logic/provertask/chunk_prover_task.go
@@ -15,7 +15,6 @@ import (
 	"scroll-tech/common/types"
 	"scroll-tech/common/types/message"
 	"scroll-tech/common/utils"
-	"scroll-tech/common/version"

 	"scroll-tech/coordinator/internal/config"
 	"scroll-tech/coordinator/internal/orm"
@@ -28,7 +27,6 @@ var ErrCoordinatorInternalFailure = fmt.Errorf("coordinator internal error")
 // ChunkProverTask the chunk prover task
 type ChunkProverTask struct {
 	BaseProverTask
-	vk string

 	chunkAttemptsExceedTotal prometheus.Counter
 	chunkTaskGetTaskTotal    prometheus.Counter
@@ -38,13 +36,13 @@ type ChunkProverTask struct {
 func NewChunkProverTask(cfg *config.Config, db *gorm.DB, vk string, reg prometheus.Registerer) *ChunkProverTask {
 	cp := &ChunkProverTask{
 		BaseProverTask: BaseProverTask{
+			vk:            vk,
 			db:            db,
 			cfg:           cfg,
 			chunkOrm:      orm.NewChunk(db),
 			blockOrm:      orm.NewL2Block(db),
 			proverTaskOrm: orm.NewProverTask(db),
 		},
-		vk: vk,
 		chunkAttemptsExceedTotal: promauto.With(reg).NewCounter(prometheus.CounterOpts{
 			Name: "coordinator_chunk_attempts_exceed_total",
 			Help: "Total number of chunk attempts exceed.",
@@ -59,74 +57,31 @@ func NewChunkProverTask(cfg *config.Config, db *gorm.DB, vk string, reg promethe

 // Assign the chunk proof which need to prove
 func (cp *ChunkProverTask) Assign(ctx *gin.Context, getTaskParameter *coordinatorType.GetTaskParameter) (*coordinatorType.GetTaskSchema, error) {
-	publicKey, publicKeyExist := ctx.Get(coordinatorType.PublicKey)
-	if !publicKeyExist {
-		return nil, fmt.Errorf("get public key from context failed")
+	taskCtx, err := cp.checkParameter(ctx, getTaskParameter)
+	if err != nil || taskCtx == nil {
+		return nil, fmt.Errorf("check prover task parameter failed, error:%w", err)
 	}

-	proverName, proverNameExist := ctx.Get(coordinatorType.ProverName)
-	if !proverNameExist {
-		return nil, fmt.Errorf("get prover name from context failed")
-	}
-
-	proverVersion, proverVersionExist := ctx.Get(coordinatorType.ProverVersion)
-	if !proverVersionExist {
-		return nil, fmt.Errorf("get prover version from context failed")
-	}
-	if getTaskParameter.VK == "" { // allow vk being empty, because for the first time the prover may not know its vk
-		if !version.CheckScrollProverVersionTag(proverVersion.(string)) { // but reject too-old provers
-			return nil, fmt.Errorf("incompatible prover version. please upgrade your prover, expect version: %s, actual version: %s", version.Version, proverVersion.(string))
-		}
-	} else if getTaskParameter.VK != cp.vk { // non-empty vk but different
-		if version.CheckScrollProverVersion(proverVersion.(string)) { // same prover version but different vks
-			return nil, fmt.Errorf("incompatible vk. please check your params files or config files")
-		}
-		// different prover versions and different vks
-		return nil, fmt.Errorf("incompatible prover version. please upgrade your prover, expect version: %s, actual version: %s", version.Version, proverVersion.(string))
-	}
-
-	isAssigned, err := cp.proverTaskOrm.IsProverAssigned(ctx, publicKey.(string))
-	if err != nil {
-		return nil, fmt.Errorf("failed to check if prover is assigned a task: %w", err)
-	}
-
-	if isAssigned {
-		return nil, fmt.Errorf("prover with publicKey %s is already assigned a task", publicKey)
-	}
-
-	// load and send chunk tasks
-	chunkTasks, err := cp.chunkOrm.UpdateUnassignedChunkReturning(ctx, getTaskParameter.ProverHeight, 1)
+	maxActiveAttempts := cp.cfg.ProverManager.ProversPerSession
+	maxTotalAttempts := cp.cfg.ProverManager.SessionAttempts
+	chunkTask, err := cp.chunkOrm.UpdateChunkAttemptsReturning(ctx, getTaskParameter.ProverHeight, maxActiveAttempts, maxTotalAttempts)
 	if err != nil {
 		log.Error("failed to get unassigned chunk proving tasks", "height", getTaskParameter.ProverHeight, "err", err)
 		return nil, ErrCoordinatorInternalFailure
 	}

-	if len(chunkTasks) == 0 {
+	if chunkTask == nil {
 		return nil, nil
 	}

-	if len(chunkTasks) != 1 {
-		log.Error("get unassigned chunk proving task len not 1", "length", len(chunkTasks), "chunk tasks", chunkTasks)
-		return nil, ErrCoordinatorInternalFailure
-	}
-
-	chunkTask := chunkTasks[0]
-
-	log.Info("start chunk generation session", "id", chunkTask.Hash, "public key", publicKey, "prover name", proverName)
-
-	if !cp.checkAttemptsExceeded(chunkTask.Hash, message.ProofTypeChunk) {
-		cp.chunkAttemptsExceedTotal.Inc()
-		// TODO: retry fetching unassigned chunk proving task
-		log.Error("chunk task proving attempts reach the maximum", "hash", chunkTask.Hash)
-		return nil, nil
-	}
+	log.Info("start chunk generation session", "id", chunkTask.Hash, "public key", taskCtx.PublicKey, "prover name", taskCtx.ProverName)

 	proverTask := orm.ProverTask{
 		TaskID:          chunkTask.Hash,
-		ProverPublicKey: publicKey.(string),
+		ProverPublicKey: taskCtx.PublicKey,
 		TaskType:        int16(message.ProofTypeChunk),
-		ProverName:      proverName.(string),
-		ProverVersion:   proverVersion.(string),
+		ProverName:      taskCtx.ProverName,
+		ProverVersion:   taskCtx.ProverVersion,
 		ProvingStatus:   int16(types.ProverAssigned),
 		FailureType:     int16(types.ProverTaskFailureTypeUndefined),
 		// here why need use UTC time. see scroll/common/databased/db.go
@@ -134,14 +89,14 @@ func (cp *ChunkProverTask) Assign(ctx *gin.Context, getTaskParameter *coordinato
 	}

 	if err = cp.proverTaskOrm.InsertProverTask(ctx, &proverTask); err != nil {
-		cp.recoverProvingStatus(ctx, chunkTask)
-		log.Error("insert chunk prover task fail", "taskID", chunkTask.Hash, "publicKey", publicKey, "err", err)
+		cp.recoverActiveAttempts(ctx, chunkTask)
+		log.Error("insert chunk prover task fail", "taskID", chunkTask.Hash, "publicKey", taskCtx.PublicKey, "err", err)
 		return nil, ErrCoordinatorInternalFailure
 	}

 	taskMsg, err := cp.formatProverTask(ctx, &proverTask)
 	if err != nil {
-		cp.recoverProvingStatus(ctx, chunkTask)
+		cp.recoverActiveAttempts(ctx, chunkTask)
 		log.Error("format prover task failure", "hash", chunkTask.Hash, "err", err)
 		return nil, ErrCoordinatorInternalFailure
 	}
@@ -181,10 +136,8 @@ func (cp *ChunkProverTask) formatProverTask(ctx context.Context, task *orm.Prove
 	return proverTaskSchema, nil
 }

-// recoverProvingStatus if not return the batch task to prover success,
-// need recover the proving status to unassigned
-func (cp *ChunkProverTask) recoverProvingStatus(ctx *gin.Context, chunkTask *orm.Chunk) {
-	if err := cp.chunkOrm.UpdateProvingStatus(ctx, chunkTask.Hash, types.ProvingTaskUnassigned); err != nil {
-		log.Warn("failed to recover chunk proving status", "hash:", chunkTask.Hash, "error", err)
+func (cp *ChunkProverTask) recoverActiveAttempts(ctx *gin.Context, chunkTask *orm.Chunk) {
+	if err := cp.chunkOrm.DecreaseActiveAttemptsByHash(ctx, chunkTask.Hash); err != nil {
+		log.Error("failed to recover chunk active attempts", "hash", chunkTask.Hash, "error", err)
 	}
 }
--- a/coordinator/internal/logic/provertask/prover_task.go
+++ b/coordinator/internal/logic/provertask/prover_task.go
@@ -1,14 +1,12 @@
 package provertask

 import (
-	"context"
+	"fmt"

 	"github.com/gin-gonic/gin"
-	"github.com/scroll-tech/go-ethereum/log"
 	"gorm.io/gorm"

-	"scroll-tech/common/types"
-	"scroll-tech/common/types/message"
+	"scroll-tech/common/version"

 	"scroll-tech/coordinator/internal/config"
 	"scroll-tech/coordinator/internal/orm"
@@ -23,8 +21,8 @@ type ProverTask interface {
 // BaseProverTask a base prover task which contain series functions
 type BaseProverTask struct {
 	cfg *config.Config
-	ctx context.Context
 	db  *gorm.DB
+	vk  string

 	batchOrm      *orm.Batch
 	chunkOrm      *orm.Chunk
@@ -32,36 +30,51 @@ type BaseProverTask struct {
 	proverTaskOrm *orm.ProverTask
 }

-// checkAttempts use the count of prover task info to check the attempts
-func (b *BaseProverTask) checkAttemptsExceeded(hash string, taskType message.ProofType) bool {
-	whereFields := make(map[string]interface{})
-	whereFields["task_id"] = hash
-	whereFields["task_type"] = int16(taskType)
-	proverTasks, err := b.proverTaskOrm.GetProverTasks(b.ctx, whereFields, nil, 0, 0)
-	if err != nil {
-		log.Error("get prover task error", "hash id", hash, "error", err)
-		return true
-	}
-
-	if len(proverTasks) >= int(b.cfg.ProverManager.SessionAttempts) {
-		log.Warn("proof generation prover task reach the max attempts", "hash", hash)
-
-		transErr := b.db.Transaction(func(tx *gorm.DB) error {
-			switch message.ProofType(proverTasks[0].TaskType) {
-			case message.ProofTypeChunk:
-				if err := b.chunkOrm.UpdateProvingStatus(b.ctx, hash, types.ProvingTaskFailed, tx); err != nil {
-					log.Error("failed to update chunk proving_status as failed", "msg.ID", hash, "error", err)
-				}
-			case message.ProofTypeBatch:
-				if err := b.batchOrm.UpdateProvingStatus(b.ctx, hash, types.ProvingTaskFailed, tx); err != nil {
-					log.Error("failed to update batch proving_status as failed", "msg.ID", hash, "error", err)
-				}
-			}
-			return nil
-		})
-		if transErr == nil {
-			return false
-		}
-	}
-	return true
+type proverTaskContext struct {
+	PublicKey     string
+	ProverName    string
+	ProverVersion string
+}
+
+// checkParameter check the prover task parameter illegal
+func (b *BaseProverTask) checkParameter(ctx *gin.Context, getTaskParameter *coordinatorType.GetTaskParameter) (*proverTaskContext, error) {
+	var ptc proverTaskContext
+
+	publicKey, publicKeyExist := ctx.Get(coordinatorType.PublicKey)
+	if !publicKeyExist {
+		return nil, fmt.Errorf("get public key from context failed")
+	}
+	ptc.PublicKey = publicKey.(string)
+
+	proverName, proverNameExist := ctx.Get(coordinatorType.ProverName)
+	if !proverNameExist {
+		return nil, fmt.Errorf("get prover name from context failed")
+	}
+	ptc.ProverName = proverName.(string)
+
+	proverVersion, proverVersionExist := ctx.Get(coordinatorType.ProverVersion)
+	if !proverVersionExist {
+		return nil, fmt.Errorf("get prover version from context failed")
+	}
+	ptc.ProverVersion = proverVersion.(string)
+
+	// if the prover has a different vk
+	if getTaskParameter.VK != b.vk {
+		// if the prover reports a different prover version
+		if !version.CheckScrollProverVersion(proverVersion.(string)) {
+			return nil, fmt.Errorf("incompatible prover version. please upgrade your prover, expect version: %s, actual version: %s", version.Version, proverVersion.(string))
+		}
+		// if the prover reports a same prover version
+		return nil, fmt.Errorf("incompatible vk. please check your params files or config files")
+	}
+
+	isAssigned, err := b.proverTaskOrm.IsProverAssigned(ctx, publicKey.(string))
+	if err != nil {
+		return nil, fmt.Errorf("failed to check if prover is assigned a task: %w", err)
+	}
+
+	if isAssigned {
+		return nil, fmt.Errorf("prover with publicKey %s is already assigned a task", publicKey)
+	}
+	return &ptc, nil
 }
--- a/coordinator/internal/logic/submitproof/proof_receiver.go
+++ b/coordinator/internal/logic/submitproof/proof_receiver.go
@@ -174,7 +174,8 @@ func (m *ProofReceiverLogic) HandleZkProof(ctx *gin.Context, proofMsg *message.P

 	if verifyErr != nil || !success {
 		m.verifierFailureTotal.WithLabelValues(pv).Inc()
-		m.proofRecover(ctx, proverTask, proofMsg)
+
+		m.proofRecover(ctx, proverTask, types.ProverTaskFailureTypeVerifiedFailed, proofMsg)

 		log.Info("proof verified by coordinator failed", "proof id", proofMsg.ID, "prover name", proverTask.ProverName,
 			"prover pk", pk, "prove type", proofMsg.Type, "proof time", proofTimeSec, "error", verifyErr)
@@ -192,7 +193,9 @@ func (m *ProofReceiverLogic) HandleZkProof(ctx *gin.Context, proofMsg *message.P

 	if err := m.closeProofTask(ctx, proverTask, proofMsg, proofTimeSec); err != nil {
 		m.proofSubmitFailure.Inc()
-		m.proofRecover(ctx, proverTask, proofMsg)
+
+		m.proofRecover(ctx, proverTask, types.ProverTaskFailureTypeServerError, proofMsg)
+
 		return ErrCoordinatorInternalFailure
 	}

@@ -226,7 +229,8 @@ func (m *ProofReceiverLogic) validator(ctx context.Context, proverTask *orm.Prov
 	}()

 	// Ensure this prover is eligible to participate in the prover task.
-	if types.ProverProveStatus(proverTask.ProvingStatus) == types.ProverProofValid {
+	if types.ProverProveStatus(proverTask.ProvingStatus) == types.ProverProofValid ||
+		types.ProverProveStatus(proverTask.ProvingStatus) == types.ProverProofInvalid {
 		m.validateFailureProverTaskSubmitTwice.Inc()
 		// In order to prevent DoS attacks, it is forbidden to repeatedly submit valid proofs.
 		// TODO: Defend invalid proof resubmissions by one of the following two methods:
@@ -247,10 +251,8 @@ func (m *ProofReceiverLogic) validator(ctx context.Context, proverTask *orm.Prov
 	if proofMsg.Status != message.StatusOk {
 		// Temporarily replace "panic" with "pa-nic" to prevent triggering the alert based on logs.
 		failureMsg := strings.Replace(proofParameter.FailureMsg, "panic", "pa-nic", -1)
-		// Verify if the proving task has already been assigned to another prover.
-		// Upon receiving an error message, it's possible the proving status has been reset by another prover
-		// and the task has been reassigned. In this case, the coordinator should avoid resetting the proving status.
-		m.processProverErr(ctx, proverTask)
+
+		m.proofRecover(ctx, proverTask, types.ProverTaskFailureTypeSubmitStatusNotOk, proofMsg)

 		m.validateFailureProverTaskStatusNotOk.Inc()

@@ -285,20 +287,20 @@ func (m *ProofReceiverLogic) validator(ctx context.Context, proverTask *orm.Prov
 	return nil
 }

-func (m *ProofReceiverLogic) proofRecover(ctx context.Context, proverTask *orm.ProverTask, proofMsg *message.ProofMsg) {
+func (m *ProofReceiverLogic) proofRecover(ctx context.Context, proverTask *orm.ProverTask, failureType types.ProverTaskFailureType, proofMsg *message.ProofMsg) {
 	log.Info("proof recover update proof status", "hash", proverTask.TaskID, "proverPublicKey", proverTask.ProverPublicKey,
-		"taskType", proofMsg.Type.String(), "status", types.ProvingTaskUnassigned.String())
+		"taskType", message.ProofType(proverTask.TaskType).String(), "status", types.ProvingTaskUnassigned.String())

-	if err := m.updateProofStatus(ctx, proverTask, proofMsg, types.ProvingTaskUnassigned, 0); err != nil {
+	if err := m.updateProofStatus(ctx, proverTask, proofMsg, types.ProverProofInvalid, failureType, 0); err != nil {
 		log.Error("failed to updated proof status ProvingTaskUnassigned", "hash", proverTask.TaskID, "pubKey", proverTask.ProverPublicKey, "error", err)
 	}
 }

 func (m *ProofReceiverLogic) closeProofTask(ctx context.Context, proverTask *orm.ProverTask, proofMsg *message.ProofMsg, proofTimeSec uint64) error {
 	log.Info("proof close task update proof status", "hash", proverTask.TaskID, "proverPublicKey", proverTask.ProverPublicKey,
-		"taskType", proofMsg.Type.String(), "status", types.ProvingTaskVerified.String())
+		"taskType", message.ProofType(proverTask.TaskType).String(), "status", types.ProvingTaskVerified.String())

-	if err := m.updateProofStatus(ctx, proverTask, proofMsg, types.ProvingTaskVerified, proofTimeSec); err != nil {
+	if err := m.updateProofStatus(ctx, proverTask, proofMsg, types.ProverProofValid, types.ProverTaskFailureTypeUndefined, proofTimeSec); err != nil {
 		log.Error("failed to updated proof status ProvingTaskVerified", "hash", proverTask.TaskID, "proverPublicKey", proverTask.ProverPublicKey, "error", err)
 		return err
 	}
@@ -306,52 +308,46 @@ func (m *ProofReceiverLogic) closeProofTask(ctx context.Context, proverTask *orm
 }

 // UpdateProofStatus update the chunk/batch task and session info status
-func (m *ProofReceiverLogic) updateProofStatus(ctx context.Context, proverTask *orm.ProverTask, proofMsg *message.ProofMsg, status types.ProvingStatus, proofTimeSec uint64) error {
-	var proverTaskStatus types.ProverProveStatus
-	switch status {
-	case types.ProvingTaskFailed, types.ProvingTaskUnassigned:
-		proverTaskStatus = types.ProverProofInvalid
-	case types.ProvingTaskVerified:
-		proverTaskStatus = types.ProverProofValid
-	}
-
+func (m *ProofReceiverLogic) updateProofStatus(ctx context.Context, proverTask *orm.ProverTask,
+	proofMsg *message.ProofMsg, status types.ProverProveStatus, failureType types.ProverTaskFailureType, proofTimeSec uint64) error {
 	err := m.db.Transaction(func(tx *gorm.DB) error {
-		if updateErr := m.proverTaskOrm.UpdateProverTaskProvingStatus(ctx, proverTask.UUID, proverTaskStatus, tx); updateErr != nil {
+		if updateErr := m.proverTaskOrm.UpdateProverTaskProvingStatusAndFailureType(ctx, proverTask.UUID, status, failureType, tx); updateErr != nil {
+			log.Error("failed to update prover task proving status and failure type", "uuid", proverTask.UUID, "error", updateErr)
 			return updateErr
 		}

-		// if the block batch has proof verified, so the failed status not update block batch proving status
-		if m.checkIsTaskSuccess(ctx, proverTask.TaskID, proofMsg.Type) {
-			log.Info("update proof status skip because this chunk / batch has been verified", "hash", proverTask.TaskID, "public key", proverTask.ProverPublicKey)
-			return nil
-		}
-
-		if status == types.ProvingTaskVerified {
-			var storeProofErr error
-			switch proofMsg.Type {
-			case message.ProofTypeChunk:
-				storeProofErr = m.chunkOrm.UpdateProofByHash(ctx, proofMsg.ID, proofMsg.ChunkProof, proofTimeSec, tx)
-			case message.ProofTypeBatch:
-				storeProofErr = m.batchOrm.UpdateProofByHash(ctx, proofMsg.ID, proofMsg.BatchProof, proofTimeSec, tx)
-			}
-			if storeProofErr != nil {
-				log.Error("failed to store chunk/batch proof into db", "hash", proverTask.TaskID, "public key", proverTask.ProverPublicKey, "error", storeProofErr)
-				return storeProofErr
-			}
-		}
-
 		switch proofMsg.Type {
 		case message.ProofTypeChunk:
-			if err := m.chunkOrm.UpdateProvingStatus(ctx, proverTask.TaskID, status, tx); err != nil {
+			if err := m.chunkOrm.DecreaseActiveAttemptsByHash(ctx, proverTask.TaskID, tx); err != nil {
 				log.Error("failed to update chunk proving_status as failed", "hash", proverTask.TaskID, "error", err)
 				return err
 			}
 		case message.ProofTypeBatch:
-			if err := m.batchOrm.UpdateProvingStatus(ctx, proverTask.TaskID, status, tx); err != nil {
+			if err := m.batchOrm.DecreaseActiveAttemptsByHash(ctx, proverTask.TaskID, tx); err != nil {
 				log.Error("failed to update batch proving_status as failed", "hash", proverTask.TaskID, "error", err)
 				return err
 			}
 		}
+
+		// if the block batch has proof verified, so the failed status not update block batch proving status
+		if m.checkIsTaskSuccess(ctx, proverTask.TaskID, proofMsg.Type) {
+			log.Info("update proof status skip because this chunk/batch has been verified", "hash", proverTask.TaskID, "public key", proverTask.ProverPublicKey)
+			return nil
+		}
+
+		if status == types.ProverProofValid {
+			var storeProofErr error
+			switch proofMsg.Type {
+			case message.ProofTypeChunk:
+				storeProofErr = m.chunkOrm.UpdateProofAndProvingStatusByHash(ctx, proofMsg.ID, proofMsg.ChunkProof, types.ProvingTaskVerified, proofTimeSec, tx)
+			case message.ProofTypeBatch:
+				storeProofErr = m.batchOrm.UpdateProofAndProvingStatusByHash(ctx, proofMsg.ID, proofMsg.BatchProof, types.ProvingTaskVerified, proofTimeSec, tx)
+			}
+			if storeProofErr != nil {
+				log.Error("failed to store chunk/batch proof and proving status", "hash", proverTask.TaskID, "public key", proverTask.ProverPublicKey, "error", storeProofErr)
+				return storeProofErr
+			}
+		}
 		return nil
 	})

@@ -359,7 +355,7 @@ func (m *ProofReceiverLogic) updateProofStatus(ctx context.Context, proverTask *
 		return err
 	}

-	if status == types.ProvingTaskVerified && proofMsg.Type == message.ProofTypeChunk {
+	if status == types.ProverProofValid && proofMsg.Type == message.ProofTypeChunk {
 		if checkReadyErr := m.checkAreAllChunkProofsReady(ctx, proverTask.TaskID); checkReadyErr != nil {
 			log.Error("failed to check are all chunk proofs ready", "error", checkReadyErr)
 			return checkReadyErr
@@ -389,34 +385,6 @@ func (m *ProofReceiverLogic) checkIsTaskSuccess(ctx context.Context, hash string
 	return provingStatus == types.ProvingTaskVerified
 }

-func (m *ProofReceiverLogic) processProverErr(ctx context.Context, proverTask *orm.ProverTask) {
-	if updateErr := m.proverTaskOrm.UpdateProverTaskProvingStatus(ctx, proverTask.UUID, types.ProverProofInvalid); updateErr != nil {
-		log.Error("update prover task proving status failure", "uuid", proverTask.UUID, "taskID", proverTask.TaskID, "proverPublicKey",
-			proverTask.ProverPublicKey, "taskType", message.ProofType(proverTask.TaskType).String(), "error", updateErr)
-	}
-
-	proverTasks, err := m.proverTaskOrm.GetAssignedTaskOfOtherProvers(ctx, message.ProofType(proverTask.TaskType), proverTask.TaskID, proverTask.ProverPublicKey)
-	if err != nil {
-		log.Warn("checkIsAssignedToOtherProver failure", "taskID", proverTask.TaskID, "proverPublicKey", proverTask.ProverPublicKey, "taskType", message.ProofType(proverTask.TaskType).String(), "error", err)
-		return
-	}
-
-	if len(proverTasks) > 0 {
-		return
-	}
-
-	switch message.ProofType(proverTask.TaskType) {
-	case message.ProofTypeChunk:
-		if err := m.chunkOrm.UpdateProvingStatusFromProverError(ctx, proverTask.TaskID, types.ProvingTaskUnassigned); err != nil {
-			log.Error("failed to update chunk proving_status as failed", proverTask.TaskID, "proverPublicKey", proverTask.ProverPublicKey, "taskType", message.ProofType(proverTask.TaskType).String(), "error", err)
-		}
-	case message.ProofTypeBatch:
-		if err := m.batchOrm.UpdateProvingStatusFromProverError(ctx, proverTask.TaskID, types.ProvingTaskUnassigned); err != nil {
-			log.Error("failed to update batch proving_status as failed", proverTask.TaskID, "proverPublicKey", proverTask.ProverPublicKey, "taskType", message.ProofType(proverTask.TaskType).String(), "error", err)
-		}
-	}
-}
-
 func (m *ProofReceiverLogic) updateProverTaskProof(ctx context.Context, proverTask *orm.ProverTask, proofMsg *message.ProofMsg) error {
 	// store the proof to prover task
 	var proofBytes []byte
--- a/coordinator/internal/orm/batch.go
+++ b/coordinator/internal/orm/batch.go
@@ -14,6 +14,7 @@ import (

 	"scroll-tech/common/types"
 	"scroll-tech/common/types/message"
+	"scroll-tech/common/utils"
 )

 const defaultBatchHeaderVersion = 0
@@ -41,6 +42,8 @@ type Batch struct {
 	ProverAssignedAt  *time.Time `json:"prover_assigned_at" gorm:"column:prover_assigned_at;default:NULL"`
 	ProvedAt          *time.Time `json:"proved_at" gorm:"column:proved_at;default:NULL"`
 	ProofTimeSec      int32      `json:"proof_time_sec" gorm:"column:proof_time_sec;default:NULL"`
+	TotalAttempts     int16      `json:"total_attempts" gorm:"column:total_attempts;default:0"`
+	ActiveAttempts    int16      `json:"active_attempts" gorm:"column:active_attempts;default:0"`

 	// rollup
 	RollupStatus   int16      `json:"rollup_status" gorm:"column:rollup_status;default:1"`
@@ -151,6 +154,18 @@ func (o *Batch) GetLatestBatch(ctx context.Context) (*Batch, error) {
 	return &latestBatch, nil
 }

+// GetAttemptsByHash get batch attempts by hash. Used by unit test
+func (o *Batch) GetAttemptsByHash(ctx context.Context, hash string) (int16, int16, error) {
+	db := o.db.WithContext(ctx)
+	db = db.Model(&Batch{})
+	db = db.Where("hash = ?", hash)
+	var batch Batch
+	if err := db.Find(&batch).Error; err != nil {
+		return 0, 0, fmt.Errorf("Batch.GetAttemptsByHash error: %w, batch hash: %v", err, hash)
+	}
+	return batch.ActiveAttempts, batch.TotalAttempts, nil
+}
+
 // InsertBatch inserts a new batch into the database.
 // for unit test
 func (o *Batch) InsertBatch(ctx context.Context, startChunkIndex, endChunkIndex uint64, startChunkHash, endChunkHash string, chunks []*types.Chunk, dbTX ...*gorm.DB) (*Batch, error) {
@@ -211,6 +226,8 @@ func (o *Batch) InsertBatch(ctx context.Context, startChunkIndex, endChunkIndex
 		BatchHeader:       batchHeader.Encode(),
 		ChunkProofsStatus: int16(types.ChunkProofsStatusPending),
 		ProvingStatus:     int16(types.ProvingTaskUnassigned),
+		TotalAttempts:     0,
+		ActiveAttempts:    0,
 		RollupStatus:      int16(types.RollupPending),
 		OracleStatus:      int16(types.GasOraclePending),
 	}
@@ -242,60 +259,25 @@ func (o *Batch) UpdateChunkProofsStatusByBatchHash(ctx context.Context, batchHas
 	return nil
 }

-// UpdateProvingStatus updates the proving status of a batch.
-func (o *Batch) UpdateProvingStatus(ctx context.Context, hash string, status types.ProvingStatus, dbTX ...*gorm.DB) error {
+// UpdateProvingStatusFailed updates the proving status failed of a batch.
+func (o *Batch) UpdateProvingStatusFailed(ctx context.Context, hash string, maxAttempts uint8, dbTX ...*gorm.DB) error {
 	db := o.db
 	if len(dbTX) > 0 && dbTX[0] != nil {
 		db = dbTX[0]
 	}
-	updateFields := make(map[string]interface{})
-	updateFields["proving_status"] = int(status)
-
-	switch status {
-	case types.ProvingTaskAssigned:
-		updateFields["prover_assigned_at"] = time.Now()
-	case types.ProvingTaskUnassigned:
-		updateFields["prover_assigned_at"] = nil
-	case types.ProvingTaskVerified:
-		updateFields["proved_at"] = time.Now()
-	}
-
 	db = db.WithContext(ctx)
 	db = db.Model(&Batch{})
 	db = db.Where("hash", hash)
-
-	if err := db.Updates(updateFields).Error; err != nil {
-		return fmt.Errorf("Batch.UpdateProvingStatus error: %w, batch hash: %v, status: %v", err, hash, status.String())
+	db = db.Where("total_attempts >= ?", maxAttempts)
+	db = db.Where("proving_status != ?", int(types.ProverProofValid))
+	if err := db.Update("proving_status", int(types.ProvingTaskFailed)).Error; err != nil {
+		return fmt.Errorf("Batch.UpdateProvingStatus error: %w, batch hash: %v, status: %v", err, hash, types.ProvingTaskFailed.String())
 	}
 	return nil
 }

-// UpdateProvingStatusFromProverError updates batch proving status when prover prove failed
-func (o *Batch) UpdateProvingStatusFromProverError(ctx context.Context, hash string, status types.ProvingStatus) error {
-	updateFields := make(map[string]interface{})
-	updateFields["proving_status"] = int(status)
-
-	switch status {
-	case types.ProvingTaskAssigned:
-		updateFields["prover_assigned_at"] = time.Now()
-	case types.ProvingTaskUnassigned:
-		updateFields["prover_assigned_at"] = nil
-	case types.ProvingTaskVerified:
-		updateFields["proved_at"] = time.Now()
-	}
-
-	db := o.db.WithContext(ctx)
-	db = db.Model(&Batch{})
-	db = db.Where("hash", hash).Where("proving_status", types.ProvingTaskAssigned)
-
-	if err := db.Updates(updateFields).Error; err != nil {
-		return fmt.Errorf("Batch.UpdateProvingStatusOptimistic error: %w, batch hash: %v, status: %v", err, hash, status.String())
-	}
-	return nil
-}
-
-// UpdateProofByHash updates the batch proof by hash.
-func (o *Batch) UpdateProofByHash(ctx context.Context, hash string, proof *message.BatchProof, proofTimeSec uint64, dbTX ...*gorm.DB) error {
+// UpdateProofAndProvingStatusByHash updates the batch proof and proving status by hash.
+func (o *Batch) UpdateProofAndProvingStatusByHash(ctx context.Context, hash string, proof *message.BatchProof, provingStatus types.ProvingStatus, proofTimeSec uint64, dbTX ...*gorm.DB) error {
 	db := o.db
 	if len(dbTX) > 0 && dbTX[0] != nil {
 		db = dbTX[0]
@@ -307,7 +289,9 @@ func (o *Batch) UpdateProofByHash(ctx context.Context, hash string, proof *messa

 	updateFields := make(map[string]interface{})
 	updateFields["proof"] = proofBytes
+	updateFields["proving_status"] = provingStatus
 	updateFields["proof_time_sec"] = proofTimeSec
+	updateFields["proved_at"] = utils.NowUTC()

 	db = db.WithContext(ctx)
 	db = db.Model(&Batch{})
@@ -319,28 +303,55 @@ func (o *Batch) UpdateProofByHash(ctx context.Context, hash string, proof *messa
 	return nil
 }

-// UpdateUnassignedBatchReturning update the unassigned batch and return the update record
-func (o *Batch) UpdateUnassignedBatchReturning(ctx context.Context, limit int) ([]*Batch, error) {
-	if limit < 0 {
-		return nil, errors.New("limit must not be smaller than zero")
-	}
-	if limit == 0 {
-		return nil, nil
-	}
-
+// UpdateBatchAttemptsReturning atomically increments the attempts count for the earliest available batch that meets the conditions.
+func (o *Batch) UpdateBatchAttemptsReturning(ctx context.Context, maxActiveAttempts, maxTotalAttempts uint8) (*Batch, error) {
 	db := o.db.WithContext(ctx)

 	subQueryDB := db.Model(&Batch{}).Select("index")
-	subQueryDB = subQueryDB.Where("proving_status = ? AND chunk_proofs_status = ?", types.ProvingTaskUnassigned, types.ChunkProofsStatusReady)
+	subQueryDB = subQueryDB.Clauses(clause.Locking{Strength: "UPDATE"})
+	subQueryDB = subQueryDB.Where("proving_status not in (?)", []int{int(types.ProvingTaskVerified), int(types.ProvingTaskFailed)})
+	subQueryDB = subQueryDB.Where("total_attempts < ?", maxTotalAttempts)
+	subQueryDB = subQueryDB.Where("active_attempts < ?", maxActiveAttempts)
+	subQueryDB = subQueryDB.Where("chunk_proofs_status = ?", int(types.ChunkProofsStatusReady))
 	subQueryDB = subQueryDB.Order("index ASC")
-	subQueryDB = subQueryDB.Limit(limit)
+	subQueryDB = subQueryDB.Limit(1)

-	var batches []*Batch
-	db = db.Model(&batches).Clauses(clause.Returning{})
+	var updatedBatch Batch
+	db = db.Model(&updatedBatch).Clauses(clause.Returning{})
 	db = db.Where("index = (?)", subQueryDB)
-	db = db.Where("proving_status = ?", types.ProvingTaskUnassigned)
-	if err := db.Update("proving_status", types.ProvingTaskAssigned).Error; err != nil {
-		return nil, fmt.Errorf("Batch.UpdateUnassignedBatchReturning error: %w", err)
+	result := db.Updates(map[string]interface{}{
+		"proving_status":  types.ProvingTaskAssigned,
+		"total_attempts":  gorm.Expr("total_attempts + 1"),
+		"active_attempts": gorm.Expr("active_attempts + 1"),
+	})
+
+	if result.Error != nil {
+		return nil, fmt.Errorf("failed to select and update batch, max active attempts: %v, max total attempts: %v, err: %w",
+			maxActiveAttempts, maxTotalAttempts, result.Error)
 	}
-	return batches, nil
+	if result.RowsAffected == 0 {
+		return nil, nil
+	}
+	return &updatedBatch, nil
+}
+
+// DecreaseActiveAttemptsByHash decrements the active_attempts of a batch given its hash.
+func (o *Batch) DecreaseActiveAttemptsByHash(ctx context.Context, batchHash string, dbTX ...*gorm.DB) error {
+	db := o.db
+	if len(dbTX) > 0 && dbTX[0] != nil {
+		db = dbTX[0]
+	}
+	db = db.WithContext(ctx)
+	db = db.Model(&Batch{})
+	db = db.Where("hash = ?", batchHash)
+	db = db.Where("proving_status != ?", int(types.ProvingTaskVerified))
+	db = db.Where("active_attempts > ?", 0)
+	result := db.UpdateColumn("active_attempts", gorm.Expr("active_attempts - 1"))
+	if result.Error != nil {
+		return fmt.Errorf("Chunk.DecreaseActiveAttemptsByHash error: %w, batch hash: %v", result.Error, batchHash)
+	}
+	if result.RowsAffected == 0 {
+		log.Warn("No rows were affected in DecreaseActiveAttemptsByHash", "batch hash", batchHash)
+	}
+	return nil
 }
--- a/coordinator/internal/orm/chunk.go
+++ b/coordinator/internal/orm/chunk.go
@@ -13,6 +13,7 @@ import (

 	"scroll-tech/common/types"
 	"scroll-tech/common/types/message"
+	"scroll-tech/common/utils"
 )

 // Chunk represents a chunk of blocks in the database.
@@ -40,6 +41,8 @@ type Chunk struct {
 	ProverAssignedAt *time.Time `json:"prover_assigned_at" gorm:"column:prover_assigned_at;default:NULL"`
 	ProvedAt         *time.Time `json:"proved_at" gorm:"column:proved_at;default:NULL"`
 	ProofTimeSec     int32      `json:"proof_time_sec" gorm:"column:proof_time_sec;default:NULL"`
+	TotalAttempts    int16      `json:"total_attempts" gorm:"column:total_attempts;default:0"`
+	ActiveAttempts   int16      `json:"active_attempts" gorm:"column:active_attempts;default:0"`

 	// batch
 	BatchHash string `json:"batch_hash" gorm:"column:batch_hash;default:NULL"`
@@ -195,6 +198,18 @@ func (o *Chunk) GetChunkBatchHash(ctx context.Context, chunkHash string) (string
 	return chunk.BatchHash, nil
 }

+// GetAttemptsByHash get chunk attempts by hash. Used by unit test
+func (o *Chunk) GetAttemptsByHash(ctx context.Context, hash string) (int16, int16, error) {
+	db := o.db.WithContext(ctx)
+	db = db.Model(&Chunk{})
+	db = db.Where("hash = ?", hash)
+	var chunk Chunk
+	if err := db.Find(&chunk).Error; err != nil {
+		return 0, 0, fmt.Errorf("Batch.GetAttemptsByHash error: %w, batch hash: %v", err, hash)
+	}
+	return chunk.ActiveAttempts, chunk.TotalAttempts, nil
+}
+
 // InsertChunk inserts a new chunk into the database.
 // for unit test
 func (o *Chunk) InsertChunk(ctx context.Context, chunk *types.Chunk, dbTX ...*gorm.DB) (*Chunk, error) {
@@ -259,6 +274,8 @@ func (o *Chunk) InsertChunk(ctx context.Context, chunk *types.Chunk, dbTX ...*go
 		ParentChunkStateRoot:         parentChunkStateRoot,
 		WithdrawRoot:                 chunk.Blocks[numBlocks-1].WithdrawRoot.Hex(),
 		ProvingStatus:                int16(types.ProvingTaskUnassigned),
+		TotalAttempts:                0,
+		ActiveAttempts:               0,
 	}

 	db := o.db
@@ -275,19 +292,8 @@ func (o *Chunk) InsertChunk(ctx context.Context, chunk *types.Chunk, dbTX ...*go
 	return &newChunk, nil
 }

-// UpdateProvingStatus updates the proving status of a chunk.
-func (o *Chunk) UpdateProvingStatus(ctx context.Context, hash string, status types.ProvingStatus, dbTX ...*gorm.DB) error {
-	updateFields := make(map[string]interface{})
-	updateFields["proving_status"] = int(status)
-
-	switch status {
-	case types.ProvingTaskAssigned:
-		updateFields["prover_assigned_at"] = time.Now()
-	case types.ProvingTaskUnassigned:
-		updateFields["prover_assigned_at"] = nil
-	case types.ProvingTaskVerified:
-		updateFields["proved_at"] = time.Now()
-	}
+// UpdateProvingStatusFailed updates the proving status failed of a batch.
+func (o *Chunk) UpdateProvingStatusFailed(ctx context.Context, hash string, maxAttempts uint8, dbTX ...*gorm.DB) error {
 	db := o.db
 	if len(dbTX) > 0 && dbTX[0] != nil {
 		db = dbTX[0]
@@ -295,31 +301,16 @@ func (o *Chunk) UpdateProvingStatus(ctx context.Context, hash string, status typ
 	db = db.WithContext(ctx)
 	db = db.Model(&Chunk{})
 	db = db.Where("hash", hash)
-
-	if err := db.Updates(updateFields).Error; err != nil {
-		return fmt.Errorf("Chunk.UpdateProvingStatus error: %w, chunk hash: %v, status: %v", err, hash, status.String())
+	db = db.Where("total_attempts >= ?", maxAttempts)
+	db = db.Where("proving_status != ?", int(types.ProverProofValid))
+	if err := db.Update("proving_status", int(types.ProvingTaskFailed)).Error; err != nil {
+		return fmt.Errorf("Batch.UpdateProvingStatus error: %w, batch hash: %v, status: %v", err, hash, types.ProvingTaskFailed.String())
 	}
 	return nil
 }

-// UpdateProvingStatusFromProverError updates chunk proving status when prover prove failed
-func (o *Chunk) UpdateProvingStatusFromProverError(ctx context.Context, hash string, status types.ProvingStatus) error {
-	updateFields := make(map[string]interface{})
-	updateFields["proving_status"] = int(status)
-	updateFields["prover_assigned_at"] = nil
-
-	db := o.db.WithContext(ctx)
-	db = db.Model(&Chunk{})
-	db = db.Where("hash", hash).Where("proving_status", types.ProvingTaskAssigned)
-
-	if err := db.Updates(updateFields).Error; err != nil {
-		return fmt.Errorf("Chunk.UpdateProvingStatusOptimistic error: %w, chunk hash: %v, status: %v", err, hash, status.String())
-	}
-	return nil
-}
-
-// UpdateProofByHash updates the chunk proof by hash.
-func (o *Chunk) UpdateProofByHash(ctx context.Context, hash string, proof *message.ChunkProof, proofTimeSec uint64, dbTX ...*gorm.DB) error {
+// UpdateProofAndProvingStatusByHash updates the chunk proof and proving_status by hash.
+func (o *Chunk) UpdateProofAndProvingStatusByHash(ctx context.Context, hash string, proof *message.ChunkProof, status types.ProvingStatus, proofTimeSec uint64, dbTX ...*gorm.DB) error {
 	db := o.db
 	if len(dbTX) > 0 && dbTX[0] != nil {
 		db = dbTX[0]
@@ -331,7 +322,9 @@ func (o *Chunk) UpdateProofByHash(ctx context.Context, hash string, proof *messa

 	updateFields := make(map[string]interface{})
 	updateFields["proof"] = proofBytes
+	updateFields["proving_status"] = int(status)
 	updateFields["proof_time_sec"] = proofTimeSec
+	updateFields["proved_at"] = utils.NowUTC()

 	db = db.WithContext(ctx)
 	db = db.Model(&Chunk{})
@@ -357,32 +350,59 @@ func (o *Chunk) UpdateBatchHashInRange(ctx context.Context, startIndex uint64, e
 	return nil
 }

-// UpdateUnassignedChunkReturning update the unassigned batch which end_block_number <= height and return the update record
-func (o *Chunk) UpdateUnassignedChunkReturning(ctx context.Context, height, limit int) ([]*Chunk, error) {
+// UpdateChunkAttemptsReturning atomically increments the attempts count for the earliest available chunk that meets the conditions.
+func (o *Chunk) UpdateChunkAttemptsReturning(ctx context.Context, height int, maxActiveAttempts, maxTotalAttempts uint8) (*Chunk, error) {
 	if height <= 0 {
-		return nil, errors.New("Chunk.UpdateUnassignedBatchReturning error: height must be larger than zero")
-	}
-	if limit < 0 {
-		return nil, errors.New("Chunk.UpdateUnassignedBatchReturning error: limit must not be smaller than zero")
-	}
-	if limit == 0 {
-		return nil, nil
+		return nil, errors.New("Chunk.UpdateChunkAttemptsReturning error: height must be larger than zero")
 	}

 	db := o.db.WithContext(ctx)

 	subQueryDB := db.Model(&Chunk{}).Select("index")
-	subQueryDB = subQueryDB.Where("proving_status = ?", types.ProvingTaskUnassigned)
+	subQueryDB = subQueryDB.Clauses(clause.Locking{Strength: "UPDATE"})
+	subQueryDB = subQueryDB.Where("proving_status not in (?)", []int{int(types.ProvingTaskVerified), int(types.ProvingTaskFailed)})
+	subQueryDB = subQueryDB.Where("total_attempts < ?", maxTotalAttempts)
+	subQueryDB = subQueryDB.Where("active_attempts < ?", maxActiveAttempts)
 	subQueryDB = subQueryDB.Where("end_block_number <= ?", height)
 	subQueryDB = subQueryDB.Order("index ASC")
-	subQueryDB = subQueryDB.Limit(limit)
+	subQueryDB = subQueryDB.Limit(1)

-	var chunks []*Chunk
-	db = db.Model(&chunks).Clauses(clause.Returning{})
+	var updatedChunk Chunk
+	db = db.Model(&updatedChunk).Clauses(clause.Returning{})
 	db = db.Where("index = (?)", subQueryDB)
-	db = db.Where("proving_status = ?", types.ProvingTaskUnassigned)
-	if err := db.Update("proving_status", types.ProvingTaskAssigned).Error; err != nil {
-		return nil, fmt.Errorf("Chunk.UpdateUnassignedBatchReturning error: %w", err)
+	result := db.Updates(map[string]interface{}{
+		"proving_status":  types.ProvingTaskAssigned,
+		"total_attempts":  gorm.Expr("total_attempts + 1"),
+		"active_attempts": gorm.Expr("active_attempts + 1"),
+	})
+
+	if result.Error != nil {
+		return nil, fmt.Errorf("failed to select and update batch, max active attempts: %v, max total attempts: %v, err: %w",
+			maxActiveAttempts, maxTotalAttempts, result.Error)
 	}
-	return chunks, nil
+	if result.RowsAffected == 0 {
+		return nil, nil
+	}
+	return &updatedChunk, nil
+}
+
+// DecreaseActiveAttemptsByHash decrements the active_attempts of a chunk given its hash.
+func (o *Chunk) DecreaseActiveAttemptsByHash(ctx context.Context, chunkHash string, dbTX ...*gorm.DB) error {
+	db := o.db
+	if len(dbTX) > 0 && dbTX[0] != nil {
+		db = dbTX[0]
+	}
+	db = db.WithContext(ctx)
+	db = db.Model(&Chunk{})
+	db = db.Where("hash = ?", chunkHash)
+	db = db.Where("proving_status != ?", int(types.ProvingTaskVerified))
+	db = db.Where("active_attempts > ?", 0)
+	result := db.UpdateColumn("active_attempts", gorm.Expr("active_attempts - 1"))
+	if result.Error != nil {
+		return fmt.Errorf("Chunk.DecreaseActiveAttemptsByHash error: %w, chunk hash: %v", result.Error, chunkHash)
+	}
+	if result.RowsAffected == 0 {
+		log.Warn("No rows were affected in DecreaseActiveAttemptsByHash", "chunk hash", chunkHash)
+	}
+	return nil
 }
--- a/coordinator/internal/orm/prover_task.go
+++ b/coordinator/internal/orm/prover_task.go
@@ -244,8 +244,8 @@ func (o *ProverTask) UpdateProverTaskProof(ctx context.Context, uuid uuid.UUID,
 	return nil
 }

-// UpdateProverTaskProvingStatus updates the proving_status of a specific ProverTask record.
-func (o *ProverTask) UpdateProverTaskProvingStatus(ctx context.Context, uuid uuid.UUID, status types.ProverProveStatus, dbTX ...*gorm.DB) error {
+// UpdateProverTaskProvingStatusAndFailureType updates the proving_status of a specific ProverTask record.
+func (o *ProverTask) UpdateProverTaskProvingStatusAndFailureType(ctx context.Context, uuid uuid.UUID, status types.ProverProveStatus, failureType types.ProverTaskFailureType, dbTX ...*gorm.DB) error {
 	db := o.db
 	if len(dbTX) > 0 && dbTX[0] != nil {
 		db = dbTX[0]
@@ -254,7 +254,12 @@ func (o *ProverTask) UpdateProverTaskProvingStatus(ctx context.Context, uuid uui
 	db = db.Model(&ProverTask{})
 	db = db.Where("uuid = ?", uuid)

-	if err := db.Update("proving_status", status).Error; err != nil {
+	updates := make(map[string]interface{})
+	updates["proving_status"] = int(status)
+	if status == types.ProverProofInvalid {
+		updates["failure_type"] = int(failureType)
+	}
+	if err := db.Updates(updates).Error; err != nil {
 		return fmt.Errorf("ProverTask.UpdateProverTaskProvingStatus error: %w, uuid:%s, status: %v", err, uuid, status.String())
 	}
 	return nil
--- a/coordinator/test/api_test.go
+++ b/coordinator/test/api_test.go
@@ -246,8 +246,14 @@ func testValidProof(t *testing.T) {
 		tickStop = time.Tick(time.Minute)
 	)

-	var chunkProofStatus types.ProvingStatus
-	var batchProofStatus types.ProvingStatus
+	var (
+		chunkProofStatus    types.ProvingStatus
+		batchProofStatus    types.ProvingStatus
+		chunkActiveAttempts int16
+		chunkMaxAttempts    int16
+		batchActiveAttempts int16
+		batchMaxAttempts    int16
+	)

 	for {
 		select {
@@ -259,6 +265,17 @@ func testValidProof(t *testing.T) {
 			if chunkProofStatus == types.ProvingTaskVerified && batchProofStatus == types.ProvingTaskVerified {
 				return
 			}
+
+			chunkActiveAttempts, chunkMaxAttempts, err = chunkOrm.GetAttemptsByHash(context.Background(), dbChunk.Hash)
+			assert.NoError(t, err)
+			assert.Equal(t, 1, int(chunkMaxAttempts))
+			assert.Equal(t, 0, int(chunkActiveAttempts))
+
+			batchActiveAttempts, batchMaxAttempts, err = batchOrm.GetAttemptsByHash(context.Background(), batch.Hash)
+			assert.NoError(t, err)
+			assert.Equal(t, 1, int(batchMaxAttempts))
+			assert.Equal(t, 0, int(batchActiveAttempts))
+
 		case <-tickStop:
 			t.Error("failed to check proof status", "chunkProofStatus", chunkProofStatus.String(), "batchProofStatus", batchProofStatus.String())
 			return
@@ -307,8 +324,14 @@ func testInvalidProof(t *testing.T) {
 		tickStop = time.Tick(time.Minute)
 	)

-	var chunkProofStatus types.ProvingStatus
-	var batchProofStatus types.ProvingStatus
+	var (
+		chunkProofStatus    types.ProvingStatus
+		batchProofStatus    types.ProvingStatus
+		chunkActiveAttempts int16
+		chunkMaxAttempts    int16
+		batchActiveAttempts int16
+		batchMaxAttempts    int16
+	)

 	for {
 		select {
@@ -317,9 +340,18 @@ func testInvalidProof(t *testing.T) {
 			assert.NoError(t, err)
 			batchProofStatus, err = batchOrm.GetProvingStatusByHash(context.Background(), batch.Hash)
 			assert.NoError(t, err)
-			if chunkProofStatus == types.ProvingTaskUnassigned && batchProofStatus == types.ProvingTaskUnassigned {
+			if chunkProofStatus == types.ProvingTaskAssigned && batchProofStatus == types.ProvingTaskAssigned {
 				return
 			}
+			chunkActiveAttempts, chunkMaxAttempts, err = chunkOrm.GetAttemptsByHash(context.Background(), dbChunk.Hash)
+			assert.NoError(t, err)
+			assert.Equal(t, 1, int(chunkMaxAttempts))
+			assert.Equal(t, 0, int(chunkActiveAttempts))
+
+			batchActiveAttempts, batchMaxAttempts, err = batchOrm.GetAttemptsByHash(context.Background(), batch.Hash)
+			assert.NoError(t, err)
+			assert.Equal(t, 1, int(batchMaxAttempts))
+			assert.Equal(t, 0, int(batchActiveAttempts))
 		case <-tickStop:
 			t.Error("failed to check proof status", "chunkProofStatus", chunkProofStatus.String(), "batchProofStatus", batchProofStatus.String())
 			return
@@ -373,6 +405,10 @@ func testProofGeneratedFailed(t *testing.T) {
 		batchProofStatus             types.ProvingStatus
 		chunkProverTaskProvingStatus types.ProverProveStatus
 		batchProverTaskProvingStatus types.ProverProveStatus
+		chunkActiveAttempts          int16
+		chunkMaxAttempts             int16
+		batchActiveAttempts          int16
+		batchMaxAttempts             int16
 	)

 	for {
@@ -386,6 +422,16 @@ func testProofGeneratedFailed(t *testing.T) {
 				return
 			}

+			chunkActiveAttempts, chunkMaxAttempts, err = chunkOrm.GetAttemptsByHash(context.Background(), dbChunk.Hash)
+			assert.NoError(t, err)
+			assert.Equal(t, 1, int(chunkMaxAttempts))
+			assert.Equal(t, 0, int(chunkActiveAttempts))
+
+			batchActiveAttempts, batchMaxAttempts, err = batchOrm.GetAttemptsByHash(context.Background(), batch.Hash)
+			assert.NoError(t, err)
+			assert.Equal(t, 1, int(batchMaxAttempts))
+			assert.Equal(t, 0, int(batchActiveAttempts))
+
 			chunkProverTaskProvingStatus, err = proverTaskOrm.GetProvingStatusByTaskID(context.Background(), message.ProofTypeChunk, dbChunk.Hash)
 			assert.NoError(t, err)
 			batchProverTaskProvingStatus, err = proverTaskOrm.GetProvingStatusByTaskID(context.Background(), message.ProofTypeBatch, batch.Hash)
@@ -409,6 +455,13 @@ func testTimeoutProof(t *testing.T) {
 		assert.NoError(t, httpHandler.Shutdown(context.Background()))
 	}()

+	var (
+		chunkActiveAttempts int16
+		chunkMaxAttempts    int16
+		batchActiveAttempts int16
+		batchMaxAttempts    int16
+	)
+
 	err := l2BlockOrm.InsertL2Blocks(context.Background(), []*types.WrappedBlock{wrappedBlock1, wrappedBlock2})
 	assert.NoError(t, err)
 	dbChunk, err := chunkOrm.InsertChunk(context.Background(), chunk)
@@ -438,6 +491,16 @@ func testTimeoutProof(t *testing.T) {
 	assert.NoError(t, err)
 	assert.Equal(t, batchProofStatus, types.ProvingTaskAssigned)

+	chunkActiveAttempts, chunkMaxAttempts, err = chunkOrm.GetAttemptsByHash(context.Background(), dbChunk.Hash)
+	assert.NoError(t, err)
+	assert.Equal(t, 1, int(chunkMaxAttempts))
+	assert.Equal(t, 1, int(chunkActiveAttempts))
+
+	batchActiveAttempts, batchMaxAttempts, err = batchOrm.GetAttemptsByHash(context.Background(), batch.Hash)
+	assert.NoError(t, err)
+	assert.Equal(t, 1, int(batchMaxAttempts))
+	assert.Equal(t, 1, int(batchActiveAttempts))
+
 	// wait coordinator to reset the prover task proving status
 	time.Sleep(time.Duration(conf.ProverManager.BatchCollectionTimeSec*2) * time.Second)

@@ -460,4 +523,14 @@ func testTimeoutProof(t *testing.T) {
 	batchProofStatus2, err := batchOrm.GetProvingStatusByHash(context.Background(), batch.Hash)
 	assert.NoError(t, err)
 	assert.Equal(t, batchProofStatus2, types.ProvingTaskVerified)
+
+	chunkActiveAttempts, chunkMaxAttempts, err = chunkOrm.GetAttemptsByHash(context.Background(), dbChunk.Hash)
+	assert.NoError(t, err)
+	assert.Equal(t, 2, int(chunkMaxAttempts))
+	assert.Equal(t, 0, int(chunkActiveAttempts))
+
+	batchActiveAttempts, batchMaxAttempts, err = batchOrm.GetAttemptsByHash(context.Background(), batch.Hash)
+	assert.NoError(t, err)
+	assert.Equal(t, 2, int(batchMaxAttempts))
+	assert.Equal(t, 0, int(batchActiveAttempts))
 }
--- a/database/migrate/migrate_test.go
+++ b/database/migrate/migrate_test.go
@@ -63,7 +63,7 @@ func testResetDB(t *testing.T) {
 	cur, err := Current(pgDB.DB)
 	assert.NoError(t, err)
 	// total number of tables.
-	assert.Equal(t, 11, int(cur))
+	assert.Equal(t, 12, int(cur))
 }

 func testMigrate(t *testing.T) {
--- a/database/migrate/migrations/00012_chunk_batch_add_attempts.sql
+++ b/database/migrate/migrations/00012_chunk_batch_add_attempts.sql
@@ -0,0 +1,28 @@
+-- +goose Up
+-- +goose StatementBegin
+
+ALTER TABLE chunk
+    ADD COLUMN total_attempts SMALLINT NOT NULL DEFAULT 0,
+    ADD COLUMN active_attempts SMALLINT NOT NULL DEFAULT 0;
+
+ALTER TABLE batch
+    ADD COLUMN total_attempts SMALLINT NOT NULL DEFAULT 0,
+    ADD COLUMN active_attempts SMALLINT NOT NULL DEFAULT 0;
+
+create index if not exists idx_total_attempts_active_attempts_end_block_number
+    on chunk (total_attempts, active_attempts, end_block_number)
+    where deleted_at IS NULL;
+
+create index if not exists idx_total_attempts_active_attempts_chunk_proofs_status
+    on batch (total_attempts, active_attempts, chunk_proofs_status)
+    where deleted_at IS NULL;
+
+-- +goose StatementEnd
+
+-- +goose Down
+-- +goose StatementBegin
+
+drop index if exists idx_total_attempts_active_attempts_end_block_number;
+drop index if exists idx_total_attempts_active_attempts_chunk_proofs_status;
+
+-- +goose StatementEnd
--- a/prover-stats-api/Makefile
+++ b/prover-stats-api/Makefile
@@ -4,6 +4,8 @@ IMAGE_NAME=prover-stats-api
 IMAGE_VERSION=latest
 REPO_ROOT_DIR=./..

+all: build
+
 build:
 	GOBIN=$(PWD)/build/bin go build -o $(PWD)/build/bin/prover-stats-api ./cmd

@@ -21,3 +23,5 @@ lint: ## Lint the files - used for CI

 docker:
 	DOCKER_BUILDKIT=1 docker build -t scrolltech/${IMAGE_NAME}:${IMAGE_VERSION} ${REPO_ROOT_DIR}/ -f ${REPO_ROOT_DIR}/build/dockerfiles/prover-stats-api.Dockerfile
+
+.PHONY: all
--- a/prover-stats-api/README.md
+++ b/prover-stats-api/README.md
@@ -1,28 +1,23 @@
 # prover-stats-api

-## how to get the prover-stats-api docs
+This directory contains the `prover-stats-api` service that provides REST APIs to query the status of proving tasks assigned to the prover.

-### 1. start the prover-stats-api server
+## Instructions

-```
-cd ./prover-stats-api
-make build
-./prover-stats --config=./conf/config.json
-```
+1. Build and start the `prover-stats-api` service.

-you will get server run log
-```
-Listening and serving HTTP on :8990
-```
+    ```
+    cd ./prover-stats-api
+    make build
+    ./build/bin/prover-stats --config=./conf/config.json
+    ```

-### 2. browse the documents
+2. Open this URL in your browser to view the API documents.
+    ```
+    http://localhost:8990/swagger/index.html
+    ```

-open this documents in your browser
-```
-http://localhost:8990/swagger/index.html
-```
-
-## how to update the prover-stats-api docs
+## How to update the prover-stats-api docs

 ```
 cd ./prover-stats-api
--- a/prover/core/mock.go
+++ b/prover/core/mock.go
@@ -16,6 +16,7 @@ import (
 // ProverCore sends block-traces to rust-prover through socket and get back the zk-proof.
 type ProverCore struct {
 	cfg *config.ProverCoreConfig
+	VK  string
 }

 // NewProverCore inits a ProverCore object.
@@ -42,7 +43,3 @@ func (p *ProverCore) ProveBatch(taskID string, chunkInfos []*message.ChunkInfo,
 		Vk:        _empty[:],
 	}, nil
 }
-
-func (p *ProverCore) GetVk() string {
-	return ""
-}
--- a/prover/core/prover.go
+++ b/prover/core/prover.go
@@ -28,7 +28,7 @@ import (
 // ProverCore sends block-traces to rust-prover through ffi and get back the zk-proof.
 type ProverCore struct {
 	cfg *config.ProverCoreConfig
-	vk  string
+	VK  string
 }

 // NewProverCore inits a ProverCore object.
@@ -40,10 +40,18 @@ func NewProverCore(cfg *config.ProverCoreConfig) (*ProverCore, error) {
 		C.free(unsafe.Pointer(assetsPathStr))
 	}()

+	var vk string
+	var rawVK *C.char
 	if cfg.ProofType == message.ProofTypeBatch {
 		C.init_batch_prover(paramsPathStr, assetsPathStr)
+		rawVK = C.get_batch_vk()
 	} else if cfg.ProofType == message.ProofTypeChunk {
 		C.init_chunk_prover(paramsPathStr, assetsPathStr)
+		rawVK = C.get_chunk_vk()
+	}
+
+	if rawVK != nil {
+		vk = C.GoString(rawVK)
 	}

 	if cfg.DumpDir != "" {
@@ -54,27 +62,7 @@ func NewProverCore(cfg *config.ProverCoreConfig) (*ProverCore, error) {
 		log.Info("Enabled dump_proof", "dir", cfg.DumpDir)
 	}

-	return &ProverCore{cfg: cfg}, nil
-}
-
-// GetVk get Base64 format of vk.
-func (p *ProverCore) GetVk() string {
-	if p.vk != "" { // cached
-		return p.vk
-	}
-
-	var raw *C.char
-	if p.cfg.ProofType == message.ProofTypeBatch {
-		raw = C.get_batch_vk()
-	} else if p.cfg.ProofType == message.ProofTypeChunk {
-		raw = C.get_chunk_vk()
-	}
-
-	if raw != nil {
-		p.vk = C.GoString(raw) // cache it
-	}
-
-	return p.vk
+	return &ProverCore{cfg: cfg, VK: vk}, nil
 }

 // ProveBatch call rust ffi to generate batch proof.
--- a/prover/core/prover_test.go
+++ b/prover/core/prover_test.go
@@ -14,7 +14,6 @@ import (
 	"github.com/scroll-tech/go-ethereum/core/types"
 	"github.com/stretchr/testify/assert"

-	scrollTypes "scroll-tech/common/types"
 	"scroll-tech/common/types/message"

 	"scroll-tech/prover/config"
@@ -44,6 +43,9 @@ func TestFFI(t *testing.T) {
 	as.NoError(err)
 	t.Log("Constructed chunk prover")

+	as.Equal(chunkProverCore.GetVk(), readVk(*chunkVkPath, as))
+	t.Log("Chunk VK must be available when init")
+
 	chunkTrace1 := readChunkTrace(*tracePath1, as)
 	chunkTrace2 := readChunkTrace(*tracePath2, as)
 	t.Log("Loaded chunk traces")
@@ -54,32 +56,6 @@ func TestFFI(t *testing.T) {
 	as.NoError(err)
 	t.Log("Converted to chunk infos")

-	wrappedBlock1 := &scrollTypes.WrappedBlock{
-		Header:       chunkTrace1[0].Header,
-		Transactions: chunkTrace1[0].Transactions,
-		WithdrawRoot: chunkTrace1[0].WithdrawTrieRoot,
-	}
-	chunk1 := &scrollTypes.Chunk{Blocks: []*scrollTypes.WrappedBlock{wrappedBlock1}}
-	chunkHash1, err := chunk1.Hash(0)
-	as.NoError(err)
-	as.Equal(chunkInfo1.PostStateRoot, wrappedBlock1.Header.Root)
-	as.Equal(chunkInfo1.WithdrawRoot, wrappedBlock1.WithdrawRoot)
-	as.Equal(chunkInfo1.DataHash, chunkHash1)
-	t.Log("Successful to check chunk info 1")
-
-	wrappedBlock2 := &scrollTypes.WrappedBlock{
-		Header:       chunkTrace2[0].Header,
-		Transactions: chunkTrace2[0].Transactions,
-		WithdrawRoot: chunkTrace2[0].WithdrawTrieRoot,
-	}
-	chunk2 := &scrollTypes.Chunk{Blocks: []*scrollTypes.WrappedBlock{wrappedBlock2}}
-	chunkHash2, err := chunk2.Hash(chunk1.NumL1Messages(0))
-	as.NoError(err)
-	as.Equal(chunkInfo2.PostStateRoot, wrappedBlock2.Header.Root)
-	as.Equal(chunkInfo2.WithdrawRoot, wrappedBlock2.WithdrawRoot)
-	as.Equal(chunkInfo2.DataHash, chunkHash2)
-	t.Log("Successful to check chunk info 2")
-
 	chunkProof1, err := chunkProverCore.ProveChunk("chunk_proof1", chunkTrace1)
 	as.NoError(err)
 	t.Log("Generated and dumped chunk proof 1")
@@ -89,7 +65,7 @@ func TestFFI(t *testing.T) {
 	t.Log("Generated and dumped chunk proof 2")

 	as.Equal(chunkProverCore.GetVk(), readVk(*chunkVkPath, as))
-	t.Log("Chunk VKs are equal")
+	t.Log("Chunk VKs must be equal after proving")

 	batchProverConfig := &config.ProverCoreConfig{
 		DumpDir:    *proofDumpPath,
@@ -100,6 +76,9 @@ func TestFFI(t *testing.T) {
 	batchProverCore, err := core.NewProverCore(batchProverConfig)
 	as.NoError(err)

+	as.Equal(batchProverCore.GetVk(), readVk(*batchVkPath, as))
+	t.Log("Batch VK must be available when init")
+
 	chunkInfos := []*message.ChunkInfo{chunkInfo1, chunkInfo2}
 	chunkProofs := []*message.ChunkProof{chunkProof1, chunkProof2}
 	_, err = batchProverCore.ProveBatch("batch_proof", chunkInfos, chunkProofs)
@@ -107,7 +86,7 @@ func TestFFI(t *testing.T) {
 	t.Log("Generated and dumped batch proof")

 	as.Equal(batchProverCore.GetVk(), readVk(*batchVkPath, as))
-	t.Log("Batch VKs are equal")
+	t.Log("Batch VKs must be equal after proving")
 }

 func readChunkTrace(filePat string, as *assert.Assertions) []*types.BlockTrace {
--- a/prover/prover.go
+++ b/prover/prover.go
@@ -181,7 +181,7 @@ func (r *Prover) fetchTaskFromCoordinator() (*store.ProvingTask, error) {
 		TaskType: r.Type(),
 		// we may not be able to get the vk at the first time, so we should pass vk to the coordinator every time we getTask
 		// instead of passing vk when we login
-		VK: r.proverCore.GetVk(),
+		VK: r.proverCore.VK,
 	}

 	if req.TaskType == message.ProofTypeChunk {
--- a/rollup/README.md
+++ b/rollup/README.md
@@ -1,15 +1,23 @@
 # Rollup

-This repo contains the Scroll rollup.
+This directory contains the three essential rollup services for the Scroll chain:
+- Event Watcher (<a href="./cmd/event_watcher/">event_watcher</a>): watches the events emitted from the L1 and L2 contracts and updates the event database.
+- Gas Oracle (<a href="./cmd/gas_oracle/">gas_oracle</a>): monitors the L1 and L2 gas price and sends transactions to update the gas price oracle contracts on L1 and L2.
+- Rollup Relayer (<a href="./cmd/rollup_relayer/">rollup_relayer</a>): consists of three components: chunk and batch proposer and a relayer.
+    - The chunk and batch proposer proposes new chunks and batches that sends Commit Transactions for data availability and Finalize Transactions for proof verification and state finalization.

 ## Dependency

-+ install `abigen`
+1. `abigen`

 ``` bash
 go install -v github.com/scroll-tech/go-ethereum/cmd/abigen
 ```

+2. `solc`
+
+See https://docs.soliditylang.org/en/latest/installing-solidity.html
+
 ## Build

 ```bash
@@ -22,18 +30,8 @@ make rollup_bins

 (Note: make sure you use different private keys for different senders in config.json.)

-* use default ports and config.json.
-
 ```bash
-./build/bin/event_watcher --http
-./build/bin/gas_oracle --http
-./build/bin/rollup_relayer --http
-```
-
-* use specified ports and config.json
-
-```bash
-./build/bin/event_watcher --config ./config.json --http --http.addr localhost --http.port 8290
-./build/bin/gas_oracle --config ./config.json --http --http.addr localhost --http.port 8290
-./build/bin/rollup_relayer --config ./config.json --http --http.addr localhost --http.port 8290
+./build/bin/event_watcher --config ./config.json
+./build/bin/gas_oracle --config ./config.json
+./build/bin/rollup_relayer --config ./config.json
 ```
--- a/rollup/cmd/gas_oracle/app/app.go
+++ b/rollup/cmd/gas_oracle/app/app.go
@@ -10,6 +10,7 @@ import (
 	"github.com/prometheus/client_golang/prometheus"
 	"github.com/scroll-tech/go-ethereum/ethclient"
 	"github.com/scroll-tech/go-ethereum/log"
+	"github.com/scroll-tech/go-ethereum/rpc"
 	"github.com/urfave/cli/v2"

 	"scroll-tech/common/database"
@@ -92,14 +93,16 @@ func action(ctx *cli.Context) error {
 	}
 	// Start l1 watcher process
 	go utils.LoopWithContext(subCtx, 10*time.Second, func(ctx context.Context) {
-		number, loopErr := butils.GetLatestConfirmedBlockNumber(ctx, l1client, cfg.L1Config.Confirmations)
+		// Fetch the latest block number to decrease the delay when fetching gas prices
+		// Use latest block number - 1 to prevent frequent reorg
+		number, loopErr := butils.GetLatestConfirmedBlockNumber(ctx, l1client, rpc.LatestBlockNumber)
 		if loopErr != nil {
 			log.Error("failed to get block number", "err", loopErr)
 			return
 		}

-		if loopErr = l1watcher.FetchBlockHeader(number); loopErr != nil {
-			log.Error("Failed to fetch L1 block header", "lastest", number, "err", loopErr)
+		if loopErr = l1watcher.FetchBlockHeader(number - 1); loopErr != nil {
+			log.Error("Failed to fetch L1 block header", "lastest", number-1, "err", loopErr)
 		}
 	})

--- a/rollup/conf/config.json
+++ b/rollup/conf/config.json
@@ -65,7 +65,8 @@
      "gas_cost_increase_multiplier": 1.2
    },
    "chunk_proposer_config": {
-      "max_tx_num_per_chunk": 1123,
+      "max_block_num_per_chunk": 100,
+      "max_tx_num_per_chunk": 100,
      "max_l1_commit_gas_per_chunk": 11234567,
      "max_l1_commit_calldata_size_per_chunk": 112345,
      "chunk_timeout_sec": 300,
--- a/rollup/internal/config/l2.go
+++ b/rollup/internal/config/l2.go
@@ -28,6 +28,7 @@ type L2Config struct {

 // ChunkProposerConfig loads chunk_proposer configuration items.
 type ChunkProposerConfig struct {
+	MaxBlockNumPerChunk             uint64  `json:"max_block_num_per_chunk"`
 	MaxTxNumPerChunk                uint64  `json:"max_tx_num_per_chunk"`
 	MaxL1CommitGasPerChunk          uint64  `json:"max_l1_commit_gas_per_chunk"`
 	MaxL1CommitCalldataSizePerChunk uint64  `json:"max_l1_commit_calldata_size_per_chunk"`
--- a/rollup/internal/controller/watcher/batch_proposer.go
+++ b/rollup/internal/controller/watcher/batch_proposer.go
@@ -154,7 +154,8 @@ func (p *BatchProposer) proposeBatchChunks() ([]*orm.Chunk, *types.BatchMeta, er
 		return nil, nil, err
 	}

-	dbChunks, err := p.chunkOrm.GetChunksGEIndex(p.ctx, unbatchedChunkIndex, int(p.maxChunkNumPerBatch)+1)
+	// select at most p.maxChunkNumPerBatch chunks
+	dbChunks, err := p.chunkOrm.GetChunksGEIndex(p.ctx, unbatchedChunkIndex, int(p.maxChunkNumPerBatch))
 	if err != nil {
 		return nil, nil, err
 	}
@@ -207,8 +208,7 @@ func (p *BatchProposer) proposeBatchChunks() ([]*orm.Chunk, *types.BatchMeta, er
 		totalL1CommitGas += types.CalldataNonZeroByteGas * (32 * (totalL1MessagePopped + 255) / 256)
 		totalL1CommitGas += types.GetKeccak256Gas(89 + 32*(totalL1MessagePopped+255)/256)
 		totalOverEstimateL1CommitGas := uint64(p.gasCostIncreaseMultiplier * float64(totalL1CommitGas))
-		if totalChunks > p.maxChunkNumPerBatch ||
-			totalL1CommitCalldataSize > p.maxL1CommitCalldataSizePerBatch ||
+		if totalL1CommitCalldataSize > p.maxL1CommitCalldataSizePerBatch ||
 			totalOverEstimateL1CommitGas > p.maxL1CommitGasPerBatch {
 			// Check if the first chunk breaks hard limits.
 			// If so, it indicates there are bugs in chunk-proposer, manual fix is needed.
@@ -234,8 +234,6 @@ func (p *BatchProposer) proposeBatchChunks() ([]*orm.Chunk, *types.BatchMeta, er
 			}

 			log.Debug("breaking limit condition in batching",
-				"currentTotalChunks", totalChunks,
-				"maxChunkNumPerBatch", p.maxChunkNumPerBatch,
 				"currentL1CommitCalldataSize", totalL1CommitCalldataSize,
 				"maxL1CommitCalldataSizePerBatch", p.maxL1CommitCalldataSizePerBatch,
 				"currentOverEstimateL1CommitGas", totalOverEstimateL1CommitGas,
@@ -249,12 +247,20 @@ func (p *BatchProposer) proposeBatchChunks() ([]*orm.Chunk, *types.BatchMeta, er
 	}

 	currentTimeSec := uint64(time.Now().Unix())
-	if dbChunks[0].StartBlockTime+p.batchTimeoutSec < currentTimeSec {
-		log.Warn("first block timeout",
-			"start block number", dbChunks[0].StartBlockNumber,
-			"first block timestamp", dbChunks[0].StartBlockTime,
-			"chunk outdated time threshold", currentTimeSec,
-		)
+	if dbChunks[0].StartBlockTime+p.batchTimeoutSec < currentTimeSec ||
+		totalChunks == p.maxChunkNumPerBatch {
+		if dbChunks[0].StartBlockTime+p.batchTimeoutSec < currentTimeSec {
+			log.Warn("first block timeout",
+				"start block number", dbChunks[0].StartBlockNumber,
+				"start block timestamp", dbChunks[0].StartBlockTime,
+				"current time", currentTimeSec,
+			)
+		} else {
+			log.Info("reached maximum number of chunks in batch",
+				"chunk count", totalChunks,
+			)
+		}
+
 		batchMeta.TotalL1CommitGas = totalL1CommitGas
 		batchMeta.TotalL1CommitCalldataSize = totalL1CommitCalldataSize
 		p.batchFirstBlockTimeoutReached.Inc()
--- a/rollup/internal/controller/watcher/batch_proposer_test.go
+++ b/rollup/internal/controller/watcher/batch_proposer_test.go
@@ -23,6 +23,7 @@ func testBatchProposer(t *testing.T) {
 	assert.NoError(t, err)

 	cp := NewChunkProposer(context.Background(), &config.ChunkProposerConfig{
+		MaxBlockNumPerChunk:             100,
 		MaxTxNumPerChunk:                10000,
 		MaxL1CommitGasPerChunk:          50000000000,
 		MaxL1CommitCalldataSizePerChunk: 1000000,
--- a/rollup/internal/controller/watcher/chunk_proposer.go
+++ b/rollup/internal/controller/watcher/chunk_proposer.go
@@ -18,10 +18,6 @@ import (
 	"scroll-tech/rollup/internal/orm"
 )

-// maxNumBlockPerChunk is the maximum number of blocks we allow per chunk.
-// Normally we will pack much fewer blocks because of other limits.
-const maxNumBlockPerChunk int = 100
-
 // chunkRowConsumption is map(sub-circuit name => sub-circuit row count)
 type chunkRowConsumption map[string]uint64

@@ -55,6 +51,7 @@ type ChunkProposer struct {
 	chunkOrm   *orm.Chunk
 	l2BlockOrm *orm.L2Block

+	maxBlockNumPerChunk             uint64
 	maxTxNumPerChunk                uint64
 	maxL1CommitGasPerChunk          uint64
 	maxL1CommitCalldataSizePerChunk uint64
@@ -90,6 +87,7 @@ func NewChunkProposer(ctx context.Context, cfg *config.ChunkProposerConfig, db *
 		db:                              db,
 		chunkOrm:                        orm.NewChunk(db),
 		l2BlockOrm:                      orm.NewL2Block(db),
+		maxBlockNumPerChunk:             cfg.MaxBlockNumPerChunk,
 		maxTxNumPerChunk:                cfg.MaxTxNumPerChunk,
 		maxL1CommitGasPerChunk:          cfg.MaxL1CommitGasPerChunk,
 		maxL1CommitCalldataSizePerChunk: cfg.MaxL1CommitCalldataSizePerChunk,
@@ -191,7 +189,8 @@ func (p *ChunkProposer) proposeChunk() (*types.Chunk, error) {
 		return nil, err
 	}

-	blocks, err := p.l2BlockOrm.GetL2WrappedBlocksGEHeight(p.ctx, unchunkedBlockHeight, maxNumBlockPerChunk)
+	// select at most p.maxBlockNumPerChunk blocks
+	blocks, err := p.l2BlockOrm.GetL2WrappedBlocksGEHeight(p.ctx, unchunkedBlockHeight, int(p.maxBlockNumPerChunk))
 	if err != nil {
 		return nil, err
 	}
@@ -293,12 +292,21 @@ func (p *ChunkProposer) proposeChunk() (*types.Chunk, error) {
 	}

 	currentTimeSec := uint64(time.Now().Unix())
-	if blocks[0].Header.Time+p.chunkTimeoutSec < currentTimeSec {
-		log.Warn("first block timeout",
-			"block number", blocks[0].Header.Number,
-			"block timestamp", blocks[0].Header.Time,
-			"block outdated time threshold", currentTimeSec,
-		)
+	if chunk.Blocks[0].Header.Time+p.chunkTimeoutSec < currentTimeSec ||
+		uint64(len(chunk.Blocks)) == p.maxBlockNumPerChunk {
+		if chunk.Blocks[0].Header.Time+p.chunkTimeoutSec < currentTimeSec {
+			log.Warn("first block timeout",
+				"block number", chunk.Blocks[0].Header.Number,
+				"block timestamp", chunk.Blocks[0].Header.Time,
+				"current time", currentTimeSec,
+			)
+		} else {
+			log.Info("reached maximum number of blocks in chunk",
+				"start block number", chunk.Blocks[0].Header.Number,
+				"block count", len(chunk.Blocks),
+			)
+		}
+
 		p.chunkFirstBlockTimeoutReached.Inc()
 		p.chunkTxNum.Set(float64(totalTxNum))
 		p.chunkEstimateL1CommitGas.Set(float64(totalL1CommitGas))
--- a/rollup/internal/controller/watcher/chunk_proposer_test.go
+++ b/rollup/internal/controller/watcher/chunk_proposer_test.go
@@ -23,6 +23,7 @@ func testChunkProposer(t *testing.T) {
 	assert.NoError(t, err)

 	cp := NewChunkProposer(context.Background(), &config.ChunkProposerConfig{
+		MaxBlockNumPerChunk:             100,
 		MaxTxNumPerChunk:                10000,
 		MaxL1CommitGasPerChunk:          50000000000,
 		MaxL1CommitCalldataSizePerChunk: 1000000,
@@ -53,6 +54,7 @@ func testChunkProposerRowConsumption(t *testing.T) {
 	assert.NoError(t, err)

 	cp := NewChunkProposer(context.Background(), &config.ChunkProposerConfig{
+		MaxBlockNumPerChunk:             100,
 		MaxTxNumPerChunk:                10000,
 		MaxL1CommitGasPerChunk:          50000000000,
 		MaxL1CommitCalldataSizePerChunk: 1000000,
--- a/rollup/internal/orm/l1_block.go
+++ b/rollup/internal/orm/l1_block.go
@@ -70,19 +70,41 @@ func (o *L1Block) GetL1Blocks(ctx context.Context, fields map[string]interface{}
 	return l1Blocks, nil
 }

-// InsertL1Blocks batch insert l1 blocks
+// InsertL1Blocks batch inserts l1 blocks.
+// If there's a block number conflict (e.g., due to reorg), soft deletes the existing block and inserts the new one.
 func (o *L1Block) InsertL1Blocks(ctx context.Context, blocks []L1Block) error {
 	if len(blocks) == 0 {
 		return nil
 	}

-	db := o.db.WithContext(ctx)
-	db = db.Model(&L1Block{})
+	return o.db.Transaction(func(tx *gorm.DB) error {
+		minBlockNumber := blocks[0].Number
+		for _, block := range blocks[1:] {
+			if block.Number < minBlockNumber {
+				minBlockNumber = block.Number
+			}
+		}

-	if err := db.Create(&blocks).Error; err != nil {
-		return fmt.Errorf("L1Block.InsertL1Blocks error: %w", err)
-	}
-	return nil
+		db := tx.WithContext(ctx)
+		db = db.Model(&L1Block{})
+		db = db.Where("number >= ?", minBlockNumber)
+		result := db.Delete(&L1Block{})
+
+		if result.Error != nil {
+			return fmt.Errorf("L1Block.InsertL1Blocks error: soft deleting blocks failed, block numbers starting from: %v, error: %w", minBlockNumber, result.Error)
+		}
+
+		// If the number of deleted blocks exceeds the limit (input length + 64), treat it as an anomaly.
+		// Because reorg with >= 64 blocks is very unlikely to happen.
+		if result.RowsAffected >= int64(len(blocks)+64) {
+			return fmt.Errorf("L1Block.InsertL1Blocks error: too many blocks were deleted, count: %d", result.RowsAffected)
+		}
+
+		if err := db.Create(&blocks).Error; err != nil {
+			return fmt.Errorf("L1Block.InsertL1Blocks error: %w", err)
+		}
+		return nil
+	})
 }

 // UpdateL1GasOracleStatusAndOracleTxHash update l1 gas oracle status and oracle tx hash
--- a/rollup/internal/orm/orm_test.go
+++ b/rollup/internal/orm/orm_test.go
@@ -89,6 +89,53 @@ func tearDownEnv(t *testing.T) {
 	base.Free()
 }

+func TestL1BlockOrm(t *testing.T) {
+	sqlDB, err := db.DB()
+	assert.NoError(t, err)
+	assert.NoError(t, migrate.ResetDB(sqlDB))
+
+	l1BlockOrm := NewL1Block(db)
+
+	// mock blocks
+	block1 := L1Block{Number: 1, Hash: "hash1"}
+	block2 := L1Block{Number: 2, Hash: "hash2"}
+	block3 := L1Block{Number: 3, Hash: "hash3"}
+	block2AfterReorg := L1Block{Number: 2, Hash: "hash2-reorg"}
+
+	err = l1BlockOrm.InsertL1Blocks(context.Background(), []L1Block{block1, block2, block3})
+	assert.NoError(t, err)
+
+	height, err := l1BlockOrm.GetLatestL1BlockHeight(context.Background())
+	assert.NoError(t, err)
+	assert.Equal(t, uint64(3), height)
+
+	blocks, err := l1BlockOrm.GetL1Blocks(context.Background(), map[string]interface{}{})
+	assert.NoError(t, err)
+	assert.Len(t, blocks, 3)
+	assert.Equal(t, "hash1", blocks[0].Hash)
+	assert.Equal(t, "hash2", blocks[1].Hash)
+	assert.Equal(t, "hash3", blocks[2].Hash)
+
+	// reorg handling: insert another block with same height and different hash
+	err = l1BlockOrm.InsertL1Blocks(context.Background(), []L1Block{block2AfterReorg})
+	assert.NoError(t, err)
+
+	blocks, err = l1BlockOrm.GetL1Blocks(context.Background(), map[string]interface{}{})
+	assert.NoError(t, err)
+	assert.Len(t, blocks, 2)
+	assert.Equal(t, "hash1", blocks[0].Hash)
+	assert.Equal(t, "hash2-reorg", blocks[1].Hash)
+
+	err = l1BlockOrm.UpdateL1GasOracleStatusAndOracleTxHash(context.Background(), "hash1", types.GasOracleImported, "txhash1")
+	assert.NoError(t, err)
+
+	updatedBlocks, err := l1BlockOrm.GetL1Blocks(context.Background(), map[string]interface{}{})
+	assert.NoError(t, err)
+	assert.Len(t, updatedBlocks, 2)
+	assert.Equal(t, types.GasOracleImported, types.GasOracleStatus(updatedBlocks[0].GasOracleStatus))
+	assert.Equal(t, "txhash1", updatedBlocks[0].OracleTxHash)
+}
+
 func TestL2BlockOrm(t *testing.T) {
 	sqlDB, err := db.DB()
 	assert.NoError(t, err)
--- a/rollup/tests/rollup_test.go
+++ b/rollup/tests/rollup_test.go
@@ -57,6 +57,7 @@ func testCommitBatchAndFinalizeBatch(t *testing.T) {
 	assert.NoError(t, err)

 	cp := watcher.NewChunkProposer(context.Background(), &config.ChunkProposerConfig{
+		MaxBlockNumPerChunk:             100,
 		MaxTxNumPerChunk:                10000,
 		MaxL1CommitGasPerChunk:          50000000000,
 		MaxL1CommitCalldataSizePerChunk: 1000000,
--- a/scripts/download_params.sh
+++ b/scripts/download_params.sh
@@ -1,10 +0,0 @@
-#!/bin/bash
-set -uex
-
-DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" > /dev/null 2>&1 && pwd)"
-PROJ_DIR=$DIR"/.."
-
-mkdir -p $PROJ_DIR/assets/params
-wget https://circuit-release.s3.us-west-2.amazonaws.com/circuit-release/release-1220/test_seed -O $PROJ_DIR/assets/seed
-wget https://circuit-release.s3.us-west-2.amazonaws.com/circuit-release/params-0320/params20 -O $PROJ_DIR/assets/params/params20
-wget https://circuit-release.s3.us-west-2.amazonaws.com/circuit-release/params-0320/params26 -O $PROJ_DIR/assets/params/params26
--- a/scripts/init_repo.sh
+++ b/scripts/init_repo.sh
@@ -1,5 +0,0 @@
-#!/bin/bash
-set -uex
-
-git submodule update --init --recursive
-# if [ -z ${CI+x} ]; then git pull --recurse-submodules; fi
Author	SHA1	Message	Date
HAOYUatHZ	cd456ee3db	feat(prover, coordinator): update vk handling logic (#931 ) Co-authored-by: Steven Gu <asongala@163.com> Co-authored-by: HAOYUatHZ <HAOYUatHZ@users.noreply.github.com>	2023-09-07 10:59:47 +08:00
Steven	2b6a3c9874	feat: libzkp `v0.8.1` (#894 ) Co-authored-by: silathdiir <silathdiir@users.noreply.github.com> Co-authored-by: HAOYUatHZ <haoyu@protonmail.com>	2023-09-07 10:34:35 +08:00
colin	d39db01c5b	fix(coordinator): decrease active attempts zero check (#934 ) Co-authored-by: HAOYUatHZ <HAOYUatHZ@users.noreply.github.com>	2023-09-06 18:22:25 +08:00
colin	90e3dddf91	fix: bump version (#933 )	2023-09-06 16:08:41 +08:00
georgehao	c6cc51bcfd	feat(coordinator): prover task assigned multiple prover (#871 ) Co-authored-by: georgehao <georgehao@users.noreply.github.com> Co-authored-by: Péter Garamvölgyi <peter@scroll.io> Co-authored-by: colin <102356659+colinlyguo@users.noreply.github.com> Co-authored-by: colinlyguo <colinlyguo@scroll.io> Co-authored-by: HAOYUatHZ <37070449+HAOYUatHZ@users.noreply.github.com>	2023-09-06 15:28:31 +08:00
Xi Lin	0b0b84a513	fix(contracts): OZ-N01 Unused Imports (#930 ) Co-authored-by: HAOYUatHZ <37070449+HAOYUatHZ@users.noreply.github.com>	2023-09-06 13:34:29 +08:00
gsovereignty	83c8071123	docs: fix link in readme (#926 ) Co-authored-by: HAOYUatHZ <37070449+HAOYUatHZ@users.noreply.github.com>	2023-09-06 12:00:07 +08:00
Xi Lin	e35de74495	docs(contracts): OZ-L01 Misleading Comment (#928 )	2023-09-06 11:59:35 +08:00
Ahmed Castro	33089b829f	docs: add code-of-conduct and contributing documents (#722 ) Co-authored-by: Péter Garamvölgyi <peter@scroll.io> Co-authored-by: Haichen Shen <shenhaichen@gmail.com>	2023-09-05 08:16:31 -07:00
colin	8e27052b36	fix(l1-watcher): soft delete blocks when reorg (#923 ) Co-authored-by: colinlyguo <colinlyguo@users.noreply.github.com>	2023-09-05 16:32:48 +08:00
colin	25b956f9b5	fix(gas-oracle): fetch base fee from the latest L1 block (#920 ) Co-authored-by: Péter Garamvölgyi <peter@scroll.io> Co-authored-by: colinlyguo <colinlyguo@users.noreply.github.com>	2023-09-05 15:21:06 +08:00
colin	f4663fd249	feat(rollup-relayer): add number of blocks per chunk limit (#880 ) Co-authored-by: colinlyguo <colinlyguo@users.noreply.github.com> Co-authored-by: Péter Garamvölgyi <peter@scroll.io> Co-authored-by: HAOYUatHZ <37070449+HAOYUatHZ@users.noreply.github.com> Co-authored-by: HAOYUatHZ <HAOYUatHZ@users.noreply.github.com>	2023-09-05 15:00:16 +08:00
Péter Garamvölgyi	c71fa5a5fc	style: use MAX_TX_IN_CHUNK env name in deployment scripts (#921 )	2023-09-04 14:35:04 +02:00
Péter Garamvölgyi	4af3834e36	test(fee-vault): add new test testCantWithdrawMoreThanBalance (#918 ) Co-authored-by: HAOYUatHZ <37070449+HAOYUatHZ@users.noreply.github.com>	2023-09-04 08:15:26 +02:00
Haichen Shen	9983585bdd	docs: Update readme (#919 )	2023-09-04 13:56:43 +08:00
Xi Lin	d288b34536	fix(contracts): OZ-L02 Anyone Can Steal ERC-20 Tokens From GasSwap (#844 ) Co-authored-by: Haichen Shen <shenhaichen@gmail.com> Co-authored-by: icemelon <icemelon@users.noreply.github.com> Co-authored-by: zimpha <zimpha@users.noreply.github.com>	2023-09-04 12:31:41 +08:00
Haichen Shen	a2fe246551	docs(rollup): improve readme (#917 ) Co-authored-by: HAOYUatHZ <haoyu@protonmail.com>	2023-09-02 23:23:17 +08:00
Péter Garamvölgyi	8699a22fa3	feat(contracts): allow setting withdraw amount in fee vault (#912 ) Co-authored-by: Haichen Shen <shenhaichen@gmail.com>	2023-09-02 02:23:53 -07:00
Xi Lin	d668180e9a	fix(contracts): OZ-M04 Use of Non-Production-Ready Trusted Forwarder (#843 ) Co-authored-by: Haichen Shen <shenhaichen@gmail.com> Co-authored-by: icemelon <icemelon@users.noreply.github.com>	2023-09-02 02:23:31 -07:00
HAOYUatHZ	d3c2e34650	docs(rollup): fix README (#916 )	2023-09-02 15:54:26 +08:00