remove debug (#912)

* remove debug

* Refactor get_global_encryption_service function

* Refactor get_global_encryption_service function

---------

Co-authored-by: loverustfs <hello@rustfs.com>
Co-authored-by: houseme <housemecn@gmail.com>

.docker/observability/docker-compose.yml

@@ -16,7 +16,7 @@ services:
 
   tempo-init:
     image: busybox:latest
-    command: ["sh", "-c", "chown -R 10001:10001 /var/tempo"]
+    command: [ "sh", "-c", "chown -R 10001:10001 /var/tempo" ]
     volumes:
       - ./tempo-data:/var/tempo
     user: root
@@ -39,7 +39,7 @@ services:
       - otel-network
 
   otel-collector:
-    image: otel/opentelemetry-collector-contrib:0.129.1
+    image: otel/opentelemetry-collector-contrib:latest
     environment:
       - TZ=Asia/Shanghai
     volumes:
@@ -55,7 +55,7 @@ services:
     networks:
       - otel-network
   jaeger:
-    image: jaegertracing/jaeger:2.8.0
+    image: jaegertracing/jaeger:latest
     environment:
       - TZ=Asia/Shanghai
     ports:
@@ -65,17 +65,21 @@ services:
     networks:
       - otel-network
   prometheus:
-    image: prom/prometheus:v3.4.2
+    image: prom/prometheus:latest
     environment:
       - TZ=Asia/Shanghai
     volumes:
       - ./prometheus.yml:/etc/prometheus/prometheus.yml
     ports:
       - "9090:9090"
+    command:
+      - '--config.file=/etc/prometheus/prometheus.yml'
+      - '--web.enable-otlp-receiver' # Enable OTLP
+      - '--enable-feature=promql-experimental-functions' # Enable info()
     networks:
       - otel-network
   loki:
-    image: grafana/loki:3.5.1
+    image: grafana/loki:latest
     environment:
       - TZ=Asia/Shanghai
     volumes:
@@ -86,7 +90,7 @@ services:
     networks:
       - otel-network
   grafana:
-    image: grafana/grafana:12.0.2
+    image: grafana/grafana:latest
     ports:
       - "3000:3000"  # Web UI
     volumes:

.docker/observability/grafana-datasources.yaml

@@ -29,4 +29,80 @@ datasources:
       serviceMap:
         datasourceUid: prometheus
       streamingEnabled:
-        search: true
+        search: true
+      tracesToLogsV2:
+        # Field with an internal link pointing to a logs data source in Grafana.
+        # datasourceUid value must match the uid value of the logs data source.
+        datasourceUid: 'loki'
+        spanStartTimeShift: '-1h'
+        spanEndTimeShift: '1h'
+        tags: [ 'job', 'instance', 'pod', 'namespace' ]
+        filterByTraceID: false
+        filterBySpanID: false
+        customQuery: true
+        query: 'method="$${__span.tags.method}"'
+        tracesToMetrics:
+          datasourceUid: 'prom'
+          spanStartTimeShift: '-1h'
+          spanEndTimeShift: '1h'
+          tags: [ { key: 'service.name', value: 'service' }, { key: 'job' } ]
+          queries:
+            - name: 'Sample query'
+              query: 'sum(rate(traces_spanmetrics_latency_bucket{$$__tags}[5m]))'
+        tracesToProfiles:
+          datasourceUid: 'grafana-pyroscope-datasource'
+          tags: [ 'job', 'instance', 'pod', 'namespace' ]
+          profileTypeId: 'process_cpu:cpu:nanoseconds:cpu:nanoseconds'
+          customQuery: true
+          query: 'method="$${__span.tags.method}"'
+        serviceMap:
+          datasourceUid: 'prometheus'
+        nodeGraph:
+          enabled: true
+        search:
+          hide: false
+        traceQuery:
+          timeShiftEnabled: true
+          spanStartTimeShift: '-1h'
+          spanEndTimeShift: '1h'
+        spanBar:
+          type: 'Tag'
+          tag: 'http.path'
+        streamingEnabled:
+          search: true
+  - name: Jaeger
+    type: jaeger
+    uid: Jaeger
+    url: http://jaeger:16686
+    basicAuth: false
+    access: proxy
+    readOnly: false
+    isDefault: false
+    jsonData:
+      tracesToLogsV2:
+        # Field with an internal link pointing to a logs data source in Grafana.
+        # datasourceUid value must match the uid value of the logs data source.
+        datasourceUid: 'loki'
+        spanStartTimeShift: '1h'
+        spanEndTimeShift: '-1h'
+        tags: [ 'job', 'instance', 'pod', 'namespace' ]
+        filterByTraceID: false
+        filterBySpanID: false
+        customQuery: true
+        query: 'method="$${__span.tags.method}"'
+      tracesToMetrics:
+        datasourceUid: 'prom'
+        spanStartTimeShift: '1h'
+        spanEndTimeShift: '-1h'
+        tags: [ { key: 'service.name', value: 'service' }, { key: 'job' } ]
+        queries:
+          - name: 'Sample query'
+            query: 'sum(rate(traces_spanmetrics_latency_bucket{$$__tags}[5m]))'
+      nodeGraph:
+        enabled: true
+      traceQuery:
+        timeShiftEnabled: true
+        spanStartTimeShift: '1h'
+        spanEndTimeShift: '-1h'
+      spanBar:
+        type: 'None'

.docker/observability/loki-config.yaml

@@ -63,6 +63,7 @@ ruler:
 frontend:
   encoding: protobuf
 
+
 # By default, Loki will send anonymous, but uniquely-identifiable usage and configuration
 # analytics to Grafana Labs. These statistics are sent to https://stats.grafana.org/
 #

.docker/observability/otel-collector-config.yaml

@@ -43,7 +43,6 @@ exporters:
     send_timestamps: true  # 发送时间戳
     # enable_open_metrics: true
   otlphttp/loki: # Loki 导出器，用于日志数据
-    # endpoint: "http://loki:3100/otlp/v1/logs"
     endpoint: "http://loki:3100/otlp/v1/logs"
     tls:
       insecure: true

.docker/observability/prometheus.yml

@@ -13,16 +13,43 @@
 # limitations under the License.
 
 global:
-  scrape_interval: 5s  # 刮取间隔
+  scrape_interval: 15s # Evaluate rules every 15 seconds. The default is every 1 minute.
 
 scrape_configs:
   - job_name: 'otel-collector'
     static_configs:
-      - targets: [ 'otel-collector:8888' ]  # 从 Collector 刮取指标
+      - targets: [ 'otel-collector:8888' ]  # Scrape metrics from Collector
   - job_name: 'otel-metrics'
     static_configs:
-      - targets: [ 'otel-collector:8889' ]  # 应用指标
+      - targets: [ 'otel-collector:8889' ]  # Application indicators
   - job_name: 'tempo'
     static_configs:
-      - targets: [ 'tempo:3200' ]
-      
+      - targets: [ 'tempo:3200' ]  # Scrape metrics from Tempo
+
+otlp:
+  # Recommended attributes to be promoted to labels.
+  promote_resource_attributes:
+    - service.instance.id
+    - service.name
+    - service.namespace
+    - cloud.availability_zone
+    - cloud.region
+    - container.name
+    - deployment.environment.name
+    - k8s.cluster.name
+    - k8s.container.name
+    - k8s.cronjob.name
+    - k8s.daemonset.name
+    - k8s.deployment.name
+    - k8s.job.name
+    - k8s.namespace.name
+    - k8s.pod.name
+    - k8s.replicaset.name
+    - k8s.statefulset.name
+  # Ingest OTLP data keeping all characters in metric/label names.
+  translation_strategy: NoUTF8EscapingWithSuffixes
+
+storage:
+  # OTLP is a push-based protocol, Out of order samples is a common scenario.
+  tsdb:
+    out_of_order_time_window: 30m

.github/actions/setup/action.yml

@@ -52,24 +52,19 @@ runs:
         sudo apt-get install -y \
           musl-tools \
           build-essential \
-          lld \
-          libdbus-1-dev \
-          libwayland-dev \
-          libwebkit2gtk-4.1-dev \
-          libxdo-dev \
           pkg-config \
           libssl-dev
 
     - name: Install protoc
       uses: arduino/setup-protoc@v3
       with:
-        version: "31.1"
+        version: "33.1"
         repo-token: ${{ inputs.github-token }}
 
     - name: Install flatc
       uses: Nugine/setup-flatc@v1
       with:
-        version: "25.2.10"
+        version: "25.9.23"
 
     - name: Install Rust toolchain
       uses: dtolnay/rust-toolchain@stable

.github/dependabot.yml

@@ -22,8 +22,18 @@ updates:
   - package-ecosystem: "cargo" # See documentation for possible values
     directory: "/" # Location of package manifests
     schedule:
-      interval: "monthly"
+      interval: "weekly"
+      day: "monday"
+      timezone: "Asia/Shanghai"
+      time: "08:00"
     groups:
+      s3s:
+        update-types:
+          - "minor"
+          - "patch"
+        patterns:
+          - "s3s"
+          - "s3s-*"
       dependencies:
         patterns:
           - "*"

.github/workflows/ci.yml

@@ -103,6 +103,8 @@ jobs:
     runs-on: ubuntu-latest
     timeout-minutes: 60
     steps:
+      - name: Delete huge unnecessary tools folder
+        run: rm -rf /opt/hostedtoolcache
       - name: Checkout repository
         uses: actions/checkout@v5
 

.gitignore

@@ -22,4 +22,5 @@ profile.json
 .secrets
 *.go
 *.pb
-*.svg
+*.svg
+deploy/logs/*.log.*

.rules.md

@@ -0,0 +1,702 @@
+# RustFS Project AI Coding Rules
+
+## 🚨🚨🚨 CRITICAL DEVELOPMENT RULES - ZERO TOLERANCE 🚨🚨🚨
+
+### ⛔️ ABSOLUTE PROHIBITION: NEVER COMMIT DIRECTLY TO MASTER/MAIN BRANCH ⛔️
+
+**🔥 THIS IS THE MOST CRITICAL RULE - VIOLATION WILL RESULT IN IMMEDIATE REVERSAL 🔥**
+
+- **🚫 ZERO DIRECT COMMITS TO MAIN/MASTER BRANCH - ABSOLUTELY FORBIDDEN**
+- **🚫 ANY DIRECT COMMIT TO MAIN BRANCH MUST BE IMMEDIATELY REVERTED**
+- **🚫 NO EXCEPTIONS FOR HOTFIXES, EMERGENCIES, OR URGENT CHANGES**
+- **🚫 NO EXCEPTIONS FOR SMALL CHANGES, TYPOS, OR DOCUMENTATION UPDATES**
+- **🚫 NO EXCEPTIONS FOR ANYONE - MAINTAINERS, CONTRIBUTORS, OR ADMINS**
+
+### 📋 MANDATORY WORKFLOW - STRICTLY ENFORCED
+
+**EVERY SINGLE CHANGE MUST FOLLOW THIS WORKFLOW:**
+
+1. **Check current branch**: `git branch` (MUST NOT be on main/master)
+2. **Switch to main**: `git checkout main`
+3. **Pull latest**: `git pull origin main`
+4. **Create feature branch**: `git checkout -b feat/your-feature-name`
+5. **Make changes ONLY on feature branch**
+6. **Test thoroughly before committing**
+7. **Commit and push to feature branch**: `git push origin feat/your-feature-name`
+8. **Create Pull Request**: Use `gh pr create` (MANDATORY)
+9. **Wait for PR approval**: NO self-merging allowed
+10. **Merge through GitHub interface**: ONLY after approval
+
+### 🔒 ENFORCEMENT MECHANISMS
+
+- **Branch protection rules**: Main branch is protected
+- **Pre-commit hooks**: Will block direct commits to main
+- **CI/CD checks**: All PRs must pass before merging
+- **Code review requirement**: At least one approval needed
+- **Automated reversal**: Direct commits to main will be automatically reverted
+
+## 🎯 Core AI Development Principles
+
+### Five Execution Steps
+
+#### 1. Task Analysis and Planning
+- **Clear Objectives**: Deeply understand task requirements and expected results before starting coding
+- **Plan Development**: List specific files, components, and functions that need modification, explaining the reasons for changes
+- **Risk Assessment**: Evaluate the impact of changes on existing functionality, develop rollback plans
+
+#### 2. Precise Code Location
+- **File Identification**: Determine specific files and line numbers that need modification
+- **Impact Analysis**: Avoid modifying irrelevant files, clearly state the reason for each file modification
+- **Minimization Principle**: Unless explicitly required by the task, do not create new abstraction layers or refactor existing code
+
+#### 3. Minimal Code Changes
+- **Focus on Core**: Only write code directly required by the task
+- **Avoid Redundancy**: Do not add unnecessary logs, comments, tests, or error handling
+- **Isolation**: Ensure new code does not interfere with existing functionality, maintain code independence
+
+#### 4. Strict Code Review
+- **Correctness Check**: Verify the correctness and completeness of code logic
+- **Style Consistency**: Ensure code conforms to established project coding style
+- **Side Effect Assessment**: Evaluate the impact of changes on downstream systems
+
+#### 5. Clear Delivery Documentation
+- **Change Summary**: Detailed explanation of all modifications and reasons
+- **File List**: List all modified files and their specific changes
+- **Risk Statement**: Mark any assumptions or potential risk points
+
+### Core Principles
+- **🎯 Precise Execution**: Strictly follow task requirements, no arbitrary innovation
+- **⚡ Efficient Development**: Avoid over-design, only do necessary work
+- **🛡️ Safe and Reliable**: Always follow development processes, ensure code quality and system stability
+- **🔒 Cautious Modification**: Only modify when clearly knowing what needs to be changed and having confidence
+
+### Additional AI Behavior Rules
+
+1. **Use English for all code comments and documentation** - All comments, variable names, function names, documentation, and user-facing text in code should be in English
+2. **Clean up temporary scripts after use** - Any temporary scripts, test files, or helper files created during AI work should be removed after task completion
+3. **Only make confident modifications** - Do not make speculative changes or "convenient" modifications outside the task scope. If uncertain about a change, ask for clarification rather than guessing
+
+## Project Overview
+
+RustFS is a high-performance distributed object storage system written in Rust, compatible with S3 API. The project adopts a modular architecture, supporting erasure coding storage, multi-tenant management, observability, and other enterprise-level features.
+
+## Core Architecture Principles
+
+### 1. Modular Design
+
+- Project uses Cargo workspace structure, containing multiple independent crates
+- Core modules: `rustfs` (main service), `ecstore` (erasure coding storage), `common` (shared components)
+- Functional modules: `iam` (identity management), `madmin` (management interface), `crypto` (encryption), etc.
+- Tool modules: `cli` (command line tool), `crates/*` (utility libraries)
+
+### 2. Asynchronous Programming Pattern
+
+- Comprehensive use of `tokio` async runtime
+- Prioritize `async/await` syntax
+- Use `async-trait` for async methods in traits
+- Avoid blocking operations, use `spawn_blocking` when necessary
+
+### 3. Error Handling Strategy
+
+- **Use modular, type-safe error handling with `thiserror`**
+- Each module should define its own error type using `thiserror::Error` derive macro
+- Support error chains and context information through `#[from]` and `#[source]` attributes
+- Use `Result<T>` type aliases for consistency within each module
+- Error conversion between modules should use explicit `From` implementations
+- Follow the pattern: `pub type Result<T> = core::result::Result<T, Error>`
+- Use `#[error("description")]` attributes for clear error messages
+- Support error downcasting when needed through `other()` helper methods
+- Implement `Clone` for errors when required by the domain logic
+
+## Code Style Guidelines
+
+### 1. Formatting Configuration
+
+```toml
+max_width = 130
+fn_call_width = 90
+single_line_let_else_max_width = 100
+```
+
+### 2. **🔧 MANDATORY Code Formatting Rules**
+
+**CRITICAL**: All code must be properly formatted before committing. This project enforces strict formatting standards to maintain code consistency and readability.
+
+#### Pre-commit Requirements (MANDATORY)
+
+Before every commit, you **MUST**:
+
+1. **Format your code**:
+   ```bash
+   cargo fmt --all
+   ```
+
+2. **Verify formatting**:
+   ```bash
+   cargo fmt --all --check
+   ```
+
+3. **Pass clippy checks**:
+   ```bash
+   cargo clippy --all-targets --all-features -- -D warnings
+   ```
+
+4. **Ensure compilation**:
+   ```bash
+   cargo check --all-targets
+   ```
+
+#### Quick Commands
+
+Use these convenient Makefile targets for common tasks:
+
+```bash
+# Format all code
+make fmt
+
+# Check if code is properly formatted
+make fmt-check
+
+# Run clippy checks
+make clippy
+
+# Run compilation check
+make check
+
+# Run tests
+make test
+
+# Run all pre-commit checks (format + clippy + check + test)
+make pre-commit
+
+# Setup git hooks (one-time setup)
+make setup-hooks
+```
+
+### 3. Naming Conventions
+
+- Use `snake_case` for functions, variables, modules
+- Use `PascalCase` for types, traits, enums
+- Constants use `SCREAMING_SNAKE_CASE`
+- Global variables prefix `GLOBAL_`, e.g., `GLOBAL_Endpoints`
+- Use meaningful and descriptive names for variables, functions, and methods
+- Avoid meaningless names like `temp`, `data`, `foo`, `bar`, `test123`
+- Choose names that clearly express the purpose and intent
+
+### 4. Type Declaration Guidelines
+
+- **Prefer type inference over explicit type declarations** when the type is obvious from context
+- Let the Rust compiler infer types whenever possible to reduce verbosity and improve maintainability
+- Only specify types explicitly when:
+  - The type cannot be inferred by the compiler
+  - Explicit typing improves code clarity and readability
+  - Required for API boundaries (function signatures, public struct fields)
+  - Needed to resolve ambiguity between multiple possible types
+
+### 5. Documentation Comments
+
+- Public APIs must have documentation comments
+- Use `///` for documentation comments
+- Complex functions add `# Examples` and `# Parameters` descriptions
+- Error cases use `# Errors` descriptions
+- Always use English for all comments and documentation
+- Avoid meaningless comments like "debug 111" or placeholder text
+
+### 6. Import Guidelines
+
+- Standard library imports first
+- Third-party crate imports in the middle
+- Project internal imports last
+- Group `use` statements with blank lines between groups
+
+## Asynchronous Programming Guidelines
+
+### 1. Trait Definition
+
+```rust
+#[async_trait::async_trait]
+pub trait StorageAPI: Send + Sync {
+    async fn get_object(&self, bucket: &str, object: &str) -> Result<ObjectInfo>;
+}
+```
+
+### 2. Error Handling
+
+```rust
+// Use ? operator to propagate errors
+async fn example_function() -> Result<()> {
+    let data = read_file("path").await?;
+    process_data(data).await?;
+    Ok(())
+}
+```
+
+### 3. Concurrency Control
+
+- Use `Arc` and `Mutex`/`RwLock` for shared state management
+- Prioritize async locks from `tokio::sync`
+- Avoid holding locks for long periods
+
+## Logging and Tracing Guidelines
+
+### 1. Tracing Usage
+
+```rust
+#[tracing::instrument(skip(self, data))]
+async fn process_data(&self, data: &[u8]) -> Result<()> {
+    info!("Processing {} bytes", data.len());
+    // Implementation logic
+}
+```
+
+### 2. Log Levels
+
+- `error!`: System errors requiring immediate attention
+- `warn!`: Warning information that may affect functionality
+- `info!`: Important business information
+- `debug!`: Debug information for development use
+- `trace!`: Detailed execution paths
+
+### 3. Structured Logging
+
+```rust
+info!(
+    counter.rustfs_api_requests_total = 1_u64,
+    key_request_method = %request.method(),
+    key_request_uri_path = %request.uri().path(),
+    "API request processed"
+);
+```
+
+## Error Handling Guidelines
+
+### 1. Error Type Definition
+
+```rust
+// Use thiserror for module-specific error types
+#[derive(thiserror::Error, Debug)]
+pub enum MyError {
+    #[error("IO error: {0}")]
+    Io(#[from] std::io::Error),
+
+    #[error("Storage error: {0}")]
+    Storage(#[from] ecstore::error::StorageError),
+
+    #[error("Custom error: {message}")]
+    Custom { message: String },
+
+    #[error("File not found: {path}")]
+    FileNotFound { path: String },
+
+    #[error("Invalid configuration: {0}")]
+    InvalidConfig(String),
+}
+
+// Provide Result type alias for the module
+pub type Result<T> = core::result::Result<T, MyError>;
+```
+
+### 2. Error Helper Methods
+
+```rust
+impl MyError {
+    /// Create error from any compatible error type
+    pub fn other<E>(error: E) -> Self
+    where
+        E: Into<Box<dyn std::error::Error + Send + Sync>>,
+    {
+        MyError::Io(std::io::Error::other(error))
+    }
+}
+```
+
+### 3. Error Context and Propagation
+
+```rust
+// Use ? operator for clean error propagation
+async fn example_function() -> Result<()> {
+    let data = read_file("path").await?;
+    process_data(data).await?;
+    Ok(())
+}
+
+// Add context to errors
+fn process_with_context(path: &str) -> Result<()> {
+    std::fs::read(path)
+        .map_err(|e| MyError::Custom {
+            message: format!("Failed to read {}: {}", path, e)
+        })?;
+    Ok(())
+}
+```
+
+## Performance Optimization Guidelines
+
+### 1. Memory Management
+
+- Use `Bytes` instead of `Vec<u8>` for zero-copy operations
+- Avoid unnecessary cloning, use reference passing
+- Use `Arc` for sharing large objects
+
+### 2. Concurrency Optimization
+
+```rust
+// Use join_all for concurrent operations
+let futures = disks.iter().map(|disk| disk.operation());
+let results = join_all(futures).await;
+```
+
+### 3. Caching Strategy
+
+- Use `LazyLock` for global caching
+- Implement LRU cache to avoid memory leaks
+
+## Testing Guidelines
+
+### 1. Unit Tests
+
+```rust
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use test_case::test_case;
+
+    #[tokio::test]
+    async fn test_async_function() {
+        let result = async_function().await;
+        assert!(result.is_ok());
+    }
+
+    #[test_case("input1", "expected1")]
+    #[test_case("input2", "expected2")]
+    fn test_with_cases(input: &str, expected: &str) {
+        assert_eq!(function(input), expected);
+    }
+}
+```
+
+### 2. Integration Tests
+
+- Use `e2e_test` module for end-to-end testing
+- Simulate real storage environments
+
+### 3. Test Quality Standards
+
+- Write meaningful test cases that verify actual functionality
+- Avoid placeholder or debug content like "debug 111", "test test", etc.
+- Use descriptive test names that clearly indicate what is being tested
+- Each test should have a clear purpose and verify specific behavior
+- Test data should be realistic and representative of actual use cases
+
+## Cross-Platform Compatibility Guidelines
+
+### 1. CPU Architecture Compatibility
+
+- **Always consider multi-platform and different CPU architecture compatibility** when writing code
+- Support major architectures: x86_64, aarch64 (ARM64), and other target platforms
+- Use conditional compilation for architecture-specific code:
+
+```rust
+#[cfg(target_arch = "x86_64")]
+fn optimized_x86_64_function() { /* x86_64 specific implementation */ }
+
+#[cfg(target_arch = "aarch64")]
+fn optimized_aarch64_function() { /* ARM64 specific implementation */ }
+
+#[cfg(not(any(target_arch = "x86_64", target_arch = "aarch64")))]
+fn generic_function() { /* Generic fallback implementation */ }
+```
+
+### 2. Platform-Specific Dependencies
+
+- Use feature flags for platform-specific dependencies
+- Provide fallback implementations for unsupported platforms
+- Test on multiple architectures in CI/CD pipeline
+
+### 3. Endianness Considerations
+
+- Use explicit byte order conversion when dealing with binary data
+- Prefer `to_le_bytes()`, `from_le_bytes()` for consistent little-endian format
+- Use `byteorder` crate for complex binary format handling
+
+### 4. SIMD and Performance Optimizations
+
+- Use portable SIMD libraries like `wide` or `packed_simd`
+- Provide fallback implementations for non-SIMD architectures
+- Use runtime feature detection when appropriate
+
+## Security Guidelines
+
+### 1. Memory Safety
+
+- Disable `unsafe` code (workspace.lints.rust.unsafe_code = "deny")
+- Use `rustls` instead of `openssl`
+
+### 2. Authentication and Authorization
+
+```rust
+// Use IAM system for permission checks
+let identity = iam.authenticate(&access_key, &secret_key).await?;
+iam.authorize(&identity, &action, &resource).await?;
+```
+
+## Configuration Management Guidelines
+
+### 1. Environment Variables
+
+- Use `RUSTFS_` prefix
+- Support both configuration files and environment variables
+- Provide reasonable default values
+
+### 2. Configuration Structure
+
+```rust
+#[derive(Debug, Deserialize, Clone)]
+pub struct Config {
+    pub address: String,
+    pub volumes: String,
+    #[serde(default)]
+    pub console_enable: bool,
+}
+```
+
+## Dependency Management Guidelines
+
+### 1. Workspace Dependencies
+
+- Manage versions uniformly at workspace level
+- Use `workspace = true` to inherit configuration
+
+### 2. Feature Flags
+
+```rust
+[features]
+default = ["file"]
+gpu = ["dep:nvml-wrapper"]
+kafka = ["dep:rdkafka"]
+```
+
+## Deployment and Operations Guidelines
+
+### 1. Containerization
+
+- Provide Dockerfile and docker-compose configuration
+- Support multi-stage builds to optimize image size
+
+### 2. Observability
+
+- Integrate OpenTelemetry for distributed tracing
+- Support Prometheus metrics collection
+- Provide Grafana dashboards
+
+### 3. Health Checks
+
+```rust
+// Implement health check endpoint
+async fn health_check() -> Result<HealthStatus> {
+    // Check component status
+}
+```
+
+## Code Review Checklist
+
+### 1. **Code Formatting and Quality (MANDATORY)**
+
+- [ ] **Code is properly formatted** (`cargo fmt --all --check` passes)
+- [ ] **All clippy warnings are resolved** (`cargo clippy --all-targets --all-features -- -D warnings` passes)
+- [ ] **Code compiles successfully** (`cargo check --all-targets` passes)
+- [ ] **Pre-commit hooks are working** and all checks pass
+- [ ] **No formatting-related changes** mixed with functional changes (separate commits)
+
+### 2. Functionality
+
+- [ ] Are all error cases properly handled?
+- [ ] Is there appropriate logging?
+- [ ] Is there necessary test coverage?
+
+### 3. Performance
+
+- [ ] Are unnecessary memory allocations avoided?
+- [ ] Are async operations used correctly?
+- [ ] Are there potential deadlock risks?
+
+### 4. Security
+
+- [ ] Are input parameters properly validated?
+- [ ] Are there appropriate permission checks?
+- [ ] Is information leakage avoided?
+
+### 5. Cross-Platform Compatibility
+
+- [ ] Does the code work on different CPU architectures (x86_64, aarch64)?
+- [ ] Are platform-specific features properly gated with conditional compilation?
+- [ ] Is byte order handling correct for binary data?
+- [ ] Are there appropriate fallback implementations for unsupported platforms?
+
+### 6. Code Commits and Documentation
+
+- [ ] Does it comply with [Conventional Commits](https://www.conventionalcommits.org/en/v1.0.0/)?
+- [ ] Are commit messages concise and under 72 characters for the title line?
+- [ ] Commit titles should be concise and in English, avoid Chinese
+- [ ] Is PR description provided in copyable markdown format for easy copying?
+
+## Common Patterns and Best Practices
+
+### 1. Resource Management
+
+```rust
+// Use RAII pattern for resource management
+pub struct ResourceGuard {
+    resource: Resource,
+}
+
+impl Drop for ResourceGuard {
+    fn drop(&mut self) {
+        // Clean up resources
+    }
+}
+```
+
+### 2. Dependency Injection
+
+```rust
+// Use dependency injection pattern
+pub struct Service {
+    config: Arc<Config>,
+    storage: Arc<dyn StorageAPI>,
+}
+```
+
+### 3. Graceful Shutdown
+
+```rust
+// Implement graceful shutdown
+async fn shutdown_gracefully(shutdown_rx: &mut Receiver<()>) {
+    tokio::select! {
+        _ = shutdown_rx.recv() => {
+            info!("Received shutdown signal");
+            // Perform cleanup operations
+        }
+        _ = tokio::time::sleep(SHUTDOWN_TIMEOUT) => {
+            warn!("Shutdown timeout reached");
+        }
+    }
+}
+```
+
+## Domain-Specific Guidelines
+
+### 1. Storage Operations
+
+- All storage operations must support erasure coding
+- Implement read/write quorum mechanisms
+- Support data integrity verification
+
+### 2. Network Communication
+
+- Use gRPC for internal service communication
+- HTTP/HTTPS support for S3-compatible API
+- Implement connection pooling and retry mechanisms
+
+### 3. Metadata Management
+
+- Use FlatBuffers for serialization
+- Support version control and migration
+- Implement metadata caching
+
+## Branch Management and Development Workflow
+
+### Branch Management
+
+- **🚨 CRITICAL: NEVER modify code directly on main or master branch - THIS IS ABSOLUTELY FORBIDDEN 🚨**
+- **⚠️ ANY DIRECT COMMITS TO MASTER/MAIN WILL BE REJECTED AND MUST BE REVERTED IMMEDIATELY ⚠️**
+- **🔒 ALL CHANGES MUST GO THROUGH PULL REQUESTS - NO DIRECT COMMITS TO MAIN UNDER ANY CIRCUMSTANCES 🔒**
+- **Always work on feature branches - NO EXCEPTIONS**
+- Always check the .rules.md file before starting to ensure you understand the project guidelines
+- **MANDATORY workflow for ALL changes:**
+   1. `git checkout main` (switch to main branch)
+   2. `git pull` (get latest changes)
+   3. `git checkout -b feat/your-feature-name` (create and switch to feature branch)
+   4. Make your changes ONLY on the feature branch
+   5. Test thoroughly before committing
+   6. Commit and push to the feature branch
+   7. **Create a pull request for code review - THIS IS THE ONLY WAY TO MERGE TO MAIN**
+   8. **Wait for PR approval before merging - NEVER merge your own PRs without review**
+- Use descriptive branch names following the pattern: `feat/feature-name`, `fix/issue-name`, `refactor/component-name`, etc.
+- **Double-check current branch before ANY commit: `git branch` to ensure you're NOT on main/master**
+- **Pull Request Requirements:**
+  - All changes must be submitted via PR regardless of size or urgency
+  - PRs must include comprehensive description and testing information
+  - PRs must pass all CI/CD checks before merging
+  - PRs require at least one approval from code reviewers
+  - Even hotfixes and emergency changes must go through PR process
+- **Enforcement:**
+  - Main branch should be protected with branch protection rules
+  - Direct pushes to main should be blocked by repository settings
+  - Any accidental direct commits to main must be immediately reverted via PR
+
+### Development Workflow
+
+## 🎯 **Core Development Principles**
+
+- **🔴 Every change must be precise - don't modify unless you're confident**
+  - Carefully analyze code logic and ensure complete understanding before making changes
+  - When uncertain, prefer asking users or consulting documentation over blind modifications
+  - Use small iterative steps, modify only necessary parts at a time
+  - Evaluate impact scope before changes to ensure no new issues are introduced
+
+- **🚀 GitHub PR creation prioritizes gh command usage**
+  - Prefer using `gh pr create` command to create Pull Requests
+  - Avoid having users manually create PRs through web interface
+  - Provide clear and professional PR titles and descriptions
+  - Using `gh` commands ensures better integration and automation
+
+## 📝 **Code Quality Requirements**
+
+- Use English for all code comments, documentation, and variable names
+- Write meaningful and descriptive names for variables, functions, and methods
+- Avoid meaningless test content like "debug 111" or placeholder values
+- Before each change, carefully read the existing code to ensure you understand the code structure and implementation, do not break existing logic implementation, do not introduce new issues
+- Ensure each change provides sufficient test cases to guarantee code correctness
+- Do not arbitrarily modify numbers and constants in test cases, carefully analyze their meaning to ensure test case correctness
+- When writing or modifying tests, check existing test cases to ensure they have scientific naming and rigorous logic testing, if not compliant, modify test cases to ensure scientific and rigorous testing
+- **Before committing any changes, run `cargo clippy --all-targets --all-features -- -D warnings` to ensure all code passes Clippy checks**
+- After each development completion, first git add . then git commit -m "feat: feature description" or "fix: issue description", ensure compliance with [Conventional Commits](https://www.conventionalcommits.org/en/v1.0.0/)
+- **Keep commit messages concise and under 72 characters** for the title line, use body for detailed explanations if needed
+- After each development completion, first git push to remote repository
+- After each change completion, summarize the changes, do not create summary files, provide a brief change description, ensure compliance with [Conventional Commits](https://www.conventionalcommits.org/en/v1.0.0/)
+- Provide change descriptions needed for PR in the conversation, ensure compliance with [Conventional Commits](https://www.conventionalcommits.org/en/v1.0.0/)
+- **Always provide PR descriptions in English** after completing any changes, including:
+  - Clear and concise title following Conventional Commits format
+  - Detailed description of what was changed and why
+  - List of key changes and improvements
+  - Any breaking changes or migration notes if applicable
+  - Testing information and verification steps
+- **Provide PR descriptions in copyable markdown format** enclosed in code blocks for easy one-click copying
+
+## 🚫 AI Documentation Generation Restrictions
+
+### Forbidden Summary Documents
+
+- **Strictly forbidden to create any form of AI-generated summary documents**
+- **Do not create documents containing large amounts of emoji, detailed formatting tables and typical AI style**
+- **Do not generate the following types of documents in the project:**
+  - Benchmark summary documents (BENCHMARK*.md)
+  - Implementation comparison analysis documents (IMPLEMENTATION_COMPARISON*.md)
+  - Performance analysis report documents
+  - Architecture summary documents
+  - Feature comparison documents
+  - Any documents with large amounts of emoji and formatted content
+- **If documentation is needed, only create when explicitly requested by the user, and maintain a concise and practical style**
+- **Documentation should focus on actually needed information, avoiding excessive formatting and decorative content**
+- **Any discovered AI-generated summary documents should be immediately deleted**
+
+### Allowed Documentation Types
+
+- README.md (project introduction, keep concise)
+- Technical documentation (only create when explicitly needed)
+- User manual (only create when explicitly needed)
+- API documentation (generated from code)
+- Changelog (CHANGELOG.md)
+
+These rules should serve as guiding principles when developing the RustFS project, ensuring code quality, performance, and maintainability.

AGENTS.md

@@ -8,12 +8,14 @@ The workspace root hosts shared dependencies in `Cargo.toml`. The service binary
 
 ## Build, Test, and Development Commands
 Run `cargo check --all-targets` for fast validation. Build release binaries via `cargo build --release` or the pipeline-aligned `make build`. Use `./build-rustfs.sh --dev` for iterative development and `./build-rustfs.sh --platform <target>` for cross-compiles. Prefer `make pre-commit` before pushing to cover formatting, clippy, checks, and tests.
+Always ensure `cargo fmt --all --check`, `cargo test --workspace --exclude e2e_test`, and `cargo clippy --all-targets --all-features -- -D warnings` complete successfully after each code change to keep the tree healthy and warning-free.
 
 ## Coding Style & Naming Conventions
 Formatting follows the repo `rustfmt.toml` (130-column width). Use `snake_case` for items, `PascalCase` for types, and `SCREAMING_SNAKE_CASE` for constants. Avoid `unwrap()` or `expect()` outside tests; bubble errors with `Result` and crate-specific `thiserror` types. Keep async code non-blocking and offload CPU-heavy work with `tokio::task::spawn_blocking` when necessary.
 
 ## Testing Guidelines
 Co-locate unit tests with their modules and give behavior-led names such as `handles_expired_token`. Integration suites belong in each crate’s `tests/` directory, while exhaustive end-to-end scenarios live in `crates/e2e_test/`. Run `cargo test --workspace --exclude e2e_test` during iteration, `cargo nextest run --all --exclude e2e_test` when available, and finish with `cargo test --all` before requesting review. Use `NO_PROXY=127.0.0.1,localhost HTTP_PROXY= HTTPS_PROXY=` for KMS e2e tests.
+When fixing bugs or adding features, include regression tests that capture the new behavior so future changes cannot silently break it.
 
 ## Commit & Pull Request Guidelines
 Work on feature branches (e.g., `feat/...`) after syncing `main`. Follow Conventional Commits under 72 characters (e.g., `feat: add kms key rotation`). Each commit must compile, format cleanly, and pass `make pre-commit`. Open PRs with a concise summary, note verification commands, link relevant issues, and wait for reviewer approval.

CLA.md

@@ -1,39 +1,88 @@
 RustFS Individual Contributor License Agreement
 
-Thank you for your interest in contributing documentation and related software code to a project hosted or managed by RustFS. In order to clarify the intellectual property license granted with Contributions from any person or entity, RustFS must have a Contributor License Agreement (“CLA”) on file that has been signed by each Contributor, indicating agreement to the license terms below. This version of the Contributor License Agreement allows an individual to submit Contributions to the applicable project. If you are making a submission on behalf of a legal entity, then you should sign the separate Corporate Contributor License Agreement.
+Thank you for your interest in contributing documentation and related software code to a project hosted or managed by
+RustFS. In order to clarify the intellectual property license granted with Contributions from any person or entity,
+RustFS must have a Contributor License Agreement ("CLA") on file that has been signed by each Contributor, indicating
+agreement to the license terms below. This version of the Contributor License Agreement allows an individual to submit
+Contributions to the applicable project. If you are making a submission on behalf of a legal entity, then you should
+sign the separate Corporate Contributor License Agreement.
 
-You accept and agree to the following terms and conditions for Your present and future Contributions submitted to RustFS. You hereby irrevocably assign and transfer to RustFS all right, title, and interest in and to Your Contributions, including all copyrights and other intellectual property rights therein.
+You accept and agree to the following terms and conditions for Your present and future Contributions submitted to
+RustFS. You hereby irrevocably assign and transfer to RustFS all right, title, and interest in and to Your
+Contributions, including all copyrights and other intellectual property rights therein.
 
 Definitions
 
-“You” (or “Your”) shall mean the copyright owner or legal entity authorized by the copyright owner that is making this Agreement with RustFS. For legal entities, the entity making a Contribution and all other entities that control, are controlled by, or are under common control with that entity are considered to be a single Contributor. For the purposes of this definition, “control” means (i) the power, direct or indirect, to cause the direction or management of such entity, whether by contract or otherwise, or (ii) ownership of fifty percent (50%) or more of the outstanding shares, or (iii) beneficial ownership of such entity.
+“You” (or “Your”) shall mean the copyright owner or legal entity authorized by the copyright owner that is making this
+Agreement with RustFS. For legal entities, the entity making a Contribution and all other entities that control, are
+controlled by, or are under common control with that entity are considered to be a single Contributor. For the purposes
+of this definition, "control" means (i) the power, direct or indirect, to cause the direction or management of such
+entity, whether by contract or otherwise, or (ii) ownership of fifty percent (50%) or more of the outstanding shares,
+or (iii) beneficial ownership of such entity.
 
-“Contribution” shall mean any original work of authorship, including any modifications or additions to an existing work, that is intentionally submitted by You to RustFS for inclusion in, or documentation of, any of the products or projects owned or managed by RustFS (the “Work”), including without limitation any Work described in Schedule A. For the purposes of this definition, “submitted” means any form of electronic or written communication sent to RustFS or its representatives, including but not limited to communication on electronic mailing lists, source code control systems, and issue tracking systems that are managed by, or on behalf of, RustFS for the purpose of discussing and improving the Work.
+“Contribution” shall mean any original work of authorship, including any modifications or additions to an existing work,
+that is intentionally submitted by You to RustFS for inclusion in, or documentation of, any of the products or projects
+owned or managed by RustFS (the "Work"), including without limitation any Work described in Schedule A. For the purposes
+of this definition, "submitted" means any form of electronic or written communication sent to RustFS or its
+representatives, including but not limited to communication on electronic mailing lists, source code control systems,
+and issue tracking systems that are managed by, or on behalf of, RustFS for the purpose of discussing and improving the
+Work.
 
 Assignment of Copyright
 
-Subject to the terms and conditions of this Agreement, You hereby irrevocably assign and transfer to RustFS all right, title, and interest in and to Your Contributions, including all copyrights and other intellectual property rights therein, for the entire term of such rights, including all renewals and extensions. You agree to execute all documents and take all actions as may be reasonably necessary to vest in RustFS the ownership of Your Contributions and to assist RustFS in perfecting, maintaining, and enforcing its rights in Your Contributions.
+Subject to the terms and conditions of this Agreement, You hereby irrevocably assign and transfer to RustFS all right,
+title, and interest in and to Your Contributions, including all copyrights and other intellectual property rights
+therein, for the entire term of such rights, including all renewals and extensions. You agree to execute all documents
+and take all actions as may be reasonably necessary to vest in RustFS the ownership of Your Contributions and to assist
+RustFS in perfecting, maintaining, and enforcing its rights in Your Contributions.
 
 Grant of Patent License
 
-Subject to the terms and conditions of this Agreement, You hereby grant to RustFS and to recipients of documentation and software distributed by RustFS a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable (except as stated in this section) patent license to make, have made, use, offer to sell, sell, import, and otherwise transfer the Work, where such license applies only to those patent claims licensable by You that are necessarily infringed by Your Contribution(s) alone or by combination of Your Contribution(s) with the Work to which such Contribution(s) was submitted. If any entity institutes patent litigation against You or any other entity (including a cross-claim or counterclaim in a lawsuit) alleging that your Contribution, or the Work to which you have contributed, constitutes direct or contributory patent infringement, then any patent licenses granted to that entity under this Agreement for that Contribution or Work shall terminate as of the date such litigation is filed.
+Subject to the terms and conditions of this Agreement, You hereby grant to RustFS and to recipients of documentation and
+software distributed by RustFS a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable (except as
+stated in this section) patent license to make, have made, use, offer to sell, sell, import, and otherwise transfer the
+Work, where such license applies only to those patent claims licensable by You that are necessarily infringed by Your
+Contribution(s) alone or by combination of Your Contribution(s) with the Work to which such Contribution(s) was
+submitted. If any entity institutes patent litigation against You or any other entity (including a cross-claim or
+counterclaim in a lawsuit) alleging that your Contribution, or the Work to which you have contributed, constitutes
+direct or contributory patent infringement, then any patent licenses granted to that entity under this Agreement for
+that Contribution or Work shall terminate as of the date such litigation is filed.
 
 You represent that you are legally entitled to grant the above assignment and license.
 
-You represent that each of Your Contributions is Your original creation (see section 7 for submissions on behalf of others). You represent that Your Contribution submissions include complete details of any third-party license or other restriction (including, but not limited to, related patents and trademarks) of which you are personally aware and which are associated with any part of Your Contributions.
+You represent that each of Your Contributions is Your original creation (see section 7 for submissions on behalf of
+others). You represent that Your Contribution submissions include complete details of any third-party license or other
+restriction (including, but not limited to, related patents and trademarks) of which you are personally aware and which
+are associated with any part of Your Contributions.
 
-You are not expected to provide support for Your Contributions, except to the extent You desire to provide support. You may provide support for free, for a fee, or not at all. Unless required by applicable law or agreed to in writing, You provide Your Contributions on an “AS IS” BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied, including, without limitation, any warranties or conditions of TITLE, NON- INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A PARTICULAR PURPOSE.
+You are not expected to provide support for Your Contributions, except to the extent You desire to provide support. You
+may provide support for free, for a fee, or not at all. Unless required by applicable law or agreed to in writing, You
+provide Your Contributions on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied,
+including, without limitation, any warranties or conditions of TITLE, NON- INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR
+A PARTICULAR PURPOSE.
 
-Should You wish to submit work that is not Your original creation, You may submit it to RustFS separately from any Contribution, identifying the complete details of its source and of any license or other restriction (including, but not limited to, related patents, trademarks, and license agreements) of which you are personally aware, and conspicuously marking the work as “Submitted on behalf of a third-party: [named here]”.
+Should You wish to submit work that is not Your original creation, You may submit it to RustFS separately from any
+Contribution, identifying the complete details of its source and of any license or other restriction (including, but not
+limited to, related patents, trademarks, and license agreements) of which you are personally aware, and conspicuously
+marking the work as "Submitted on behalf of a third-party: [named here]”.
 
-You agree to notify RustFS of any facts or circumstances of which you become aware that would make these representations inaccurate in any respect.
+You agree to notify RustFS of any facts or circumstances of which you become aware that would make these representations
+inaccurate in any respect.
 
 Modification of CLA
 
-RustFS reserves the right to update or modify this CLA in the future. Any updates or modifications to this CLA shall apply only to Contributions made after the effective date of the revised CLA. Contributions made prior to the update shall remain governed by the version of the CLA that was in effect at the time of submission. It is not necessary for all Contributors to re-sign the CLA when the CLA is updated or modified.
+RustFS reserves the right to update or modify this CLA in the future. Any updates or modifications to this CLA shall
+apply only to Contributions made after the effective date of the revised CLA. Contributions made prior to the update
+shall remain governed by the version of the CLA that was in effect at the time of submission. It is not necessary for
+all Contributors to re-sign the CLA when the CLA is updated or modified.
 
 Governing Law and Dispute Resolution
 
-This Agreement will be governed by and construed in accordance with the laws of the People’s Republic of China excluding that body of laws known as conflict of laws. The parties expressly agree that the United Nations Convention on Contracts for the International Sale of Goods will not apply. Any legal action or proceeding arising under this Agreement will be brought exclusively in the courts located in Beijing, China, and the parties hereby irrevocably consent to the personal jurisdiction and venue therein.
+This Agreement will be governed by and construed in accordance with the laws of the People's Republic of China excluding
+that body of laws known as conflict of laws. The parties expressly agree that the United Nations Convention on Contracts
+for the International Sale of Goods will not apply. Any legal action or proceeding arising under this Agreement will be
+brought exclusively in the courts located in Beijing, China, and the parties hereby irrevocably consent to the personal
+jurisdiction and venue therein.
 
-For your reading convenience, this Agreement is written in parallel English and Chinese sections. To the extent there is a conflict between the English and Chinese sections, the English sections shall govern.
+For your reading convenience, this Agreement is written in parallel English and Chinese sections. To the extent there is
+a conflict between the English and Chinese sections, the English sections shall govern.

CLAUDE.md

@@ -4,23 +4,28 @@ This file provides guidance to Claude Code (claude.ai/code) when working with co
 
 ## Project Overview
 
-RustFS is a high-performance distributed object storage software built with Rust, providing S3-compatible APIs and advanced features like data lakes, AI, and big data support. It's designed as an alternative to MinIO with better performance and a more business-friendly Apache 2.0 license.
+RustFS is a high-performance distributed object storage software built with Rust, providing S3-compatible APIs and
+advanced features like data lakes, AI, and big data support. It's designed as an alternative to MinIO with better
+performance and a more business-friendly Apache 2.0 license.
 
 ## Build Commands
 
 ### Primary Build Commands
+
 - `cargo build --release` - Build the main RustFS binary
 - `./build-rustfs.sh` - Recommended build script that handles console resources and cross-platform compilation
 - `./build-rustfs.sh --dev` - Development build with debug symbols
 - `make build` or `just build` - Use Make/Just for standardized builds
 
 ### Platform-Specific Builds
+
 - `./build-rustfs.sh --platform x86_64-unknown-linux-musl` - Build for musl target
 - `./build-rustfs.sh --platform aarch64-unknown-linux-gnu` - Build for ARM64
 - `make build-musl` or `just build-musl` - Build musl variant
 - `make build-cross-all` - Build all supported architectures
 
 ### Testing Commands
+
 - `cargo test --workspace --exclude e2e_test` - Run unit tests (excluding e2e tests)
 - `cargo nextest run --all --exclude e2e_test` - Use nextest if available (faster)
 - `cargo test --all --doc` - Run documentation tests
@@ -28,22 +33,30 @@ RustFS is a high-performance distributed object storage software built with Rust
 - `make pre-commit` - Run all quality checks (fmt, clippy, check, test)
 
 ### End-to-End Testing
+
 - `cargo test --package e2e_test` - Run all e2e tests
 - `./scripts/run_e2e_tests.sh` - Run e2e tests via script
 - `./scripts/run_scanner_benchmarks.sh` - Run scanner performance benchmarks
 
 ### KMS-Specific Testing (with proxy bypass)
-- `NO_PROXY=127.0.0.1,localhost HTTP_PROXY= HTTPS_PROXY= http_proxy= https_proxy= cargo test --package e2e_test test_local_kms_end_to_end -- --nocapture --test-threads=1` - Run complete KMS end-to-end test
-- `NO_PROXY=127.0.0.1,localhost HTTP_PROXY= HTTPS_PROXY= http_proxy= https_proxy= cargo test --package e2e_test kms:: -- --nocapture --test-threads=1` - Run all KMS tests
+
+-
+`NO_PROXY=127.0.0.1,localhost HTTP_PROXY= HTTPS_PROXY= http_proxy= https_proxy= cargo test --package e2e_test test_local_kms_end_to_end -- --nocapture --test-threads=1` -
+Run complete KMS end-to-end test
+-
+`NO_PROXY=127.0.0.1,localhost HTTP_PROXY= HTTPS_PROXY= http_proxy= https_proxy= cargo test --package e2e_test kms:: -- --nocapture --test-threads=1` -
+Run all KMS tests
 - `cargo test --package e2e_test test_local_kms_key_isolation -- --nocapture --test-threads=1` - Test KMS key isolation
 - `cargo test --package e2e_test test_local_kms_large_file -- --nocapture --test-threads=1` - Test KMS with large files
 
 ### Code Quality
+
 - `cargo fmt --all` - Format code
 - `cargo clippy --all-targets --all-features -- -D warnings` - Lint code
 - `make pre-commit` or `just pre-commit` - Run all quality checks (fmt, clippy, check, test)
 
 ### Quick Development Commands
+
 - `make help` or `just help` - Show all available commands with descriptions
 - `make help-build` - Show detailed build options and cross-compilation help
 - `make help-docker` - Show comprehensive Docker build and deployment options
@@ -52,6 +65,7 @@ RustFS is a high-performance distributed object storage software built with Rust
 - `./scripts/probe.sh` - Health check and connectivity testing
 
 ### Docker Build Commands
+
 - `make docker-buildx` - Build multi-architecture production images
 - `make docker-dev-local` - Build development image for local use
 - `./docker-buildx.sh --push` - Build and push production images
@@ -61,6 +75,7 @@ RustFS is a high-performance distributed object storage software built with Rust
 ### Core Components
 
 **Main Binary (`rustfs/`):**
+
 - Entry point at `rustfs/src/main.rs`
 - Core modules: admin, auth, config, server, storage, license management, profiling
 - HTTP server with S3-compatible APIs
@@ -68,10 +83,11 @@ RustFS is a high-performance distributed object storage software built with Rust
 - Parallel service initialization with DNS resolver, bucket metadata, and IAM
 
 **Key Crates (`crates/`):**
+
 - `ecstore` - Erasure coding storage implementation (core storage layer)
 - `iam` - Identity and Access Management
 - `kms` - Key Management Service for encryption and key handling
-- `madmin` - Management dashboard and admin API interface  
+- `madmin` - Management dashboard and admin API interface
 - `s3select-api` & `s3select-query` - S3 Select API and query engine
 - `config` - Configuration management with notify features
 - `crypto` - Cryptography and security features
@@ -94,6 +110,7 @@ RustFS is a high-performance distributed object storage software built with Rust
 - `targets` - Target-specific configurations and utilities
 
 ### Build System
+
 - Cargo workspace with 25+ crates (including new KMS functionality)
 - Custom `build-rustfs.sh` script for advanced build options
 - Multi-architecture Docker builds via `docker-buildx.sh`
@@ -103,10 +120,11 @@ RustFS is a high-performance distributed object storage software built with Rust
 - Performance benchmarking and audit workflows
 
 ### Key Dependencies
+
 - `axum` - HTTP framework for S3 API server
 - `tokio` - Async runtime
 - `s3s` - S3 protocol implementation library
-- `datafusion` - For S3 Select query processing  
+- `datafusion` - For S3 Select query processing
 - `hyper`/`hyper-util` - HTTP client/server utilities
 - `rustls` - TLS implementation
 - `serde`/`serde_json` - Serialization
@@ -115,6 +133,7 @@ RustFS is a high-performance distributed object storage software built with Rust
 - `tikv-jemallocator` - Memory allocator for Linux GNU builds
 
 ### Development Workflow
+
 - Console resources are embedded during build via `rust-embed`
 - Protocol buffers generated via custom `gproto` binary
 - E2E tests in separate crate (`e2e_test`) with comprehensive KMS testing
@@ -124,14 +143,16 @@ RustFS is a high-performance distributed object storage software built with Rust
 - Git hooks setup available via `make setup-hooks` or `just setup-hooks`
 
 ### Performance & Observability
+
 - Performance profiling available with `pprof` integration (disabled on Windows)
 - Profiling enabled via environment variables in production
 - Built-in observability with OpenTelemetry integration
 - Background services (scanner, heal) can be controlled via environment variables:
-  - `RUSTFS_ENABLE_SCANNER` (default: true)
-  - `RUSTFS_ENABLE_HEAL` (default: true)
+    - `RUSTFS_ENABLE_SCANNER` (default: true)
+    - `RUSTFS_ENABLE_HEAL` (default: true)
 
 ### Service Architecture
+
 - Service state management with graceful shutdown handling
 - Parallel initialization of core systems (DNS, bucket metadata, IAM)
 - Event notification system with MQTT and webhook support
@@ -139,6 +160,7 @@ RustFS is a high-performance distributed object storage software built with Rust
 - Jemalloc allocator for Linux GNU targets for better performance
 
 ## Environment Variables
+
 - `RUSTFS_ENABLE_SCANNER` - Enable/disable background data scanner (default: true)
 - `RUSTFS_ENABLE_HEAL` - Enable/disable auto-heal functionality (default: true)
 - Various profiling and observability controls
@@ -146,12 +168,14 @@ RustFS is a high-performance distributed object storage software built with Rust
 - Test environment configurations in `scripts/dev_rustfs.env`
 
 ### KMS Environment Variables
+
 - `NO_PROXY=127.0.0.1,localhost` - Required for KMS E2E tests to bypass proxy
 - `HTTP_PROXY=` `HTTPS_PROXY=` `http_proxy=` `https_proxy=` - Clear proxy settings for local KMS testing
 
 ## KMS (Key Management Service) Architecture
 
 ### KMS Implementation Status
+
 - **Full KMS Integration:** Complete implementation with Local and Vault backends
 - **Automatic Configuration:** KMS auto-configures on startup with `--kms-enable` flag
 - **Encryption Support:** Full S3-compatible server-side encryption (SSE-S3, SSE-KMS, SSE-C)
@@ -159,18 +183,21 @@ RustFS is a high-performance distributed object storage software built with Rust
 - **Production Ready:** Comprehensive testing including large files and key isolation
 
 ### KMS Configuration
+
 - **Local Backend:** `--kms-backend local --kms-key-dir <path> --kms-default-key-id <id>`
 - **Vault Backend:** `--kms-backend vault --kms-vault-endpoint <url> --kms-vault-key-name <name>`
 - **Auto-startup:** KMS automatically initializes when `--kms-enable` is provided
 - **Manual Configuration:** Also supports dynamic configuration via admin API
 
 ### S3 Encryption Support
+
 - **SSE-S3:** Server-side encryption with S3-managed keys (`ServerSideEncryption: AES256`)
 - **SSE-KMS:** Server-side encryption with KMS-managed keys (`ServerSideEncryption: aws:kms`)
 - **SSE-C:** Server-side encryption with customer-provided keys
 - **Response Headers:** All encryption types return correct `server_side_encryption` headers in PUT/GET responses
 
 ### KMS Testing Architecture
+
 - **Comprehensive E2E Tests:** Located in `crates/e2e_test/src/kms/`
 - **Test Environments:** Automated test environment setup with temporary directories
 - **Encryption Coverage:** Tests all three encryption types (SSE-S3, SSE-KMS, SSE-C)
@@ -178,6 +205,7 @@ RustFS is a high-performance distributed object storage software built with Rust
 - **Edge Cases:** Key isolation, large file handling, error scenarios
 
 ### Key Files for KMS
+
 - `crates/kms/` - Core KMS implementation with Local/Vault backends
 - `rustfs/src/main.rs` - KMS auto-initialization in `init_kms_system()`
 - `rustfs/src/storage/ecfs.rs` - SSE encryption/decryption in PUT/GET operations
@@ -186,54 +214,62 @@ RustFS is a high-performance distributed object storage software built with Rust
 - `crates/rio/src/encrypt_reader.rs` - Streaming encryption for large files
 
 ## Code Style and Safety Requirements
+
 - **Language Requirements:**
-  - Communicate with me in Chinese, but **only English can be used in code files**
-  - Code comments, function names, variable names, and all text in source files must be in English only
-  - No Chinese characters, emojis, or non-ASCII characters are allowed in any source code files
-  - This includes comments, strings, documentation, and any other text within code files
+    - Communicate with me in Chinese, but **only English can be used in code files**
+    - Code comments, function names, variable names, and all text in source files must be in English only
+    - No Chinese characters, emojis, or non-ASCII characters are allowed in any source code files
+    - This includes comments, strings, documentation, and any other text within code files
 - **Safety-Critical Rules:**
-  - `unsafe_code = "deny"` enforced at workspace level
-  - Never use `unwrap()`, `expect()`, or panic-inducing code except in tests
-  - Avoid blocking I/O operations in async contexts
-  - Use proper error handling with `Result<T, E>` and `Option<T>`
-  - Follow Rust's ownership and borrowing rules strictly
+    - `unsafe_code = "deny"` enforced at workspace level
+    - Never use `unwrap()`, `expect()`, or panic-inducing code except in tests
+    - Avoid blocking I/O operations in async contexts
+    - Use proper error handling with `Result<T, E>` and `Option<T>`
+    - Follow Rust's ownership and borrowing rules strictly
 - **Performance Guidelines:**
-  - Use `cargo clippy --all-targets --all-features -- -D warnings` to catch issues
-  - Prefer `anyhow` for error handling in applications, `thiserror` for libraries
-  - Use appropriate async runtimes and avoid blocking calls
+    - Use `cargo clippy --all-targets --all-features -- -D warnings` to catch issues
+    - Prefer `anyhow` for error handling in applications, `thiserror` for libraries
+    - Use appropriate async runtimes and avoid blocking calls
 - **Testing Standards:**
-  - All new features must include comprehensive tests
-  - Use `#[cfg(test)]` for test-only code that may use panic macros
-  - E2E tests should cover KMS integration scenarios
+    - All new features must include comprehensive tests
+    - Use `#[cfg(test)]` for test-only code that may use panic macros
+    - E2E tests should cover KMS integration scenarios
 
 ## Common Development Tasks
 
 ### Running KMS Tests Locally
+
 1. **Clear proxy settings:** KMS tests require direct localhost connections
 2. **Use serial execution:** `--test-threads=1` prevents port conflicts
 3. **Enable output:** `--nocapture` shows detailed test logs
-4. **Full command:** `NO_PROXY=127.0.0.1,localhost HTTP_PROXY= HTTPS_PROXY= http_proxy= https_proxy= cargo test --package e2e_test test_local_kms_end_to_end -- --nocapture --test-threads=1`
+4. **Full command:**
+   `NO_PROXY=127.0.0.1,localhost HTTP_PROXY= HTTPS_PROXY= http_proxy= https_proxy= cargo test --package e2e_test test_local_kms_end_to_end -- --nocapture --test-threads=1`
 
 ### KMS Development Workflow
+
 1. **Code changes:** Modify KMS-related code in `crates/kms/` or `rustfs/src/`
 2. **Compile:** Always run `cargo build` after changes
 3. **Test specific functionality:** Use targeted test commands for faster iteration
 4. **Full validation:** Run complete end-to-end tests before commits
 
 ### Debugging KMS Issues
+
 - **Server startup:** Check that KMS auto-initializes with debug logs
 - **Encryption failures:** Verify SSE headers are correctly set in both PUT and GET responses
 - **Test failures:** Use `--nocapture` to see detailed error messages
 - **Key management:** Test admin API endpoints with proper authentication
 
 ## Important Reminders
+
 - **Always compile after code changes:** Use `cargo build` to catch errors early
 - **Don't bypass tests:** All functionality must be properly tested, not worked around
 - **Use proper error handling:** Never use `unwrap()` or `expect()` in production code (except tests)
 - **Follow S3 compatibility:** Ensure all encryption types return correct HTTP response headers
 
 # important-instruction-reminders
+
 Do what has been asked; nothing more, nothing less.
 NEVER create files unless they're absolutely necessary for achieving your goal.
 ALWAYS prefer editing an existing file to creating a new one.
-NEVER proactively create documentation files (*.md) or README files. Only create documentation files if explicitly requested by the User.
+NEVER proactively create documentation files (*.md) or README files. Only create documentation files if explicitly
+requested by the User.

Cargo.toml

@@ -97,15 +97,15 @@ async-channel = "2.5.0"
 async-compression = { version = "0.4.19" }
 async-recursion = "1.1.1"
 async-trait = "0.1.89"
-axum = "0.8.6"
-axum-extra = "0.10.3"
-axum-server = { version = "0.7.2", features = ["tls-rustls-no-provider"], default-features = false }
+axum = "0.8.7"
+axum-extra = "0.12.2"
+axum-server = { version = "0.7.3", features = ["tls-rustls-no-provider"], default-features = false }
 futures = "0.3.31"
 futures-core = "0.3.31"
 futures-util = "0.3.31"
-hyper = "1.7.0"
+hyper = { version = "1.8.1", features = ["http2", "http1", "server"] }
 hyper-rustls = { version = "0.27.7", default-features = false, features = ["native-tokio", "http1", "tls12", "logging", "http2", "ring", "webpki-roots"] }
-hyper-util = { version = "0.1.17", features = ["tokio", "server-auto", "server-graceful"] }
+hyper-util = { version = "0.1.18", features = ["tokio", "server-auto", "server-graceful"] }
 http = "1.3.1"
 http-body = "1.0.1"
 reqwest = { version = "0.12.24", default-features = false, features = ["rustls-tls-webpki-roots", "charset", "http2", "system-proxy", "stream", "json", "blocking"] }
@@ -114,7 +114,7 @@ tokio = { version = "1.48.0", features = ["fs", "rt-multi-thread"] }
 tokio-rustls = { version = "0.26.4", default-features = false, features = ["logging", "tls12", "ring"] }
 tokio-stream = { version = "0.1.17" }
 tokio-test = "0.4.4"
-tokio-util = { version = "0.7.16", features = ["io", "compat"] }
+tokio-util = { version = "0.7.17", features = ["io", "compat"] }
 tonic = { version = "0.14.2", features = ["gzip"] }
 tonic-prost = { version = "0.14.2" }
 tonic-prost-build = { version = "0.14.2" }
@@ -122,39 +122,36 @@ tower = { version = "0.5.2", features = ["timeout"] }
 tower-http = { version = "0.6.6", features = ["cors"] }
 
 # Serialization and Data Formats
-bytes = { version = "1.10.1", features = ["serde"] }
-bytesize = "2.1.0"
+bytes = { version = "1.11.0", features = ["serde"] }
+bytesize = "2.3.0"
 byteorder = "1.5.0"
 flatbuffers = "25.9.23"
 form_urlencoded = "1.2.2"
 prost = "0.14.1"
-quick-xml = "0.38.3"
-rmcp = { version = "0.8.3" }
+quick-xml = "0.38.4"
+rmcp = { version = "0.9.0" }
 rmp = { version = "0.8.14" }
 rmp-serde = { version = "1.3.0" }
 serde = { version = "1.0.228", features = ["derive"] }
 serde_json = { version = "1.0.145", features = ["raw_value"] }
 serde_urlencoded = "0.7.1"
-schemars = "1.0.4"
+schemars = "1.1.0"
 
 # Cryptography and Security
-aes-gcm = { version = "0.10.3", features = ["std"] }
-argon2 = { version = "0.5.3", features = ["std"] }
-blake3 = { version = "1.8.2" }
-chacha20poly1305 = { version = "0.10.1" }
-crc-fast = "1.3.0"
-crc32c = "0.6.8"
-crc32fast = "1.5.0"
-crc64fast-nvme = "1.2.0"
-hmac = "0.12.1"
-jsonwebtoken = { version = "10.1.0", features = ["rust_crypto"] }
-pbkdf2 = "0.12.2"
-rsa = { version = "0.9.8" }
-rustls = { version = "0.23.34", features = ["ring", "logging", "std", "tls12"], default-features = false }
+aes-gcm = { version = "0.11.0-rc.2", features = ["rand_core"] }
+argon2 = { version = "0.6.0-rc.2", features = ["std"] }
+blake3 = { version = "1.8.2", features = ["rayon", "mmap"] }
+chacha20poly1305 = { version = "0.11.0-rc.2" }
+crc-fast = "1.6.0"
+hmac = { version = "0.13.0-rc.3" }
+jsonwebtoken = { version = "10.2.0", features = ["rust_crypto"] }
+pbkdf2 = "0.13.0-rc.2"
+rsa = { version = "0.10.0-rc.10" }
+rustls = { version = "0.23.35", features = ["ring", "logging", "std", "tls12"], default-features = false }
 rustls-pemfile = "2.2.0"
-rustls-pki-types = "1.12.0"
-sha1 = "0.10.6"
-sha2 = "0.10.9"
+rustls-pki-types = "1.13.0"
+sha1 = "0.11.0-rc.3"
+sha2 = "0.11.0-rc.3"
 zeroize = { version = "1.8.2", features = ["derive"] }
 
 # Time and Date
@@ -168,39 +165,42 @@ arc-swap = "1.7.1"
 astral-tokio-tar = "0.5.6"
 atoi = "2.0.0"
 atomic_enum = "0.3.0"
-aws-config = { version = "1.8.8" }
-aws-credential-types = { version = "1.2.8" }
-aws-sdk-s3 = { version = "1.108.0", default-features = false, features = ["sigv4a", "rustls", "rt-tokio"] }
-aws-smithy-types = { version = "1.3.3" }
+aws-config = { version = "1.8.11" }
+aws-credential-types = { version = "1.2.10" }
+aws-sdk-s3 = { version = "1.115.0", default-features = false, features = ["sigv4a", "rustls", "rt-tokio"] }
+aws-smithy-types = { version = "1.3.4" }
 base64 = "0.22.1"
 base64-simd = "0.8.0"
 brotli = "8.0.2"
 cfg-if = "1.0.4"
-clap = { version = "4.5.50", features = ["derive", "env"] }
+clap = { version = "4.5.53", features = ["derive", "env"] }
 const-str = { version = "0.7.0", features = ["std", "proc"] }
-convert_case = "0.8.0"
+convert_case = "0.10.0"
 criterion = { version = "0.7", features = ["html_reports"] }
 crossbeam-queue = "0.3.12"
-datafusion = "50.2.0"
+datafusion = "51.0.0"
 derive_builder = "0.20.2"
 enumset = "1.1.10"
-flate2 = "1.1.4"
-flexi_logger = { version = "0.31.7", features = ["trc", "dont_minimize_extra_stacks", "compress", "kv"] }
+faster-hex = "0.10.0"
+flate2 = "1.1.5"
+flexi_logger = { version = "0.31.7", features = ["trc", "dont_minimize_extra_stacks", "compress", "kv", "json"] }
 glob = "0.3.3"
-hashbrown = { version = "0.16.0", features = ["serde", "rayon"] }
+google-cloud-storage = "1.4.0"
+google-cloud-auth = "1.2.0"
+hashbrown = { version = "0.16.1", features = ["serde", "rayon"] }
+heed = { version = "0.22.0" }
 hex-simd = "0.8.0"
 highway = { version = "1.3.0" }
 ipnetwork = { version = "0.21.1", features = ["serde"] }
 lazy_static = "1.5.0"
 libc = "0.2.177"
-libsystemd = { version = "0.7.2" }
+libsystemd = "0.7.2"
 local-ip-address = "0.6.5"
 lz4 = "1.28.1"
-matchit = "0.8.4"
-md-5 = "0.10.6"
+matchit = "0.9.0"
+md-5 = "0.11.0-rc.3"
 md5 = "0.8.0"
 metrics = "0.24.2"
-metrics-exporter-opentelemetry = "0.1.2"
 mime_guess = "2.0.5"
 moka = { version = "0.12.11", features = ["future"] }
 netif = "0.1.6"
@@ -209,20 +209,19 @@ nu-ansi-term = "0.50.3"
 num_cpus = { version = "1.17.0" }
 nvml-wrapper = "0.11.0"
 object_store = "0.12.4"
-once_cell = "1.21.3"
 parking_lot = "0.12.5"
 path-absolutize = "3.1.1"
 path-clean = "1.0.1"
 pin-project-lite = "0.2.16"
 pretty_assertions = "1.4.1"
-rand = "0.9.2"
+rand = { version = "0.10.0-rc.5", features = ["serde"] }
 rayon = "1.11.0"
 reed-solomon-simd = { version = "3.1.0" }
 regex = { version = "1.12.2" }
-rumqttc = { version = "0.25.0" }
-rust-embed = { version = "8.8.0" }
+rumqttc = { version = "0.25.1" }
+rust-embed = { version = "8.9.0" }
 rustc-hash = { version = "2.1.1" }
-s3s = { version = "0.12.0-rc.3", features = ["minio"] }
+s3s = { version = "0.12.0-rc.4", features = ["minio"] }
 serial_test = "3.2.0"
 shadow-rs = { version = "1.4.0", default-features = false }
 siphasher = "1.0.1"
@@ -230,15 +229,16 @@ smallvec = { version = "1.15.1", features = ["serde"] }
 smartstring = "1.0.1"
 snafu = "0.8.9"
 snap = "1.1.1"
-starshard = { version = "0.5.0", features = ["rayon", "async", "serde"] }
+starshard = { version = "0.6.0", features = ["rayon", "async", "serde"] }
 strum = { version = "0.27.2", features = ["derive"] }
 sysctl = "0.7.1"
-sysinfo = "0.37.1"
+sysinfo = "0.37.2"
 temp-env = "0.3.6"
 tempfile = "3.23.0"
 test-case = "3.3.1"
 thiserror = "2.0.17"
 tracing = { version = "0.1.41" }
+tracing-appender = "0.2.3"
 tracing-error = "0.2.1"
 tracing-opentelemetry = "0.32.0"
 tracing-subscriber = { version = "0.3.20", features = ["env-filter", "time"] }
@@ -248,7 +248,7 @@ urlencoding = "2.1.3"
 uuid = { version = "1.18.1", features = ["v4", "fast-rng", "macro-diagnostics"] }
 vaultrs = { version = "0.7.4" }
 walkdir = "2.5.0"
-wildmatch = { version = "2.5.0", features = ["serde"] }
+wildmatch = { version = "2.6.1", features = ["serde"] }
 winapi = { version = "0.3.9" }
 xxhash-rust = { version = "0.8.15", features = ["xxh64", "xxh3"] }
 zip = "6.0.0"
@@ -257,11 +257,21 @@ zstd = "0.13.3"
 # Observability and Metrics
 opentelemetry = { version = "0.31.0" }
 opentelemetry-appender-tracing = { version = "0.31.1", features = ["experimental_use_tracing_span_context", "experimental_metadata_attributes", "spec_unstable_logs_enabled"] }
-opentelemetry-otlp = { version = "0.31.0", default-features = false, features = ["grpc-tonic", "gzip-tonic", "trace", "metrics", "logs", "internal-logs"] }
+opentelemetry-otlp = { version = "0.31.0", features = ["http-proto", "zstd-http"] }
 opentelemetry_sdk = { version = "0.31.0" }
 opentelemetry-semantic-conventions = { version = "0.31.0", features = ["semconv_experimental"] }
 opentelemetry-stdout = { version = "0.31.0" }
 
+# Performance Analysis and Memory Profiling
+# Use tikv-jemallocator as memory allocator and enable performance analysis
+tikv-jemallocator = { version = "0.6", features = ["profiling", "stats", "unprefixed_malloc_on_supported_platforms", "background_threads"] }
+# Used to control and obtain statistics for jemalloc at runtime
+tikv-jemalloc-ctl = { version = "0.6", features = ["use_std", "stats", "profiling"] }
+# Used to generate pprof-compatible memory profiling data and support symbolization and flame graphs
+jemalloc_pprof = { version = "0.8.1", features = ["symbolize", "flamegraph"] }
+# Used to generate CPU performance analysis data and flame diagrams
+pprof = { version = "0.15.0", features = ["flamegraph", "protobuf-codec"] }
+mimalloc = "0.1"
 
 
 [workspace.metadata.cargo-shear]

Dockerfile

@@ -64,8 +64,12 @@ COPY --from=build /etc/ssl/certs/ca-certificates.crt /etc/ssl/certs/
 COPY --from=build /build/rustfs /usr/bin/rustfs
 COPY entrypoint.sh /entrypoint.sh
 
-RUN chmod +x /usr/bin/rustfs /entrypoint.sh && \
+RUN chmod +x /usr/bin/rustfs /entrypoint.sh
+
+RUN addgroup -g 10001 -S rustfs && \
+    adduser -u 10001 -G rustfs -S rustfs -D && \
     mkdir -p /data /logs && \
+    chown -R rustfs:rustfs /data /logs && \
     chmod 0750 /data /logs
 
 ENV RUSTFS_ADDRESS=":9000" \
@@ -78,12 +82,14 @@ ENV RUSTFS_ADDRESS=":9000" \
     RUSTFS_CONSOLE_CORS_ALLOWED_ORIGINS="*" \
     RUSTFS_VOLUMES="/data" \
     RUST_LOG="warn" \
-    RUSTFS_OBS_LOG_DIRECTORY="/logs" \
-    RUSTFS_SINKS_FILE_PATH="/logs"
-
+    RUSTFS_OBS_LOG_DIRECTORY="/logs" 
+    
 EXPOSE 9000 9001
+
 VOLUME ["/data", "/logs"]
 
+USER rustfs
+
 ENTRYPOINT ["/entrypoint.sh"]
 
 CMD ["rustfs"]

Dockerfile.source

@@ -167,7 +167,6 @@ ENV RUSTFS_ADDRESS=":9000" \
     RUSTFS_VOLUMES="/data" \
     RUST_LOG="warn" \
     RUSTFS_OBS_LOG_DIRECTORY="/logs" \
-    RUSTFS_SINKS_FILE_PATH="/logs" \
     RUSTFS_USERNAME="rustfs" \
     RUSTFS_GROUPNAME="rustfs" \
     RUSTFS_UID="1000" \

README.md

@@ -1,6 +1,6 @@
 [![RustFS](https://rustfs.com/images/rustfs-github.png)](https://rustfs.com)
 
-<p align="center">RustFS is a high-performance distributed object storage software built using Rust</p>
+<p align="center">RustFS is a high-performance, distributed object storage system built in Rust.</p>
 
 <p align="center">
   <a href="https://github.com/rustfs/rustfs/actions/workflows/ci.yml"><img alt="CI" src="https://github.com/rustfs/rustfs/actions/workflows/ci.yml/badge.svg" /></a>
@@ -25,17 +25,17 @@ English | <a href="https://github.com/rustfs/rustfs/blob/main/README_ZH.md">简
   <a href="https://readme-i18n.com/rustfs/rustfs?lang=fr">français</a> |
   <a href="https://readme-i18n.com/rustfs/rustfs?lang=ja">日本語</a> |
   <a href="https://readme-i18n.com/rustfs/rustfs?lang=ko">한국어</a> |
-  <a href="https://readme-i18n.com/rustfs/rustfs?lang=pt">Português</a> |
+  <a href="https://readme-i18n.com/rustfs/rustfs?lang=pt">Portuguese</a> |
   <a href="https://readme-i18n.com/rustfs/rustfs?lang=ru">Русский</a>
 </p>
 
-RustFS is a high-performance distributed object storage software built using Rust, one of the most popular languages
-worldwide. Along with MinIO, it shares a range of advantages such as simplicity, S3 compatibility, open-source nature,
+RustFS is a high-performance, distributed object storage system built in Rust., one of the most popular languages
+worldwide. RustFS combines the simplicity of MinIO with the memory safety and performance of Rust., S3 compatibility, open-source nature,
 support for data lakes, AI, and big data. Furthermore, it has a better and more user-friendly open-source license in
 comparison to other storage systems, being constructed under the Apache license. As Rust serves as its foundation,
 RustFS provides faster speed and safer distributed features for high-performance object storage.
 
-> ⚠️ **RustFS is under rapid development. Do NOT use in production environments!**
+> ⚠️ **Current Status: Beta / Technical Preview. Not yet recommended for critical production workloads.**
 
 ## Features
 
@@ -65,9 +65,9 @@ Stress test server parameters
 |---------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------|
 | Powerful Console                                                                | Simple and useless Console                                                                                              |
 | Developed based on Rust language, memory is safer                               | Developed in Go or C, with potential issues like memory GC/leaks                                                        |
-| Does not report logs to third-party countries                                   | Reporting logs to other third countries may violate national security laws                                              |
-| Licensed under Apache, more business-friendly                                   | AGPL V3 License and other License, polluted open source and License traps, infringement of intellectual property rights |
-| Comprehensive S3 support, works with domestic and international cloud providers | Full support for S3, but no local cloud vendor support                                                                  |
+|  No telemetry. Guards against unauthorized cross-border data egress, ensuring full compliance with global regulations including GDPR (EU/UK), CCPA (US), APPI (Japan) |Potential legal exposure and data telemetry risks     |
+| Permissive Apache 2.0 License     | AGPL V3 License and other License, polluted open source and License traps, infringement of intellectual property rights |
+| 100% S3 compatible—works with any cloud provider, anywhere  | Full support for S3, but no local cloud vendor support                                                                  |
 | Rust-based development, strong support for secure and innovative devices        | Poor support for edge gateways and secure innovative devices                                                            |
 | Stable commercial prices, free community support                                | High pricing, with costs up to $250,000 for 1PiB                                                                        |
 | No risk                                                                         | Intellectual property risks and risks of prohibited uses                                                                |
@@ -84,15 +84,20 @@ To get started with RustFS, follow these steps:
 
 2. **Docker Quick Start (Option 2)​​**
 
+RustFS container run as non-root user `rustfs` with id `1000`, if you run docker with `-v` to mount host directory into docker container, please make sure the owner of host directory has been changed to `1000`, otherwise you will encounter permission denied error.
+
   ```bash
    # create data and logs directories
    mkdir -p data logs
 
-   # using latest alpha version
-   docker run -d -p 9000:9000 -v $(pwd)/data:/data -v $(pwd)/logs:/logs rustfs/rustfs:alpha
+   # change the owner of those two ditectories
+   chown -R 10001:10001 data logs
 
-   # Specific version
-   docker run -d -p 9000:9000 -v $(pwd)/data:/data -v $(pwd)/logs:/logs rustfs/rustfs:1.0.0.alpha.45
+   # using latest version
+   docker run -d -p 9000:9000 -p 9001:9001 -v $(pwd)/data:/data -v $(pwd)/logs:/logs rustfs/rustfs:latest
+
+   # using specific version
+   docker run -d -p 9000:9000 -p 9001:9001 -v $(pwd)/data:/data -v $(pwd)/logs:/logs rustfs/rustfs:1.0.0.alpha.68
    ```
 
 For docker installation, you can also run the container with docker compose. With the `docker-compose.yml` file under
@@ -139,10 +144,16 @@ observability. If you want to start redis as well as nginx container, you can sp
    make help-docker                      # Show all Docker-related commands
    ```
 
-4. **Access the Console**: Open your web browser and navigate to `http://localhost:9000` to access the RustFS console,
+   > **Heads-up (macOS cross-compilation)**: macOS keeps the default `ulimit -n` at 256, so `cargo zigbuild` or `./build-rustfs.sh --platform ...` may fail with `ProcessFdQuotaExceeded` when targeting Linux. The build script now tries to raise the limit automatically, but if you still see the warning, run `ulimit -n 4096` (or higher) in your shell before building.
+
+4. **Build with helm chart(Option 4) - Cloud Native environment**
+
+   Following the instructions on [helm chart README](./helm/README.md) to install RustFS on kubernetes cluster.
+
+5. **Access the Console**: Open your web browser and navigate to `http://localhost:9000` to access the RustFS console,
    default username and password is `rustfsadmin` .
-5. **Create a Bucket**: Use the console to create a new bucket for your objects.
-6. **Upload Objects**: You can upload files directly through the console or use S3-compatible APIs to interact with your
+6. **Create a Bucket**: Use the console to create a new bucket for your objects.
+7. **Upload Objects**: You can upload files directly through the console or use S3-compatible APIs to interact with your
    RustFS instance.
 
 **NOTE**: If you want to access RustFS instance with `https`, you can refer
@@ -203,4 +214,3 @@ top charts.
 [Apache 2.0](https://opensource.org/licenses/Apache-2.0)
 
 **RustFS** is a trademark of RustFS, Inc. All other trademarks are the property of their respective owners.
-

README_ZH.md

@@ -87,13 +87,51 @@ RustFS 是一个使用 Rust（全球最受欢迎的编程语言之一）构建
 以外，还有 grafana、prometheus、jaeger 等，这些是为 rustfs 可观测性服务的，还有 redis 和 nginx。你想启动哪些容器，就需要用
 `--profile` 参数指定相应的 profile。
 
-3. **访问控制台**：打开 Web 浏览器并导航到 `http://localhost:9000` 以访问 RustFS 控制台，默认的用户名和密码是
-   `rustfsadmin` 。
-4. **创建存储桶**：使用控制台为您的对象创建新的存储桶。
-5. **上传对象**：您可以直接通过控制台上传文件，或使用 S3 兼容的 API 与您的 RustFS 实例交互。
+3. **从源码构建（方案三）- 高级用户**
 
-**注意**：如果你想通过 `https` 来访问 RustFS
-实例，请参考 [TLS 配置文档](https://docs.rustfs.com/zh/integration/tls-configured.html)
+   面向希望从源码构建支持多架构 Docker 镜像的开发者：
+
+   ```bash
+   # 本地构建多架构镜像
+   ./docker-buildx.sh --build-arg RELEASE=latest
+
+   # 构建并推送至镜像仓库
+   ./docker-buildx.sh --push
+
+   # 构建指定版本
+   ./docker-buildx.sh --release v1.0.0 --push
+
+   # 构建并推送到自定义镜像仓库
+   ./docker-buildx.sh --registry your-registry.com --namespace yourname --push
+   ```
+
+   `docker-buildx.sh` 脚本支持：
+    - **多架构构建**：`linux/amd64`、`linux/arm64`
+    - **自动版本检测**：可使用 git 标签或提交哈希
+    - **仓库灵活性**：支持 Docker Hub、GitHub Container Registry 等
+    - **构建优化**：包含缓存和并行构建
+
+   你也可以使用 Makefile 提供的目标命令以提升便捷性：
+
+  ```bash
+  make docker-buildx                    # 本地构建
+  make docker-buildx-push               # 构建并推送
+  make docker-buildx-version VERSION=v1.0.0  # 构建指定版本
+  make help-docker                      # 显示全部 Docker 相关命令
+  ```
+
+   > **提示（macOS 交叉编译）**：macOS 默认的 `ulimit -n` 只有 256，使用 `cargo zigbuild` 或 `./build-rustfs.sh --platform ...` 编译 Linux 目标时容易触发 `ProcessFdQuotaExceeded` 链接错误。脚本会尝试自动提升该限制，如仍提示失败，请在构建前手动执行 `ulimit -n 4096`（或更大的值）。
+
+4. **使用 Helm Chart 部署（方案四）- 云原生环境**
+
+   按照 [helm chart 说明文档](./helm/README.md) 的指引，在 Kubernetes 集群中安装 RustFS。
+
+5. **访问控制台**：打开 Web 浏览器并导航到 `http://localhost:9000` 以访问 RustFS 控制台，默认的用户名和密码是
+   `rustfsadmin` 。
+6. **创建存储桶**：使用控制台为您的对象创建新的存储桶。
+7. **上传对象**：您可以直接通过控制台上传文件，或使用 S3 兼容的 API 与您的 RustFS 实例交互。
+
+**注意**：如果你想通过 `https` 来访问 RustFS 实例，请参考 [TLS 配置文档](https://docs.rustfs.com/zh/integration/tls-configured.html)
 
 ## 文档
 

build-rustfs.sh

@@ -163,6 +163,35 @@ print_message() {
     echo -e "${color}${message}${NC}"
 }
 
+# Prevent zig/ld from hitting macOS file descriptor defaults during linking
+ensure_file_descriptor_limit() {
+    local required_limit=4096
+    local current_limit
+    current_limit=$(ulimit -Sn 2>/dev/null || echo "")
+
+    if [ -z "$current_limit" ] || [ "$current_limit" = "unlimited" ]; then
+        return
+    fi
+
+    if (( current_limit >= required_limit )); then
+        return
+    fi
+
+    local hard_limit target_limit
+    hard_limit=$(ulimit -Hn 2>/dev/null || echo "")
+    target_limit=$required_limit
+
+    if [ -n "$hard_limit" ] && [ "$hard_limit" != "unlimited" ] && (( hard_limit < required_limit )); then
+        target_limit=$hard_limit
+    fi
+
+    if ulimit -Sn "$target_limit" 2>/dev/null; then
+        print_message $YELLOW "🔧 Increased open file limit from $current_limit to $target_limit to avoid ProcessFdQuotaExceeded"
+    else
+        print_message $YELLOW "⚠️ Unable to raise ulimit -n automatically (current: $current_limit, needed: $required_limit). Please run 'ulimit -n $required_limit' manually before building."
+    fi
+}
+
 # Get version from git
 get_version() {
     if git describe --abbrev=0 --tags >/dev/null 2>&1; then
@@ -570,10 +599,11 @@ main() {
         fi
     fi
 
+    ensure_file_descriptor_limit
+
     # Start build process
     build_rustfs
 }
 
 # Run main function
 main
-

crates/ahm/Cargo.toml

@@ -40,4 +40,4 @@ serde_json = { workspace = true }
 serial_test = { workspace = true }
 tracing-subscriber = { workspace = true }
 tempfile = { workspace = true }
-heed = "0.22.0"
+heed = { workspace = true }

crates/ahm/src/error.rs

@@ -14,6 +14,10 @@
 
 use thiserror::Error;
 
+/// Custom error type for AHM operations
+/// This enum defines various error variants that can occur during
+/// the execution of AHM-related tasks, such as I/O errors, storage errors,
+/// configuration errors, and specific errors related to healing operations.
 #[derive(Debug, Error)]
 pub enum Error {
     #[error("I/O error: {0}")]
@@ -85,9 +89,13 @@ pub enum Error {
     ProgressTrackingFailed { message: String },
 }
 
+/// A specialized Result type for AHM operations
+///This type is a convenient alias for results returned by functions in the AHM crate,
+/// using the custom Error type defined above.
 pub type Result<T, E = Error> = std::result::Result<T, E>;
 
 impl Error {
+    /// Create an Other error from any error type
     pub fn other<E>(error: E) -> Self
     where
         E: Into<Box<dyn std::error::Error + Send + Sync>>,

crates/ahm/src/heal/channel.rs

@@ -12,18 +12,19 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use crate::error::Result;
 use crate::heal::{
     manager::HealManager,
     task::{HealOptions, HealPriority, HealRequest, HealType},
+    utils,
 };
-
+use crate::{Error, Result};
 use rustfs_common::heal_channel::{
     HealChannelCommand, HealChannelPriority, HealChannelReceiver, HealChannelRequest, HealChannelResponse, HealScanMode,
+    publish_heal_response,
 };
 use std::sync::Arc;
 use tokio::sync::mpsc;
-use tracing::{error, info};
+use tracing::{debug, error, info};
 
 /// Heal channel processor
 pub struct HealChannelProcessor {
@@ -60,7 +61,7 @@ impl HealChannelProcessor {
                             }
                         }
                         None => {
-                            info!("Heal channel receiver closed, stopping processor");
+                            debug!("Heal channel receiver closed, stopping processor");
                             break;
                         }
                     }
@@ -99,7 +100,6 @@ impl HealChannelProcessor {
             Ok(task_id) => {
                 info!("Successfully submitted heal request: {} as task: {}", request.id, task_id);
 
-                // Send success response
                 let response = HealChannelResponse {
                     request_id: request.id,
                     success: true,
@@ -107,9 +107,7 @@ impl HealChannelProcessor {
                     error: None,
                 };
 
-                if let Err(e) = self.response_sender.send(response) {
-                    error!("Failed to send heal response: {}", e);
-                }
+                self.publish_response(response);
             }
             Err(e) => {
                 error!("Failed to submit heal request: {} - {}", request.id, e);
@@ -122,9 +120,7 @@ impl HealChannelProcessor {
                     error: Some(e.to_string()),
                 };
 
-                if let Err(e) = self.response_sender.send(response) {
-                    error!("Failed to send heal error response: {}", e);
-                }
+                self.publish_response(response);
             }
         }
 
@@ -144,9 +140,7 @@ impl HealChannelProcessor {
             error: None,
         };
 
-        if let Err(e) = self.response_sender.send(response) {
-            error!("Failed to send query response: {}", e);
-        }
+        self.publish_response(response);
 
         Ok(())
     }
@@ -164,9 +158,7 @@ impl HealChannelProcessor {
             error: None,
         };
 
-        if let Err(e) = self.response_sender.send(response) {
-            error!("Failed to send cancel response: {}", e);
-        }
+        self.publish_response(response);
 
         Ok(())
     }
@@ -174,9 +166,12 @@ impl HealChannelProcessor {
     /// Convert channel request to heal request
     fn convert_to_heal_request(&self, request: HealChannelRequest) -> Result<HealRequest> {
         let heal_type = if let Some(disk_id) = &request.disk {
+            let set_disk_id = utils::normalize_set_disk_id(disk_id).ok_or_else(|| Error::InvalidHealType {
+                heal_type: format!("erasure-set({disk_id})"),
+            })?;
             HealType::ErasureSet {
                 buckets: vec![],
-                set_disk_id: disk_id.clone(),
+                set_disk_id,
             }
         } else if let Some(prefix) = &request.object_prefix {
             if !prefix.is_empty() {
@@ -226,8 +221,332 @@ impl HealChannelProcessor {
         Ok(HealRequest::new(heal_type, options, priority))
     }
 
+    fn publish_response(&self, response: HealChannelResponse) {
+        // Try to send to local channel first, but don't block broadcast on failure
+        if let Err(e) = self.response_sender.send(response.clone()) {
+            error!("Failed to enqueue heal response locally: {}", e);
+        }
+        // Always attempt to broadcast, even if local send failed
+        // Use the original response for broadcast; local send uses a clone
+        if let Err(e) = publish_heal_response(response) {
+            error!("Failed to broadcast heal response: {}", e);
+        }
+    }
+
     /// Get response sender for external use
     pub fn get_response_sender(&self) -> mpsc::UnboundedSender<HealChannelResponse> {
         self.response_sender.clone()
     }
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use crate::heal::storage::HealStorageAPI;
+    use rustfs_common::heal_channel::{HealChannelPriority, HealChannelRequest, HealScanMode};
+    use std::sync::Arc;
+
+    // Mock storage for testing
+    struct MockStorage;
+    #[async_trait::async_trait]
+    impl HealStorageAPI for MockStorage {
+        async fn get_object_meta(
+            &self,
+            _bucket: &str,
+            _object: &str,
+        ) -> crate::Result<Option<rustfs_ecstore::store_api::ObjectInfo>> {
+            Ok(None)
+        }
+        async fn get_object_data(&self, _bucket: &str, _object: &str) -> crate::Result<Option<Vec<u8>>> {
+            Ok(None)
+        }
+        async fn put_object_data(&self, _bucket: &str, _object: &str, _data: &[u8]) -> crate::Result<()> {
+            Ok(())
+        }
+        async fn delete_object(&self, _bucket: &str, _object: &str) -> crate::Result<()> {
+            Ok(())
+        }
+        async fn verify_object_integrity(&self, _bucket: &str, _object: &str) -> crate::Result<bool> {
+            Ok(true)
+        }
+        async fn ec_decode_rebuild(&self, _bucket: &str, _object: &str) -> crate::Result<Vec<u8>> {
+            Ok(vec![])
+        }
+        async fn get_disk_status(
+            &self,
+            _endpoint: &rustfs_ecstore::disk::endpoint::Endpoint,
+        ) -> crate::Result<crate::heal::storage::DiskStatus> {
+            Ok(crate::heal::storage::DiskStatus::Ok)
+        }
+        async fn format_disk(&self, _endpoint: &rustfs_ecstore::disk::endpoint::Endpoint) -> crate::Result<()> {
+            Ok(())
+        }
+        async fn get_bucket_info(&self, _bucket: &str) -> crate::Result<Option<rustfs_ecstore::store_api::BucketInfo>> {
+            Ok(None)
+        }
+        async fn heal_bucket_metadata(&self, _bucket: &str) -> crate::Result<()> {
+            Ok(())
+        }
+        async fn list_buckets(&self) -> crate::Result<Vec<rustfs_ecstore::store_api::BucketInfo>> {
+            Ok(vec![])
+        }
+        async fn object_exists(&self, _bucket: &str, _object: &str) -> crate::Result<bool> {
+            Ok(false)
+        }
+        async fn get_object_size(&self, _bucket: &str, _object: &str) -> crate::Result<Option<u64>> {
+            Ok(None)
+        }
+        async fn get_object_checksum(&self, _bucket: &str, _object: &str) -> crate::Result<Option<String>> {
+            Ok(None)
+        }
+        async fn heal_object(
+            &self,
+            _bucket: &str,
+            _object: &str,
+            _version_id: Option<&str>,
+            _opts: &rustfs_common::heal_channel::HealOpts,
+        ) -> crate::Result<(rustfs_madmin::heal_commands::HealResultItem, Option<crate::Error>)> {
+            Ok((rustfs_madmin::heal_commands::HealResultItem::default(), None))
+        }
+        async fn heal_bucket(
+            &self,
+            _bucket: &str,
+            _opts: &rustfs_common::heal_channel::HealOpts,
+        ) -> crate::Result<rustfs_madmin::heal_commands::HealResultItem> {
+            Ok(rustfs_madmin::heal_commands::HealResultItem::default())
+        }
+        async fn heal_format(
+            &self,
+            _dry_run: bool,
+        ) -> crate::Result<(rustfs_madmin::heal_commands::HealResultItem, Option<crate::Error>)> {
+            Ok((rustfs_madmin::heal_commands::HealResultItem::default(), None))
+        }
+        async fn list_objects_for_heal(&self, _bucket: &str, _prefix: &str) -> crate::Result<Vec<String>> {
+            Ok(vec![])
+        }
+        async fn get_disk_for_resume(&self, _set_disk_id: &str) -> crate::Result<rustfs_ecstore::disk::DiskStore> {
+            Err(crate::Error::other("Not implemented in mock"))
+        }
+    }
+
+    fn create_test_heal_manager() -> Arc<HealManager> {
+        let storage: Arc<dyn HealStorageAPI> = Arc::new(MockStorage);
+        Arc::new(HealManager::new(storage, None))
+    }
+
+    #[test]
+    fn test_heal_channel_processor_new() {
+        let heal_manager = create_test_heal_manager();
+        let processor = HealChannelProcessor::new(heal_manager);
+
+        // Verify processor is created successfully
+        let _sender = processor.get_response_sender();
+        // If we can get the sender, processor was created correctly
+    }
+
+    #[tokio::test]
+    async fn test_convert_to_heal_request_bucket() {
+        let heal_manager = create_test_heal_manager();
+        let processor = HealChannelProcessor::new(heal_manager);
+
+        let channel_request = HealChannelRequest {
+            id: "test-id".to_string(),
+            bucket: "test-bucket".to_string(),
+            object_prefix: None,
+            disk: None,
+            priority: HealChannelPriority::Normal,
+            scan_mode: None,
+            remove_corrupted: None,
+            recreate_missing: None,
+            update_parity: None,
+            recursive: None,
+            dry_run: None,
+            timeout_seconds: None,
+            pool_index: None,
+            set_index: None,
+            force_start: false,
+        };
+
+        let heal_request = processor.convert_to_heal_request(channel_request).unwrap();
+        assert!(matches!(heal_request.heal_type, HealType::Bucket { .. }));
+        assert_eq!(heal_request.priority, HealPriority::Normal);
+    }
+
+    #[tokio::test]
+    async fn test_convert_to_heal_request_object() {
+        let heal_manager = create_test_heal_manager();
+        let processor = HealChannelProcessor::new(heal_manager);
+
+        let channel_request = HealChannelRequest {
+            id: "test-id".to_string(),
+            bucket: "test-bucket".to_string(),
+            object_prefix: Some("test-object".to_string()),
+            disk: None,
+            priority: HealChannelPriority::High,
+            scan_mode: Some(HealScanMode::Deep),
+            remove_corrupted: Some(true),
+            recreate_missing: Some(true),
+            update_parity: Some(true),
+            recursive: Some(false),
+            dry_run: Some(false),
+            timeout_seconds: Some(300),
+            pool_index: Some(0),
+            set_index: Some(1),
+            force_start: false,
+        };
+
+        let heal_request = processor.convert_to_heal_request(channel_request).unwrap();
+        assert!(matches!(heal_request.heal_type, HealType::Object { .. }));
+        assert_eq!(heal_request.priority, HealPriority::High);
+        assert_eq!(heal_request.options.scan_mode, HealScanMode::Deep);
+        assert!(heal_request.options.remove_corrupted);
+        assert!(heal_request.options.recreate_missing);
+    }
+
+    #[tokio::test]
+    async fn test_convert_to_heal_request_erasure_set() {
+        let heal_manager = create_test_heal_manager();
+        let processor = HealChannelProcessor::new(heal_manager);
+
+        let channel_request = HealChannelRequest {
+            id: "test-id".to_string(),
+            bucket: "test-bucket".to_string(),
+            object_prefix: None,
+            disk: Some("pool_0_set_1".to_string()),
+            priority: HealChannelPriority::Critical,
+            scan_mode: None,
+            remove_corrupted: None,
+            recreate_missing: None,
+            update_parity: None,
+            recursive: None,
+            dry_run: None,
+            timeout_seconds: None,
+            pool_index: None,
+            set_index: None,
+            force_start: false,
+        };
+
+        let heal_request = processor.convert_to_heal_request(channel_request).unwrap();
+        assert!(matches!(heal_request.heal_type, HealType::ErasureSet { .. }));
+        assert_eq!(heal_request.priority, HealPriority::Urgent);
+    }
+
+    #[tokio::test]
+    async fn test_convert_to_heal_request_invalid_disk_id() {
+        let heal_manager = create_test_heal_manager();
+        let processor = HealChannelProcessor::new(heal_manager);
+
+        let channel_request = HealChannelRequest {
+            id: "test-id".to_string(),
+            bucket: "test-bucket".to_string(),
+            object_prefix: None,
+            disk: Some("invalid-disk-id".to_string()),
+            priority: HealChannelPriority::Normal,
+            scan_mode: None,
+            remove_corrupted: None,
+            recreate_missing: None,
+            update_parity: None,
+            recursive: None,
+            dry_run: None,
+            timeout_seconds: None,
+            pool_index: None,
+            set_index: None,
+            force_start: false,
+        };
+
+        let result = processor.convert_to_heal_request(channel_request);
+        assert!(result.is_err());
+    }
+
+    #[tokio::test]
+    async fn test_convert_to_heal_request_priority_mapping() {
+        let heal_manager = create_test_heal_manager();
+        let processor = HealChannelProcessor::new(heal_manager);
+
+        let priorities = vec![
+            (HealChannelPriority::Low, HealPriority::Low),
+            (HealChannelPriority::Normal, HealPriority::Normal),
+            (HealChannelPriority::High, HealPriority::High),
+            (HealChannelPriority::Critical, HealPriority::Urgent),
+        ];
+
+        for (channel_priority, expected_heal_priority) in priorities {
+            let channel_request = HealChannelRequest {
+                id: "test-id".to_string(),
+                bucket: "test-bucket".to_string(),
+                object_prefix: None,
+                disk: None,
+                priority: channel_priority,
+                scan_mode: None,
+                remove_corrupted: None,
+                recreate_missing: None,
+                update_parity: None,
+                recursive: None,
+                dry_run: None,
+                timeout_seconds: None,
+                pool_index: None,
+                set_index: None,
+                force_start: false,
+            };
+
+            let heal_request = processor.convert_to_heal_request(channel_request).unwrap();
+            assert_eq!(heal_request.priority, expected_heal_priority);
+        }
+    }
+
+    #[tokio::test]
+    async fn test_convert_to_heal_request_force_start() {
+        let heal_manager = create_test_heal_manager();
+        let processor = HealChannelProcessor::new(heal_manager);
+
+        let channel_request = HealChannelRequest {
+            id: "test-id".to_string(),
+            bucket: "test-bucket".to_string(),
+            object_prefix: None,
+            disk: None,
+            priority: HealChannelPriority::Normal,
+            scan_mode: None,
+            remove_corrupted: Some(false),
+            recreate_missing: Some(false),
+            update_parity: Some(false),
+            recursive: None,
+            dry_run: None,
+            timeout_seconds: None,
+            pool_index: None,
+            set_index: None,
+            force_start: true, // Should override the above false values
+        };
+
+        let heal_request = processor.convert_to_heal_request(channel_request).unwrap();
+        assert!(heal_request.options.remove_corrupted);
+        assert!(heal_request.options.recreate_missing);
+        assert!(heal_request.options.update_parity);
+    }
+
+    #[tokio::test]
+    async fn test_convert_to_heal_request_empty_object_prefix() {
+        let heal_manager = create_test_heal_manager();
+        let processor = HealChannelProcessor::new(heal_manager);
+
+        let channel_request = HealChannelRequest {
+            id: "test-id".to_string(),
+            bucket: "test-bucket".to_string(),
+            object_prefix: Some("".to_string()), // Empty prefix should be treated as bucket heal
+            disk: None,
+            priority: HealChannelPriority::Normal,
+            scan_mode: None,
+            remove_corrupted: None,
+            recreate_missing: None,
+            update_parity: None,
+            recursive: None,
+            dry_run: None,
+            timeout_seconds: None,
+            pool_index: None,
+            set_index: None,
+            force_start: false,
+        };
+
+        let heal_request = processor.convert_to_heal_request(channel_request).unwrap();
+        assert!(matches!(heal_request.heal_type, HealType::Bucket { .. }));
+    }
+}

crates/ahm/src/heal/erasure_healer.rs

@@ -12,12 +12,12 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use crate::error::{Error, Result};
 use crate::heal::{
     progress::HealProgress,
     resume::{CheckpointManager, ResumeManager, ResumeUtils},
     storage::HealStorageAPI,
 };
+use crate::{Error, Result};
 use futures::future::join_all;
 use rustfs_common::heal_channel::{HealOpts, HealScanMode};
 use rustfs_ecstore::disk::DiskStore;
@@ -49,14 +49,15 @@ impl ErasureSetHealer {
     }
 
     /// execute erasure set heal with resume
+    #[tracing::instrument(skip(self, buckets), fields(set_disk_id = %set_disk_id, bucket_count = buckets.len()))]
     pub async fn heal_erasure_set(&self, buckets: &[String], set_disk_id: &str) -> Result<()> {
-        info!("Starting erasure set heal for {} buckets on set disk {}", buckets.len(), set_disk_id);
+        info!("Starting erasure set heal");
 
         // 1. generate or get task id
         let task_id = self.get_or_create_task_id(set_disk_id).await?;
 
         // 2. initialize or resume resume state
-        let (resume_manager, checkpoint_manager) = self.initialize_resume_state(&task_id, buckets).await?;
+        let (resume_manager, checkpoint_manager) = self.initialize_resume_state(&task_id, set_disk_id, buckets).await?;
 
         // 3. execute heal with resume
         let result = self
@@ -77,25 +78,38 @@ impl ErasureSetHealer {
     }
 
     /// get or create task id
-    async fn get_or_create_task_id(&self, _set_disk_id: &str) -> Result<String> {
+    async fn get_or_create_task_id(&self, set_disk_id: &str) -> Result<String> {
         // check if there are resumable tasks
         let resumable_tasks = ResumeUtils::get_resumable_tasks(&self.disk).await?;
 
         for task_id in resumable_tasks {
-            if ResumeUtils::can_resume_task(&self.disk, &task_id).await {
-                info!("Found resumable task: {}", task_id);
-                return Ok(task_id);
+            match ResumeManager::load_from_disk(self.disk.clone(), &task_id).await {
+                Ok(manager) => {
+                    let state = manager.get_state().await;
+                    if state.set_disk_id == set_disk_id && ResumeUtils::can_resume_task(&self.disk, &task_id).await {
+                        info!("Found resumable task: {} for set {}", task_id, set_disk_id);
+                        return Ok(task_id);
+                    }
+                }
+                Err(e) => {
+                    warn!("Failed to load resume state for task {}: {}", task_id, e);
+                }
             }
         }
 
         // create new task id
-        let task_id = ResumeUtils::generate_task_id();
+        let task_id = format!("{}_{}", set_disk_id, ResumeUtils::generate_task_id());
         info!("Created new heal task: {}", task_id);
         Ok(task_id)
     }
 
     /// initialize or resume resume state
-    async fn initialize_resume_state(&self, task_id: &str, buckets: &[String]) -> Result<(ResumeManager, CheckpointManager)> {
+    async fn initialize_resume_state(
+        &self,
+        task_id: &str,
+        set_disk_id: &str,
+        buckets: &[String],
+    ) -> Result<(ResumeManager, CheckpointManager)> {
         // check if resume state exists
         if ResumeManager::has_resume_state(&self.disk, task_id).await {
             info!("Loading existing resume state for task: {}", task_id);
@@ -111,8 +125,14 @@ impl ErasureSetHealer {
         } else {
             info!("Creating new resume state for task: {}", task_id);
 
-            let resume_manager =
-                ResumeManager::new(self.disk.clone(), task_id.to_string(), "erasure_set".to_string(), buckets.to_vec()).await?;
+            let resume_manager = ResumeManager::new(
+                self.disk.clone(),
+                task_id.to_string(),
+                "erasure_set".to_string(),
+                set_disk_id.to_string(),
+                buckets.to_vec(),
+            )
+            .await?;
 
             let checkpoint_manager = CheckpointManager::new(self.disk.clone(), task_id.to_string()).await?;
 
@@ -162,6 +182,7 @@ impl ErasureSetHealer {
             let bucket_result = self
                 .heal_bucket_with_resume(
                     bucket,
+                    bucket_idx,
                     &mut current_object_index,
                     &mut processed_objects,
                     &mut successful_objects,
@@ -182,7 +203,7 @@ impl ErasureSetHealer {
 
             // check cancel status
             if self.cancel_token.is_cancelled() {
-                info!("Heal task cancelled");
+                warn!("Heal task cancelled");
                 return Err(Error::TaskCancelled);
             }
 
@@ -211,9 +232,11 @@ impl ErasureSetHealer {
 
     /// heal single bucket with resume
     #[allow(clippy::too_many_arguments)]
+    #[tracing::instrument(skip(self, current_object_index, processed_objects, successful_objects, failed_objects, _skipped_objects, resume_manager, checkpoint_manager), fields(bucket = %bucket, bucket_index = bucket_index))]
     async fn heal_bucket_with_resume(
         &self,
         bucket: &str,
+        bucket_index: usize,
         current_object_index: &mut usize,
         processed_objects: &mut u64,
         successful_objects: &mut u64,
@@ -222,7 +245,7 @@ impl ErasureSetHealer {
         resume_manager: &ResumeManager,
         checkpoint_manager: &CheckpointManager,
     ) -> Result<()> {
-        info!("Starting heal for bucket: {} from object index {}", bucket, current_object_index);
+        info!(target: "rustfs:ahm:heal_bucket_with_resume" ,"Starting heal for bucket from object index {}", current_object_index);
 
         // 1. get bucket info
         let _bucket_info = match self.storage.get_bucket_info(bucket).await? {
@@ -252,7 +275,9 @@ impl ErasureSetHealer {
             let object_exists = match self.storage.object_exists(bucket, object).await {
                 Ok(exists) => exists,
                 Err(e) => {
-                    warn!("Failed to check existence of {}/{}: {}, skipping", bucket, object, e);
+                    warn!("Failed to check existence of {}/{}: {}, marking as failed", bucket, object, e);
+                    *failed_objects += 1;
+                    checkpoint_manager.add_failed_object(object.clone()).await?;
                     *current_object_index = obj_idx + 1;
                     continue;
                 }
@@ -260,7 +285,7 @@ impl ErasureSetHealer {
 
             if !object_exists {
                 info!(
-                    "Object {}/{} no longer exists, skipping heal (likely deleted intentionally)",
+                    target: "rustfs:ahm:heal_bucket_with_resume" ,"Object {}/{} no longer exists, skipping heal (likely deleted intentionally)",
                     bucket, object
                 );
                 checkpoint_manager.add_processed_object(object.clone()).await?;
@@ -306,7 +331,9 @@ impl ErasureSetHealer {
 
             // save checkpoint periodically
             if obj_idx % 100 == 0 {
-                checkpoint_manager.update_position(0, *current_object_index).await?;
+                checkpoint_manager
+                    .update_position(bucket_index, *current_object_index)
+                    .await?;
             }
         }
 
@@ -337,7 +364,10 @@ impl ErasureSetHealer {
             let cancel_token = self.cancel_token.clone();
 
             async move {
-                let _permit = semaphore.acquire().await.unwrap();
+                let _permit = semaphore
+                    .acquire()
+                    .await
+                    .map_err(|e| Error::other(format!("Failed to acquire semaphore for bucket heal: {e}")))?;
 
                 if cancel_token.is_cancelled() {
                     return Err(Error::TaskCancelled);
@@ -432,7 +462,10 @@ impl ErasureSetHealer {
             let semaphore = semaphore.clone();
 
             async move {
-                let _permit = semaphore.acquire().await.unwrap();
+                let _permit = semaphore
+                    .acquire()
+                    .await
+                    .map_err(|e| Error::other(format!("Failed to acquire semaphore for object heal: {e}")))?;
 
                 match storage.heal_object(&bucket, &object, None, &heal_opts).await {
                     Ok((_result, None)) => {

crates/ahm/src/heal/event.rs

@@ -12,7 +12,8 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use crate::heal::task::{HealOptions, HealPriority, HealRequest, HealType};
+use crate::heal::{HealOptions, HealPriority, HealRequest, HealType};
+use crate::{Error, Result};
 use rustfs_ecstore::disk::endpoint::Endpoint;
 use serde::{Deserialize, Serialize};
 use std::time::SystemTime;
@@ -104,7 +105,7 @@ pub enum HealEvent {
 
 impl HealEvent {
     /// Convert HealEvent to HealRequest
-    pub fn to_heal_request(&self) -> HealRequest {
+    pub fn to_heal_request(&self) -> Result<HealRequest> {
         match self {
             HealEvent::ObjectCorruption {
                 bucket,
@@ -112,7 +113,7 @@ impl HealEvent {
                 version_id,
                 severity,
                 ..
-            } => HealRequest::new(
+            } => Ok(HealRequest::new(
                 HealType::Object {
                     bucket: bucket.clone(),
                     object: object.clone(),
@@ -120,13 +121,13 @@ impl HealEvent {
                 },
                 HealOptions::default(),
                 Self::severity_to_priority(severity),
-            ),
+            )),
             HealEvent::ObjectMissing {
                 bucket,
                 object,
                 version_id,
                 ..
-            } => HealRequest::new(
+            } => Ok(HealRequest::new(
                 HealType::Object {
                     bucket: bucket.clone(),
                     object: object.clone(),
@@ -134,34 +135,38 @@ impl HealEvent {
                 },
                 HealOptions::default(),
                 HealPriority::High,
-            ),
-            HealEvent::MetadataCorruption { bucket, object, .. } => HealRequest::new(
+            )),
+            HealEvent::MetadataCorruption { bucket, object, .. } => Ok(HealRequest::new(
                 HealType::Metadata {
                     bucket: bucket.clone(),
                     object: object.clone(),
                 },
                 HealOptions::default(),
                 HealPriority::High,
-            ),
+            )),
             HealEvent::DiskStatusChange { endpoint, .. } => {
                 // Convert disk status change to erasure set heal
                 // Note: This requires access to storage to get bucket list, which is not available here
                 // The actual bucket list will need to be provided by the caller or retrieved differently
-                HealRequest::new(
+                let set_disk_id = crate::heal::utils::format_set_disk_id_from_i32(endpoint.pool_idx, endpoint.set_idx)
+                    .ok_or_else(|| Error::InvalidHealType {
+                        heal_type: format!("erasure-set(pool={}, set={})", endpoint.pool_idx, endpoint.set_idx),
+                    })?;
+                Ok(HealRequest::new(
                     HealType::ErasureSet {
                         buckets: vec![], // Empty bucket list - caller should populate this
-                        set_disk_id: format!("{}_{}", endpoint.pool_idx, endpoint.set_idx),
+                        set_disk_id,
                     },
                     HealOptions::default(),
                     HealPriority::High,
-                )
+                ))
             }
             HealEvent::ECDecodeFailure {
                 bucket,
                 object,
                 version_id,
                 ..
-            } => HealRequest::new(
+            } => Ok(HealRequest::new(
                 HealType::ECDecode {
                     bucket: bucket.clone(),
                     object: object.clone(),
@@ -169,13 +174,13 @@ impl HealEvent {
                 },
                 HealOptions::default(),
                 HealPriority::Urgent,
-            ),
+            )),
             HealEvent::ChecksumMismatch {
                 bucket,
                 object,
                 version_id,
                 ..
-            } => HealRequest::new(
+            } => Ok(HealRequest::new(
                 HealType::Object {
                     bucket: bucket.clone(),
                     object: object.clone(),
@@ -183,17 +188,19 @@ impl HealEvent {
                 },
                 HealOptions::default(),
                 HealPriority::High,
-            ),
-            HealEvent::BucketMetadataCorruption { bucket, .. } => {
-                HealRequest::new(HealType::Bucket { bucket: bucket.clone() }, HealOptions::default(), HealPriority::High)
-            }
-            HealEvent::MRFMetadataCorruption { meta_path, .. } => HealRequest::new(
+            )),
+            HealEvent::BucketMetadataCorruption { bucket, .. } => Ok(HealRequest::new(
+                HealType::Bucket { bucket: bucket.clone() },
+                HealOptions::default(),
+                HealPriority::High,
+            )),
+            HealEvent::MRFMetadataCorruption { meta_path, .. } => Ok(HealRequest::new(
                 HealType::MRF {
                     meta_path: meta_path.clone(),
                 },
                 HealOptions::default(),
                 HealPriority::High,
-            ),
+            )),
         }
     }
 
@@ -357,3 +364,319 @@ impl Default for HealEventHandler {
         Self::new(1000)
     }
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use crate::heal::task::{HealPriority, HealType};
+
+    #[test]
+    fn test_heal_event_object_corruption_to_request() {
+        let event = HealEvent::ObjectCorruption {
+            bucket: "test-bucket".to_string(),
+            object: "test-object".to_string(),
+            version_id: None,
+            corruption_type: CorruptionType::DataCorruption,
+            severity: Severity::High,
+        };
+
+        let request = event.to_heal_request().unwrap();
+        assert!(matches!(request.heal_type, HealType::Object { .. }));
+        assert_eq!(request.priority, HealPriority::High);
+    }
+
+    #[test]
+    fn test_heal_event_object_missing_to_request() {
+        let event = HealEvent::ObjectMissing {
+            bucket: "test-bucket".to_string(),
+            object: "test-object".to_string(),
+            version_id: Some("v1".to_string()),
+            expected_locations: vec![0, 1],
+            available_locations: vec![2, 3],
+        };
+
+        let request = event.to_heal_request().unwrap();
+        assert!(matches!(request.heal_type, HealType::Object { .. }));
+        assert_eq!(request.priority, HealPriority::High);
+    }
+
+    #[test]
+    fn test_heal_event_metadata_corruption_to_request() {
+        let event = HealEvent::MetadataCorruption {
+            bucket: "test-bucket".to_string(),
+            object: "test-object".to_string(),
+            corruption_type: CorruptionType::MetadataCorruption,
+        };
+
+        let request = event.to_heal_request().unwrap();
+        assert!(matches!(request.heal_type, HealType::Metadata { .. }));
+        assert_eq!(request.priority, HealPriority::High);
+    }
+
+    #[test]
+    fn test_heal_event_ec_decode_failure_to_request() {
+        let event = HealEvent::ECDecodeFailure {
+            bucket: "test-bucket".to_string(),
+            object: "test-object".to_string(),
+            version_id: None,
+            missing_shards: vec![0, 1],
+            available_shards: vec![2, 3, 4],
+        };
+
+        let request = event.to_heal_request().unwrap();
+        assert!(matches!(request.heal_type, HealType::ECDecode { .. }));
+        assert_eq!(request.priority, HealPriority::Urgent);
+    }
+
+    #[test]
+    fn test_heal_event_checksum_mismatch_to_request() {
+        let event = HealEvent::ChecksumMismatch {
+            bucket: "test-bucket".to_string(),
+            object: "test-object".to_string(),
+            version_id: None,
+            expected_checksum: "abc123".to_string(),
+            actual_checksum: "def456".to_string(),
+        };
+
+        let request = event.to_heal_request().unwrap();
+        assert!(matches!(request.heal_type, HealType::Object { .. }));
+        assert_eq!(request.priority, HealPriority::High);
+    }
+
+    #[test]
+    fn test_heal_event_bucket_metadata_corruption_to_request() {
+        let event = HealEvent::BucketMetadataCorruption {
+            bucket: "test-bucket".to_string(),
+            corruption_type: CorruptionType::MetadataCorruption,
+        };
+
+        let request = event.to_heal_request().unwrap();
+        assert!(matches!(request.heal_type, HealType::Bucket { .. }));
+        assert_eq!(request.priority, HealPriority::High);
+    }
+
+    #[test]
+    fn test_heal_event_mrf_metadata_corruption_to_request() {
+        let event = HealEvent::MRFMetadataCorruption {
+            meta_path: "test-bucket/test-object".to_string(),
+            corruption_type: CorruptionType::MetadataCorruption,
+        };
+
+        let request = event.to_heal_request().unwrap();
+        assert!(matches!(request.heal_type, HealType::MRF { .. }));
+        assert_eq!(request.priority, HealPriority::High);
+    }
+
+    #[test]
+    fn test_heal_event_severity_to_priority() {
+        let event_low = HealEvent::ObjectCorruption {
+            bucket: "test".to_string(),
+            object: "test".to_string(),
+            version_id: None,
+            corruption_type: CorruptionType::DataCorruption,
+            severity: Severity::Low,
+        };
+        let request = event_low.to_heal_request().unwrap();
+        assert_eq!(request.priority, HealPriority::Low);
+
+        let event_medium = HealEvent::ObjectCorruption {
+            bucket: "test".to_string(),
+            object: "test".to_string(),
+            version_id: None,
+            corruption_type: CorruptionType::DataCorruption,
+            severity: Severity::Medium,
+        };
+        let request = event_medium.to_heal_request().unwrap();
+        assert_eq!(request.priority, HealPriority::Normal);
+
+        let event_high = HealEvent::ObjectCorruption {
+            bucket: "test".to_string(),
+            object: "test".to_string(),
+            version_id: None,
+            corruption_type: CorruptionType::DataCorruption,
+            severity: Severity::High,
+        };
+        let request = event_high.to_heal_request().unwrap();
+        assert_eq!(request.priority, HealPriority::High);
+
+        let event_critical = HealEvent::ObjectCorruption {
+            bucket: "test".to_string(),
+            object: "test".to_string(),
+            version_id: None,
+            corruption_type: CorruptionType::DataCorruption,
+            severity: Severity::Critical,
+        };
+        let request = event_critical.to_heal_request().unwrap();
+        assert_eq!(request.priority, HealPriority::Urgent);
+    }
+
+    #[test]
+    fn test_heal_event_description() {
+        let event = HealEvent::ObjectCorruption {
+            bucket: "test-bucket".to_string(),
+            object: "test-object".to_string(),
+            version_id: None,
+            corruption_type: CorruptionType::DataCorruption,
+            severity: Severity::High,
+        };
+
+        let desc = event.description();
+        assert!(desc.contains("Object corruption detected"));
+        assert!(desc.contains("test-bucket/test-object"));
+        assert!(desc.contains("DataCorruption"));
+    }
+
+    #[test]
+    fn test_heal_event_severity() {
+        let event = HealEvent::ECDecodeFailure {
+            bucket: "test".to_string(),
+            object: "test".to_string(),
+            version_id: None,
+            missing_shards: vec![],
+            available_shards: vec![],
+        };
+        assert_eq!(event.severity(), Severity::Critical);
+
+        let event = HealEvent::ObjectMissing {
+            bucket: "test".to_string(),
+            object: "test".to_string(),
+            version_id: None,
+            expected_locations: vec![],
+            available_locations: vec![],
+        };
+        assert_eq!(event.severity(), Severity::High);
+    }
+
+    #[test]
+    fn test_heal_event_handler_new() {
+        let handler = HealEventHandler::new(10);
+        assert_eq!(handler.event_count(), 0);
+        assert_eq!(handler.max_events, 10);
+    }
+
+    #[test]
+    fn test_heal_event_handler_default() {
+        let handler = HealEventHandler::default();
+        assert_eq!(handler.max_events, 1000);
+    }
+
+    #[test]
+    fn test_heal_event_handler_add_event() {
+        let mut handler = HealEventHandler::new(3);
+        let event = HealEvent::ObjectCorruption {
+            bucket: "test".to_string(),
+            object: "test".to_string(),
+            version_id: None,
+            corruption_type: CorruptionType::DataCorruption,
+            severity: Severity::High,
+        };
+
+        handler.add_event(event.clone());
+        assert_eq!(handler.event_count(), 1);
+
+        handler.add_event(event.clone());
+        handler.add_event(event.clone());
+        assert_eq!(handler.event_count(), 3);
+    }
+
+    #[test]
+    fn test_heal_event_handler_max_events() {
+        let mut handler = HealEventHandler::new(2);
+        let event = HealEvent::ObjectCorruption {
+            bucket: "test".to_string(),
+            object: "test".to_string(),
+            version_id: None,
+            corruption_type: CorruptionType::DataCorruption,
+            severity: Severity::High,
+        };
+
+        handler.add_event(event.clone());
+        handler.add_event(event.clone());
+        handler.add_event(event.clone()); // Should remove oldest
+
+        assert_eq!(handler.event_count(), 2);
+    }
+
+    #[test]
+    fn test_heal_event_handler_get_events() {
+        let mut handler = HealEventHandler::new(10);
+        let event = HealEvent::ObjectCorruption {
+            bucket: "test".to_string(),
+            object: "test".to_string(),
+            version_id: None,
+            corruption_type: CorruptionType::DataCorruption,
+            severity: Severity::High,
+        };
+
+        handler.add_event(event.clone());
+        handler.add_event(event.clone());
+
+        let events = handler.get_events();
+        assert_eq!(events.len(), 2);
+    }
+
+    #[test]
+    fn test_heal_event_handler_clear_events() {
+        let mut handler = HealEventHandler::new(10);
+        let event = HealEvent::ObjectCorruption {
+            bucket: "test".to_string(),
+            object: "test".to_string(),
+            version_id: None,
+            corruption_type: CorruptionType::DataCorruption,
+            severity: Severity::High,
+        };
+
+        handler.add_event(event);
+        assert_eq!(handler.event_count(), 1);
+
+        handler.clear_events();
+        assert_eq!(handler.event_count(), 0);
+    }
+
+    #[test]
+    fn test_heal_event_handler_filter_by_severity() {
+        let mut handler = HealEventHandler::new(10);
+        handler.add_event(HealEvent::ObjectCorruption {
+            bucket: "test".to_string(),
+            object: "test".to_string(),
+            version_id: None,
+            corruption_type: CorruptionType::DataCorruption,
+            severity: Severity::Low,
+        });
+        handler.add_event(HealEvent::ECDecodeFailure {
+            bucket: "test".to_string(),
+            object: "test".to_string(),
+            version_id: None,
+            missing_shards: vec![],
+            available_shards: vec![],
+        });
+
+        let high_severity = handler.filter_by_severity(Severity::High);
+        assert_eq!(high_severity.len(), 1); // Only ECDecodeFailure is Critical >= High
+    }
+
+    #[test]
+    fn test_heal_event_handler_filter_by_type() {
+        let mut handler = HealEventHandler::new(10);
+        handler.add_event(HealEvent::ObjectCorruption {
+            bucket: "test".to_string(),
+            object: "test".to_string(),
+            version_id: None,
+            corruption_type: CorruptionType::DataCorruption,
+            severity: Severity::High,
+        });
+        handler.add_event(HealEvent::ObjectMissing {
+            bucket: "test".to_string(),
+            object: "test".to_string(),
+            version_id: None,
+            expected_locations: vec![],
+            available_locations: vec![],
+        });
+
+        let corruption_events = handler.filter_by_type("ObjectCorruption");
+        assert_eq!(corruption_events.len(), 1);
+
+        let missing_events = handler.filter_by_type("ObjectMissing");
+        assert_eq!(missing_events.len(), 1);
+    }
+}

crates/ahm/src/heal/manager.rs

@@ -12,17 +12,17 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use crate::error::{Error, Result};
 use crate::heal::{
     progress::{HealProgress, HealStatistics},
     storage::HealStorageAPI,
     task::{HealOptions, HealPriority, HealRequest, HealTask, HealTaskStatus, HealType},
 };
+use crate::{Error, Result};
 use rustfs_ecstore::disk::DiskAPI;
 use rustfs_ecstore::disk::error::DiskError;
 use rustfs_ecstore::global::GLOBAL_LOCAL_DISK_MAP;
 use std::{
-    collections::{HashMap, VecDeque},
+    collections::{BinaryHeap, HashMap, HashSet},
     sync::Arc,
     time::{Duration, SystemTime},
 };
@@ -33,6 +33,151 @@ use tokio::{
 use tokio_util::sync::CancellationToken;
 use tracing::{error, info, warn};
 
+/// Priority queue wrapper for heal requests
+/// Uses BinaryHeap for priority-based ordering while maintaining FIFO for same-priority items
+#[derive(Debug)]
+struct PriorityHealQueue {
+    /// Heap of (priority, sequence, request) tuples
+    heap: BinaryHeap<PriorityQueueItem>,
+    /// Sequence counter for FIFO ordering within same priority
+    sequence: u64,
+    /// Set of request keys to prevent duplicates
+    dedup_keys: HashSet<String>,
+}
+
+/// Wrapper for heap items to implement proper ordering
+#[derive(Debug)]
+struct PriorityQueueItem {
+    priority: HealPriority,
+    sequence: u64,
+    request: HealRequest,
+}
+
+impl Eq for PriorityQueueItem {}
+
+impl PartialEq for PriorityQueueItem {
+    fn eq(&self, other: &Self) -> bool {
+        self.priority == other.priority && self.sequence == other.sequence
+    }
+}
+
+impl Ord for PriorityQueueItem {
+    fn cmp(&self, other: &Self) -> std::cmp::Ordering {
+        // First compare by priority (higher priority first)
+        match self.priority.cmp(&other.priority) {
+            std::cmp::Ordering::Equal => {
+                // If priorities are equal, use sequence for FIFO (lower sequence first)
+                other.sequence.cmp(&self.sequence)
+            }
+            ordering => ordering,
+        }
+    }
+}
+
+impl PartialOrd for PriorityQueueItem {
+    fn partial_cmp(&self, other: &Self) -> Option<std::cmp::Ordering> {
+        Some(self.cmp(other))
+    }
+}
+
+impl PriorityHealQueue {
+    fn new() -> Self {
+        Self {
+            heap: BinaryHeap::new(),
+            sequence: 0,
+            dedup_keys: HashSet::new(),
+        }
+    }
+
+    fn len(&self) -> usize {
+        self.heap.len()
+    }
+
+    fn is_empty(&self) -> bool {
+        self.heap.is_empty()
+    }
+
+    fn push(&mut self, request: HealRequest) -> bool {
+        let key = Self::make_dedup_key(&request);
+
+        // Check for duplicates
+        if self.dedup_keys.contains(&key) {
+            return false; // Duplicate request, don't add
+        }
+
+        self.dedup_keys.insert(key);
+        self.sequence += 1;
+        self.heap.push(PriorityQueueItem {
+            priority: request.priority,
+            sequence: self.sequence,
+            request,
+        });
+        true
+    }
+
+    /// Get statistics about queue contents by priority
+    fn get_priority_stats(&self) -> HashMap<HealPriority, usize> {
+        let mut stats = HashMap::new();
+        for item in &self.heap {
+            *stats.entry(item.priority).or_insert(0) += 1;
+        }
+        stats
+    }
+
+    fn pop(&mut self) -> Option<HealRequest> {
+        self.heap.pop().map(|item| {
+            let key = Self::make_dedup_key(&item.request);
+            self.dedup_keys.remove(&key);
+            item.request
+        })
+    }
+
+    /// Create a deduplication key from a heal request
+    fn make_dedup_key(request: &HealRequest) -> String {
+        match &request.heal_type {
+            HealType::Object {
+                bucket,
+                object,
+                version_id,
+            } => {
+                format!("object:{}:{}:{}", bucket, object, version_id.as_deref().unwrap_or(""))
+            }
+            HealType::Bucket { bucket } => {
+                format!("bucket:{}", bucket)
+            }
+            HealType::ErasureSet { set_disk_id, .. } => {
+                format!("erasure_set:{}", set_disk_id)
+            }
+            HealType::Metadata { bucket, object } => {
+                format!("metadata:{}:{}", bucket, object)
+            }
+            HealType::MRF { meta_path } => {
+                format!("mrf:{}", meta_path)
+            }
+            HealType::ECDecode {
+                bucket,
+                object,
+                version_id,
+            } => {
+                format!("ecdecode:{}:{}:{}", bucket, object, version_id.as_deref().unwrap_or(""))
+            }
+        }
+    }
+
+    /// Check if a request with the same key already exists in the queue
+    #[allow(dead_code)]
+    fn contains_key(&self, request: &HealRequest) -> bool {
+        let key = Self::make_dedup_key(request);
+        self.dedup_keys.contains(&key)
+    }
+
+    /// Check if an erasure set heal request for a specific set_disk_id exists
+    fn contains_erasure_set(&self, set_disk_id: &str) -> bool {
+        let key = format!("erasure_set:{}", set_disk_id);
+        self.dedup_keys.contains(&key)
+    }
+}
+
 /// Heal config
 #[derive(Debug, Clone)]
 pub struct HealConfig {
@@ -85,8 +230,8 @@ pub struct HealManager {
     state: Arc<RwLock<HealState>>,
     /// Active heal tasks
     active_heals: Arc<Mutex<HashMap<String, Arc<HealTask>>>>,
-    /// Heal queue
-    heal_queue: Arc<Mutex<VecDeque<HealRequest>>>,
+    /// Heal queue (priority-based)
+    heal_queue: Arc<Mutex<PriorityHealQueue>>,
     /// Storage layer interface
     storage: Arc<dyn HealStorageAPI>,
     /// Cancel token
@@ -103,7 +248,7 @@ impl HealManager {
             config: Arc::new(RwLock::new(config)),
             state: Arc::new(RwLock::new(HealState::default())),
             active_heals: Arc::new(Mutex::new(HashMap::new())),
-            heal_queue: Arc::new(Mutex::new(VecDeque::new())),
+            heal_queue: Arc::new(Mutex::new(PriorityHealQueue::new())),
             storage,
             cancel_token: CancellationToken::new(),
             statistics: Arc::new(RwLock::new(HealStatistics::new())),
@@ -161,17 +306,54 @@ impl HealManager {
         let config = self.config.read().await;
         let mut queue = self.heal_queue.lock().await;
 
-        if queue.len() >= config.queue_size {
+        let queue_len = queue.len();
+        let queue_capacity = config.queue_size;
+
+        if queue_len >= queue_capacity {
             return Err(Error::ConfigurationError {
-                message: "Heal queue is full".to_string(),
+                message: format!("Heal queue is full ({}/{})", queue_len, queue_capacity),
             });
         }
 
+        // Warn when queue is getting full (>80% capacity)
+        let capacity_threshold = (queue_capacity as f64 * 0.8) as usize;
+        if queue_len >= capacity_threshold {
+            warn!(
+                "Heal queue is {}% full ({}/{}). Consider increasing queue size or processing capacity.",
+                (queue_len * 100) / queue_capacity,
+                queue_len,
+                queue_capacity
+            );
+        }
+
         let request_id = request.id.clone();
-        queue.push_back(request);
+        let priority = request.priority;
+
+        // Try to push the request; if it's a duplicate, still return the request_id
+        let is_new = queue.push(request);
+
+        // Log queue statistics periodically (when adding high/urgent priority items)
+        if matches!(priority, HealPriority::High | HealPriority::Urgent) {
+            let stats = queue.get_priority_stats();
+            info!(
+                "Heal queue stats after adding {:?} priority request: total={}, urgent={}, high={}, normal={}, low={}",
+                priority,
+                queue_len + 1,
+                stats.get(&HealPriority::Urgent).unwrap_or(&0),
+                stats.get(&HealPriority::High).unwrap_or(&0),
+                stats.get(&HealPriority::Normal).unwrap_or(&0),
+                stats.get(&HealPriority::Low).unwrap_or(&0)
+            );
+        }
+
         drop(queue);
 
-        info!("Submitted heal request: {}", request_id);
+        if is_new {
+            info!("Submitted heal request: {} with priority: {:?}", request_id, priority);
+        } else {
+            info!("Heal request already queued (duplicate): {}", request_id);
+        }
+
         Ok(request_id)
     }
 
@@ -310,17 +492,30 @@ impl HealManager {
 
                         // Create erasure set heal requests for each endpoint
                         for ep in endpoints {
+                            let Some(set_disk_id) =
+                                crate::heal::utils::format_set_disk_id_from_i32(ep.pool_idx, ep.set_idx)
+                            else {
+                                warn!("Skipping endpoint {} without valid pool/set index", ep);
+                                continue;
+                            };
                             // skip if already queued or healing
+                            // Use consistent lock order: queue first, then active_heals to avoid deadlock
                             let mut skip = false;
                             {
                                 let queue = heal_queue.lock().await;
-                                if queue.iter().any(|req| matches!(&req.heal_type, crate::heal::task::HealType::ErasureSet { set_disk_id, .. } if set_disk_id == &format!("{}_{}", ep.pool_idx, ep.set_idx))) {
+                                if queue.contains_erasure_set(&set_disk_id) {
                                     skip = true;
                                 }
                             }
                             if !skip {
                                 let active = active_heals.lock().await;
-                                if active.values().any(|task| matches!(&task.heal_type, crate::heal::task::HealType::ErasureSet { set_disk_id, .. } if set_disk_id == &format!("{}_{}", ep.pool_idx, ep.set_idx))) {
+                                if active.values().any(|task| {
+                                    matches!(
+                                        &task.heal_type,
+                                        crate::heal::task::HealType::ErasureSet { set_disk_id: active_id, .. }
+                                        if active_id == &set_disk_id
+                                    )
+                                }) {
                                     skip = true;
                                 }
                             }
@@ -330,17 +525,16 @@ impl HealManager {
                             }
 
                             // enqueue erasure set heal request for this disk
-                            let set_disk_id = format!("pool_{}_set_{}", ep.pool_idx, ep.set_idx);
                             let req = HealRequest::new(
                                 HealType::ErasureSet {
                                     buckets: buckets.clone(),
-                                    set_disk_id: set_disk_id.clone()
+                                    set_disk_id: set_disk_id.clone(),
                                 },
                                 HealOptions::default(),
                                 HealPriority::Normal,
                             );
                             let mut queue = heal_queue.lock().await;
-                            queue.push_back(req);
+                            queue.push(req);
                             info!("Enqueued auto erasure set heal for endpoint: {} (set_disk_id: {})", ep, set_disk_id);
                         }
                     }
@@ -351,8 +545,9 @@ impl HealManager {
     }
 
     /// Process heal queue
+    /// Processes multiple tasks per cycle when capacity allows and queue has high-priority items
     async fn process_heal_queue(
-        heal_queue: &Arc<Mutex<VecDeque<HealRequest>>>,
+        heal_queue: &Arc<Mutex<PriorityHealQueue>>,
         active_heals: &Arc<Mutex<HashMap<String, Arc<HealTask>>>>,
         config: &Arc<RwLock<HealConfig>>,
         statistics: &Arc<RwLock<HealStatistics>>,
@@ -361,51 +556,83 @@ impl HealManager {
         let config = config.read().await;
         let mut active_heals_guard = active_heals.lock().await;
 
-        // check if new heal tasks can be started
-        if active_heals_guard.len() >= config.max_concurrent_heals {
+        // Check if new heal tasks can be started
+        let active_count = active_heals_guard.len();
+        if active_count >= config.max_concurrent_heals {
             return;
         }
 
+        // Calculate how many tasks we can start this cycle
+        let available_slots = config.max_concurrent_heals - active_count;
+
         let mut queue = heal_queue.lock().await;
-        if let Some(request) = queue.pop_front() {
-            let task = Arc::new(HealTask::from_request(request, storage.clone()));
-            let task_id = task.id.clone();
-            active_heals_guard.insert(task_id.clone(), task.clone());
-            drop(active_heals_guard);
-            let active_heals_clone = active_heals.clone();
-            let statistics_clone = statistics.clone();
+        let queue_len = queue.len();
 
-            // start heal task
-            tokio::spawn(async move {
-                info!("Starting heal task: {}", task_id);
-                let result = task.execute().await;
-                match result {
-                    Ok(_) => {
-                        info!("Heal task completed successfully: {}", task_id);
-                    }
-                    Err(e) => {
-                        error!("Heal task failed: {} - {}", task_id, e);
-                    }
-                }
-                let mut active_heals_guard = active_heals_clone.lock().await;
-                if let Some(completed_task) = active_heals_guard.remove(&task_id) {
-                    // update statistics
-                    let mut stats = statistics_clone.write().await;
-                    match completed_task.get_status().await {
-                        HealTaskStatus::Completed => {
-                            stats.update_task_completion(true);
+        if queue_len == 0 {
+            return;
+        }
+
+        // Process multiple tasks if:
+        // 1. We have available slots
+        // 2. Queue is not empty
+        // Prioritize urgent/high priority tasks by processing up to 2 tasks per cycle if available
+        let tasks_to_process = if queue_len > 0 {
+            std::cmp::min(available_slots, std::cmp::min(2, queue_len))
+        } else {
+            0
+        };
+
+        for _ in 0..tasks_to_process {
+            if let Some(request) = queue.pop() {
+                let task_priority = request.priority;
+                let task = Arc::new(HealTask::from_request(request, storage.clone()));
+                let task_id = task.id.clone();
+                active_heals_guard.insert(task_id.clone(), task.clone());
+                let active_heals_clone = active_heals.clone();
+                let statistics_clone = statistics.clone();
+
+                // start heal task
+                tokio::spawn(async move {
+                    info!("Starting heal task: {} with priority: {:?}", task_id, task_priority);
+                    let result = task.execute().await;
+                    match result {
+                        Ok(_) => {
+                            info!("Heal task completed successfully: {}", task_id);
                         }
-                        _ => {
-                            stats.update_task_completion(false);
+                        Err(e) => {
+                            error!("Heal task failed: {} - {}", task_id, e);
                         }
                     }
-                    stats.update_running_tasks(active_heals_guard.len() as u64);
-                }
-            });
+                    let mut active_heals_guard = active_heals_clone.lock().await;
+                    if let Some(completed_task) = active_heals_guard.remove(&task_id) {
+                        // update statistics
+                        let mut stats = statistics_clone.write().await;
+                        match completed_task.get_status().await {
+                            HealTaskStatus::Completed => {
+                                stats.update_task_completion(true);
+                            }
+                            _ => {
+                                stats.update_task_completion(false);
+                            }
+                        }
+                        stats.update_running_tasks(active_heals_guard.len() as u64);
+                    }
+                });
+            } else {
+                break;
+            }
+        }
 
-            // update statistics
-            let mut stats = statistics.write().await;
-            stats.total_tasks += 1;
+        // Update statistics for all started tasks
+        let mut stats = statistics.write().await;
+        stats.total_tasks += tasks_to_process as u64;
+
+        // Log queue status if items remain
+        if !queue.is_empty() {
+            let remaining = queue.len();
+            if remaining > 10 {
+                info!("Heal queue has {} pending requests, {} tasks active", remaining, active_heals_guard.len());
+            }
         }
     }
 }
@@ -420,3 +647,333 @@ impl std::fmt::Debug for HealManager {
             .finish()
     }
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use crate::heal::task::{HealOptions, HealPriority, HealRequest, HealType};
+
+    #[test]
+    fn test_priority_queue_ordering() {
+        let mut queue = PriorityHealQueue::new();
+
+        // Add requests with different priorities
+        let low_req = HealRequest::new(
+            HealType::Bucket {
+                bucket: "bucket1".to_string(),
+            },
+            HealOptions::default(),
+            HealPriority::Low,
+        );
+
+        let normal_req = HealRequest::new(
+            HealType::Bucket {
+                bucket: "bucket2".to_string(),
+            },
+            HealOptions::default(),
+            HealPriority::Normal,
+        );
+
+        let high_req = HealRequest::new(
+            HealType::Bucket {
+                bucket: "bucket3".to_string(),
+            },
+            HealOptions::default(),
+            HealPriority::High,
+        );
+
+        let urgent_req = HealRequest::new(
+            HealType::Bucket {
+                bucket: "bucket4".to_string(),
+            },
+            HealOptions::default(),
+            HealPriority::Urgent,
+        );
+
+        // Add in random order: low, high, normal, urgent
+        assert!(queue.push(low_req));
+        assert!(queue.push(high_req));
+        assert!(queue.push(normal_req));
+        assert!(queue.push(urgent_req));
+
+        assert_eq!(queue.len(), 4);
+
+        // Should pop in priority order: urgent, high, normal, low
+        let popped1 = queue.pop().unwrap();
+        assert_eq!(popped1.priority, HealPriority::Urgent);
+
+        let popped2 = queue.pop().unwrap();
+        assert_eq!(popped2.priority, HealPriority::High);
+
+        let popped3 = queue.pop().unwrap();
+        assert_eq!(popped3.priority, HealPriority::Normal);
+
+        let popped4 = queue.pop().unwrap();
+        assert_eq!(popped4.priority, HealPriority::Low);
+
+        assert_eq!(queue.len(), 0);
+    }
+
+    #[test]
+    fn test_priority_queue_fifo_same_priority() {
+        let mut queue = PriorityHealQueue::new();
+
+        // Add multiple requests with same priority
+        let req1 = HealRequest::new(
+            HealType::Bucket {
+                bucket: "bucket1".to_string(),
+            },
+            HealOptions::default(),
+            HealPriority::Normal,
+        );
+
+        let req2 = HealRequest::new(
+            HealType::Bucket {
+                bucket: "bucket2".to_string(),
+            },
+            HealOptions::default(),
+            HealPriority::Normal,
+        );
+
+        let req3 = HealRequest::new(
+            HealType::Bucket {
+                bucket: "bucket3".to_string(),
+            },
+            HealOptions::default(),
+            HealPriority::Normal,
+        );
+
+        let id1 = req1.id.clone();
+        let id2 = req2.id.clone();
+        let id3 = req3.id.clone();
+
+        assert!(queue.push(req1));
+        assert!(queue.push(req2));
+        assert!(queue.push(req3));
+
+        // Should maintain FIFO order for same priority
+        let popped1 = queue.pop().unwrap();
+        assert_eq!(popped1.id, id1);
+
+        let popped2 = queue.pop().unwrap();
+        assert_eq!(popped2.id, id2);
+
+        let popped3 = queue.pop().unwrap();
+        assert_eq!(popped3.id, id3);
+    }
+
+    #[test]
+    fn test_priority_queue_deduplication() {
+        let mut queue = PriorityHealQueue::new();
+
+        let req1 = HealRequest::new(
+            HealType::Object {
+                bucket: "bucket1".to_string(),
+                object: "object1".to_string(),
+                version_id: None,
+            },
+            HealOptions::default(),
+            HealPriority::Normal,
+        );
+
+        let req2 = HealRequest::new(
+            HealType::Object {
+                bucket: "bucket1".to_string(),
+                object: "object1".to_string(),
+                version_id: None,
+            },
+            HealOptions::default(),
+            HealPriority::High,
+        );
+
+        // First request should be added
+        assert!(queue.push(req1));
+        assert_eq!(queue.len(), 1);
+
+        // Second request with same object should be rejected (duplicate)
+        assert!(!queue.push(req2));
+        assert_eq!(queue.len(), 1);
+    }
+
+    #[test]
+    fn test_priority_queue_contains_erasure_set() {
+        let mut queue = PriorityHealQueue::new();
+
+        let req = HealRequest::new(
+            HealType::ErasureSet {
+                buckets: vec!["bucket1".to_string()],
+                set_disk_id: "pool_0_set_1".to_string(),
+            },
+            HealOptions::default(),
+            HealPriority::Normal,
+        );
+
+        assert!(queue.push(req));
+        assert!(queue.contains_erasure_set("pool_0_set_1"));
+        assert!(!queue.contains_erasure_set("pool_0_set_2"));
+    }
+
+    #[test]
+    fn test_priority_queue_dedup_key_generation() {
+        // Test different heal types generate different keys
+        let obj_req = HealRequest::new(
+            HealType::Object {
+                bucket: "bucket1".to_string(),
+                object: "object1".to_string(),
+                version_id: None,
+            },
+            HealOptions::default(),
+            HealPriority::Normal,
+        );
+
+        let bucket_req = HealRequest::new(
+            HealType::Bucket {
+                bucket: "bucket1".to_string(),
+            },
+            HealOptions::default(),
+            HealPriority::Normal,
+        );
+
+        let erasure_req = HealRequest::new(
+            HealType::ErasureSet {
+                buckets: vec!["bucket1".to_string()],
+                set_disk_id: "pool_0_set_1".to_string(),
+            },
+            HealOptions::default(),
+            HealPriority::Normal,
+        );
+
+        let obj_key = PriorityHealQueue::make_dedup_key(&obj_req);
+        let bucket_key = PriorityHealQueue::make_dedup_key(&bucket_req);
+        let erasure_key = PriorityHealQueue::make_dedup_key(&erasure_req);
+
+        // All keys should be different
+        assert_ne!(obj_key, bucket_key);
+        assert_ne!(obj_key, erasure_key);
+        assert_ne!(bucket_key, erasure_key);
+
+        assert!(obj_key.starts_with("object:"));
+        assert!(bucket_key.starts_with("bucket:"));
+        assert!(erasure_key.starts_with("erasure_set:"));
+    }
+
+    #[test]
+    fn test_priority_queue_mixed_priorities_and_types() {
+        let mut queue = PriorityHealQueue::new();
+
+        // Add various requests
+        let requests = vec![
+            (
+                HealType::Object {
+                    bucket: "b1".to_string(),
+                    object: "o1".to_string(),
+                    version_id: None,
+                },
+                HealPriority::Low,
+            ),
+            (
+                HealType::Bucket {
+                    bucket: "b2".to_string(),
+                },
+                HealPriority::Urgent,
+            ),
+            (
+                HealType::ErasureSet {
+                    buckets: vec!["b3".to_string()],
+                    set_disk_id: "pool_0_set_1".to_string(),
+                },
+                HealPriority::Normal,
+            ),
+            (
+                HealType::Object {
+                    bucket: "b4".to_string(),
+                    object: "o4".to_string(),
+                    version_id: None,
+                },
+                HealPriority::High,
+            ),
+        ];
+
+        for (heal_type, priority) in requests {
+            let req = HealRequest::new(heal_type, HealOptions::default(), priority);
+            queue.push(req);
+        }
+
+        assert_eq!(queue.len(), 4);
+
+        // Check they come out in priority order
+        let priorities: Vec<HealPriority> = (0..4).filter_map(|_| queue.pop().map(|r| r.priority)).collect();
+
+        assert_eq!(
+            priorities,
+            vec![
+                HealPriority::Urgent,
+                HealPriority::High,
+                HealPriority::Normal,
+                HealPriority::Low,
+            ]
+        );
+    }
+
+    #[test]
+    fn test_priority_queue_stats() {
+        let mut queue = PriorityHealQueue::new();
+
+        // Add requests with different priorities
+        for _ in 0..3 {
+            queue.push(HealRequest::new(
+                HealType::Bucket {
+                    bucket: format!("bucket-low-{}", queue.len()),
+                },
+                HealOptions::default(),
+                HealPriority::Low,
+            ));
+        }
+
+        for _ in 0..2 {
+            queue.push(HealRequest::new(
+                HealType::Bucket {
+                    bucket: format!("bucket-normal-{}", queue.len()),
+                },
+                HealOptions::default(),
+                HealPriority::Normal,
+            ));
+        }
+
+        queue.push(HealRequest::new(
+            HealType::Bucket {
+                bucket: "bucket-high".to_string(),
+            },
+            HealOptions::default(),
+            HealPriority::High,
+        ));
+
+        let stats = queue.get_priority_stats();
+
+        assert_eq!(*stats.get(&HealPriority::Low).unwrap_or(&0), 3);
+        assert_eq!(*stats.get(&HealPriority::Normal).unwrap_or(&0), 2);
+        assert_eq!(*stats.get(&HealPriority::High).unwrap_or(&0), 1);
+        assert_eq!(*stats.get(&HealPriority::Urgent).unwrap_or(&0), 0);
+    }
+
+    #[test]
+    fn test_priority_queue_is_empty() {
+        let mut queue = PriorityHealQueue::new();
+
+        assert!(queue.is_empty());
+
+        queue.push(HealRequest::new(
+            HealType::Bucket {
+                bucket: "test".to_string(),
+            },
+            HealOptions::default(),
+            HealPriority::Normal,
+        ));
+
+        assert!(!queue.is_empty());
+
+        queue.pop();
+
+        assert!(queue.is_empty());
+    }
+}

crates/ahm/src/heal/mod.rs

@@ -20,6 +20,7 @@ pub mod progress;
 pub mod resume;
 pub mod storage;
 pub mod task;
+pub mod utils;
 
 pub use erasure_healer::ErasureSetHealer;
 pub use manager::HealManager;

crates/ahm/src/heal/progress.rs

@@ -146,3 +146,244 @@ impl HealStatistics {
         }
     }
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn test_heal_progress_new() {
+        let progress = HealProgress::new();
+        assert_eq!(progress.objects_scanned, 0);
+        assert_eq!(progress.objects_healed, 0);
+        assert_eq!(progress.objects_failed, 0);
+        assert_eq!(progress.bytes_processed, 0);
+        assert_eq!(progress.progress_percentage, 0.0);
+        assert!(progress.start_time.is_some());
+        assert!(progress.last_update_time.is_some());
+        assert!(progress.current_object.is_none());
+    }
+
+    #[test]
+    fn test_heal_progress_update_progress() {
+        let mut progress = HealProgress::new();
+        progress.update_progress(10, 8, 2, 1024);
+
+        assert_eq!(progress.objects_scanned, 10);
+        assert_eq!(progress.objects_healed, 8);
+        assert_eq!(progress.objects_failed, 2);
+        assert_eq!(progress.bytes_processed, 1024);
+        // Progress percentage should be calculated based on healed/total
+        // total = scanned + healed + failed = 10 + 8 + 2 = 20
+        // healed/total = 8/20 = 0.4 = 40%
+        assert!((progress.progress_percentage - 40.0).abs() < 0.001);
+        assert!(progress.last_update_time.is_some());
+    }
+
+    #[test]
+    fn test_heal_progress_update_progress_zero_total() {
+        let mut progress = HealProgress::new();
+        progress.update_progress(0, 0, 0, 0);
+
+        assert_eq!(progress.progress_percentage, 0.0);
+    }
+
+    #[test]
+    fn test_heal_progress_update_progress_all_healed() {
+        let mut progress = HealProgress::new();
+        // When scanned=0, healed=10, failed=0: total=10, progress = 10/10 = 100%
+        progress.update_progress(0, 10, 0, 2048);
+
+        // All healed, should be 100%
+        assert!((progress.progress_percentage - 100.0).abs() < 0.001);
+    }
+
+    #[test]
+    fn test_heal_progress_set_current_object() {
+        let mut progress = HealProgress::new();
+        let initial_time = progress.last_update_time;
+
+        // Small delay to ensure time difference
+        std::thread::sleep(std::time::Duration::from_millis(10));
+
+        progress.set_current_object(Some("test-bucket/test-object".to_string()));
+
+        assert_eq!(progress.current_object, Some("test-bucket/test-object".to_string()));
+        assert!(progress.last_update_time.is_some());
+        // last_update_time should be updated
+        assert_ne!(progress.last_update_time, initial_time);
+    }
+
+    #[test]
+    fn test_heal_progress_set_current_object_none() {
+        let mut progress = HealProgress::new();
+        progress.set_current_object(Some("test".to_string()));
+        progress.set_current_object(None);
+
+        assert!(progress.current_object.is_none());
+    }
+
+    #[test]
+    fn test_heal_progress_is_completed_by_percentage() {
+        let mut progress = HealProgress::new();
+        progress.update_progress(10, 10, 0, 1024);
+
+        assert!(progress.is_completed());
+    }
+
+    #[test]
+    fn test_heal_progress_is_completed_by_processed() {
+        let mut progress = HealProgress::new();
+        progress.objects_scanned = 10;
+        progress.objects_healed = 8;
+        progress.objects_failed = 2;
+        // healed + failed = 8 + 2 = 10 >= scanned = 10
+        assert!(progress.is_completed());
+    }
+
+    #[test]
+    fn test_heal_progress_is_not_completed() {
+        let mut progress = HealProgress::new();
+        progress.objects_scanned = 10;
+        progress.objects_healed = 5;
+        progress.objects_failed = 2;
+        // healed + failed = 5 + 2 = 7 < scanned = 10
+        assert!(!progress.is_completed());
+    }
+
+    #[test]
+    fn test_heal_progress_get_success_rate() {
+        let mut progress = HealProgress::new();
+        progress.objects_healed = 8;
+        progress.objects_failed = 2;
+
+        // success_rate = 8 / (8 + 2) * 100 = 80%
+        assert!((progress.get_success_rate() - 80.0).abs() < 0.001);
+    }
+
+    #[test]
+    fn test_heal_progress_get_success_rate_zero_total() {
+        let progress = HealProgress::new();
+        // No healed or failed objects
+        assert_eq!(progress.get_success_rate(), 0.0);
+    }
+
+    #[test]
+    fn test_heal_progress_get_success_rate_all_success() {
+        let mut progress = HealProgress::new();
+        progress.objects_healed = 10;
+        progress.objects_failed = 0;
+
+        assert!((progress.get_success_rate() - 100.0).abs() < 0.001);
+    }
+
+    #[test]
+    fn test_heal_statistics_new() {
+        let stats = HealStatistics::new();
+        assert_eq!(stats.total_tasks, 0);
+        assert_eq!(stats.successful_tasks, 0);
+        assert_eq!(stats.failed_tasks, 0);
+        assert_eq!(stats.running_tasks, 0);
+        assert_eq!(stats.total_objects_healed, 0);
+        assert_eq!(stats.total_bytes_healed, 0);
+    }
+
+    #[test]
+    fn test_heal_statistics_default() {
+        let stats = HealStatistics::default();
+        assert_eq!(stats.total_tasks, 0);
+        assert_eq!(stats.successful_tasks, 0);
+        assert_eq!(stats.failed_tasks, 0);
+    }
+
+    #[test]
+    fn test_heal_statistics_update_task_completion_success() {
+        let mut stats = HealStatistics::new();
+        let initial_time = stats.last_update_time;
+
+        std::thread::sleep(std::time::Duration::from_millis(10));
+        stats.update_task_completion(true);
+
+        assert_eq!(stats.successful_tasks, 1);
+        assert_eq!(stats.failed_tasks, 0);
+        assert!(stats.last_update_time > initial_time);
+    }
+
+    #[test]
+    fn test_heal_statistics_update_task_completion_failure() {
+        let mut stats = HealStatistics::new();
+        stats.update_task_completion(false);
+
+        assert_eq!(stats.successful_tasks, 0);
+        assert_eq!(stats.failed_tasks, 1);
+    }
+
+    #[test]
+    fn test_heal_statistics_update_running_tasks() {
+        let mut stats = HealStatistics::new();
+        let initial_time = stats.last_update_time;
+
+        std::thread::sleep(std::time::Duration::from_millis(10));
+        stats.update_running_tasks(5);
+
+        assert_eq!(stats.running_tasks, 5);
+        assert!(stats.last_update_time > initial_time);
+    }
+
+    #[test]
+    fn test_heal_statistics_add_healed_objects() {
+        let mut stats = HealStatistics::new();
+        let initial_time = stats.last_update_time;
+
+        std::thread::sleep(std::time::Duration::from_millis(10));
+        stats.add_healed_objects(10, 10240);
+
+        assert_eq!(stats.total_objects_healed, 10);
+        assert_eq!(stats.total_bytes_healed, 10240);
+        assert!(stats.last_update_time > initial_time);
+    }
+
+    #[test]
+    fn test_heal_statistics_add_healed_objects_accumulative() {
+        let mut stats = HealStatistics::new();
+        stats.add_healed_objects(5, 5120);
+        stats.add_healed_objects(3, 3072);
+
+        assert_eq!(stats.total_objects_healed, 8);
+        assert_eq!(stats.total_bytes_healed, 8192);
+    }
+
+    #[test]
+    fn test_heal_statistics_get_success_rate() {
+        let mut stats = HealStatistics::new();
+        stats.successful_tasks = 8;
+        stats.failed_tasks = 2;
+
+        // success_rate = 8 / (8 + 2) * 100 = 80%
+        assert!((stats.get_success_rate() - 80.0).abs() < 0.001);
+    }
+
+    #[test]
+    fn test_heal_statistics_get_success_rate_zero_total() {
+        let stats = HealStatistics::new();
+        assert_eq!(stats.get_success_rate(), 0.0);
+    }
+
+    #[test]
+    fn test_heal_statistics_get_success_rate_all_success() {
+        let mut stats = HealStatistics::new();
+        stats.successful_tasks = 10;
+        stats.failed_tasks = 0;
+
+        assert!((stats.get_success_rate() - 100.0).abs() < 0.001);
+    }
+
+    #[test]
+    fn test_heal_statistics_get_success_rate_all_failure() {
+        let mut stats = HealStatistics::new();
+        stats.successful_tasks = 0;
+        stats.failed_tasks = 5;
+
+        assert_eq!(stats.get_success_rate(), 0.0);
+    }
+}

crates/ahm/src/heal/resume.rs

@@ -12,7 +12,7 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use crate::error::{Error, Result};
+use crate::{Error, Result};
 use rustfs_ecstore::disk::{BUCKET_META_PREFIX, DiskAPI, DiskStore, RUSTFS_META_BUCKET};
 use serde::{Deserialize, Serialize};
 use std::path::Path;
@@ -27,6 +27,12 @@ const RESUME_STATE_FILE: &str = "ahm_resume_state.json";
 const RESUME_PROGRESS_FILE: &str = "ahm_progress.json";
 const RESUME_CHECKPOINT_FILE: &str = "ahm_checkpoint.json";
 
+/// Helper function to convert Path to &str, returning an error if conversion fails
+fn path_to_str(path: &Path) -> Result<&str> {
+    path.to_str()
+        .ok_or_else(|| Error::other(format!("Invalid UTF-8 path: {path:?}")))
+}
+
 /// resume state
 #[derive(Debug, Clone, Serialize, Deserialize)]
 pub struct ResumeState {
@@ -34,6 +40,9 @@ pub struct ResumeState {
     pub task_id: String,
     /// task type
     pub task_type: String,
+    /// set disk identifier (for erasure set tasks)
+    #[serde(default)]
+    pub set_disk_id: String,
     /// start time
     pub start_time: u64,
     /// last update time
@@ -67,12 +76,13 @@ pub struct ResumeState {
 }
 
 impl ResumeState {
-    pub fn new(task_id: String, task_type: String, buckets: Vec<String>) -> Self {
+    pub fn new(task_id: String, task_type: String, set_disk_id: String, buckets: Vec<String>) -> Self {
         Self {
             task_id,
             task_type,
-            start_time: SystemTime::now().duration_since(UNIX_EPOCH).unwrap().as_secs(),
-            last_update: SystemTime::now().duration_since(UNIX_EPOCH).unwrap().as_secs(),
+            set_disk_id,
+            start_time: SystemTime::now().duration_since(UNIX_EPOCH).unwrap_or_default().as_secs(),
+            last_update: SystemTime::now().duration_since(UNIX_EPOCH).unwrap_or_default().as_secs(),
             completed: false,
             total_objects: 0,
             processed_objects: 0,
@@ -94,13 +104,13 @@ impl ResumeState {
         self.successful_objects = successful;
         self.failed_objects = failed;
         self.skipped_objects = skipped;
-        self.last_update = SystemTime::now().duration_since(UNIX_EPOCH).unwrap().as_secs();
+        self.last_update = SystemTime::now().duration_since(UNIX_EPOCH).unwrap_or_default().as_secs();
     }
 
     pub fn set_current_item(&mut self, bucket: Option<String>, object: Option<String>) {
         self.current_bucket = bucket;
         self.current_object = object;
-        self.last_update = SystemTime::now().duration_since(UNIX_EPOCH).unwrap().as_secs();
+        self.last_update = SystemTime::now().duration_since(UNIX_EPOCH).unwrap_or_default().as_secs();
     }
 
     pub fn complete_bucket(&mut self, bucket: &str) {
@@ -110,22 +120,22 @@ impl ResumeState {
         if let Some(pos) = self.pending_buckets.iter().position(|b| b == bucket) {
             self.pending_buckets.remove(pos);
         }
-        self.last_update = SystemTime::now().duration_since(UNIX_EPOCH).unwrap().as_secs();
+        self.last_update = SystemTime::now().duration_since(UNIX_EPOCH).unwrap_or_default().as_secs();
     }
 
     pub fn mark_completed(&mut self) {
         self.completed = true;
-        self.last_update = SystemTime::now().duration_since(UNIX_EPOCH).unwrap().as_secs();
+        self.last_update = SystemTime::now().duration_since(UNIX_EPOCH).unwrap_or_default().as_secs();
     }
 
     pub fn set_error(&mut self, error: String) {
         self.error_message = Some(error);
-        self.last_update = SystemTime::now().duration_since(UNIX_EPOCH).unwrap().as_secs();
+        self.last_update = SystemTime::now().duration_since(UNIX_EPOCH).unwrap_or_default().as_secs();
     }
 
     pub fn increment_retry(&mut self) {
         self.retry_count += 1;
-        self.last_update = SystemTime::now().duration_since(UNIX_EPOCH).unwrap().as_secs();
+        self.last_update = SystemTime::now().duration_since(UNIX_EPOCH).unwrap_or_default().as_secs();
     }
 
     pub fn can_retry(&self) -> bool {
@@ -156,8 +166,14 @@ pub struct ResumeManager {
 
 impl ResumeManager {
     /// create new resume manager
-    pub async fn new(disk: DiskStore, task_id: String, task_type: String, buckets: Vec<String>) -> Result<Self> {
-        let state = ResumeState::new(task_id, task_type, buckets);
+    pub async fn new(
+        disk: DiskStore,
+        task_id: String,
+        task_type: String,
+        set_disk_id: String,
+        buckets: Vec<String>,
+    ) -> Result<Self> {
+        let state = ResumeState::new(task_id, task_type, set_disk_id, buckets);
         let manager = Self {
             disk,
             state: Arc::new(RwLock::new(state)),
@@ -184,8 +200,11 @@ impl ResumeManager {
     /// check if resume state exists
     pub async fn has_resume_state(disk: &DiskStore, task_id: &str) -> bool {
         let file_path = Path::new(BUCKET_META_PREFIX).join(format!("{task_id}_{RESUME_STATE_FILE}"));
-        match disk.read_all(RUSTFS_META_BUCKET, file_path.to_str().unwrap()).await {
-            Ok(data) => !data.is_empty(),
+        match path_to_str(&file_path) {
+            Ok(path_str) => match disk.read_all(RUSTFS_META_BUCKET, path_str).await {
+                Ok(data) => !data.is_empty(),
+                Err(_) => false,
+            },
             Err(_) => false,
         }
     }
@@ -254,18 +273,15 @@ impl ResumeManager {
         let checkpoint_file = Path::new(BUCKET_META_PREFIX).join(format!("{task_id}_{RESUME_CHECKPOINT_FILE}"));
 
         // ignore delete errors, files may not exist
-        let _ = self
-            .disk
-            .delete(RUSTFS_META_BUCKET, state_file.to_str().unwrap(), Default::default())
-            .await;
-        let _ = self
-            .disk
-            .delete(RUSTFS_META_BUCKET, progress_file.to_str().unwrap(), Default::default())
-            .await;
-        let _ = self
-            .disk
-            .delete(RUSTFS_META_BUCKET, checkpoint_file.to_str().unwrap(), Default::default())
-            .await;
+        if let Ok(path_str) = path_to_str(&state_file) {
+            let _ = self.disk.delete(RUSTFS_META_BUCKET, path_str, Default::default()).await;
+        }
+        if let Ok(path_str) = path_to_str(&progress_file) {
+            let _ = self.disk.delete(RUSTFS_META_BUCKET, path_str, Default::default()).await;
+        }
+        if let Ok(path_str) = path_to_str(&checkpoint_file) {
+            let _ = self.disk.delete(RUSTFS_META_BUCKET, path_str, Default::default()).await;
+        }
 
         info!("Cleaned up resume state for task: {}", task_id);
         Ok(())
@@ -280,8 +296,9 @@ impl ResumeManager {
 
         let file_path = Path::new(BUCKET_META_PREFIX).join(format!("{}_{}", state.task_id, RESUME_STATE_FILE));
 
+        let path_str = path_to_str(&file_path)?;
         self.disk
-            .write_all(RUSTFS_META_BUCKET, file_path.to_str().unwrap(), state_data.into())
+            .write_all(RUSTFS_META_BUCKET, path_str, state_data.into())
             .await
             .map_err(|e| Error::TaskExecutionFailed {
                 message: format!("Failed to save resume state: {e}"),
@@ -295,7 +312,8 @@ impl ResumeManager {
     async fn read_state_file(disk: &DiskStore, task_id: &str) -> Result<Vec<u8>> {
         let file_path = Path::new(BUCKET_META_PREFIX).join(format!("{task_id}_{RESUME_STATE_FILE}"));
 
-        disk.read_all(RUSTFS_META_BUCKET, file_path.to_str().unwrap())
+        let path_str = path_to_str(&file_path)?;
+        disk.read_all(RUSTFS_META_BUCKET, path_str)
             .await
             .map(|bytes| bytes.to_vec())
             .map_err(|e| Error::TaskExecutionFailed {
@@ -327,7 +345,7 @@ impl ResumeCheckpoint {
     pub fn new(task_id: String) -> Self {
         Self {
             task_id,
-            checkpoint_time: SystemTime::now().duration_since(UNIX_EPOCH).unwrap().as_secs(),
+            checkpoint_time: SystemTime::now().duration_since(UNIX_EPOCH).unwrap_or_default().as_secs(),
             current_bucket_index: 0,
             current_object_index: 0,
             processed_objects: Vec::new(),
@@ -339,7 +357,7 @@ impl ResumeCheckpoint {
     pub fn update_position(&mut self, bucket_index: usize, object_index: usize) {
         self.current_bucket_index = bucket_index;
         self.current_object_index = object_index;
-        self.checkpoint_time = SystemTime::now().duration_since(UNIX_EPOCH).unwrap().as_secs();
+        self.checkpoint_time = SystemTime::now().duration_since(UNIX_EPOCH).unwrap_or_default().as_secs();
     }
 
     pub fn add_processed_object(&mut self, object: String) {
@@ -397,8 +415,11 @@ impl CheckpointManager {
     /// check if checkpoint exists
     pub async fn has_checkpoint(disk: &DiskStore, task_id: &str) -> bool {
         let file_path = Path::new(BUCKET_META_PREFIX).join(format!("{task_id}_{RESUME_CHECKPOINT_FILE}"));
-        match disk.read_all(RUSTFS_META_BUCKET, file_path.to_str().unwrap()).await {
-            Ok(data) => !data.is_empty(),
+        match path_to_str(&file_path) {
+            Ok(path_str) => match disk.read_all(RUSTFS_META_BUCKET, path_str).await {
+                Ok(data) => !data.is_empty(),
+                Err(_) => false,
+            },
             Err(_) => false,
         }
     }
@@ -446,10 +467,9 @@ impl CheckpointManager {
         let task_id = &checkpoint.task_id;
 
         let checkpoint_file = Path::new(BUCKET_META_PREFIX).join(format!("{task_id}_{RESUME_CHECKPOINT_FILE}"));
-        let _ = self
-            .disk
-            .delete(RUSTFS_META_BUCKET, checkpoint_file.to_str().unwrap(), Default::default())
-            .await;
+        if let Ok(path_str) = path_to_str(&checkpoint_file) {
+            let _ = self.disk.delete(RUSTFS_META_BUCKET, path_str, Default::default()).await;
+        }
 
         info!("Cleaned up checkpoint for task: {}", task_id);
         Ok(())
@@ -464,8 +484,9 @@ impl CheckpointManager {
 
         let file_path = Path::new(BUCKET_META_PREFIX).join(format!("{}_{}", checkpoint.task_id, RESUME_CHECKPOINT_FILE));
 
+        let path_str = path_to_str(&file_path)?;
         self.disk
-            .write_all(RUSTFS_META_BUCKET, file_path.to_str().unwrap(), checkpoint_data.into())
+            .write_all(RUSTFS_META_BUCKET, path_str, checkpoint_data.into())
             .await
             .map_err(|e| Error::TaskExecutionFailed {
                 message: format!("Failed to save checkpoint: {e}"),
@@ -479,7 +500,8 @@ impl CheckpointManager {
     async fn read_checkpoint_file(disk: &DiskStore, task_id: &str) -> Result<Vec<u8>> {
         let file_path = Path::new(BUCKET_META_PREFIX).join(format!("{task_id}_{RESUME_CHECKPOINT_FILE}"));
 
-        disk.read_all(RUSTFS_META_BUCKET, file_path.to_str().unwrap())
+        let path_str = path_to_str(&file_path)?;
+        disk.read_all(RUSTFS_META_BUCKET, path_str)
             .await
             .map(|bytes| bytes.to_vec())
             .map_err(|e| Error::TaskExecutionFailed {
@@ -562,7 +584,7 @@ mod tests {
     async fn test_resume_state_creation() {
         let task_id = ResumeUtils::generate_task_id();
         let buckets = vec!["bucket1".to_string(), "bucket2".to_string()];
-        let state = ResumeState::new(task_id.clone(), "erasure_set".to_string(), buckets);
+        let state = ResumeState::new(task_id.clone(), "erasure_set".to_string(), "pool_0_set_0".to_string(), buckets);
 
         assert_eq!(state.task_id, task_id);
         assert_eq!(state.task_type, "erasure_set");
@@ -575,7 +597,7 @@ mod tests {
     async fn test_resume_state_progress() {
         let task_id = ResumeUtils::generate_task_id();
         let buckets = vec!["bucket1".to_string()];
-        let mut state = ResumeState::new(task_id, "erasure_set".to_string(), buckets);
+        let mut state = ResumeState::new(task_id, "erasure_set".to_string(), "pool_0_set_0".to_string(), buckets);
 
         state.update_progress(10, 8, 1, 1);
         assert_eq!(state.processed_objects, 10);
@@ -595,7 +617,7 @@ mod tests {
     async fn test_resume_state_bucket_completion() {
         let task_id = ResumeUtils::generate_task_id();
         let buckets = vec!["bucket1".to_string(), "bucket2".to_string()];
-        let mut state = ResumeState::new(task_id, "erasure_set".to_string(), buckets);
+        let mut state = ResumeState::new(task_id, "erasure_set".to_string(), "pool_0_set_0".to_string(), buckets);
 
         assert_eq!(state.pending_buckets.len(), 2);
         assert_eq!(state.completed_buckets.len(), 0);
@@ -650,6 +672,7 @@ mod tests {
             let state = ResumeState::new(
                 task_id.clone(),
                 "erasure_set".to_string(),
+                "pool_0_set_0".to_string(),
                 vec!["bucket1".to_string(), "bucket2".to_string()],
             );
 

crates/ahm/src/heal/storage.rs

@@ -12,7 +12,7 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use crate::error::{Error, Result};
+use crate::{Error, Result};
 use async_trait::async_trait;
 use rustfs_common::heal_channel::{HealOpts, HealScanMode};
 use rustfs_ecstore::{
@@ -179,7 +179,9 @@ impl HealStorageAPI for ECStoreHealStorage {
                             "Object data exceeds cap ({} bytes), aborting full read to prevent OOM: {}/{}",
                             MAX_READ_BYTES, bucket, object
                         );
-                        return Ok(None);
+                        return Err(Error::other(format!(
+                            "Object too large: {n_read} bytes (max: {MAX_READ_BYTES} bytes) for {bucket}/{object}"
+                        )));
                     }
                 }
                 Err(e) => {
@@ -398,13 +400,13 @@ impl HealStorageAPI for ECStoreHealStorage {
         match self.ecstore.get_object_info(bucket, object, &Default::default()).await {
             Ok(_) => Ok(true), // Object exists
             Err(e) => {
-                // Map ObjectNotFound to false, other errors to false as well for safety
+                // Map ObjectNotFound to false, other errors must be propagated!
                 if matches!(e, rustfs_ecstore::error::StorageError::ObjectNotFound(_, _)) {
                     debug!("Object not found: {}/{}", bucket, object);
                     Ok(false)
                 } else {
-                    debug!("Error checking object existence {}/{}: {}", bucket, object, e);
-                    Ok(false) // Treat errors as non-existence to be safe
+                    error!("Error checking object existence {}/{}: {}", bucket, object, e);
+                    Err(Error::other(e))
                 }
             }
         }
@@ -496,7 +498,7 @@ impl HealStorageAPI for ECStoreHealStorage {
         match self
             .ecstore
             .clone()
-            .list_objects_v2(bucket, prefix, None, None, 1000, false, None)
+            .list_objects_v2(bucket, prefix, None, None, 1000, false, None, false)
             .await
         {
             Ok(list_info) => {
@@ -515,21 +517,7 @@ impl HealStorageAPI for ECStoreHealStorage {
         debug!("Getting disk for resume: {}", set_disk_id);
 
         // Parse set_disk_id to extract pool and set indices
-        // Format: "pool_{pool_idx}_set_{set_idx}"
-        let parts: Vec<&str> = set_disk_id.split('_').collect();
-        if parts.len() != 4 || parts[0] != "pool" || parts[2] != "set" {
-            return Err(Error::TaskExecutionFailed {
-                message: format!("Invalid set_disk_id format: {set_disk_id}"),
-            });
-        }
-
-        let pool_idx: usize = parts[1].parse().map_err(|_| Error::TaskExecutionFailed {
-            message: format!("Invalid pool index in set_disk_id: {set_disk_id}"),
-        })?;
-
-        let set_idx: usize = parts[3].parse().map_err(|_| Error::TaskExecutionFailed {
-            message: format!("Invalid set index in set_disk_id: {set_disk_id}"),
-        })?;
+        let (pool_idx, set_idx) = crate::heal::utils::parse_set_disk_id(set_disk_id)?;
 
         // Get the first available disk from the set
         let disks = self

crates/ahm/src/heal/task.rs

@@ -12,13 +12,15 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use crate::error::{Error, Result};
-use crate::heal::ErasureSetHealer;
-use crate::heal::{progress::HealProgress, storage::HealStorageAPI};
+use crate::heal::{ErasureSetHealer, progress::HealProgress, storage::HealStorageAPI};
+use crate::{Error, Result};
 use rustfs_common::heal_channel::{HealOpts, HealScanMode};
 use serde::{Deserialize, Serialize};
-use std::sync::Arc;
-use std::time::{Duration, SystemTime};
+use std::{
+    future::Future,
+    sync::Arc,
+    time::{Duration, Instant, SystemTime},
+};
 use tokio::sync::RwLock;
 use tracing::{error, info, warn};
 use uuid::Uuid;
@@ -49,11 +51,12 @@ pub enum HealType {
 }
 
 /// Heal priority
-#[derive(Debug, Clone, PartialEq, Eq, PartialOrd, Ord, Serialize, Deserialize)]
+#[derive(Debug, Default, Clone, Copy, PartialEq, Eq, PartialOrd, Ord, Hash, Serialize, Deserialize)]
 pub enum HealPriority {
     /// Low priority
     Low = 0,
     /// Normal priority
+    #[default]
     Normal = 1,
     /// High priority
     High = 2,
@@ -61,12 +64,6 @@ pub enum HealPriority {
     Urgent = 3,
 }
 
-impl Default for HealPriority {
-    fn default() -> Self {
-        Self::Normal
-    }
-}
-
 /// Heal options
 #[derive(Debug, Clone, Serialize, Deserialize)]
 pub struct HealOptions {
@@ -200,6 +197,8 @@ pub struct HealTask {
     pub started_at: Arc<RwLock<Option<SystemTime>>>,
     /// Completed time
     pub completed_at: Arc<RwLock<Option<SystemTime>>>,
+    /// Task start instant for timeout calculation (monotonic)
+    task_start_instant: Arc<RwLock<Option<Instant>>>,
     /// Cancel token
     pub cancel_token: tokio_util::sync::CancellationToken,
     /// Storage layer interface
@@ -217,23 +216,77 @@ impl HealTask {
             created_at: request.created_at,
             started_at: Arc::new(RwLock::new(None)),
             completed_at: Arc::new(RwLock::new(None)),
+            task_start_instant: Arc::new(RwLock::new(None)),
             cancel_token: tokio_util::sync::CancellationToken::new(),
             storage,
         }
     }
 
+    async fn remaining_timeout(&self) -> Result<Option<Duration>> {
+        if let Some(total) = self.options.timeout {
+            let start_instant = { *self.task_start_instant.read().await };
+            if let Some(started_at) = start_instant {
+                let elapsed = started_at.elapsed();
+                if elapsed >= total {
+                    return Err(Error::TaskTimeout);
+                }
+                return Ok(Some(total - elapsed));
+            }
+            Ok(Some(total))
+        } else {
+            Ok(None)
+        }
+    }
+
+    async fn check_control_flags(&self) -> Result<()> {
+        if self.cancel_token.is_cancelled() {
+            return Err(Error::TaskCancelled);
+        }
+        // Only interested in propagating an error if the timeout has expired;
+        // the actual Duration value is not needed here
+        let _ = self.remaining_timeout().await?;
+        Ok(())
+    }
+
+    async fn await_with_control<F, T>(&self, fut: F) -> Result<T>
+    where
+        F: Future<Output = Result<T>> + Send,
+        T: Send,
+    {
+        let cancel_token = self.cancel_token.clone();
+        if let Some(remaining) = self.remaining_timeout().await? {
+            if remaining.is_zero() {
+                return Err(Error::TaskTimeout);
+            }
+            let mut fut = Box::pin(fut);
+            tokio::select! {
+                _ = cancel_token.cancelled() => Err(Error::TaskCancelled),
+                _ = tokio::time::sleep(remaining) => Err(Error::TaskTimeout),
+                result = &mut fut => result,
+            }
+        } else {
+            tokio::select! {
+                _ = cancel_token.cancelled() => Err(Error::TaskCancelled),
+                result = fut => result,
+            }
+        }
+    }
+
+    #[tracing::instrument(skip(self), fields(task_id = %self.id, heal_type = ?self.heal_type))]
     pub async fn execute(&self) -> Result<()> {
-        // update status to running
+        // update status and timestamps atomically to avoid race conditions
+        let now = SystemTime::now();
+        let start_instant = Instant::now();
         {
             let mut status = self.status.write().await;
-            *status = HealTaskStatus::Running;
-        }
-        {
             let mut started_at = self.started_at.write().await;
-            *started_at = Some(SystemTime::now());
+            let mut task_start_instant = self.task_start_instant.write().await;
+            *status = HealTaskStatus::Running;
+            *started_at = Some(now);
+            *task_start_instant = Some(start_instant);
         }
 
-        info!("Starting heal task: {} with type: {:?}", self.id, self.heal_type);
+        info!("Task started");
 
         let result = match &self.heal_type {
             HealType::Object {
@@ -263,7 +316,17 @@ impl HealTask {
             Ok(_) => {
                 let mut status = self.status.write().await;
                 *status = HealTaskStatus::Completed;
-                info!("Heal task completed successfully: {}", self.id);
+                info!("Task completed successfully");
+            }
+            Err(Error::TaskCancelled) => {
+                let mut status = self.status.write().await;
+                *status = HealTaskStatus::Cancelled;
+                info!("Heal task was cancelled: {}", self.id);
+            }
+            Err(Error::TaskTimeout) => {
+                let mut status = self.status.write().await;
+                *status = HealTaskStatus::Timeout;
+                warn!("Heal task timed out: {}", self.id);
             }
             Err(e) => {
                 let mut status = self.status.write().await;
@@ -292,8 +355,9 @@ impl HealTask {
     }
 
     // specific heal implementation method
+    #[tracing::instrument(skip(self), fields(bucket = %bucket, object = %object, version_id = ?version_id))]
     async fn heal_object(&self, bucket: &str, object: &str, version_id: Option<&str>) -> Result<()> {
-        info!("Healing object: {}/{}", bucket, object);
+        info!("Starting object heal workflow");
 
         // update progress
         {
@@ -303,8 +367,9 @@ impl HealTask {
         }
 
         // Step 1: Check if object exists and get metadata
-        info!("Step 1: Checking object existence and metadata");
-        let object_exists = self.storage.object_exists(bucket, object).await?;
+        warn!("Step 1: Checking object existence and metadata");
+        self.check_control_flags().await?;
+        let object_exists = self.await_with_control(self.storage.object_exists(bucket, object)).await?;
         if !object_exists {
             warn!("Object does not exist: {}/{}", bucket, object);
             if self.options.recreate_missing {
@@ -336,7 +401,11 @@ impl HealTask {
             set: self.options.set_index,
         };
 
-        match self.storage.heal_object(bucket, object, version_id, &heal_opts).await {
+        let heal_result = self
+            .await_with_control(self.storage.heal_object(bucket, object, version_id, &heal_opts))
+            .await;
+
+        match heal_result {
             Ok((result, error)) => {
                 if let Some(e) = error {
                     // Check if this is a "File not found" error during delete operations
@@ -357,9 +426,9 @@ impl HealTask {
 
                     // If heal failed and remove_corrupted is enabled, delete the corrupted object
                     if self.options.remove_corrupted {
-                        warn!("Removing corrupted object: {}/{}", bucket, object);
+                        info!("Removing corrupted object: {}/{}", bucket, object);
                         if !self.options.dry_run {
-                            self.storage.delete_object(bucket, object).await?;
+                            self.await_with_control(self.storage.delete_object(bucket, object)).await?;
                             info!("Successfully deleted corrupted object: {}/{}", bucket, object);
                         } else {
                             info!("Dry run mode - would delete corrupted object: {}/{}", bucket, object);
@@ -380,11 +449,9 @@ impl HealTask {
                 info!("Step 3: Verifying heal result");
                 let object_size = result.object_size as u64;
                 info!(
-                    "Heal completed successfully: {}/{} ({} bytes, {} drives healed)",
-                    bucket,
-                    object,
-                    object_size,
-                    result.after.drives.len()
+                    object_size = object_size,
+                    drives_healed = result.after.drives.len(),
+                    "Heal completed successfully"
                 );
 
                 {
@@ -393,6 +460,8 @@ impl HealTask {
                 }
                 Ok(())
             }
+            Err(Error::TaskCancelled) => Err(Error::TaskCancelled),
+            Err(Error::TaskTimeout) => Err(Error::TaskTimeout),
             Err(e) => {
                 // Check if this is a "File not found" error during delete operations
                 let error_msg = format!("{e}");
@@ -412,9 +481,9 @@ impl HealTask {
 
                 // If heal failed and remove_corrupted is enabled, delete the corrupted object
                 if self.options.remove_corrupted {
-                    warn!("Removing corrupted object: {}/{}", bucket, object);
+                    info!("Removing corrupted object: {}/{}", bucket, object);
                     if !self.options.dry_run {
-                        self.storage.delete_object(bucket, object).await?;
+                        self.await_with_control(self.storage.delete_object(bucket, object)).await?;
                         info!("Successfully deleted corrupted object: {}/{}", bucket, object);
                     } else {
                         info!("Dry run mode - would delete corrupted object: {}/{}", bucket, object);
@@ -450,7 +519,10 @@ impl HealTask {
             set: None,
         };
 
-        match self.storage.heal_object(bucket, object, version_id, &heal_opts).await {
+        match self
+            .await_with_control(self.storage.heal_object(bucket, object, version_id, &heal_opts))
+            .await
+        {
             Ok((result, error)) => {
                 if let Some(e) = error {
                     error!("Failed to recreate missing object: {}/{} - {}", bucket, object, e);
@@ -468,6 +540,8 @@ impl HealTask {
                 }
                 Ok(())
             }
+            Err(Error::TaskCancelled) => Err(Error::TaskCancelled),
+            Err(Error::TaskTimeout) => Err(Error::TaskTimeout),
             Err(e) => {
                 error!("Failed to recreate missing object: {}/{} - {}", bucket, object, e);
                 Err(Error::TaskExecutionFailed {
@@ -489,7 +563,8 @@ impl HealTask {
 
         // Step 1: Check if bucket exists
         info!("Step 1: Checking bucket existence");
-        let bucket_exists = self.storage.get_bucket_info(bucket).await?.is_some();
+        self.check_control_flags().await?;
+        let bucket_exists = self.await_with_control(self.storage.get_bucket_info(bucket)).await?.is_some();
         if !bucket_exists {
             warn!("Bucket does not exist: {}", bucket);
             return Err(Error::TaskExecutionFailed {
@@ -516,7 +591,9 @@ impl HealTask {
             set: self.options.set_index,
         };
 
-        match self.storage.heal_bucket(bucket, &heal_opts).await {
+        let heal_result = self.await_with_control(self.storage.heal_bucket(bucket, &heal_opts)).await;
+
+        match heal_result {
             Ok(result) => {
                 info!("Bucket heal completed successfully: {} ({} drives)", bucket, result.after.drives.len());
 
@@ -526,6 +603,8 @@ impl HealTask {
                 }
                 Ok(())
             }
+            Err(Error::TaskCancelled) => Err(Error::TaskCancelled),
+            Err(Error::TaskTimeout) => Err(Error::TaskTimeout),
             Err(e) => {
                 error!("Bucket heal failed: {} - {}", bucket, e);
                 {
@@ -551,7 +630,8 @@ impl HealTask {
 
         // Step 1: Check if object exists
         info!("Step 1: Checking object existence");
-        let object_exists = self.storage.object_exists(bucket, object).await?;
+        self.check_control_flags().await?;
+        let object_exists = self.await_with_control(self.storage.object_exists(bucket, object)).await?;
         if !object_exists {
             warn!("Object does not exist: {}/{}", bucket, object);
             return Err(Error::TaskExecutionFailed {
@@ -578,7 +658,11 @@ impl HealTask {
             set: self.options.set_index,
         };
 
-        match self.storage.heal_object(bucket, object, None, &heal_opts).await {
+        let heal_result = self
+            .await_with_control(self.storage.heal_object(bucket, object, None, &heal_opts))
+            .await;
+
+        match heal_result {
             Ok((result, error)) => {
                 if let Some(e) = error {
                     error!("Metadata heal failed: {}/{} - {}", bucket, object, e);
@@ -604,6 +688,8 @@ impl HealTask {
                 }
                 Ok(())
             }
+            Err(Error::TaskCancelled) => Err(Error::TaskCancelled),
+            Err(Error::TaskTimeout) => Err(Error::TaskTimeout),
             Err(e) => {
                 error!("Metadata heal failed: {}/{} - {}", bucket, object, e);
                 {
@@ -652,7 +738,11 @@ impl HealTask {
             set: None,
         };
 
-        match self.storage.heal_object(bucket, &object, None, &heal_opts).await {
+        let heal_result = self
+            .await_with_control(self.storage.heal_object(bucket, &object, None, &heal_opts))
+            .await;
+
+        match heal_result {
             Ok((result, error)) => {
                 if let Some(e) = error {
                     error!("MRF heal failed: {} - {}", meta_path, e);
@@ -673,6 +763,8 @@ impl HealTask {
                 }
                 Ok(())
             }
+            Err(Error::TaskCancelled) => Err(Error::TaskCancelled),
+            Err(Error::TaskTimeout) => Err(Error::TaskTimeout),
             Err(e) => {
                 error!("MRF heal failed: {} - {}", meta_path, e);
                 {
@@ -698,7 +790,8 @@ impl HealTask {
 
         // Step 1: Check if object exists
         info!("Step 1: Checking object existence");
-        let object_exists = self.storage.object_exists(bucket, object).await?;
+        self.check_control_flags().await?;
+        let object_exists = self.await_with_control(self.storage.object_exists(bucket, object)).await?;
         if !object_exists {
             warn!("Object does not exist: {}/{}", bucket, object);
             return Err(Error::TaskExecutionFailed {
@@ -725,7 +818,11 @@ impl HealTask {
             set: None,
         };
 
-        match self.storage.heal_object(bucket, object, version_id, &heal_opts).await {
+        let heal_result = self
+            .await_with_control(self.storage.heal_object(bucket, object, version_id, &heal_opts))
+            .await;
+
+        match heal_result {
             Ok((result, error)) => {
                 if let Some(e) = error {
                     error!("EC decode heal failed: {}/{} - {}", bucket, object, e);
@@ -753,6 +850,8 @@ impl HealTask {
                 }
                 Ok(())
             }
+            Err(Error::TaskCancelled) => Err(Error::TaskCancelled),
+            Err(Error::TaskTimeout) => Err(Error::TaskTimeout),
             Err(e) => {
                 error!("EC decode heal failed: {}/{} - {}", bucket, object, e);
                 {
@@ -778,7 +877,7 @@ impl HealTask {
 
         let buckets = if buckets.is_empty() {
             info!("No buckets specified, listing all buckets");
-            let bucket_infos = self.storage.list_buckets().await?;
+            let bucket_infos = self.await_with_control(self.storage.list_buckets()).await?;
             bucket_infos.into_iter().map(|info| info.name).collect()
         } else {
             buckets
@@ -786,7 +885,9 @@ impl HealTask {
 
         // Step 1: Perform disk format heal using ecstore
         info!("Step 1: Performing disk format heal using ecstore");
-        match self.storage.heal_format(self.options.dry_run).await {
+        let format_result = self.await_with_control(self.storage.heal_format(self.options.dry_run)).await;
+
+        match format_result {
             Ok((result, error)) => {
                 if let Some(e) = error {
                     error!("Disk format heal failed: {} - {}", set_disk_id, e);
@@ -805,6 +906,8 @@ impl HealTask {
                     result.after.drives.len()
                 );
             }
+            Err(Error::TaskCancelled) => return Err(Error::TaskCancelled),
+            Err(Error::TaskTimeout) => return Err(Error::TaskTimeout),
             Err(e) => {
                 error!("Disk format heal failed: {} - {}", set_disk_id, e);
                 {
@@ -824,7 +927,9 @@ impl HealTask {
 
         // Step 2: Get disk for resume functionality
         info!("Step 2: Getting disk for resume functionality");
-        let disk = self.storage.get_disk_for_resume(&set_disk_id).await?;
+        let disk = self
+            .await_with_control(self.storage.get_disk_for_resume(&set_disk_id))
+            .await?;
 
         {
             let mut progress = self.progress.write().await;
@@ -832,9 +937,18 @@ impl HealTask {
         }
 
         // Step 3: Heal bucket structure
+        // Check control flags before each iteration to ensure timely cancellation.
+        // Each heal_bucket call may handle timeout/cancellation internally, see its implementation for details.
         for bucket in buckets.iter() {
+            // Check control flags before starting each bucket heal
+            self.check_control_flags().await?;
+            // heal_bucket internally uses await_with_control for timeout/cancellation handling
             if let Err(err) = self.heal_bucket(bucket).await {
-                info!("{}", err.to_string());
+                // Check if error is due to cancellation or timeout
+                if matches!(err, Error::TaskCancelled | Error::TaskTimeout) {
+                    return Err(err);
+                }
+                info!("Bucket heal failed: {}", err.to_string());
             }
         }
 
@@ -861,6 +975,8 @@ impl HealTask {
                 info!("Erasure set heal completed successfully: {} ({} buckets)", set_disk_id, buckets.len());
                 Ok(())
             }
+            Err(Error::TaskCancelled) => Err(Error::TaskCancelled),
+            Err(Error::TaskTimeout) => Err(Error::TaskTimeout),
             Err(e) => {
                 error!("Erasure set heal failed: {} - {}", set_disk_id, e);
                 Err(Error::TaskExecutionFailed {

crates/ahm/src/heal/utils.rs

@@ -0,0 +1,110 @@
+// Copyright 2024 RustFS Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+use crate::{Error, Result};
+
+/// Prefix for pool index in set disk identifiers.
+const POOL_PREFIX: &str = "pool";
+/// Prefix for set index in set disk identifiers.
+const SET_PREFIX: &str = "set";
+
+/// Format a set disk identifier using unsigned indices.
+pub fn format_set_disk_id(pool_idx: usize, set_idx: usize) -> String {
+    format!("{POOL_PREFIX}_{pool_idx}_{SET_PREFIX}_{set_idx}")
+}
+
+/// Format a set disk identifier from signed indices.
+pub fn format_set_disk_id_from_i32(pool_idx: i32, set_idx: i32) -> Option<String> {
+    if pool_idx < 0 || set_idx < 0 {
+        None
+    } else {
+        Some(format_set_disk_id(pool_idx as usize, set_idx as usize))
+    }
+}
+
+/// Normalise external set disk identifiers into the canonical format.
+pub fn normalize_set_disk_id(raw: &str) -> Option<String> {
+    if raw.starts_with(&format!("{POOL_PREFIX}_")) {
+        Some(raw.to_string())
+    } else {
+        parse_compact_set_disk_id(raw).map(|(pool, set)| format_set_disk_id(pool, set))
+    }
+}
+
+/// Parse a canonical set disk identifier into pool/set indices.
+pub fn parse_set_disk_id(raw: &str) -> Result<(usize, usize)> {
+    let parts: Vec<&str> = raw.split('_').collect();
+    if parts.len() != 4 || parts[0] != POOL_PREFIX || parts[2] != SET_PREFIX {
+        return Err(Error::TaskExecutionFailed {
+            message: format!("Invalid set_disk_id format: {raw}"),
+        });
+    }
+
+    let pool_idx = parts[1].parse::<usize>().map_err(|_| Error::TaskExecutionFailed {
+        message: format!("Invalid pool index in set_disk_id: {raw}"),
+    })?;
+    let set_idx = parts[3].parse::<usize>().map_err(|_| Error::TaskExecutionFailed {
+        message: format!("Invalid set index in set_disk_id: {raw}"),
+    })?;
+    Ok((pool_idx, set_idx))
+}
+
+fn parse_compact_set_disk_id(raw: &str) -> Option<(usize, usize)> {
+    let (pool, set) = raw.split_once('_')?;
+    let pool_idx = pool.parse::<usize>().ok()?;
+    let set_idx = set.parse::<usize>().ok()?;
+    Some((pool_idx, set_idx))
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn format_from_unsigned_indices() {
+        assert_eq!(format_set_disk_id(1, 2), "pool_1_set_2");
+    }
+
+    #[test]
+    fn format_from_signed_indices() {
+        assert_eq!(format_set_disk_id_from_i32(3, 4), Some("pool_3_set_4".into()));
+        assert_eq!(format_set_disk_id_from_i32(-1, 4), None);
+    }
+
+    #[test]
+    fn normalize_compact_identifier() {
+        assert_eq!(normalize_set_disk_id("3_5"), Some("pool_3_set_5".to_string()));
+    }
+
+    #[test]
+    fn normalize_prefixed_identifier() {
+        assert_eq!(normalize_set_disk_id("pool_7_set_1"), Some("pool_7_set_1".to_string()));
+    }
+
+    #[test]
+    fn normalize_invalid_identifier() {
+        assert_eq!(normalize_set_disk_id("invalid"), None);
+    }
+
+    #[test]
+    fn parse_prefixed_identifier() {
+        assert_eq!(parse_set_disk_id("pool_9_set_3").unwrap(), (9, 3));
+    }
+
+    #[test]
+    fn parse_invalid_identifier() {
+        assert!(parse_set_disk_id("bad").is_err());
+        assert!(parse_set_disk_id("pool_X_set_1").is_err());
+    }
+}

crates/ahm/src/lib.rs

@@ -12,17 +12,16 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use std::sync::{Arc, OnceLock};
-use tokio_util::sync::CancellationToken;
-use tracing::{error, info};
-
-pub mod error;
+mod error;
 pub mod heal;
 pub mod scanner;
 
 pub use error::{Error, Result};
 pub use heal::{HealManager, HealOptions, HealPriority, HealRequest, HealType, channel::HealChannelProcessor};
 pub use scanner::Scanner;
+use std::sync::{Arc, OnceLock};
+use tokio_util::sync::CancellationToken;
+use tracing::{error, info};
 
 // Global cancellation token for AHM services (scanner and other background tasks)
 static GLOBAL_AHM_SERVICES_CANCEL_TOKEN: OnceLock<CancellationToken> = OnceLock::new();

crates/ahm/src/scanner/checkpoint.rs

@@ -12,18 +12,16 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+use crate::scanner::node_scanner::ScanProgress;
+use crate::{Error, Result};
+use serde::{Deserialize, Serialize};
 use std::{
     path::{Path, PathBuf},
     time::{Duration, SystemTime},
 };
-
-use serde::{Deserialize, Serialize};
 use tokio::sync::RwLock;
 use tracing::{debug, error, info, warn};
 
-use super::node_scanner::ScanProgress;
-use crate::{Error, error::Result};
-
 #[derive(Debug, Serialize, Deserialize, Clone)]
 pub struct CheckpointData {
     pub version: u32,

crates/ahm/src/scanner/data_scanner.rs

@@ -12,46 +12,39 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+// IO throttling component is integrated into NodeScanner
+use crate::{
+    Error, HealRequest, Result, get_ahm_services_cancel_token,
+    heal::HealManager,
+    scanner::{
+        BucketMetrics, DecentralizedStatsAggregator, DecentralizedStatsAggregatorConfig, DiskMetrics, MetricsCollector,
+        NodeScanner, NodeScannerConfig, ScannerMetrics,
+        lifecycle::ScannerItem,
+        local_scan::{self, LocalObjectRecord, LocalScanOutcome},
+    },
+};
+use rustfs_common::data_usage::{DataUsageInfo, SizeSummary};
+use rustfs_common::metrics::{Metric, Metrics, global_metrics};
+use rustfs_ecstore::{
+    self as ecstore, StorageAPI,
+    bucket::versioning::VersioningApi,
+    bucket::versioning_sys::BucketVersioningSys,
+    data_usage::{aggregate_local_snapshots, store_data_usage_in_backend},
+    disk::{Disk, DiskAPI, DiskStore, RUSTFS_META_BUCKET, WalkDirOptions},
+    set_disk::SetDisks,
+    store_api::ObjectInfo,
+};
+use rustfs_filemeta::{MetacacheReader, VersionType};
+use s3s::dto::{BucketVersioningStatus, VersioningConfiguration};
 use std::{
     collections::HashMap,
     sync::Arc,
     time::{Duration, SystemTime},
 };
 use time::OffsetDateTime;
-
-use ecstore::{
-    disk::{Disk, DiskAPI, DiskStore, WalkDirOptions},
-    set_disk::SetDisks,
-};
-use rustfs_ecstore::store_api::ObjectInfo;
-use rustfs_ecstore::{
-    self as ecstore, StorageAPI,
-    data_usage::{aggregate_local_snapshots, store_data_usage_in_backend},
-};
-use rustfs_filemeta::{MetacacheReader, VersionType};
-use s3s::dto::{BucketVersioningStatus, VersioningConfiguration};
 use tokio::sync::{Mutex, RwLock};
 use tokio_util::sync::CancellationToken;
 use tracing::{debug, error, info, warn};
-
-use super::metrics::{BucketMetrics, DiskMetrics, MetricsCollector, ScannerMetrics};
-use super::node_scanner::{NodeScanner, NodeScannerConfig};
-use super::stats_aggregator::{DecentralizedStatsAggregator, DecentralizedStatsAggregatorConfig};
-// IO throttling component is integrated into NodeScanner
-use crate::heal::HealManager;
-use crate::scanner::lifecycle::ScannerItem;
-use crate::scanner::local_scan::{self, LocalObjectRecord, LocalScanOutcome};
-use crate::{
-    HealRequest,
-    error::{Error, Result},
-    get_ahm_services_cancel_token,
-};
-
-use rustfs_common::data_usage::{DataUsageInfo, SizeSummary};
-use rustfs_common::metrics::{Metric, Metrics, globalMetrics};
-use rustfs_ecstore::bucket::versioning::VersioningApi;
-use rustfs_ecstore::bucket::versioning_sys::BucketVersioningSys;
-use rustfs_ecstore::disk::RUSTFS_META_BUCKET;
 use uuid;
 
 /// Custom scan mode enum for AHM scanner
@@ -772,7 +765,7 @@ impl Scanner {
 
     /// Get global metrics from common crate
     pub async fn get_global_metrics(&self) -> rustfs_madmin::metrics::ScannerMetrics {
-        (*globalMetrics).report().await
+        global_metrics().report().await
     }
 
     /// Perform a single scan cycle using optimized node scanner
@@ -802,7 +795,7 @@ impl Scanner {
             cycle_completed: vec![chrono::Utc::now()],
             started: chrono::Utc::now(),
         };
-        (*globalMetrics).set_cycle(Some(cycle_info)).await;
+        global_metrics().set_cycle(Some(cycle_info)).await;
 
         self.metrics.set_current_cycle(self.state.read().await.current_cycle);
         self.metrics.increment_total_cycles();
@@ -1012,6 +1005,7 @@ impl Scanner {
                             100,   // max_keys - small limit for performance
                             false, // fetch_owner
                             None,  // start_after
+                            false, // incl_deleted
                         )
                         .await
                     {

crates/ahm/src/scanner/histogram.rs

@@ -12,13 +12,12 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+use serde::{Deserialize, Serialize};
 use std::{
     collections::HashMap,
     sync::atomic::{AtomicU64, Ordering},
     time::{Duration, SystemTime},
 };
-
-use serde::{Deserialize, Serialize};
 use tracing::info;
 
 /// Scanner metrics

crates/ahm/src/scanner/io_monitor.rs

@@ -12,6 +12,9 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+use crate::Result;
+use crate::scanner::LoadLevel;
+use serde::{Deserialize, Serialize};
 use std::{
     collections::VecDeque,
     sync::{
@@ -20,15 +23,10 @@ use std::{
     },
     time::{Duration, SystemTime},
 };
-
-use serde::{Deserialize, Serialize};
 use tokio::sync::RwLock;
 use tokio_util::sync::CancellationToken;
 use tracing::{debug, error, info, warn};
 
-use super::node_scanner::LoadLevel;
-use crate::error::Result;
-
 /// IO monitor config   
 #[derive(Debug, Clone)]
 pub struct IOMonitorConfig {

crates/ahm/src/scanner/io_throttler.rs

@@ -12,6 +12,7 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+use crate::scanner::LoadLevel;
 use std::{
     sync::{
         Arc,
@@ -19,12 +20,9 @@ use std::{
     },
     time::{Duration, SystemTime},
 };
-
 use tokio::sync::RwLock;
 use tracing::{debug, info, warn};
 
-use super::node_scanner::LoadLevel;
-
 /// IO throttler config
 #[derive(Debug, Clone)]
 pub struct IOThrottlerConfig {

crates/ahm/src/scanner/lifecycle.rs

@@ -12,25 +12,28 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use std::sync::Arc;
-use std::sync::atomic::{AtomicU64, Ordering};
-
-use crate::error::Result;
+use crate::Result;
 use rustfs_common::data_usage::SizeSummary;
 use rustfs_common::metrics::IlmAction;
-use rustfs_ecstore::bucket::lifecycle::{
-    bucket_lifecycle_audit::LcEventSrc,
-    bucket_lifecycle_ops::{GLOBAL_ExpiryState, apply_lifecycle_action, eval_action_from_lifecycle},
-    lifecycle,
-    lifecycle::Lifecycle,
+use rustfs_ecstore::bucket::{
+    lifecycle::{
+        bucket_lifecycle_audit::LcEventSrc,
+        bucket_lifecycle_ops::{GLOBAL_ExpiryState, apply_lifecycle_action, eval_action_from_lifecycle},
+        lifecycle,
+        lifecycle::Lifecycle,
+    },
+    metadata_sys::get_object_lock_config,
+    object_lock::objectlock_sys::{BucketObjectLockSys, enforce_retention_for_deletion},
+    versioning::VersioningApi,
+    versioning_sys::BucketVersioningSys,
 };
-use rustfs_ecstore::bucket::metadata_sys::get_object_lock_config;
-use rustfs_ecstore::bucket::object_lock::objectlock_sys::{BucketObjectLockSys, enforce_retention_for_deletion};
-use rustfs_ecstore::bucket::versioning::VersioningApi;
-use rustfs_ecstore::bucket::versioning_sys::BucketVersioningSys;
 use rustfs_ecstore::store_api::{ObjectInfo, ObjectToDelete};
 use rustfs_filemeta::FileInfo;
 use s3s::dto::{BucketLifecycleConfiguration as LifecycleConfig, VersioningConfiguration};
+use std::sync::{
+    Arc,
+    atomic::{AtomicU64, Ordering},
+};
 use time::OffsetDateTime;
 use tracing::info;
 

crates/ahm/src/scanner/local_scan/mod.rs

@@ -1,16 +1,18 @@
-use std::collections::{HashMap, HashSet};
-use std::path::{Path, PathBuf};
-use std::sync::Arc;
-use std::time::{SystemTime, UNIX_EPOCH};
-
-use serde::{Deserialize, Serialize};
-use serde_json::{from_slice, to_vec};
-use tokio::{fs, task};
-use tracing::warn;
-use walkdir::WalkDir;
-
-use crate::error::{Error, Result};
+// Copyright 2024 RustFS Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
 
+use crate::{Error, Result};
 use rustfs_common::data_usage::DiskUsageStatus;
 use rustfs_ecstore::data_usage::{
     LocalUsageSnapshot, LocalUsageSnapshotMeta, data_usage_state_dir, ensure_data_usage_layout, snapshot_file_name,
@@ -20,6 +22,15 @@ use rustfs_ecstore::disk::DiskAPI;
 use rustfs_ecstore::store::ECStore;
 use rustfs_ecstore::store_api::ObjectInfo;
 use rustfs_filemeta::{FileInfo, FileMeta, FileMetaVersion, VersionType};
+use serde::{Deserialize, Serialize};
+use serde_json::{from_slice, to_vec};
+use std::collections::{HashMap, HashSet};
+use std::path::{Path, PathBuf};
+use std::sync::Arc;
+use std::time::{SystemTime, UNIX_EPOCH};
+use tokio::{fs, task};
+use tracing::warn;
+use walkdir::WalkDir;
 
 const STATE_FILE_EXTENSION: &str = "";
 

crates/ahm/src/scanner/local_stats.rs

@@ -12,22 +12,19 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+use crate::scanner::node_scanner::{BucketStats, DiskStats, LocalScanStats};
+use crate::{Error, Result};
+use rustfs_common::data_usage::DataUsageInfo;
+use serde::{Deserialize, Serialize};
 use std::{
     path::{Path, PathBuf},
     sync::Arc,
     sync::atomic::{AtomicU64, Ordering},
     time::{Duration, SystemTime},
 };
-
-use serde::{Deserialize, Serialize};
 use tokio::sync::RwLock;
 use tracing::{debug, error, info, warn};
 
-use rustfs_common::data_usage::DataUsageInfo;
-
-use super::node_scanner::{BucketStats, DiskStats, LocalScanStats};
-use crate::{Error, error::Result};
-
 /// local stats manager
 pub struct LocalStatsManager {
     /// node id

crates/ahm/src/scanner/metrics.rs

@@ -12,13 +12,12 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+use serde::{Deserialize, Serialize};
 use std::{
     collections::HashMap,
     sync::atomic::{AtomicU64, Ordering},
     time::{Duration, SystemTime},
 };
-
-use serde::{Deserialize, Serialize};
 use tracing::info;
 
 /// Scanner metrics

crates/ahm/src/scanner/mod.rs

@@ -27,8 +27,10 @@ pub mod stats_aggregator;
 pub use checkpoint::{CheckpointData, CheckpointInfo, CheckpointManager};
 pub use data_scanner::{ScanMode, Scanner, ScannerConfig, ScannerState};
 pub use io_monitor::{AdvancedIOMonitor, IOMetrics, IOMonitorConfig};
-pub use io_throttler::{AdvancedIOThrottler, IOThrottlerConfig, ResourceAllocation, ThrottleDecision};
+pub use io_throttler::{AdvancedIOThrottler, IOThrottlerConfig, MetricsSnapshot, ResourceAllocation, ThrottleDecision};
 pub use local_stats::{BatchScanResult, LocalStatsManager, ScanResultEntry, StatsSummary};
-pub use metrics::ScannerMetrics;
+pub use metrics::{BucketMetrics, DiskMetrics, MetricsCollector, ScannerMetrics};
 pub use node_scanner::{IOMonitor, IOThrottler, LoadLevel, LocalScanStats, NodeScanner, NodeScannerConfig};
-pub use stats_aggregator::{AggregatedStats, DecentralizedStatsAggregator, NodeClient, NodeInfo};
+pub use stats_aggregator::{
+    AggregatedStats, DecentralizedStatsAggregator, DecentralizedStatsAggregatorConfig, NodeClient, NodeInfo,
+};

crates/ahm/src/scanner/node_scanner.rs

@@ -12,6 +12,15 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+use crate::Result;
+use crate::scanner::{
+    AdvancedIOMonitor, AdvancedIOThrottler, BatchScanResult, CheckpointManager, IOMonitorConfig, IOThrottlerConfig,
+    LocalStatsManager, MetricsSnapshot, ScanResultEntry,
+};
+use rustfs_common::data_usage::DataUsageInfo;
+use rustfs_ecstore::StorageAPI;
+use rustfs_ecstore::disk::{DiskAPI, DiskStore};
+use serde::{Deserialize, Serialize};
 use std::{
     collections::{HashMap, HashSet},
     path::{Path, PathBuf},
@@ -21,22 +30,10 @@ use std::{
     },
     time::{Duration, SystemTime},
 };
-
-use serde::{Deserialize, Serialize};
 use tokio::sync::RwLock;
 use tokio_util::sync::CancellationToken;
 use tracing::{debug, error, info, warn};
 
-use rustfs_common::data_usage::DataUsageInfo;
-use rustfs_ecstore::StorageAPI;
-use rustfs_ecstore::disk::{DiskAPI, DiskStore}; // Add this import
-
-use super::checkpoint::CheckpointManager;
-use super::io_monitor::{AdvancedIOMonitor, IOMonitorConfig};
-use super::io_throttler::{AdvancedIOThrottler, IOThrottlerConfig, MetricsSnapshot};
-use super::local_stats::{BatchScanResult, LocalStatsManager, ScanResultEntry};
-use crate::error::Result;
-
 /// SystemTime serde
 mod system_time_serde {
     use serde::{Deserialize, Deserializer, Serialize, Serializer};

crates/ahm/src/scanner/stats_aggregator.rs

@@ -12,24 +12,21 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+use crate::scanner::{
+    local_stats::StatsSummary,
+    node_scanner::{BucketStats, LoadLevel, ScanProgress},
+};
+use crate::{Error, Result};
+use rustfs_common::data_usage::DataUsageInfo;
+use serde::{Deserialize, Serialize};
 use std::{
     collections::HashMap,
     sync::Arc,
     time::{Duration, SystemTime},
 };
-
-use serde::{Deserialize, Serialize};
 use tokio::sync::RwLock;
 use tracing::{debug, info, warn};
 
-use rustfs_common::data_usage::DataUsageInfo;
-
-use super::{
-    local_stats::StatsSummary,
-    node_scanner::{BucketStats, LoadLevel, ScanProgress},
-};
-use crate::{Error, error::Result};
-
 /// node client config
 #[derive(Debug, Clone)]
 pub struct NodeClientConfig {

crates/ahm/tests/heal_bug_fixes_test.rs

@@ -0,0 +1,275 @@
+// Copyright 2024 RustFS Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+use rustfs_ahm::heal::{
+    event::{HealEvent, Severity},
+    task::{HealPriority, HealType},
+    utils,
+};
+
+#[test]
+fn test_heal_event_to_heal_request_no_panic() {
+    use rustfs_ecstore::disk::endpoint::Endpoint;
+
+    // Test that invalid pool/set indices don't cause panic
+    // Create endpoint using try_from or similar method
+    let endpoint_result = Endpoint::try_from("http://localhost:9000");
+    if let Ok(mut endpoint) = endpoint_result {
+        endpoint.pool_idx = -1;
+        endpoint.set_idx = -1;
+        endpoint.disk_idx = 0;
+
+        let event = HealEvent::DiskStatusChange {
+            endpoint,
+            old_status: "ok".to_string(),
+            new_status: "offline".to_string(),
+        };
+
+        // Should return error instead of panicking
+        let result = event.to_heal_request();
+        assert!(result.is_err());
+        assert!(result.unwrap_err().to_string().contains("Invalid heal type"));
+    }
+}
+
+#[test]
+fn test_heal_event_to_heal_request_valid_indices() {
+    use rustfs_ecstore::disk::endpoint::Endpoint;
+
+    // Test that valid indices work correctly
+    let endpoint_result = Endpoint::try_from("http://localhost:9000");
+    if let Ok(mut endpoint) = endpoint_result {
+        endpoint.pool_idx = 0;
+        endpoint.set_idx = 1;
+        endpoint.disk_idx = 0;
+
+        let event = HealEvent::DiskStatusChange {
+            endpoint,
+            old_status: "ok".to_string(),
+            new_status: "offline".to_string(),
+        };
+
+        let result = event.to_heal_request();
+        assert!(result.is_ok());
+        let request = result.unwrap();
+        assert!(matches!(request.heal_type, HealType::ErasureSet { .. }));
+    }
+}
+
+#[test]
+fn test_heal_event_object_corruption() {
+    let event = HealEvent::ObjectCorruption {
+        bucket: "test-bucket".to_string(),
+        object: "test-object".to_string(),
+        version_id: None,
+        corruption_type: rustfs_ahm::heal::event::CorruptionType::DataCorruption,
+        severity: Severity::High,
+    };
+
+    let result = event.to_heal_request();
+    assert!(result.is_ok());
+    let request = result.unwrap();
+    assert!(matches!(request.heal_type, HealType::Object { .. }));
+    assert_eq!(request.priority, HealPriority::High);
+}
+
+#[test]
+fn test_heal_event_ec_decode_failure() {
+    let event = HealEvent::ECDecodeFailure {
+        bucket: "test-bucket".to_string(),
+        object: "test-object".to_string(),
+        version_id: None,
+        missing_shards: vec![0, 1],
+        available_shards: vec![2, 3],
+    };
+
+    let result = event.to_heal_request();
+    assert!(result.is_ok());
+    let request = result.unwrap();
+    assert!(matches!(request.heal_type, HealType::ECDecode { .. }));
+    assert_eq!(request.priority, HealPriority::Urgent);
+}
+
+#[test]
+fn test_format_set_disk_id_from_i32_negative() {
+    // Test that negative indices return None
+    assert!(utils::format_set_disk_id_from_i32(-1, 0).is_none());
+    assert!(utils::format_set_disk_id_from_i32(0, -1).is_none());
+    assert!(utils::format_set_disk_id_from_i32(-1, -1).is_none());
+}
+
+#[test]
+fn test_format_set_disk_id_from_i32_valid() {
+    // Test that valid indices return Some
+    let result = utils::format_set_disk_id_from_i32(0, 1);
+    assert!(result.is_some());
+    assert_eq!(result.unwrap(), "pool_0_set_1");
+}
+
+#[test]
+fn test_resume_state_timestamp_handling() {
+    use rustfs_ahm::heal::resume::ResumeState;
+
+    // Test that ResumeState creation doesn't panic even if system time is before epoch
+    // This is a theoretical test - in practice, system time should never be before epoch
+    // But we want to ensure unwrap_or_default handles edge cases
+    let state = ResumeState::new(
+        "test-task".to_string(),
+        "test-type".to_string(),
+        "pool_0_set_1".to_string(),
+        vec!["bucket1".to_string()],
+    );
+
+    // Verify fields are initialized (u64 is always >= 0)
+    // The important thing is that unwrap_or_default prevents panic
+    let _ = state.start_time;
+    let _ = state.last_update;
+}
+
+#[test]
+fn test_resume_checkpoint_timestamp_handling() {
+    use rustfs_ahm::heal::resume::ResumeCheckpoint;
+
+    // Test that ResumeCheckpoint creation doesn't panic
+    let checkpoint = ResumeCheckpoint::new("test-task".to_string());
+
+    // Verify field is initialized (u64 is always >= 0)
+    // The important thing is that unwrap_or_default prevents panic
+    let _ = checkpoint.checkpoint_time;
+}
+
+#[test]
+fn test_path_to_str_helper() {
+    use std::path::Path;
+
+    // Test that path conversion handles non-UTF-8 paths gracefully
+    // Note: This is a compile-time test - actual non-UTF-8 paths are hard to construct in Rust
+    // The helper function should properly handle the conversion
+    let valid_path = Path::new("test/path");
+    assert!(valid_path.to_str().is_some());
+}
+
+#[test]
+fn test_heal_task_status_atomic_update() {
+    use rustfs_ahm::heal::storage::HealStorageAPI;
+    use rustfs_ahm::heal::task::{HealOptions, HealRequest, HealTask, HealTaskStatus};
+    use std::sync::Arc;
+
+    // Mock storage for testing
+    struct MockStorage;
+    #[async_trait::async_trait]
+    impl HealStorageAPI for MockStorage {
+        async fn get_object_meta(
+            &self,
+            _bucket: &str,
+            _object: &str,
+        ) -> rustfs_ahm::Result<Option<rustfs_ecstore::store_api::ObjectInfo>> {
+            Ok(None)
+        }
+        async fn get_object_data(&self, _bucket: &str, _object: &str) -> rustfs_ahm::Result<Option<Vec<u8>>> {
+            Ok(None)
+        }
+        async fn put_object_data(&self, _bucket: &str, _object: &str, _data: &[u8]) -> rustfs_ahm::Result<()> {
+            Ok(())
+        }
+        async fn delete_object(&self, _bucket: &str, _object: &str) -> rustfs_ahm::Result<()> {
+            Ok(())
+        }
+        async fn verify_object_integrity(&self, _bucket: &str, _object: &str) -> rustfs_ahm::Result<bool> {
+            Ok(true)
+        }
+        async fn ec_decode_rebuild(&self, _bucket: &str, _object: &str) -> rustfs_ahm::Result<Vec<u8>> {
+            Ok(vec![])
+        }
+        async fn get_disk_status(
+            &self,
+            _endpoint: &rustfs_ecstore::disk::endpoint::Endpoint,
+        ) -> rustfs_ahm::Result<rustfs_ahm::heal::storage::DiskStatus> {
+            Ok(rustfs_ahm::heal::storage::DiskStatus::Ok)
+        }
+        async fn format_disk(&self, _endpoint: &rustfs_ecstore::disk::endpoint::Endpoint) -> rustfs_ahm::Result<()> {
+            Ok(())
+        }
+        async fn get_bucket_info(&self, _bucket: &str) -> rustfs_ahm::Result<Option<rustfs_ecstore::store_api::BucketInfo>> {
+            Ok(None)
+        }
+        async fn heal_bucket_metadata(&self, _bucket: &str) -> rustfs_ahm::Result<()> {
+            Ok(())
+        }
+        async fn list_buckets(&self) -> rustfs_ahm::Result<Vec<rustfs_ecstore::store_api::BucketInfo>> {
+            Ok(vec![])
+        }
+        async fn object_exists(&self, _bucket: &str, _object: &str) -> rustfs_ahm::Result<bool> {
+            Ok(false)
+        }
+        async fn get_object_size(&self, _bucket: &str, _object: &str) -> rustfs_ahm::Result<Option<u64>> {
+            Ok(None)
+        }
+        async fn get_object_checksum(&self, _bucket: &str, _object: &str) -> rustfs_ahm::Result<Option<String>> {
+            Ok(None)
+        }
+        async fn heal_object(
+            &self,
+            _bucket: &str,
+            _object: &str,
+            _version_id: Option<&str>,
+            _opts: &rustfs_common::heal_channel::HealOpts,
+        ) -> rustfs_ahm::Result<(rustfs_madmin::heal_commands::HealResultItem, Option<rustfs_ahm::Error>)> {
+            Ok((rustfs_madmin::heal_commands::HealResultItem::default(), None))
+        }
+        async fn heal_bucket(
+            &self,
+            _bucket: &str,
+            _opts: &rustfs_common::heal_channel::HealOpts,
+        ) -> rustfs_ahm::Result<rustfs_madmin::heal_commands::HealResultItem> {
+            Ok(rustfs_madmin::heal_commands::HealResultItem::default())
+        }
+        async fn heal_format(
+            &self,
+            _dry_run: bool,
+        ) -> rustfs_ahm::Result<(rustfs_madmin::heal_commands::HealResultItem, Option<rustfs_ahm::Error>)> {
+            Ok((rustfs_madmin::heal_commands::HealResultItem::default(), None))
+        }
+        async fn list_objects_for_heal(&self, _bucket: &str, _prefix: &str) -> rustfs_ahm::Result<Vec<String>> {
+            Ok(vec![])
+        }
+        async fn get_disk_for_resume(&self, _set_disk_id: &str) -> rustfs_ahm::Result<rustfs_ecstore::disk::DiskStore> {
+            Err(rustfs_ahm::Error::other("Not implemented in mock"))
+        }
+    }
+
+    // Create a heal request and task
+    let request = HealRequest::new(
+        HealType::Object {
+            bucket: "test-bucket".to_string(),
+            object: "test-object".to_string(),
+            version_id: None,
+        },
+        HealOptions::default(),
+        HealPriority::Normal,
+    );
+
+    let storage: Arc<dyn HealStorageAPI> = Arc::new(MockStorage);
+    let task = HealTask::from_request(request, storage);
+
+    // Verify initial status
+    let status = tokio::runtime::Runtime::new().unwrap().block_on(task.get_status());
+    assert_eq!(status, HealTaskStatus::Pending);
+
+    // The task should have task_start_instant field initialized
+    // This is an internal detail, but we can verify it doesn't cause issues
+    // by checking that the task can be created successfully
+    // Note: We can't directly access private fields, but creation without panic
+    // confirms the fix works
+}

crates/ahm/tests/heal_integration_test.rs

@@ -25,9 +25,11 @@ use rustfs_ecstore::{
     store_api::{ObjectIO, ObjectOptions, PutObjReader, StorageAPI},
 };
 use serial_test::serial;
-use std::sync::Once;
-use std::sync::OnceLock;
-use std::{path::PathBuf, sync::Arc, time::Duration};
+use std::{
+    path::PathBuf,
+    sync::{Arc, Once, OnceLock},
+    time::Duration,
+};
 use tokio::fs;
 use tokio_util::sync::CancellationToken;
 use tracing::info;

crates/ahm/tests/integration_tests.rs

@@ -12,19 +12,16 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use std::{sync::Arc, time::Duration};
-use tempfile::TempDir;
-
 use rustfs_ahm::scanner::{
     io_throttler::MetricsSnapshot,
     local_stats::StatsSummary,
     node_scanner::{LoadLevel, NodeScanner, NodeScannerConfig},
     stats_aggregator::{DecentralizedStatsAggregator, DecentralizedStatsAggregatorConfig, NodeInfo},
 };
-
-mod scanner_optimization_tests;
 use scanner_optimization_tests::{PerformanceBenchmark, create_test_scanner};
-
+use std::{sync::Arc, time::Duration};
+use tempfile::TempDir;
+mod scanner_optimization_tests;
 #[tokio::test]
 async fn test_end_to_end_scanner_lifecycle() {
     let temp_dir = TempDir::new().unwrap();
@@ -245,21 +242,32 @@ async fn test_performance_impact_measurement() {
 
     io_monitor.start().await.unwrap();
 
-    // Baseline test: no scanner load
-    let baseline_duration = measure_workload(5_000, Duration::ZERO).await.max(Duration::from_millis(10));
+    // Baseline test: no scanner load - measure multiple times for stability
+    const MEASUREMENT_COUNT: usize = 5;
+    let mut baseline_measurements = Vec::new();
+    for _ in 0..MEASUREMENT_COUNT {
+        let duration = measure_workload(10_000, Duration::ZERO).await;
+        baseline_measurements.push(duration);
+    }
+    // Use median to reduce impact of outliers
+    baseline_measurements.sort();
+    let median_idx = baseline_measurements.len() / 2;
+    let baseline_duration = baseline_measurements[median_idx].max(Duration::from_millis(20));
 
     // Simulate scanner activity
     scanner.update_business_metrics(50, 500, 0, 25).await;
 
-    tokio::time::sleep(Duration::from_millis(100)).await;
+    tokio::time::sleep(Duration::from_millis(200)).await;
 
-    // Performance test: with scanner load
-    let with_scanner_duration_raw = measure_workload(5_000, Duration::from_millis(2)).await;
-    let with_scanner_duration = if with_scanner_duration_raw <= baseline_duration {
-        baseline_duration + Duration::from_millis(2)
-    } else {
-        with_scanner_duration_raw
-    };
+    // Performance test: with scanner load - measure multiple times for stability
+    let mut scanner_measurements = Vec::new();
+    for _ in 0..MEASUREMENT_COUNT {
+        let duration = measure_workload(10_000, Duration::ZERO).await;
+        scanner_measurements.push(duration);
+    }
+    scanner_measurements.sort();
+    let median_idx = scanner_measurements.len() / 2;
+    let with_scanner_duration = scanner_measurements[median_idx].max(baseline_duration);
 
     // Calculate performance impact
     let baseline_ns = baseline_duration.as_nanos().max(1) as f64;
@@ -281,8 +289,9 @@ async fn test_performance_impact_measurement() {
     println!("  Impact percentage: {impact_percentage:.2}%");
     println!("  Meets optimization goals: {}", benchmark.meets_optimization_goals());
 
-    // Verify optimization target (business impact < 10%)
-    // Note: In real environment this test may need longer time and real load
+    // Verify optimization target (business impact < 50%)
+    // Note: In test environment, allow higher threshold due to system load variability
+    // In production, the actual impact should be much lower (< 10%)
     assert!(impact_percentage < 50.0, "Performance impact too high: {impact_percentage:.2}%");
 
     io_monitor.stop().await;

crates/ahm/tests/lifecycle_cache_test.rs

@@ -23,16 +23,16 @@ use rustfs_ecstore::{
     store_api::{MakeBucketOptions, ObjectIO, ObjectInfo, ObjectOptions, PutObjReader, StorageAPI},
 };
 use serial_test::serial;
-use std::borrow::Cow;
-use std::sync::Once;
-use std::sync::OnceLock;
-use std::{path::PathBuf, sync::Arc};
-use tokio::fs;
-use tokio_util::sync::CancellationToken;
-use tracing::warn;
-use tracing::{debug, info};
+use std::{
+    borrow::Cow,
+    path::PathBuf,
+    sync::{Arc, Once, OnceLock},
+};
 //use heed_traits::Comparator;
 use time::OffsetDateTime;
+use tokio::fs;
+use tokio_util::sync::CancellationToken;
+use tracing::{debug, info, warn};
 use uuid::Uuid;
 
 static GLOBAL_ENV: OnceLock<(Vec<PathBuf>, Arc<ECStore>)> = OnceLock::new();
@@ -144,7 +144,7 @@ async fn setup_test_env() -> (Vec<PathBuf>, Arc<ECStore>) {
     let mut wtxn = lmdb_env.write_txn().unwrap();
     let db = match lmdb_env
         .database_options()
-        .name(&format!("bucket_{}", bucket_name))
+        .name(&format!("bucket_{bucket_name}"))
         .types::<I64<BigEndian>, LifecycleContentCodec>()
         .flags(DatabaseFlags::DUP_SORT)
         //.dup_sort_comparator::<>()
@@ -152,7 +152,7 @@ async fn setup_test_env() -> (Vec<PathBuf>, Arc<ECStore>) {
     {
         Ok(db) => db,
         Err(err) => {
-            panic!("lmdb error: {}", err);
+            panic!("lmdb error: {err}");
         }
     };
     let _ = wtxn.commit();
@@ -199,7 +199,7 @@ async fn upload_test_object(ecstore: &Arc<ECStore>, bucket: &str, object: &str,
         .await
         .expect("Failed to upload test object");
 
-    println!("object_info1: {:?}", object_info);
+    println!("object_info1: {object_info:?}");
 
     info!("Uploaded test object: {}/{} ({} bytes)", bucket, object, object_info.size);
 }
@@ -456,7 +456,7 @@ mod serial_tests {
                     }
 
                     let object_info = convert_record_to_object_info(record);
-                    println!("object_info2: {:?}", object_info);
+                    println!("object_info2: {object_info:?}");
                     let mod_time = object_info.mod_time.unwrap_or(OffsetDateTime::now_utc());
                     let expiry_time = rustfs_ecstore::bucket::lifecycle::lifecycle::expected_expiry_time(mod_time, 1);
 
@@ -494,9 +494,9 @@ mod serial_tests {
                             type_,
                             object_name,
                         } = &elm.1;
-                        println!("cache row:{} {} {} {:?} {}", ver_no, ver_id, mod_time, type_, object_name);
+                        println!("cache row:{ver_no} {ver_id} {mod_time} {type_:?} {object_name}");
                     }
-                    println!("row:{:?}", row);
+                    println!("row:{row:?}");
                 }
                 //drop(iter);
                 wtxn.commit().unwrap();

crates/ahm/tests/lifecycle_integration_test.rs

@@ -24,9 +24,11 @@ use rustfs_ecstore::{
     tier::tier_config::{TierConfig, TierMinIO, TierType},
 };
 use serial_test::serial;
-use std::sync::Once;
-use std::sync::OnceLock;
-use std::{path::PathBuf, sync::Arc, time::Duration};
+use std::{
+    path::PathBuf,
+    sync::{Arc, Once, OnceLock},
+    time::Duration,
+};
 use tokio::fs;
 use tokio_util::sync::CancellationToken;
 use tracing::info;
@@ -275,11 +277,11 @@ async fn create_test_tier(server: u32) {
     };
     let mut tier_config_mgr = GLOBAL_TierConfigMgr.write().await;
     if let Err(err) = tier_config_mgr.add(args, false).await {
-        println!("tier_config_mgr add failed, e: {:?}", err);
+        println!("tier_config_mgr add failed, e: {err:?}");
         panic!("tier add failed. {err}");
     }
     if let Err(e) = tier_config_mgr.save().await {
-        println!("tier_config_mgr save failed, e: {:?}", e);
+        println!("tier_config_mgr save failed, e: {e:?}");
         panic!("tier save failed");
     }
     println!("Created test tier: COLDTIER44");
@@ -297,7 +299,7 @@ async fn object_exists(ecstore: &Arc<ECStore>, bucket: &str, object: &str) -> bo
 #[allow(dead_code)]
 async fn object_is_delete_marker(ecstore: &Arc<ECStore>, bucket: &str, object: &str) -> bool {
     if let Ok(oi) = (**ecstore).get_object_info(bucket, object, &ObjectOptions::default()).await {
-        println!("oi: {:?}", oi);
+        println!("oi: {oi:?}");
         oi.delete_marker
     } else {
         println!("object_is_delete_marker is error");
@@ -309,7 +311,7 @@ async fn object_is_delete_marker(ecstore: &Arc<ECStore>, bucket: &str, object: &
 #[allow(dead_code)]
 async fn object_is_transitioned(ecstore: &Arc<ECStore>, bucket: &str, object: &str) -> bool {
     if let Ok(oi) = (**ecstore).get_object_info(bucket, object, &ObjectOptions::default()).await {
-        println!("oi: {:?}", oi);
+        println!("oi: {oi:?}");
         !oi.transitioned_object.status.is_empty()
     } else {
         println!("object_is_transitioned is error");

crates/ahm/tests/optimized_scanner_tests.rs

@@ -12,26 +12,23 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use std::{fs, net::SocketAddr, sync::Arc, sync::OnceLock, time::Duration};
-use tempfile::TempDir;
-use tokio_util::sync::CancellationToken;
-
-use serial_test::serial;
-
 use rustfs_ahm::heal::manager::HealConfig;
 use rustfs_ahm::scanner::{
     Scanner,
     data_scanner::ScanMode,
     node_scanner::{LoadLevel, NodeScanner, NodeScannerConfig},
 };
-
-use rustfs_ecstore::disk::endpoint::Endpoint;
-use rustfs_ecstore::endpoints::{EndpointServerPools, Endpoints, PoolEndpoints};
-use rustfs_ecstore::store::ECStore;
 use rustfs_ecstore::{
     StorageAPI,
+    disk::endpoint::Endpoint,
+    endpoints::{EndpointServerPools, Endpoints, PoolEndpoints},
+    store::ECStore,
     store_api::{MakeBucketOptions, ObjectIO, PutObjReader},
 };
+use serial_test::serial;
+use std::{fs, net::SocketAddr, sync::Arc, sync::OnceLock, time::Duration};
+use tempfile::TempDir;
+use tokio_util::sync::CancellationToken;
 
 // Global test environment cache to avoid repeated initialization
 static GLOBAL_TEST_ENV: OnceLock<(Vec<std::path::PathBuf>, Arc<ECStore>)> = OnceLock::new();

crates/ahm/tests/scanner_optimization_tests.rs

@@ -12,9 +12,6 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use std::time::Duration;
-use tempfile::TempDir;
-
 use rustfs_ahm::scanner::{
     checkpoint::{CheckpointData, CheckpointManager},
     io_monitor::{AdvancedIOMonitor, IOMonitorConfig},
@@ -23,6 +20,8 @@ use rustfs_ahm::scanner::{
     node_scanner::{LoadLevel, NodeScanner, NodeScannerConfig, ScanProgress},
     stats_aggregator::{DecentralizedStatsAggregator, DecentralizedStatsAggregatorConfig},
 };
+use std::time::Duration;
+use tempfile::TempDir;
 
 #[tokio::test]
 async fn test_checkpoint_manager_save_and_load() {

crates/appauth/Cargo.toml

@@ -29,6 +29,7 @@ base64-simd = { workspace = true }
 rsa = { workspace = true }
 serde.workspace = true
 serde_json.workspace = true
+rand.workspace = true
 
 [lints]
 workspace = true

crates/appauth/src/token.rs

@@ -12,11 +12,9 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use rsa::Pkcs1v15Encrypt;
 use rsa::{
-    RsaPrivateKey, RsaPublicKey,
+    Pkcs1v15Encrypt, RsaPrivateKey, RsaPublicKey,
     pkcs8::{DecodePrivateKey, DecodePublicKey},
-    rand_core::OsRng,
 };
 use serde::{Deserialize, Serialize};
 use std::io::{Error, Result};
@@ -33,8 +31,9 @@ pub struct Token {
 /// Returns the encrypted string processed by base64
 pub fn gencode(token: &Token, key: &str) -> Result<String> {
     let data = serde_json::to_vec(token)?;
+    let mut rng = rand::rng();
     let public_key = RsaPublicKey::from_public_key_pem(key).map_err(Error::other)?;
-    let encrypted_data = public_key.encrypt(&mut OsRng, Pkcs1v15Encrypt, &data).map_err(Error::other)?;
+    let encrypted_data = public_key.encrypt(&mut rng, Pkcs1v15Encrypt, &data).map_err(Error::other)?;
     Ok(base64_simd::URL_SAFE_NO_PAD.encode_to_string(&encrypted_data))
 }
 
@@ -76,9 +75,10 @@ mod tests {
         pkcs8::{EncodePrivateKey, EncodePublicKey, LineEnding},
     };
     use std::time::{SystemTime, UNIX_EPOCH};
+
     #[test]
     fn test_gencode_and_parse() {
-        let mut rng = OsRng;
+        let mut rng = rand::rng();
         let bits = 2048;
         let private_key = RsaPrivateKey::new(&mut rng, bits).expect("Failed to generate private key");
         let public_key = RsaPublicKey::from(&private_key);
@@ -101,7 +101,8 @@ mod tests {
 
     #[test]
     fn test_parse_invalid_token() {
-        let private_key_pem = RsaPrivateKey::new(&mut OsRng, 2048)
+        let mut rng = rand::rng();
+        let private_key_pem = RsaPrivateKey::new(&mut rng, 2048)
             .expect("Failed to generate private key")
             .to_pkcs8_pem(LineEnding::LF)
             .unwrap();

crates/audit/Cargo.toml

@@ -30,7 +30,10 @@ rustfs-targets = { workspace = true }
 rustfs-config = { workspace = true, features = ["audit", "constants"] }
 rustfs-ecstore = { workspace = true }
 chrono = { workspace = true }
+const-str = { workspace = true }
 futures = { workspace = true }
+hashbrown = { workspace = true }
+metrics = { workspace = true }
 serde = { workspace = true }
 serde_json = { workspace = true }
 thiserror = { workspace = true }
@@ -39,5 +42,6 @@ tracing = { workspace = true, features = ["std", "attributes"] }
 url = { workspace = true }
 rumqttc = { workspace = true }
 
+
 [lints]
 workspace = true

crates/audit/src/entity.rs

@@ -13,18 +13,10 @@
 // limitations under the License.
 
 use chrono::{DateTime, Utc};
+use hashbrown::HashMap;
 use rustfs_targets::EventName;
 use serde::{Deserialize, Serialize};
 use serde_json::Value;
-use std::collections::HashMap;
-
-/// Trait for types that can be serialized to JSON and have a timestamp
-pub trait LogRecord {
-    /// Serialize the record to a JSON string
-    fn to_json(&self) -> String;
-    /// Get the timestamp of the record
-    fn get_timestamp(&self) -> chrono::DateTime<chrono::Utc>;
-}
 
 /// ObjectVersion represents an object version with key and versionId
 #[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Eq, Default)]
@@ -36,19 +28,12 @@ pub struct ObjectVersion {
 }
 
 impl ObjectVersion {
-    /// Set the object name (chainable)
-    pub fn set_object_name(&mut self, name: String) -> &mut Self {
-        self.object_name = name;
-        self
-    }
-    /// Set the version ID (chainable)
-    pub fn set_version_id(&mut self, version_id: Option<String>) -> &mut Self {
-        self.version_id = version_id;
-        self
+    pub fn new(object_name: String, version_id: Option<String>) -> Self {
+        Self { object_name, version_id }
     }
 }
 
-/// ApiDetails contains API information for the audit entry
+/// `ApiDetails` contains API information for the audit entry.
 #[derive(Debug, Clone, Serialize, Deserialize, Default)]
 pub struct ApiDetails {
     #[serde(skip_serializing_if = "Option::is_none")]
@@ -79,75 +64,86 @@ pub struct ApiDetails {
     pub time_to_response_in_ns: Option<String>,
 }
 
-impl ApiDetails {
-    /// Set API name (chainable)
-    pub fn set_name(&mut self, name: Option<String>) -> &mut Self {
-        self.name = name;
+/// Builder for `ApiDetails`.
+#[derive(Default, Clone)]
+pub struct ApiDetailsBuilder(pub ApiDetails);
+
+impl ApiDetailsBuilder {
+    pub fn new() -> Self {
+        Self::default()
+    }
+
+    pub fn name(mut self, name: impl Into<String>) -> Self {
+        self.0.name = Some(name.into());
         self
     }
-    /// Set bucket name (chainable)
-    pub fn set_bucket(&mut self, bucket: Option<String>) -> &mut Self {
-        self.bucket = bucket;
+
+    pub fn bucket(mut self, bucket: impl Into<String>) -> Self {
+        self.0.bucket = Some(bucket.into());
         self
     }
-    /// Set object name (chainable)
-    pub fn set_object(&mut self, object: Option<String>) -> &mut Self {
-        self.object = object;
+
+    pub fn object(mut self, object: impl Into<String>) -> Self {
+        self.0.object = Some(object.into());
         self
     }
-    /// Set objects list (chainable)
-    pub fn set_objects(&mut self, objects: Option<Vec<ObjectVersion>>) -> &mut Self {
-        self.objects = objects;
+
+    pub fn objects(mut self, objects: Vec<ObjectVersion>) -> Self {
+        self.0.objects = Some(objects);
         self
     }
-    /// Set status (chainable)
-    pub fn set_status(&mut self, status: Option<String>) -> &mut Self {
-        self.status = status;
+
+    pub fn status(mut self, status: impl Into<String>) -> Self {
+        self.0.status = Some(status.into());
         self
     }
-    /// Set status code (chainable)
-    pub fn set_status_code(&mut self, code: Option<i32>) -> &mut Self {
-        self.status_code = code;
+
+    pub fn status_code(mut self, code: i32) -> Self {
+        self.0.status_code = Some(code);
         self
     }
-    /// Set input bytes (chainable)
-    pub fn set_input_bytes(&mut self, bytes: Option<i64>) -> &mut Self {
-        self.input_bytes = bytes;
+
+    pub fn input_bytes(mut self, bytes: i64) -> Self {
+        self.0.input_bytes = Some(bytes);
         self
     }
-    /// Set output bytes (chainable)
-    pub fn set_output_bytes(&mut self, bytes: Option<i64>) -> &mut Self {
-        self.output_bytes = bytes;
+
+    pub fn output_bytes(mut self, bytes: i64) -> Self {
+        self.0.output_bytes = Some(bytes);
         self
     }
-    /// Set header bytes (chainable)
-    pub fn set_header_bytes(&mut self, bytes: Option<i64>) -> &mut Self {
-        self.header_bytes = bytes;
+
+    pub fn header_bytes(mut self, bytes: i64) -> Self {
+        self.0.header_bytes = Some(bytes);
         self
     }
-    /// Set time to first byte (chainable)
-    pub fn set_time_to_first_byte(&mut self, t: Option<String>) -> &mut Self {
-        self.time_to_first_byte = t;
+
+    pub fn time_to_first_byte(mut self, t: impl Into<String>) -> Self {
+        self.0.time_to_first_byte = Some(t.into());
         self
     }
-    /// Set time to first byte in nanoseconds (chainable)
-    pub fn set_time_to_first_byte_in_ns(&mut self, t: Option<String>) -> &mut Self {
-        self.time_to_first_byte_in_ns = t;
+
+    pub fn time_to_first_byte_in_ns(mut self, t: impl Into<String>) -> Self {
+        self.0.time_to_first_byte_in_ns = Some(t.into());
         self
     }
-    /// Set time to response (chainable)
-    pub fn set_time_to_response(&mut self, t: Option<String>) -> &mut Self {
-        self.time_to_response = t;
+
+    pub fn time_to_response(mut self, t: impl Into<String>) -> Self {
+        self.0.time_to_response = Some(t.into());
         self
     }
-    /// Set time to response in nanoseconds (chainable)
-    pub fn set_time_to_response_in_ns(&mut self, t: Option<String>) -> &mut Self {
-        self.time_to_response_in_ns = t;
+
+    pub fn time_to_response_in_ns(mut self, t: impl Into<String>) -> Self {
+        self.0.time_to_response_in_ns = Some(t.into());
         self
     }
+
+    pub fn build(self) -> ApiDetails {
+        self.0
+    }
 }
 
-/// AuditEntry represents an audit log entry
+/// `AuditEntry` represents an audit log entry.
 #[derive(Debug, Clone, Serialize, Deserialize, Default)]
 pub struct AuditEntry {
     pub version: String,
@@ -155,6 +151,7 @@ pub struct AuditEntry {
     pub deployment_id: Option<String>,
     #[serde(rename = "siteName", skip_serializing_if = "Option::is_none")]
     pub site_name: Option<String>,
+    #[serde(with = "chrono::serde::ts_milliseconds")]
     pub time: DateTime<Utc>,
     pub event: EventName,
     #[serde(rename = "type", skip_serializing_if = "Option::is_none")]
@@ -191,200 +188,130 @@ pub struct AuditEntry {
     pub error: Option<String>,
 }
 
-impl AuditEntry {
-    /// Create a new AuditEntry with required fields
-    #[allow(clippy::too_many_arguments)]
-    pub fn new(
-        version: String,
-        deployment_id: Option<String>,
-        site_name: Option<String>,
-        time: DateTime<Utc>,
-        event: EventName,
-        entry_type: Option<String>,
-        trigger: String,
-        api: ApiDetails,
-    ) -> Self {
-        AuditEntry {
-            version,
-            deployment_id,
-            site_name,
-            time,
+/// Constructor for `AuditEntry`.
+pub struct AuditEntryBuilder(AuditEntry);
+
+impl AuditEntryBuilder {
+    /// Create a new builder with all required fields.
+    pub fn new(version: impl Into<String>, event: EventName, trigger: impl Into<String>, api: ApiDetails) -> Self {
+        Self(AuditEntry {
+            version: version.into(),
+            time: Utc::now(),
             event,
-            entry_type,
-            trigger,
+            trigger: trigger.into(),
             api,
-            remote_host: None,
-            request_id: None,
-            user_agent: None,
-            req_path: None,
-            req_host: None,
-            req_node: None,
-            req_claims: None,
-            req_query: None,
-            req_header: None,
-            resp_header: None,
-            tags: None,
-            access_key: None,
-            parent_user: None,
-            error: None,
-        }
+            ..Default::default()
+        })
     }
 
-    /// Set version (chainable)
-    pub fn set_version(&mut self, version: String) -> &mut Self {
-        self.version = version;
-        self
-    }
-    /// Set deployment ID (chainable)
-    pub fn set_deployment_id(&mut self, id: Option<String>) -> &mut Self {
-        self.deployment_id = id;
-        self
-    }
-    /// Set site name (chainable)
-    pub fn set_site_name(&mut self, name: Option<String>) -> &mut Self {
-        self.site_name = name;
-        self
-    }
-    /// Set time (chainable)
-    pub fn set_time(&mut self, time: DateTime<Utc>) -> &mut Self {
-        self.time = time;
-        self
-    }
-    /// Set event (chainable)
-    pub fn set_event(&mut self, event: EventName) -> &mut Self {
-        self.event = event;
-        self
-    }
-    /// Set entry type (chainable)
-    pub fn set_entry_type(&mut self, entry_type: Option<String>) -> &mut Self {
-        self.entry_type = entry_type;
-        self
-    }
-    /// Set trigger (chainable)
-    pub fn set_trigger(&mut self, trigger: String) -> &mut Self {
-        self.trigger = trigger;
-        self
-    }
-    /// Set API details (chainable)
-    pub fn set_api(&mut self, api: ApiDetails) -> &mut Self {
-        self.api = api;
-        self
-    }
-    /// Set remote host (chainable)
-    pub fn set_remote_host(&mut self, host: Option<String>) -> &mut Self {
-        self.remote_host = host;
-        self
-    }
-    /// Set request ID (chainable)
-    pub fn set_request_id(&mut self, id: Option<String>) -> &mut Self {
-        self.request_id = id;
-        self
-    }
-    /// Set user agent (chainable)
-    pub fn set_user_agent(&mut self, agent: Option<String>) -> &mut Self {
-        self.user_agent = agent;
-        self
-    }
-    /// Set request path (chainable)
-    pub fn set_req_path(&mut self, path: Option<String>) -> &mut Self {
-        self.req_path = path;
-        self
-    }
-    /// Set request host (chainable)
-    pub fn set_req_host(&mut self, host: Option<String>) -> &mut Self {
-        self.req_host = host;
-        self
-    }
-    /// Set request node (chainable)
-    pub fn set_req_node(&mut self, node: Option<String>) -> &mut Self {
-        self.req_node = node;
-        self
-    }
-    /// Set request claims (chainable)
-    pub fn set_req_claims(&mut self, claims: Option<HashMap<String, Value>>) -> &mut Self {
-        self.req_claims = claims;
-        self
-    }
-    /// Set request query (chainable)
-    pub fn set_req_query(&mut self, query: Option<HashMap<String, String>>) -> &mut Self {
-        self.req_query = query;
-        self
-    }
-    /// Set request header (chainable)
-    pub fn set_req_header(&mut self, header: Option<HashMap<String, String>>) -> &mut Self {
-        self.req_header = header;
-        self
-    }
-    /// Set response header (chainable)
-    pub fn set_resp_header(&mut self, header: Option<HashMap<String, String>>) -> &mut Self {
-        self.resp_header = header;
-        self
-    }
-    /// Set tags (chainable)
-    pub fn set_tags(&mut self, tags: Option<HashMap<String, Value>>) -> &mut Self {
-        self.tags = tags;
-        self
-    }
-    /// Set access key (chainable)
-    pub fn set_access_key(&mut self, key: Option<String>) -> &mut Self {
-        self.access_key = key;
-        self
-    }
-    /// Set parent user (chainable)
-    pub fn set_parent_user(&mut self, user: Option<String>) -> &mut Self {
-        self.parent_user = user;
-        self
-    }
-    /// Set error message (chainable)
-    pub fn set_error(&mut self, error: Option<String>) -> &mut Self {
-        self.error = error;
+    // event
+    pub fn version(mut self, version: impl Into<String>) -> Self {
+        self.0.version = version.into();
         self
     }
 
-    /// Build AuditEntry from context or parameters (example, can be extended)
-    pub fn from_context(
-        version: String,
-        deployment_id: Option<String>,
-        time: DateTime<Utc>,
-        event: EventName,
-        trigger: String,
-        api: ApiDetails,
-        tags: Option<HashMap<String, Value>>,
-    ) -> Self {
-        AuditEntry {
-            version,
-            deployment_id,
-            site_name: None,
-            time,
-            event,
-            entry_type: None,
-            trigger,
-            api,
-            remote_host: None,
-            request_id: None,
-            user_agent: None,
-            req_path: None,
-            req_host: None,
-            req_node: None,
-            req_claims: None,
-            req_query: None,
-            req_header: None,
-            resp_header: None,
-            tags,
-            access_key: None,
-            parent_user: None,
-            error: None,
-        }
-    }
-}
-
-impl LogRecord for AuditEntry {
-    /// Serialize AuditEntry to JSON string
-    fn to_json(&self) -> String {
-        serde_json::to_string(self).unwrap_or_else(|_| String::from("{}"))
-    }
-    /// Get the timestamp of the audit entry
-    fn get_timestamp(&self) -> DateTime<Utc> {
-        self.time
+    pub fn event(mut self, event: EventName) -> Self {
+        self.0.event = event;
+        self
+    }
+
+    pub fn api(mut self, api_details: ApiDetails) -> Self {
+        self.0.api = api_details;
+        self
+    }
+
+    pub fn deployment_id(mut self, id: impl Into<String>) -> Self {
+        self.0.deployment_id = Some(id.into());
+        self
+    }
+
+    pub fn site_name(mut self, name: impl Into<String>) -> Self {
+        self.0.site_name = Some(name.into());
+        self
+    }
+
+    pub fn time(mut self, time: DateTime<Utc>) -> Self {
+        self.0.time = time;
+        self
+    }
+
+    pub fn entry_type(mut self, entry_type: impl Into<String>) -> Self {
+        self.0.entry_type = Some(entry_type.into());
+        self
+    }
+
+    pub fn remote_host(mut self, host: impl Into<String>) -> Self {
+        self.0.remote_host = Some(host.into());
+        self
+    }
+
+    pub fn request_id(mut self, id: impl Into<String>) -> Self {
+        self.0.request_id = Some(id.into());
+        self
+    }
+
+    pub fn user_agent(mut self, agent: impl Into<String>) -> Self {
+        self.0.user_agent = Some(agent.into());
+        self
+    }
+
+    pub fn req_path(mut self, path: impl Into<String>) -> Self {
+        self.0.req_path = Some(path.into());
+        self
+    }
+
+    pub fn req_host(mut self, host: impl Into<String>) -> Self {
+        self.0.req_host = Some(host.into());
+        self
+    }
+
+    pub fn req_node(mut self, node: impl Into<String>) -> Self {
+        self.0.req_node = Some(node.into());
+        self
+    }
+
+    pub fn req_claims(mut self, claims: HashMap<String, Value>) -> Self {
+        self.0.req_claims = Some(claims);
+        self
+    }
+
+    pub fn req_query(mut self, query: HashMap<String, String>) -> Self {
+        self.0.req_query = Some(query);
+        self
+    }
+
+    pub fn req_header(mut self, header: HashMap<String, String>) -> Self {
+        self.0.req_header = Some(header);
+        self
+    }
+
+    pub fn resp_header(mut self, header: HashMap<String, String>) -> Self {
+        self.0.resp_header = Some(header);
+        self
+    }
+
+    pub fn tags(mut self, tags: HashMap<String, Value>) -> Self {
+        self.0.tags = Some(tags);
+        self
+    }
+
+    pub fn access_key(mut self, key: impl Into<String>) -> Self {
+        self.0.access_key = Some(key.into());
+        self
+    }
+
+    pub fn parent_user(mut self, user: impl Into<String>) -> Self {
+        self.0.parent_user = Some(user.into());
+        self
+    }
+
+    pub fn error(mut self, error: impl Into<String>) -> Self {
+        self.0.error = Some(error.into());
+        self
+    }
+
+    /// Construct the final `AuditEntry`.
+    pub fn build(self) -> AuditEntry {
+        self.0
     }
 }

crates/audit/src/error.rs

@@ -21,7 +21,7 @@ pub type AuditResult<T> = Result<T, AuditError>;
 #[derive(Error, Debug)]
 pub enum AuditError {
     #[error("Configuration error: {0}")]
-    Configuration(String),
+    Configuration(String, #[source] Option<Box<dyn std::error::Error + Send + Sync>>),
 
     #[error("config not loaded")]
     ConfigNotLoaded,
@@ -35,11 +35,14 @@ pub enum AuditError {
     #[error("System already initialized")]
     AlreadyInitialized,
 
+    #[error("Storage not available: {0}")]
+    StorageNotAvailable(String),
+
     #[error("Failed to save configuration: {0}")]
-    SaveConfig(String),
+    SaveConfig(#[source] Box<dyn std::error::Error + Send + Sync>),
 
     #[error("Failed to load configuration: {0}")]
-    LoadConfig(String),
+    LoadConfig(#[source] Box<dyn std::error::Error + Send + Sync>),
 
     #[error("Serialization error: {0}")]
     Serialization(#[from] serde_json::Error),
@@ -49,7 +52,4 @@ pub enum AuditError {
 
     #[error("Join error: {0}")]
     Join(#[from] tokio::task::JoinError),
-
-    #[error("Server storage not initialized: {0}")]
-    ServerNotInitialized(String),
 }

crates/audit/src/global.rs

@@ -15,7 +15,7 @@
 use crate::{AuditEntry, AuditResult, AuditSystem};
 use rustfs_ecstore::config::Config;
 use std::sync::{Arc, OnceLock};
-use tracing::{error, warn};
+use tracing::{debug, error, trace, warn};
 
 /// Global audit system instance
 static AUDIT_SYSTEM: OnceLock<Arc<AuditSystem>> = OnceLock::new();
@@ -30,6 +30,19 @@ pub fn audit_system() -> Option<Arc<AuditSystem>> {
     AUDIT_SYSTEM.get().cloned()
 }
 
+/// A helper macro for executing closures if the global audit system is initialized.
+/// If not initialized, log a warning and return `Ok(())`.
+macro_rules! with_audit_system {
+    ($async_closure:expr) => {
+        if let Some(system) = audit_system() {
+            (async move { $async_closure(system).await }).await
+        } else {
+            warn!("Audit system not initialized, operation skipped.");
+            Ok(())
+        }
+    };
+}
+
 /// Start the global audit system with configuration
 pub async fn start_audit_system(config: Config) -> AuditResult<()> {
     let system = init_audit_system();
@@ -38,32 +51,17 @@ pub async fn start_audit_system(config: Config) -> AuditResult<()> {
 
 /// Stop the global audit system
 pub async fn stop_audit_system() -> AuditResult<()> {
-    if let Some(system) = audit_system() {
-        system.close().await
-    } else {
-        warn!("Audit system not initialized, cannot stop");
-        Ok(())
-    }
+    with_audit_system!(|system: Arc<AuditSystem>| async move { system.close().await })
 }
 
 /// Pause the global audit system
 pub async fn pause_audit_system() -> AuditResult<()> {
-    if let Some(system) = audit_system() {
-        system.pause().await
-    } else {
-        warn!("Audit system not initialized, cannot pause");
-        Ok(())
-    }
+    with_audit_system!(|system: Arc<AuditSystem>| async move { system.pause().await })
 }
 
 /// Resume the global audit system
 pub async fn resume_audit_system() -> AuditResult<()> {
-    if let Some(system) = audit_system() {
-        system.resume().await
-    } else {
-        warn!("Audit system not initialized, cannot resume");
-        Ok(())
-    }
+    with_audit_system!(|system: Arc<AuditSystem>| async move { system.resume().await })
 }
 
 /// Dispatch an audit log entry to all targets
@@ -72,23 +70,23 @@ pub async fn dispatch_audit_log(entry: Arc<AuditEntry>) -> AuditResult<()> {
         if system.is_running().await {
             system.dispatch(entry).await
         } else {
-            // System not running, just drop the log entry without error
+            // The system is initialized but not running (for example, it is suspended). Silently discard log entries based on original logic.
+            // For debugging purposes, it can be useful to add a trace log here.
+            trace!("Audit system is not running, dropping audit entry.");
             Ok(())
         }
     } else {
-        // System not initialized, just drop the log entry without error
+        // The system is not initialized at all. This is a more important state.
+        // It might be better to return an error or log a warning.
+        debug!("Audit system not initialized, dropping audit entry.");
+        // If this should be a hard failure, you can return Err(AuditError::NotInitialized("..."))
         Ok(())
     }
 }
 
 /// Reload the global audit system configuration
 pub async fn reload_audit_config(config: Config) -> AuditResult<()> {
-    if let Some(system) = audit_system() {
-        system.reload_config(config).await
-    } else {
-        warn!("Audit system not initialized, cannot reload config");
-        Ok(())
-    }
+    with_audit_system!(|system: Arc<AuditSystem>| async move { system.reload_config(config).await })
 }
 
 /// Check if the global audit system is running

crates/audit/src/lib.rs

@@ -25,7 +25,7 @@ pub mod observability;
 pub mod registry;
 pub mod system;
 
-pub use entity::{ApiDetails, AuditEntry, LogRecord, ObjectVersion};
+pub use entity::{ApiDetails, AuditEntry, ObjectVersion};
 pub use error::{AuditError, AuditResult};
 pub use global::*;
 pub use observability::{AuditMetrics, AuditMetricsReport, PerformanceValidation};

crates/audit/src/observability.rs

@@ -21,12 +21,47 @@
 //! - Error rate monitoring
 //! - Queue depth monitoring
 
+use metrics::{counter, describe_counter, describe_gauge, describe_histogram, gauge, histogram};
 use std::sync::atomic::{AtomicU64, Ordering};
 use std::sync::{Arc, OnceLock};
 use std::time::{Duration, Instant};
 use tokio::sync::RwLock;
 use tracing::info;
 
+const RUSTFS_AUDIT_METRICS_NAMESPACE: &str = "rustfs.audit.";
+
+const M_AUDIT_EVENTS_TOTAL: &str = const_str::concat!(RUSTFS_AUDIT_METRICS_NAMESPACE, "events.total");
+const M_AUDIT_EVENTS_FAILED: &str = const_str::concat!(RUSTFS_AUDIT_METRICS_NAMESPACE, "events.failed");
+const M_AUDIT_DISPATCH_NS: &str = const_str::concat!(RUSTFS_AUDIT_METRICS_NAMESPACE, "dispatch.ns");
+const M_AUDIT_EPS: &str = const_str::concat!(RUSTFS_AUDIT_METRICS_NAMESPACE, "eps");
+const M_AUDIT_TARGET_OPS: &str = const_str::concat!(RUSTFS_AUDIT_METRICS_NAMESPACE, "target.ops");
+const M_AUDIT_CONFIG_RELOADS: &str = const_str::concat!(RUSTFS_AUDIT_METRICS_NAMESPACE, "config.reloads");
+const M_AUDIT_SYSTEM_STARTS: &str = const_str::concat!(RUSTFS_AUDIT_METRICS_NAMESPACE, "system.starts");
+
+const L_RESULT: &str = "result";
+const L_STATUS: &str = "status";
+
+const V_SUCCESS: &str = "success";
+const V_FAILURE: &str = "failure";
+
+/// One-time registration of indicator meta information
+/// This function ensures that metric descriptors are registered only once.
+pub fn init_observability_metrics() {
+    static METRICS_DESC_INIT: OnceLock<()> = OnceLock::new();
+    METRICS_DESC_INIT.get_or_init(|| {
+        // Event/Time-consuming
+        describe_counter!(M_AUDIT_EVENTS_TOTAL, "Total audit events (labeled by result).");
+        describe_counter!(M_AUDIT_EVENTS_FAILED, "Total failed audit events.");
+        describe_histogram!(M_AUDIT_DISPATCH_NS, "Dispatch time per event (ns).");
+        describe_gauge!(M_AUDIT_EPS, "Events per second since last reset.");
+
+        // Target operation/system event
+        describe_counter!(M_AUDIT_TARGET_OPS, "Total target operations (labeled by status).");
+        describe_counter!(M_AUDIT_CONFIG_RELOADS, "Total configuration reloads.");
+        describe_counter!(M_AUDIT_SYSTEM_STARTS, "Total system starts.");
+    });
+}
+
 /// Metrics collector for audit system observability
 #[derive(Debug)]
 pub struct AuditMetrics {
@@ -56,6 +91,7 @@ impl Default for AuditMetrics {
 impl AuditMetrics {
     /// Creates a new metrics collector
     pub fn new() -> Self {
+        init_observability_metrics();
         Self {
             total_events_processed: AtomicU64::new(0),
             total_events_failed: AtomicU64::new(0),
@@ -68,11 +104,28 @@ impl AuditMetrics {
         }
     }
 
+    // Suggestion: Call this auxiliary function in the existing "Successful Event Recording" method body to complete the instrumentation
+    #[inline]
+    fn emit_event_success_metrics(&self, dispatch_time: Duration) {
+        // count + histogram
+        counter!(M_AUDIT_EVENTS_TOTAL, L_RESULT => V_SUCCESS).increment(1);
+        histogram!(M_AUDIT_DISPATCH_NS).record(dispatch_time.as_nanos() as f64);
+    }
+
+    // Suggestion: Call this auxiliary function in the existing "Failure Event Recording" method body to complete the instrumentation
+    #[inline]
+    fn emit_event_failure_metrics(&self, dispatch_time: Duration) {
+        counter!(M_AUDIT_EVENTS_TOTAL, L_RESULT => V_FAILURE).increment(1);
+        counter!(M_AUDIT_EVENTS_FAILED).increment(1);
+        histogram!(M_AUDIT_DISPATCH_NS).record(dispatch_time.as_nanos() as f64);
+    }
+
     /// Records a successful event dispatch
     pub fn record_event_success(&self, dispatch_time: Duration) {
         self.total_events_processed.fetch_add(1, Ordering::Relaxed);
         self.total_dispatch_time_ns
             .fetch_add(dispatch_time.as_nanos() as u64, Ordering::Relaxed);
+        self.emit_event_success_metrics(dispatch_time);
     }
 
     /// Records a failed event dispatch
@@ -80,27 +133,32 @@ impl AuditMetrics {
         self.total_events_failed.fetch_add(1, Ordering::Relaxed);
         self.total_dispatch_time_ns
             .fetch_add(dispatch_time.as_nanos() as u64, Ordering::Relaxed);
+        self.emit_event_failure_metrics(dispatch_time);
     }
 
     /// Records a successful target operation
     pub fn record_target_success(&self) {
         self.target_success_count.fetch_add(1, Ordering::Relaxed);
+        counter!(M_AUDIT_TARGET_OPS, L_STATUS => V_SUCCESS).increment(1);
     }
 
     /// Records a failed target operation
     pub fn record_target_failure(&self) {
         self.target_failure_count.fetch_add(1, Ordering::Relaxed);
+        counter!(M_AUDIT_TARGET_OPS, L_STATUS => V_FAILURE).increment(1);
     }
 
     /// Records a configuration reload
     pub fn record_config_reload(&self) {
         self.config_reload_count.fetch_add(1, Ordering::Relaxed);
+        counter!(M_AUDIT_CONFIG_RELOADS).increment(1);
         info!("Audit configuration reloaded");
     }
 
     /// Records a system start
     pub fn record_system_start(&self) {
         self.system_start_count.fetch_add(1, Ordering::Relaxed);
+        counter!(M_AUDIT_SYSTEM_STARTS).increment(1);
         info!("Audit system started");
     }
 
@@ -110,11 +168,14 @@ impl AuditMetrics {
         let elapsed = reset_time.elapsed();
         let total_events = self.total_events_processed.load(Ordering::Relaxed) + self.total_events_failed.load(Ordering::Relaxed);
 
-        if elapsed.as_secs_f64() > 0.0 {
+        let eps = if elapsed.as_secs_f64() > 0.0 {
             total_events as f64 / elapsed.as_secs_f64()
         } else {
             0.0
-        }
+        };
+        // EPS is reported in gauge
+        gauge!(M_AUDIT_EPS).set(eps);
+        eps
     }
 
     /// Gets the average dispatch latency in milliseconds
@@ -166,6 +227,8 @@ impl AuditMetrics {
         let mut reset_time = self.last_reset_time.write().await;
         *reset_time = Instant::now();
 
+        // Reset EPS to zero after reset
+        gauge!(M_AUDIT_EPS).set(0.0);
         info!("Audit metrics reset");
     }
 

crates/audit/src/registry.rs

@@ -14,6 +14,7 @@
 
 use crate::{AuditEntry, AuditError, AuditResult};
 use futures::{StreamExt, stream::FuturesUnordered};
+use hashbrown::{HashMap, HashSet};
 use rustfs_config::{
     DEFAULT_DELIMITER, ENABLE_KEY, ENV_PREFIX, MQTT_BROKER, MQTT_KEEP_ALIVE_INTERVAL, MQTT_PASSWORD, MQTT_QOS, MQTT_QUEUE_DIR,
     MQTT_QUEUE_LIMIT, MQTT_RECONNECT_INTERVAL, MQTT_TOPIC, MQTT_USERNAME, WEBHOOK_AUTH_TOKEN, WEBHOOK_BATCH_SIZE,
@@ -25,7 +26,6 @@ use rustfs_targets::{
     Target, TargetError,
     target::{ChannelTargetType, TargetType, mqtt::MQTTArgs, webhook::WebhookArgs},
 };
-use std::collections::{HashMap, HashSet};
 use std::sync::Arc;
 use std::time::Duration;
 use tracing::{debug, error, info, warn};
@@ -251,7 +251,7 @@ impl AuditRegistry {
             sections.extend(successes_by_section.keys().cloned());
 
             for section_name in sections {
-                let mut section_map: HashMap<String, KVS> = HashMap::new();
+                let mut section_map: std::collections::HashMap<String, KVS> = std::collections::HashMap::new();
 
                 // The default entry (if present) is written back to `_`
                 if let Some(default_cfg) = section_defaults.get(&section_name) {
@@ -277,7 +277,7 @@ impl AuditRegistry {
 
             // 7. Save the new configuration to the system
             let Some(store) = rustfs_ecstore::new_object_layer_fn() else {
-                return Err(AuditError::ServerNotInitialized(
+                return Err(AuditError::StorageNotAvailable(
                     "Failed to save target configuration: server storage not initialized".to_string(),
                 ));
             };
@@ -286,7 +286,7 @@ impl AuditRegistry {
                 Ok(_) => info!("New audit configuration saved to system successfully"),
                 Err(e) => {
                     error!(error = %e, "Failed to save new audit configuration");
-                    return Err(AuditError::SaveConfig(e.to_string()));
+                    return Err(AuditError::SaveConfig(Box::new(e)));
                 }
             }
         }

crates/audit/src/system.rs

@@ -59,7 +59,7 @@ impl AuditSystem {
 
     /// Starts the audit system with the given configuration
     pub async fn start(&self, config: Config) -> AuditResult<()> {
-        let mut state = self.state.write().await;
+        let state = self.state.write().await;
 
         match *state {
             AuditSystemState::Running => {
@@ -72,7 +72,6 @@ impl AuditSystem {
             _ => {}
         }
 
-        *state = AuditSystemState::Starting;
         drop(state);
 
         info!("Starting audit system");
@@ -90,6 +89,17 @@ impl AuditSystem {
         let mut registry = self.registry.lock().await;
         match registry.create_targets_from_config(&config).await {
             Ok(targets) => {
+                if targets.is_empty() {
+                    info!("No enabled audit targets found, keeping audit system stopped");
+                    drop(registry);
+                    return Ok(());
+                }
+
+                {
+                    let mut state = self.state.write().await;
+                    *state = AuditSystemState::Starting;
+                }
+
                 info!(target_count = targets.len(), "Created audit targets successfully");
 
                 // Initialize all targets
@@ -146,7 +156,7 @@ impl AuditSystem {
                 warn!("Audit system is already paused");
                 Ok(())
             }
-            _ => Err(AuditError::Configuration("Cannot pause audit system in current state".to_string())),
+            _ => Err(AuditError::Configuration("Cannot pause audit system in current state".to_string(), None)),
         }
     }
 
@@ -164,7 +174,7 @@ impl AuditSystem {
                 warn!("Audit system is already running");
                 Ok(())
             }
-            _ => Err(AuditError::Configuration("Cannot resume audit system in current state".to_string())),
+            _ => Err(AuditError::Configuration("Cannot resume audit system in current state".to_string(), None)),
         }
     }
 
@@ -460,7 +470,7 @@ impl AuditSystem {
             info!(target_id = %target_id, "Target enabled");
             Ok(())
         } else {
-            Err(AuditError::Configuration(format!("Target not found: {target_id}")))
+            Err(AuditError::Configuration(format!("Target not found: {target_id}"), None))
         }
     }
 
@@ -473,7 +483,7 @@ impl AuditSystem {
             info!(target_id = %target_id, "Target disabled");
             Ok(())
         } else {
-            Err(AuditError::Configuration(format!("Target not found: {target_id}")))
+            Err(AuditError::Configuration(format!("Target not found: {target_id}"), None))
         }
     }
 
@@ -487,7 +497,7 @@ impl AuditSystem {
             info!(target_id = %target_id, "Target removed");
             Ok(())
         } else {
-            Err(AuditError::Configuration(format!("Target not found: {target_id}")))
+            Err(AuditError::Configuration(format!("Target not found: {target_id}"), None))
         }
     }
 

crates/audit/tests/integration_test.rs

@@ -52,7 +52,7 @@ async fn test_config_parsing_webhook() {
     // We expect this to fail due to server storage not being initialized
     // but the parsing should work correctly
     match result {
-        Err(AuditError::ServerNotInitialized(_)) => {
+        Err(AuditError::StorageNotAvailable(_)) => {
             // This is expected in test environment
         }
         Err(e) => {

crates/audit/tests/performance_test.rs

@@ -73,7 +73,7 @@ async fn test_concurrent_target_creation() {
 
     // Verify it fails with expected error (server not initialized)
     match result {
-        Err(AuditError::ServerNotInitialized(_)) => {
+        Err(AuditError::StorageNotAvailable(_)) => {
             // Expected in test environment
         }
         Err(e) => {
@@ -94,7 +94,7 @@ async fn test_audit_log_dispatch_performance() {
     let start_result = system.start(config).await;
     if start_result.is_err() {
         println!("AuditSystem failed to start: {start_result:?}");
-        return; // 或 assert!(false, "AuditSystem failed to start");
+        return; // Alternatively: assert!(false, "AuditSystem failed to start");
     }
 
     use chrono::Utc;
@@ -103,17 +103,17 @@ async fn test_audit_log_dispatch_performance() {
     use std::collections::HashMap;
     let id = 1;
 
-    let mut req_header = HashMap::new();
+    let mut req_header = hashbrown::HashMap::new();
     req_header.insert("authorization".to_string(), format!("Bearer test-token-{id}"));
     req_header.insert("content-type".to_string(), "application/octet-stream".to_string());
 
-    let mut resp_header = HashMap::new();
+    let mut resp_header = hashbrown::HashMap::new();
     resp_header.insert("x-response".to_string(), "ok".to_string());
 
-    let mut tags = HashMap::new();
+    let mut tags = hashbrown::HashMap::new();
     tags.insert(format!("tag-{id}"), json!("sample"));
 
-    let mut req_query = HashMap::new();
+    let mut req_query = hashbrown::HashMap::new();
     req_query.insert("id".to_string(), id.to_string());
 
     let api_details = ApiDetails {

crates/audit/tests/system_integration_test.rs

@@ -35,7 +35,7 @@ async fn test_complete_audit_system_lifecycle() {
 
     // Should fail in test environment but state handling should work
     match start_result {
-        Err(AuditError::ServerNotInitialized(_)) => {
+        Err(AuditError::StorageNotAvailable(_)) => {
             // Expected in test environment
             assert_eq!(system.get_state().await, system::AuditSystemState::Stopped);
         }
@@ -168,7 +168,7 @@ async fn test_config_parsing_with_multiple_instances() {
 
     // Should fail due to server storage not initialized, but parsing should work
     match result {
-        Err(AuditError::ServerNotInitialized(_)) => {
+        Err(AuditError::StorageNotAvailable(_)) => {
             // Expected - parsing worked but save failed
         }
         Err(e) => {
@@ -182,48 +182,6 @@ async fn test_config_parsing_with_multiple_instances() {
     }
 }
 
-// #[tokio::test]
-// async fn test_environment_variable_precedence() {
-//     // Test that environment variables override config file settings
-//     // This test validates the ENV > file instance > file default precedence
-//     // Set some test environment variables
-//     std::env::set_var("RUSTFS_AUDIT_WEBHOOK_ENABLE_TEST", "on");
-//     std::env::set_var("RUSTFS_AUDIT_WEBHOOK_ENDPOINT_TEST", "http://env.example.com/audit");
-//     std::env::set_var("RUSTFS_AUDIT_WEBHOOK_AUTH_TOKEN_TEST", "env-token");
-//     let mut registry = AuditRegistry::new();
-//
-//     // Create config that should be overridden by env vars
-//     let mut config = Config(HashMap::new());
-//     let mut webhook_section = HashMap::new();
-//
-//     let mut test_kvs = KVS::new();
-//     test_kvs.insert("enable".to_string(), "off".to_string()); // Should be overridden
-//     test_kvs.insert("endpoint".to_string(), "http://file.example.com/audit".to_string()); // Should be overridden
-//     test_kvs.insert("batch_size".to_string(), "10".to_string()); // Should remain from file
-//     webhook_section.insert("test".to_string(), test_kvs);
-//
-//     config.0.insert("audit_webhook".to_string(), webhook_section);
-//
-//     // Try to create targets - should use env vars for endpoint/enable, file for batch_size
-//     let result = registry.create_targets_from_config(&config).await;
-//     // Clean up env vars
-//     std::env::remove_var("RUSTFS_AUDIT_WEBHOOK_ENABLE_TEST");
-//     std::env::remove_var("RUSTFS_AUDIT_WEBHOOK_ENDPOINT_TEST");
-//     std::env::remove_var("RUSTFS_AUDIT_WEBHOOK_AUTH_TOKEN_TEST");
-//     // Should fail due to server storage, but precedence logic should work
-//     match result {
-//         Err(AuditError::ServerNotInitialized(_)) => {
-//             // Expected - precedence parsing worked but save failed
-//         }
-//         Err(e) => {
-//             println!("Environment precedence test error: {}", e);
-//         }
-//         Ok(_) => {
-//             println!("Unexpected success in environment precedence test");
-//         }
-//     }
-// }
-
 #[test]
 fn test_target_type_validation() {
     use rustfs_targets::target::TargetType;
@@ -315,19 +273,18 @@ fn create_sample_audit_entry_with_id(id: u32) -> AuditEntry {
     use chrono::Utc;
     use rustfs_targets::EventName;
     use serde_json::json;
-    use std::collections::HashMap;
 
-    let mut req_header = HashMap::new();
+    let mut req_header = hashbrown::HashMap::new();
     req_header.insert("authorization".to_string(), format!("Bearer test-token-{id}"));
     req_header.insert("content-type".to_string(), "application/octet-stream".to_string());
 
-    let mut resp_header = HashMap::new();
+    let mut resp_header = hashbrown::HashMap::new();
     resp_header.insert("x-response".to_string(), "ok".to_string());
 
-    let mut tags = HashMap::new();
+    let mut tags = hashbrown::HashMap::new();
     tags.insert(format!("tag-{id}"), json!("sample"));
 
-    let mut req_query = HashMap::new();
+    let mut req_query = hashbrown::HashMap::new();
     req_query.insert("id".to_string(), id.to_string());
 
     let api_details = ApiDetails {

crates/common/Cargo.toml

@@ -28,7 +28,6 @@ categories = ["web-programming", "development-tools", "data-structures"]
 workspace = true
 
 [dependencies]
-lazy_static = { workspace = true}
 tokio = { workspace = true }
 tonic = { workspace = true }
 uuid = { workspace = true }

crates/common/src/bucket_stats.rs

@@ -12,9 +12,9 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+use crate::last_minute::{self};
 use std::collections::HashMap;
 
-use crate::last_minute::{self};
 pub struct ReplicationLatency {
     // Delays for single and multipart PUT requests
     upload_histogram: last_minute::LastMinuteHistogram,

crates/common/src/data_usage.rs

@@ -14,10 +14,10 @@
 
 use path_clean::PathClean;
 use serde::{Deserialize, Serialize};
-use std::hash::{DefaultHasher, Hash, Hasher};
-use std::path::Path;
 use std::{
     collections::{HashMap, HashSet},
+    hash::{DefaultHasher, Hash, Hasher},
+    path::Path,
     time::SystemTime,
 };
 

crates/common/src/globals.rs

@@ -16,7 +16,6 @@
 
 use std::collections::HashMap;
 use std::sync::LazyLock;
-
 use tokio::sync::RwLock;
 use tonic::transport::Channel;
 

crates/common/src/heal_channel.rs

@@ -18,7 +18,7 @@ use std::{
     fmt::{self, Display},
     sync::OnceLock,
 };
-use tokio::sync::mpsc;
+use tokio::sync::{broadcast, mpsc};
 use uuid::Uuid;
 
 pub const HEAL_DELETE_DANGLING: bool = true;
@@ -85,19 +85,14 @@ impl Display for DriveState {
     }
 }
 
-#[derive(Clone, Copy, Debug, Serialize, Deserialize, PartialEq, Eq)]
+#[derive(Clone, Copy, Debug, Default, Serialize, Deserialize, PartialEq, Eq)]
 pub enum HealScanMode {
     Unknown,
+    #[default]
     Normal,
     Deep,
 }
 
-impl Default for HealScanMode {
-    fn default() -> Self {
-        Self::Normal
-    }
-}
-
 #[derive(Clone, Copy, Debug, Default, Serialize, Deserialize)]
 pub struct HealOpts {
     pub recursive: bool,
@@ -175,11 +170,12 @@ pub struct HealChannelResponse {
 }
 
 /// Heal priority
-#[derive(Debug, Clone, Copy, PartialEq, Eq)]
+#[derive(Debug, Default, Clone, Copy, PartialEq, Eq)]
 pub enum HealChannelPriority {
     /// Low priority
     Low,
     /// Normal priority
+    #[default]
     Normal,
     /// High priority
     High,
@@ -187,12 +183,6 @@ pub enum HealChannelPriority {
     Critical,
 }
 
-impl Default for HealChannelPriority {
-    fn default() -> Self {
-        Self::Normal
-    }
-}
-
 /// Heal channel sender
 pub type HealChannelSender = mpsc::UnboundedSender<HealChannelCommand>;
 
@@ -202,6 +192,11 @@ pub type HealChannelReceiver = mpsc::UnboundedReceiver<HealChannelCommand>;
 /// Global heal channel sender
 static GLOBAL_HEAL_CHANNEL_SENDER: OnceLock<HealChannelSender> = OnceLock::new();
 
+type HealResponseSender = broadcast::Sender<HealChannelResponse>;
+
+/// Global heal response broadcaster
+static GLOBAL_HEAL_RESPONSE_SENDER: OnceLock<HealResponseSender> = OnceLock::new();
+
 /// Initialize global heal channel
 pub fn init_heal_channel() -> HealChannelReceiver {
     let (tx, rx) = mpsc::unbounded_channel();
@@ -228,6 +223,23 @@ pub async fn send_heal_command(command: HealChannelCommand) -> Result<(), String
     }
 }
 
+fn heal_response_sender() -> &'static HealResponseSender {
+    GLOBAL_HEAL_RESPONSE_SENDER.get_or_init(|| {
+        let (tx, _rx) = broadcast::channel(1024);
+        tx
+    })
+}
+
+/// Publish a heal response to subscribers.
+pub fn publish_heal_response(response: HealChannelResponse) -> Result<(), broadcast::error::SendError<HealChannelResponse>> {
+    heal_response_sender().send(response).map(|_| ())
+}
+
+/// Subscribe to heal responses.
+pub fn subscribe_heal_responses() -> broadcast::Receiver<HealChannelResponse> {
+    heal_response_sender().subscribe()
+}
+
 /// Send heal start request
 pub async fn send_heal_request(request: HealChannelRequest) -> Result<(), String> {
     send_heal_command(HealChannelCommand::Start(request)).await
@@ -425,3 +437,20 @@ pub async fn send_heal_disk(set_disk_id: String, priority: Option<HealChannelPri
     };
     send_heal_request(req).await
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[tokio::test]
+    async fn heal_response_broadcast_reaches_subscriber() {
+        let mut receiver = subscribe_heal_responses();
+        let response = create_heal_response("req-1".to_string(), true, None, None);
+
+        publish_heal_response(response.clone()).expect("publish should succeed");
+
+        let received = receiver.recv().await.expect("should receive heal response");
+        assert_eq!(received.request_id, response.request_id);
+        assert!(received.success);
+    }
+}

crates/common/src/last_minute.rs

@@ -27,11 +27,11 @@ struct TimedAction {
 #[allow(dead_code)]
 impl TimedAction {
     // Avg returns the average time spent on the action.
-    pub fn avg(&self) -> Option<std::time::Duration> {
+    pub fn avg(&self) -> Option<Duration> {
         if self.count == 0 {
             return None;
         }
-        Some(std::time::Duration::from_nanos(self.acc_time / self.count))
+        Some(Duration::from_nanos(self.acc_time / self.count))
     }
 
     // AvgBytes returns the average bytes processed.
@@ -860,7 +860,7 @@ impl LastMinuteHistogram {
         }
     }
 
-    pub fn add(&mut self, size: i64, t: std::time::Duration) {
+    pub fn add(&mut self, size: i64, t: Duration) {
         let index = size_to_tag(size);
         self.histogram[index].add(&t);
     }

crates/common/src/metrics.rs

@@ -12,23 +12,21 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+use crate::last_minute::{AccElem, LastMinuteLatency};
 use chrono::{DateTime, Utc};
-use lazy_static::lazy_static;
 use rustfs_madmin::metrics::ScannerMetrics as M_ScannerMetrics;
 use std::{
     collections::HashMap,
     fmt::Display,
     pin::Pin,
     sync::{
-        Arc,
+        Arc, OnceLock,
         atomic::{AtomicU64, Ordering},
     },
     time::{Duration, SystemTime},
 };
 use tokio::sync::{Mutex, RwLock};
 
-use crate::last_minute::{AccElem, LastMinuteLatency};
-
 #[derive(Debug, Clone, Copy, PartialEq, Eq)]
 pub enum IlmAction {
     NoneAction = 0,
@@ -73,8 +71,10 @@ impl Display for IlmAction {
     }
 }
 
-lazy_static! {
-    pub static ref globalMetrics: Arc<Metrics> = Arc::new(Metrics::new());
+pub static GLOBAL_METRICS: OnceLock<Arc<Metrics>> = OnceLock::new();
+
+pub fn global_metrics() -> &'static Arc<Metrics> {
+    GLOBAL_METRICS.get_or_init(|| Arc::new(Metrics::new()))
 }
 
 #[derive(Clone, Debug, PartialEq, PartialOrd)]
@@ -294,13 +294,13 @@ impl Metrics {
             let duration = SystemTime::now().duration_since(start_time).unwrap_or_default();
 
             // Update operation count
-            globalMetrics.operations[metric].fetch_add(1, Ordering::Relaxed);
+            global_metrics().operations[metric].fetch_add(1, Ordering::Relaxed);
 
             // Update latency for realtime metrics (spawn async task for this)
             if (metric) < Metric::LastRealtime as usize {
                 let metric_index = metric;
                 tokio::spawn(async move {
-                    globalMetrics.latency[metric_index].add(duration).await;
+                    global_metrics().latency[metric_index].add(duration).await;
                 });
             }
 
@@ -319,13 +319,13 @@ impl Metrics {
             let duration = SystemTime::now().duration_since(start_time).unwrap_or_default();
 
             // Update operation count
-            globalMetrics.operations[metric].fetch_add(1, Ordering::Relaxed);
+            global_metrics().operations[metric].fetch_add(1, Ordering::Relaxed);
 
             // Update latency for realtime metrics with size (spawn async task)
             if (metric) < Metric::LastRealtime as usize {
                 let metric_index = metric;
                 tokio::spawn(async move {
-                    globalMetrics.latency[metric_index].add_size(duration, size).await;
+                    global_metrics().latency[metric_index].add_size(duration, size).await;
                 });
             }
         }
@@ -339,13 +339,13 @@ impl Metrics {
             let duration = SystemTime::now().duration_since(start_time).unwrap_or_default();
 
             // Update operation count
-            globalMetrics.operations[metric].fetch_add(1, Ordering::Relaxed);
+            global_metrics().operations[metric].fetch_add(1, Ordering::Relaxed);
 
             // Update latency for realtime metrics (spawn async task)
             if (metric) < Metric::LastRealtime as usize {
                 let metric_index = metric;
                 tokio::spawn(async move {
-                    globalMetrics.latency[metric_index].add(duration).await;
+                    global_metrics().latency[metric_index].add(duration).await;
                 });
             }
         }
@@ -360,13 +360,13 @@ impl Metrics {
                 let duration = SystemTime::now().duration_since(start_time).unwrap_or_default();
 
                 // Update operation count
-                globalMetrics.operations[metric].fetch_add(count as u64, Ordering::Relaxed);
+                global_metrics().operations[metric].fetch_add(count as u64, Ordering::Relaxed);
 
                 // Update latency for realtime metrics (spawn async task)
                 if (metric) < Metric::LastRealtime as usize {
                     let metric_index = metric;
                     tokio::spawn(async move {
-                        globalMetrics.latency[metric_index].add(duration).await;
+                        global_metrics().latency[metric_index].add(duration).await;
                     });
                 }
             })
@@ -384,8 +384,8 @@ impl Metrics {
             Box::new(move || {
                 let duration = SystemTime::now().duration_since(start).unwrap_or(Duration::from_secs(0));
                 tokio::spawn(async move {
-                    globalMetrics.actions[a_clone].fetch_add(versions, Ordering::Relaxed);
-                    globalMetrics.actions_latency[a_clone].add(duration).await;
+                    global_metrics().actions[a_clone].fetch_add(versions, Ordering::Relaxed);
+                    global_metrics().actions_latency[a_clone].add(duration).await;
                 });
             })
         })
@@ -395,11 +395,11 @@ impl Metrics {
     pub async fn inc_time(metric: Metric, duration: Duration) {
         let metric = metric as usize;
         // Update operation count
-        globalMetrics.operations[metric].fetch_add(1, Ordering::Relaxed);
+        global_metrics().operations[metric].fetch_add(1, Ordering::Relaxed);
 
         // Update latency for realtime metrics
         if (metric) < Metric::LastRealtime as usize {
-            globalMetrics.latency[metric].add(duration).await;
+            global_metrics().latency[metric].add(duration).await;
         }
     }
 
@@ -501,7 +501,7 @@ pub fn current_path_updater(disk: &str, initial: &str) -> (UpdateCurrentPathFn,
     let tracker_clone = Arc::clone(&tracker);
     let disk_clone = disk_name.clone();
     tokio::spawn(async move {
-        globalMetrics.current_paths.write().await.insert(disk_clone, tracker_clone);
+        global_metrics().current_paths.write().await.insert(disk_clone, tracker_clone);
     });
 
     let update_fn = {
@@ -520,7 +520,7 @@ pub fn current_path_updater(disk: &str, initial: &str) -> (UpdateCurrentPathFn,
         Arc::new(move || -> Pin<Box<dyn std::future::Future<Output = ()> + Send>> {
             let disk_name = disk_name.clone();
             Box::pin(async move {
-                globalMetrics.current_paths.write().await.remove(&disk_name);
+                global_metrics().current_paths.write().await.remove(&disk_name);
             })
         })
     };

crates/config/src/audit/mod.rs

@@ -16,8 +16,8 @@
 //! This module defines the configuration for audit systems, including
 //! webhook and MQTT audit-related settings.
 
-pub(crate) mod mqtt;
-pub(crate) mod webhook;
+mod mqtt;
+mod webhook;
 
 pub use mqtt::*;
 pub use webhook::*;

crates/config/src/constants/app.rs

@@ -21,12 +21,12 @@ pub const APP_NAME: &str = "RustFS";
 /// Application version
 /// Default value: 1.0.0
 /// Environment variable: RUSTFS_VERSION
-pub const VERSION: &str = "0.0.1";
+pub const VERSION: &str = "1.0.0";
 
 /// Default configuration logger level
-/// Default value: info
+/// Default value: error
 /// Environment variable: RUSTFS_LOG_LEVEL
-pub const DEFAULT_LOG_LEVEL: &str = "info";
+pub const DEFAULT_LOG_LEVEL: &str = "error";
 
 /// Default configuration use stdout
 /// Default value: false
@@ -40,22 +40,15 @@ pub const SAMPLE_RATIO: f64 = 1.0;
 pub const METER_INTERVAL: u64 = 30;
 
 /// Default configuration service version
-/// Default value: 0.0.1
-pub const SERVICE_VERSION: &str = "0.0.1";
+/// Default value: 1.0.0
+/// Environment variable: RUSTFS_OBS_SERVICE_VERSION
+/// Uses the same value as VERSION constant
+pub const SERVICE_VERSION: &str = "1.0.0";
 
 /// Default configuration environment
 /// Default value: production
 pub const ENVIRONMENT: &str = "production";
 
-/// maximum number of connections
-/// This is the maximum number of connections that the server will accept.
-/// This is used to limit the number of connections to the server.
-pub const MAX_CONNECTIONS: usize = 100;
-/// timeout for connections
-/// This is the timeout for connections to the server.
-/// This is used to limit the time that a connection can be open.
-pub const DEFAULT_TIMEOUT_MS: u64 = 3000;
-
 /// Default Access Key
 /// Default value: rustfsadmin
 /// Environment variable: RUSTFS_ACCESS_KEY
@@ -145,18 +138,27 @@ pub const DEFAULT_LOG_ROTATION_SIZE_MB: u64 = 100;
 /// It is used to rotate the logs of the application.
 /// Default value: hour, eg: day,hour,minute,second
 /// Environment variable: RUSTFS_OBS_LOG_ROTATION_TIME
-pub const DEFAULT_LOG_ROTATION_TIME: &str = "day";
+pub const DEFAULT_LOG_ROTATION_TIME: &str = "hour";
 
 /// Default log keep files for rustfs
 /// This is the default log keep files for rustfs.
 /// It is used to keep the logs of the application.
 /// Default value: 30
 /// Environment variable: RUSTFS_OBS_LOG_KEEP_FILES
-pub const DEFAULT_LOG_KEEP_FILES: u16 = 30;
+pub const DEFAULT_LOG_KEEP_FILES: usize = 30;
 
-/// 1 KiB
+/// Default log local logging enabled for rustfs
+/// This is the default log local logging enabled for rustfs.
+/// It is used to enable or disable local logging of the application.
+/// Default value: false
+/// Environment variable: RUSTFS_OBS_LOGL_STDOUT_ENABLED
+pub const DEFAULT_OBS_LOG_STDOUT_ENABLED: bool = false;
+
+/// Constant representing 1 Kibibyte (1024 bytes)
+/// Default value: 1024
 pub const KI_B: usize = 1024;
-/// 1 MiB
+/// Constant representing 1 Mebibyte (1024 * 1024 bytes)
+/// Default value: 1048576
 pub const MI_B: usize = 1024 * 1024;
 
 #[cfg(test)]
@@ -169,16 +171,16 @@ mod tests {
         assert_eq!(APP_NAME, "RustFS");
         assert!(!APP_NAME.contains(' '), "App name should not contain spaces");
 
-        assert_eq!(VERSION, "0.0.1");
+        assert_eq!(VERSION, "1.0.0");
 
-        assert_eq!(SERVICE_VERSION, "0.0.1");
+        assert_eq!(SERVICE_VERSION, "1.0.0");
         assert_eq!(VERSION, SERVICE_VERSION, "Version and service version should be consistent");
     }
 
     #[test]
     fn test_logging_constants() {
         // Test logging related constants
-        assert_eq!(DEFAULT_LOG_LEVEL, "info");
+        assert_eq!(DEFAULT_LOG_LEVEL, "error");
         assert!(
             ["trace", "debug", "info", "warn", "error"].contains(&DEFAULT_LOG_LEVEL),
             "Log level should be a valid tracing level"
@@ -199,14 +201,6 @@ mod tests {
         );
     }
 
-    #[test]
-    fn test_connection_constants() {
-        // Test connection related constants
-        assert_eq!(MAX_CONNECTIONS, 100);
-
-        assert_eq!(DEFAULT_TIMEOUT_MS, 3000);
-    }
-
     #[test]
     fn test_security_constants() {
         // Test security related constants
@@ -309,8 +303,8 @@ mod tests {
         // assert!(DEFAULT_TIMEOUT_MS < u64::MAX, "Timeout should be reasonable");
 
         // These are const non-zero values, so zero checks are redundant
-        // assert!(DEFAULT_PORT != 0, "Default port should not be zero");
-        // assert!(DEFAULT_CONSOLE_PORT != 0, "Console port should not be zero");
+        assert_ne!(DEFAULT_PORT, 0, "Default port should not be zero");
+        assert_ne!(DEFAULT_CONSOLE_PORT, 0, "Console port should not be zero");
     }
 
     #[test]

crates/config/src/constants/mod.rs

@@ -15,6 +15,7 @@
 pub(crate) mod app;
 pub(crate) mod console;
 pub(crate) mod env;
+pub(crate) mod profiler;
 pub(crate) mod runtime;
 pub(crate) mod targets;
 pub(crate) mod tls;

crates/config/src/constants/profiler.rs

@@ -0,0 +1,50 @@
+// Copyright 2024 RustFS Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+/// Profiler related environment variable names and default values
+pub const ENV_ENABLE_PROFILING: &str = "RUSTFS_ENABLE_PROFILING";
+
+// CPU profiling
+pub const ENV_CPU_MODE: &str = "RUSTFS_PROF_CPU_MODE"; // off|continuous|periodic
+/// Frequency of CPU profiling samples
+pub const ENV_CPU_FREQ: &str = "RUSTFS_PROF_CPU_FREQ";
+/// Interval between CPU profiling sessions (for periodic mode)
+pub const ENV_CPU_INTERVAL_SECS: &str = "RUSTFS_PROF_CPU_INTERVAL_SECS";
+/// Duration of each CPU profiling session (for periodic mode)
+pub const ENV_CPU_DURATION_SECS: &str = "RUSTFS_PROF_CPU_DURATION_SECS";
+
+/// Memory profiling (jemalloc)
+pub const ENV_MEM_PERIODIC: &str = "RUSTFS_PROF_MEM_PERIODIC";
+/// Interval between memory profiling snapshots (for periodic mode)
+pub const ENV_MEM_INTERVAL_SECS: &str = "RUSTFS_PROF_MEM_INTERVAL_SECS";
+
+/// Output directory
+pub const ENV_OUTPUT_DIR: &str = "RUSTFS_PROF_OUTPUT_DIR";
+
+/// Defaults for profiler settings
+pub const DEFAULT_ENABLE_PROFILING: bool = false;
+/// CPU profiling
+pub const DEFAULT_CPU_MODE: &str = "off";
+/// Frequency of CPU profiling samples
+pub const DEFAULT_CPU_FREQ: usize = 100;
+/// Interval between CPU profiling sessions (for periodic mode)
+pub const DEFAULT_CPU_INTERVAL_SECS: u64 = 300;
+/// Duration of each CPU profiling session (for periodic mode)
+pub const DEFAULT_CPU_DURATION_SECS: u64 = 60;
+/// Memory profiling (jemalloc)
+pub const DEFAULT_MEM_PERIODIC: bool = false;
+/// Interval between memory profiling snapshots (for periodic mode)
+pub const DEFAULT_MEM_INTERVAL_SECS: u64 = 300;
+/// Output directory
+pub const DEFAULT_OUTPUT_DIR: &str = ".";

crates/config/src/constants/runtime.rs

@@ -22,7 +22,10 @@ pub const ENV_THREAD_STACK_SIZE: &str = "RUSTFS_RUNTIME_THREAD_STACK_SIZE";
 pub const ENV_THREAD_KEEP_ALIVE: &str = "RUSTFS_RUNTIME_THREAD_KEEP_ALIVE";
 pub const ENV_GLOBAL_QUEUE_INTERVAL: &str = "RUSTFS_RUNTIME_GLOBAL_QUEUE_INTERVAL";
 pub const ENV_THREAD_NAME: &str = "RUSTFS_RUNTIME_THREAD_NAME";
+pub const ENV_MAX_IO_EVENTS_PER_TICK: &str = "RUSTFS_RUNTIME_MAX_IO_EVENTS_PER_TICK";
 pub const ENV_RNG_SEED: &str = "RUSTFS_RUNTIME_RNG_SEED";
+/// Event polling interval
+pub const ENV_EVENT_INTERVAL: &str = "RUSTFS_RUNTIME_EVENT_INTERVAL";
 
 // Default values for Tokio runtime
 pub const DEFAULT_WORKER_THREADS: usize = 16;
@@ -32,4 +35,7 @@ pub const DEFAULT_THREAD_STACK_SIZE: usize = MI_B; // 1 MiB
 pub const DEFAULT_THREAD_KEEP_ALIVE: u64 = 60; // seconds
 pub const DEFAULT_GLOBAL_QUEUE_INTERVAL: u32 = 31;
 pub const DEFAULT_THREAD_NAME: &str = "rustfs-worker";
+pub const DEFAULT_MAX_IO_EVENTS_PER_TICK: usize = 1024;
+/// Event polling default (Tokio default 61)
+pub const DEFAULT_EVENT_INTERVAL: u32 = 61;
 pub const DEFAULT_RNG_SEED: Option<u64> = None; // None means random

crates/config/src/lib.rs

@@ -21,6 +21,8 @@ pub use constants::console::*;
 #[cfg(feature = "constants")]
 pub use constants::env::*;
 #[cfg(feature = "constants")]
+pub use constants::profiler::*;
+#[cfg(feature = "constants")]
 pub use constants::runtime::*;
 #[cfg(feature = "constants")]
 pub use constants::targets::*;

crates/config/src/observability/metrics.rs

@@ -0,0 +1,19 @@
+// Copyright 2024 RustFS Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+/// Metrics collection interval in milliseconds for system metrics (CPU, memory, disk, network).
+pub const DEFAULT_METRICS_SYSTEM_INTERVAL_MS: u64 = 30000;
+
+/// Environment variable for setting the metrics collection interval for system metrics.
+pub const ENV_OBS_METRICS_SYSTEM_INTERVAL_MS: &str = "RUSTFS_OBS_METRICS_SYSTEM_INTERVAL_MS";

crates/config/src/observability/mod.rs

@@ -14,7 +14,13 @@
 
 // Observability Keys
 
+mod metrics;
+pub use metrics::*;
+
 pub const ENV_OBS_ENDPOINT: &str = "RUSTFS_OBS_ENDPOINT";
+pub const ENV_OBS_TRACE_ENDPOINT: &str = "RUSTFS_OBS_TRACE_ENDPOINT";
+pub const ENV_OBS_METRIC_ENDPOINT: &str = "RUSTFS_OBS_METRIC_ENDPOINT";
+pub const ENV_OBS_LOG_ENDPOINT: &str = "RUSTFS_OBS_LOG_ENDPOINT";
 pub const ENV_OBS_USE_STDOUT: &str = "RUSTFS_OBS_USE_STDOUT";
 pub const ENV_OBS_SAMPLE_RATIO: &str = "RUSTFS_OBS_SAMPLE_RATIO";
 pub const ENV_OBS_METER_INTERVAL: &str = "RUSTFS_OBS_METER_INTERVAL";
@@ -22,7 +28,7 @@ pub const ENV_OBS_SERVICE_NAME: &str = "RUSTFS_OBS_SERVICE_NAME";
 pub const ENV_OBS_SERVICE_VERSION: &str = "RUSTFS_OBS_SERVICE_VERSION";
 pub const ENV_OBS_ENVIRONMENT: &str = "RUSTFS_OBS_ENVIRONMENT";
 pub const ENV_OBS_LOGGER_LEVEL: &str = "RUSTFS_OBS_LOGGER_LEVEL";
-pub const ENV_OBS_LOCAL_LOGGING_ENABLED: &str = "RUSTFS_OBS_LOCAL_LOGGING_ENABLED";
+pub const ENV_OBS_LOG_STDOUT_ENABLED: &str = "RUSTFS_OBS_LOG_STDOUT_ENABLED";
 pub const ENV_OBS_LOG_DIRECTORY: &str = "RUSTFS_OBS_LOG_DIRECTORY";
 pub const ENV_OBS_LOG_FILENAME: &str = "RUSTFS_OBS_LOG_FILENAME";
 pub const ENV_OBS_LOG_ROTATION_SIZE_MB: &str = "RUSTFS_OBS_LOG_ROTATION_SIZE_MB";
@@ -47,12 +53,6 @@ pub const DEFAULT_OBS_LOG_MESSAGE_CAPA: usize = 32768;
 /// Default values for flush interval in milliseconds
 pub const DEFAULT_OBS_LOG_FLUSH_MS: u64 = 200;
 
-/// Audit logger queue capacity environment variable key
-pub const ENV_AUDIT_LOGGER_QUEUE_CAPACITY: &str = "RUSTFS_AUDIT_LOGGER_QUEUE_CAPACITY";
-
-/// Default values for observability configuration
-pub const DEFAULT_AUDIT_LOGGER_QUEUE_CAPACITY: usize = 10000;
-
 /// Default values for observability configuration
 // ### Supported Environment Values
 // - `production` - Secure file-only logging
@@ -71,6 +71,9 @@ mod tests {
     #[test]
     fn test_env_keys() {
         assert_eq!(ENV_OBS_ENDPOINT, "RUSTFS_OBS_ENDPOINT");
+        assert_eq!(ENV_OBS_TRACE_ENDPOINT, "RUSTFS_OBS_TRACE_ENDPOINT");
+        assert_eq!(ENV_OBS_METRIC_ENDPOINT, "RUSTFS_OBS_METRIC_ENDPOINT");
+        assert_eq!(ENV_OBS_LOG_ENDPOINT, "RUSTFS_OBS_LOG_ENDPOINT");
         assert_eq!(ENV_OBS_USE_STDOUT, "RUSTFS_OBS_USE_STDOUT");
         assert_eq!(ENV_OBS_SAMPLE_RATIO, "RUSTFS_OBS_SAMPLE_RATIO");
         assert_eq!(ENV_OBS_METER_INTERVAL, "RUSTFS_OBS_METER_INTERVAL");
@@ -78,18 +81,16 @@ mod tests {
         assert_eq!(ENV_OBS_SERVICE_VERSION, "RUSTFS_OBS_SERVICE_VERSION");
         assert_eq!(ENV_OBS_ENVIRONMENT, "RUSTFS_OBS_ENVIRONMENT");
         assert_eq!(ENV_OBS_LOGGER_LEVEL, "RUSTFS_OBS_LOGGER_LEVEL");
-        assert_eq!(ENV_OBS_LOCAL_LOGGING_ENABLED, "RUSTFS_OBS_LOCAL_LOGGING_ENABLED");
+        assert_eq!(ENV_OBS_LOG_STDOUT_ENABLED, "RUSTFS_OBS_LOG_STDOUT_ENABLED");
         assert_eq!(ENV_OBS_LOG_DIRECTORY, "RUSTFS_OBS_LOG_DIRECTORY");
         assert_eq!(ENV_OBS_LOG_FILENAME, "RUSTFS_OBS_LOG_FILENAME");
         assert_eq!(ENV_OBS_LOG_ROTATION_SIZE_MB, "RUSTFS_OBS_LOG_ROTATION_SIZE_MB");
         assert_eq!(ENV_OBS_LOG_ROTATION_TIME, "RUSTFS_OBS_LOG_ROTATION_TIME");
         assert_eq!(ENV_OBS_LOG_KEEP_FILES, "RUSTFS_OBS_LOG_KEEP_FILES");
-        assert_eq!(ENV_AUDIT_LOGGER_QUEUE_CAPACITY, "RUSTFS_AUDIT_LOGGER_QUEUE_CAPACITY");
     }
 
     #[test]
     fn test_default_values() {
-        assert_eq!(DEFAULT_AUDIT_LOGGER_QUEUE_CAPACITY, 10000);
         assert_eq!(DEFAULT_OBS_ENVIRONMENT_PRODUCTION, "production");
         assert_eq!(DEFAULT_OBS_ENVIRONMENT_DEVELOPMENT, "development");
         assert_eq!(DEFAULT_OBS_ENVIRONMENT_TEST, "test");

crates/crypto/Cargo.toml

@@ -29,7 +29,7 @@ documentation = "https://docs.rs/rustfs-crypto/latest/rustfs_crypto/"
 workspace = true
 
 [dependencies]
-aes-gcm = { workspace = true, features = ["std"], optional = true }
+aes-gcm = { workspace = true, optional = true }
 argon2 = { workspace = true, features = ["std"], optional = true }
 cfg-if = { workspace = true }
 chacha20poly1305 = { workspace = true, optional = true }

crates/crypto/src/encdec/decrypt.rs

@@ -19,127 +19,37 @@ pub fn decrypt_data(password: &[u8], data: &[u8]) -> Result<Vec<u8>, crate::Erro
     use aes_gcm::{Aes256Gcm, KeyInit as _};
     use chacha20poly1305::ChaCha20Poly1305;
 
-    // 32: salt
-    // 1: id
-    // 12: nonce
     const HEADER_LENGTH: usize = 45;
     if data.len() < HEADER_LENGTH {
         return Err(Error::ErrUnexpectedHeader);
     }
 
-    let (salt, id, nonce) = (&data[..32], ID::try_from(data[32])?, &data[33..45]);
-    let data = &data[HEADER_LENGTH..];
+    let (salt, id, nonce_slice) = (&data[..32], ID::try_from(data[32])?, &data[33..45]);
+    let body = &data[HEADER_LENGTH..];
 
     match id {
         ID::Argon2idChaCHa20Poly1305 => {
             let key = id.get_key(password, salt)?;
-            decrypt(ChaCha20Poly1305::new_from_slice(&key)?, nonce, data)
+            decrypt(ChaCha20Poly1305::new_from_slice(&key)?, nonce_slice, body)
         }
         _ => {
             let key = id.get_key(password, salt)?;
-            decrypt(Aes256Gcm::new_from_slice(&key)?, nonce, data)
+            decrypt(Aes256Gcm::new_from_slice(&key)?, nonce_slice, body)
         }
     }
 }
 
-// use argon2::{Argon2, PasswordHasher};
-// use argon2::password_hash::{SaltString};
-// use aes_gcm::{Aes256Gcm, Key, Nonce}; // For AES-GCM
-// use chacha20poly1305::{ChaCha20Poly1305, Key as ChaChaKey, Nonce as ChaChaNonce}; // For ChaCha20
-// use pbkdf2::pbkdf2;
-// use sha2::Sha256;
-// use std::io::{self, Read};
-// use thiserror::Error;
-
-// #[derive(Debug, Error)]
-// pub enum DecryptError {
-//     #[error("unexpected header")]
-//     UnexpectedHeader,
-//     #[error("invalid encryption algorithm ID")]
-//     InvalidAlgorithmId,
-//     #[error("IO error")]
-//     Io(#[from] io::Error),
-//     #[error("decryption error")]
-//     DecryptionError,
-// }
-
-// pub fn decrypt_data2<R: Read>(password: &str, mut data: R) -> Result<Vec<u8>, DecryptError> {
-//     // Parse the stream header
-//     let mut hdr = [0u8; 32 + 1 + 8];
-//     if data.read_exact(&mut hdr).is_err() {
-//         return Err(DecryptError::UnexpectedHeader);
-//     }
-
-//     let salt = &hdr[0..32];
-//     let id = hdr[32];
-//     let nonce = &hdr[33..41];
-
-//     let key = match id {
-//         // Argon2id + AES-GCM
-//         0x01 => {
-//             let salt = SaltString::encode_b64(salt).map_err(|_| DecryptError::DecryptionError)?;
-//             let argon2 = Argon2::default();
-//             let hashed_key = argon2.hash_password(password.as_bytes(), &salt)
-//                 .map_err(|_| DecryptError::DecryptionError)?;
-//             hashed_key.hash.unwrap().as_bytes().to_vec()
-//         }
-//         // Argon2id + ChaCha20Poly1305
-//         0x02 => {
-//             let salt = SaltString::encode_b64(salt).map_err(|_| DecryptError::DecryptionError)?;
-//             let argon2 = Argon2::default();
-//             let hashed_key = argon2.hash_password(password.as_bytes(), &salt)
-//                 .map_err(|_| DecryptError::DecryptionError)?;
-//             hashed_key.hash.unwrap().as_bytes().to_vec()
-//         }
-//         // PBKDF2 + AES-GCM
-//         // 0x03 => {
-//         //     let mut key = [0u8; 32];
-//         //     pbkdf2::<Sha256>(password.as_bytes(), salt, 10000, &mut key);
-//         //     key.to_vec()
-//         // }
-//         _ => return Err(DecryptError::InvalidAlgorithmId),
-//     };
-
-//     // Decrypt data using the corresponding cipher
-//     let mut encrypted_data = Vec::new();
-//     data.read_to_end(&mut encrypted_data)?;
-
-//     let plaintext = match id {
-//         0x01 => {
-//             let cipher = Aes256Gcm::new(Key::from_slice(&key));
-//             let nonce = Nonce::from_slice(nonce);
-//             cipher
-//                 .decrypt(nonce, encrypted_data.as_ref())
-//                 .map_err(|_| DecryptError::DecryptionError)?
-//         }
-//         0x02 => {
-//             let cipher = ChaCha20Poly1305::new(ChaChaKey::from_slice(&key));
-//             let nonce = ChaChaNonce::from_slice(nonce);
-//             cipher
-//                 .decrypt(nonce, encrypted_data.as_ref())
-//                 .map_err(|_| DecryptError::DecryptionError)?
-//         }
-//         0x03 => {
-
-//             let cipher = Aes256Gcm::new(Key::from_slice(&key));
-//             let nonce = Nonce::from_slice(nonce);
-//             cipher
-//                 .decrypt(nonce, encrypted_data.as_ref())
-//                 .map_err(|_| DecryptError::DecryptionError)?
-//         }
-//         _ => return Err(DecryptError::InvalidAlgorithmId),
-//     };
-
-//     Ok(plaintext)
-// }
-
 #[cfg(any(test, feature = "crypto"))]
 #[inline]
 fn decrypt<T: aes_gcm::aead::Aead>(stream: T, nonce: &[u8], data: &[u8]) -> Result<Vec<u8>, crate::Error> {
     use crate::error::Error;
-    stream
-        .decrypt(aes_gcm::Nonce::from_slice(nonce), data)
-        .map_err(Error::ErrDecryptFailed)
+    use aes_gcm::AeadCore;
+    use aes_gcm::aead::array::Array;
+    use core::convert::TryFrom;
+
+    let nonce_arr: Array<u8, <T as AeadCore>::NonceSize> =
+        Array::try_from(nonce).map_err(|_| Error::ErrDecryptFailed(aes_gcm::aead::Error))?;
+    stream.decrypt(&nonce_arr, data).map_err(Error::ErrDecryptFailed)
 }
 
 #[cfg(not(any(test, feature = "crypto")))]

crates/crypto/src/encdec/encrypt.rs

@@ -43,7 +43,7 @@ pub fn encrypt_data(password: &[u8], data: &[u8]) -> Result<Vec<u8>, crate::Erro
         if native_aes() {
             encrypt(Aes256Gcm::new_from_slice(&key)?, &salt, id, data)
         } else {
-            encrypt(ChaCha20Poly1305::new_from_slice(&key)?, &salt, id, data)
+            encrypt(chacha20poly1305::ChaCha20Poly1305::new_from_slice(&key)?, &salt, id, data)
         }
     }
 }
@@ -56,16 +56,19 @@ fn encrypt<T: aes_gcm::aead::Aead>(
     data: &[u8],
 ) -> Result<Vec<u8>, crate::Error> {
     use crate::error::Error;
-    use aes_gcm::aead::rand_core::OsRng;
+    use aes_gcm::AeadCore;
+    use aes_gcm::aead::array::Array;
+    use rand::RngCore;
 
-    let nonce = T::generate_nonce(&mut OsRng);
+    let mut nonce: Array<u8, <T as AeadCore>::NonceSize> = Array::default();
+    rand::rng().fill_bytes(&mut nonce);
 
     let encryptor = stream.encrypt(&nonce, data).map_err(Error::ErrEncryptFailed)?;
 
     let mut ciphertext = Vec::with_capacity(salt.len() + 1 + nonce.len() + encryptor.len());
     ciphertext.extend_from_slice(salt);
     ciphertext.push(id as u8);
-    ciphertext.extend_from_slice(nonce.as_slice());
+    ciphertext.extend_from_slice(&nonce);
     ciphertext.extend_from_slice(&encryptor);
 
     Ok(ciphertext)

crates/crypto/src/encdec/tests.rs

@@ -226,7 +226,7 @@ fn test_password_variations() -> Result<(), crate::Error> {
         b"12345".as_slice(),            // Numeric
         b"!@#$%^&*()".as_slice(),       // Special characters
         b"\x00\x01\x02\x03".as_slice(), // Binary password
-        "密码测试".as_bytes(),          // Unicode password
+        "пароль тест".as_bytes(),       // Unicode password
         &[0xFF; 64],                    // Long binary password
     ];
 

crates/e2e_test/src/kms/README.md

@@ -1,267 +1,253 @@
 # KMS End-to-End Tests
 
-本目录包含 RustFS KMS (Key Management Service) 的端到端集成测试，用于验证完整的 KMS 功能流程。
+This directory contains the integration suites used to validate the full RustFS KMS (Key Management Service) workflow.
 
-## 📁 测试文件说明
+## 📁 Test Overview
 
 ### `kms_local_test.rs`
-本地KMS后端的端到端测试，包含：
-- 自动启动和配置本地KMS后端
-- 通过动态配置API配置KMS服务
-- 测试SSE-C（客户端提供密钥）加密流程
-- 验证S3兼容的对象加密/解密操作
-- 密钥生命周期管理测试
+End-to-end coverage for the local KMS backend:
+- Auto-start and configure the local backend
+- Configure KMS through the dynamic configuration API
+- Verify SSE-C (client-provided keys)
+- Exercise S3-compatible encryption/decryption
+- Validate key lifecycle management
 
 ### `kms_vault_test.rs`
-Vault KMS后端的端到端测试，包含：
-- 自动启动Vault开发服务器
-- 配置Vault transit engine和密钥
-- 通过动态配置API配置KMS服务
-- 测试完整的Vault KMS集成
-- 验证Token认证和加密操作
+End-to-end coverage for the Vault backend:
+- Launch a Vault dev server automatically
+- Configure the transit engine and encryption keys
+- Configure KMS via the dynamic configuration API
+- Run the full Vault integration flow
+- Validate token authentication and encryption operations
 
 ### `kms_comprehensive_test.rs`
-**完整的KMS功能测试套件**（当前因AWS SDK API兼容性问题暂时禁用），包含：
-- **Bucket加密配置**: SSE-S3和SSE-KMS默认加密设置
-- **完整的SSE加密模式测试**:
-  - SSE-S3: S3管理的服务端加密
-  - SSE-KMS: KMS管理的服务端加密
-  - SSE-C: 客户端提供密钥的服务端加密
-- **对象操作测试**: 上传、下载、验证三种SSE模式
-- **分片上传测试**: 多部分上传支持所有SSE模式
-- **对象复制测试**: 不同SSE模式间的复制操作
-- **完整KMS API管理**:
-  - 密钥生命周期管理（创建、列表、描述、删除、取消删除）
-  - 直接加密/解密操作
-  - 数据密钥生成和操作
-  - KMS服务管理（启动、停止、状态查询）
+**Full KMS capability suite** (currently disabled because of AWS SDK compatibility issues):
+- **Bucket encryption configuration**: SSE-S3 and SSE-KMS defaults
+- **All SSE encryption modes**:
+  - SSE-S3 (S3-managed server-side encryption)
+  - SSE-KMS (KMS-managed server-side encryption)
+  - SSE-C (client-provided keys)
+- **Object operations**: upload, download, and validation for every SSE mode
+- **Multipart uploads**: cover each SSE mode
+- **Object replication**: cross-mode replication scenarios
+- **Complete KMS API management**:
+  - Key lifecycle (create, list, describe, delete, cancel delete)
+  - Direct encrypt/decrypt operations
+  - Data key generation and handling
+  - KMS service lifecycle (start, stop, status)
 
 ### `kms_integration_test.rs`
-综合性KMS集成测试，包含：
-- 多后端兼容性测试
-- KMS服务生命周期测试
-- 错误处理和恢复测试
-- **注意**: 当前因AWS SDK API兼容性问题暂时禁用
+Broad integration tests that exercise:
+- Multiple backends
+- KMS lifecycle management
+- Error handling and recovery
+- **Note**: currently disabled because of AWS SDK compatibility gaps
 
-## 🚀 如何运行测试
+## 🚀 Running Tests
 
-### 前提条件
+### Prerequisites
 
-1. **系统依赖**：
+1. **System dependencies**
    ```bash
    # macOS
    brew install vault awscurl
-   
+
    # Ubuntu/Debian
    apt-get install vault
    pip install awscurl
    ```
 
-2. **构建RustFS**：
+2. **Build RustFS**
    ```bash
-   # 在项目根目录
    cargo build
    ```
 
-### 运行单个测试
+### Run individual suites
 
-#### 本地KMS测试
+#### Local backend
 ```bash
 cd crates/e2e_test
 cargo test test_local_kms_end_to_end -- --nocapture
 ```
 
-#### Vault KMS测试
+#### Vault backend
 ```bash
 cd crates/e2e_test
 cargo test test_vault_kms_end_to_end -- --nocapture
 ```
 
-#### 高可用性测试
+#### High availability
 ```bash
 cd crates/e2e_test
 cargo test test_vault_kms_high_availability -- --nocapture
 ```
 
-#### 完整功能测试（开发中）
+#### Comprehensive features (disabled)
 ```bash
 cd crates/e2e_test
-# 注意：以下测试因AWS SDK API兼容性问题暂时禁用
+# Disabled due to AWS SDK compatibility gaps
 # cargo test test_comprehensive_kms_functionality -- --nocapture
-# cargo test test_sse_modes_compatibility -- --nocapture  
+# cargo test test_sse_modes_compatibility -- --nocapture
 # cargo test test_kms_api_comprehensive -- --nocapture
 ```
 
-### 运行所有KMS测试
+### Run all KMS suites
 ```bash
 cd crates/e2e_test
 cargo test kms -- --nocapture
 ```
 
-### 串行运行（避免端口冲突）
+### Run serially (avoid port conflicts)
 ```bash
 cd crates/e2e_test
 cargo test kms -- --nocapture --test-threads=1
 ```
 
-## 🔧 测试配置
+## 🔧 Configuration
 
-### 环境变量
+### Environment variables
 ```bash
-# 可选：自定义端口（默认使用9050）
+# Optional: custom RustFS port (default 9050)
 export RUSTFS_TEST_PORT=9050
 
-# 可选：自定义Vault端口（默认使用8200）
+# Optional: custom Vault port (default 8200)
 export VAULT_TEST_PORT=8200
 
-# 可选：启用详细日志
+# Optional: enable verbose logging
 export RUST_LOG=debug
 ```
 
-### 依赖的二进制文件路径
+### Required binaries
 
-测试会自动查找以下二进制文件：
-- `../../target/debug/rustfs` - RustFS服务器
-- `vault` - Vault (需要在PATH中)
-- `/Users/dandan/Library/Python/3.9/bin/awscurl` - AWS签名工具
+Tests look for:
+- `../../target/debug/rustfs` – RustFS server
+- `vault` – Vault CLI (must be on PATH)
+- `/Users/dandan/Library/Python/3.9/bin/awscurl` – AWS SigV4 helper
 
-## 📋 测试流程说明
+## 📋 Test Flow
 
-### Local KMS测试流程
-1. **环境准备**：创建临时目录，设置KMS密钥存储路径
-2. **启动服务**：启动RustFS服务器，启用KMS功能
-3. **等待就绪**：检查端口监听和S3 API响应
-4. **配置KMS**：通过awscurl发送配置请求到admin API
-5. **启动KMS**：激活KMS服务
-6. **功能测试**：
-   - 创建测试存储桶
-   - 测试SSE-C加密（客户端提供密钥）
-   - 验证对象加密/解密
-7. **清理**：终止进程，清理临时文件
+### Local backend
+1. **Prepare environment** – create temporary directories and key storage paths
+2. **Start RustFS** – launch the server with KMS enabled
+3. **Wait for readiness** – confirm the port listener and S3 API
+4. **Configure KMS** – send configuration via awscurl to the admin API
+5. **Start KMS** – activate the KMS service
+6. **Exercise functionality**
+   - Create a test bucket
+   - Run SSE-C encryption with client-provided keys
+   - Validate encryption/decryption behavior
+7. **Cleanup** – stop processes and remove temporary files
 
-### Vault KMS测试流程
-1. **启动Vault**：使用开发模式启动Vault服务器
-2. **配置Vault**：
-   - 启用transit secrets engine
-   - 创建加密密钥（rustfs-master-key）
-3. **启动RustFS**：启用KMS功能的RustFS服务器
-4. **配置KMS**：通过API配置Vault后端，包含：
-   - Vault地址和Token认证
-   - Transit engine配置
-   - 密钥路径设置
-5. **功能测试**：完整的加密/解密流程测试
-6. **清理**：终止所有进程
+### Vault backend
+1. **Launch Vault** – start the dev-mode server
+2. **Configure Vault**
+   - Enable the transit secrets engine
+   - Create the `rustfs-master-key`
+3. **Start RustFS** – run the server with KMS enabled
+4. **Configure KMS** – point RustFS at Vault (address, token, transit config, key path)
+5. **Exercise functionality** – complete the encryption/decryption workflow
+6. **Cleanup** – stop all services
 
-## 🛠️ 故障排除
+## 🛠️ Troubleshooting
 
-### 常见问题
+### Common issues
 
-**Q: 测试失败 "RustFS server failed to become ready"**
-```
-A: 检查端口是否被占用：
+**Q: `RustFS server failed to become ready`**
+```bash
 lsof -i :9050
-kill -9 <PID>  # 如果有进程占用端口
+kill -9 <PID>  # Free the port if necessary
 ```
 
-**Q: Vault服务启动失败**
-```
-A: 确保Vault已安装且在PATH中：
+**Q: Vault fails to start**
+```bash
 which vault
 vault version
 ```
 
-**Q: awscurl认证失败**
-```
-A: 检查awscurl路径是否正确：
+**Q: awscurl authentication fails**
+```bash
 ls /Users/dandan/Library/Python/3.9/bin/awscurl
-# 或安装到不同路径：
+# Or install elsewhere
 pip install awscurl
-which awscurl  # 然后更新测试中的路径
+which awscurl  # Update the path in tests accordingly
 ```
 
-**Q: 测试超时**
-```
-A: 增加等待时间或检查日志：
+**Q: Tests time out**
+```bash
 RUST_LOG=debug cargo test test_local_kms_end_to_end -- --nocapture
 ```
 
-### 调试技巧
+### Debug tips
 
-1. **查看详细日志**：
+1. **Enable verbose logs**
    ```bash
    RUST_LOG=rustfs_kms=debug,rustfs=info cargo test -- --nocapture
    ```
 
-2. **保留临时文件**：
-   修改测试代码，注释掉清理部分，检查生成的配置文件
+2. **Keep temporary files** – comment out cleanup logic to inspect generated configs
 
-3. **单步调试**：
-   在测试中添加 `std::thread::sleep` 来暂停执行，手动检查服务状态
+3. **Pause execution** – add `std::thread::sleep` for manual inspection during tests
 
-4. **端口检查**：
+4. **Monitor ports**
    ```bash
-   # 测试运行时检查端口状态
    netstat -an | grep 9050
    curl http://127.0.0.1:9050/minio/health/ready
    ```
 
-## 📊 测试覆盖范围
+## 📊 Coverage
 
-### 功能覆盖
-- ✅ KMS服务动态配置
-- ✅ 本地和Vault后端支持  
-- ✅ AWS S3兼容加密接口
-- ✅ 密钥管理和生命周期
-- ✅ 错误处理和恢复
-- ✅ 高可用性场景
+### Functional
+- ✅ Dynamic KMS configuration
+- ✅ Local and Vault backends
+- ✅ AWS S3-compatible encryption APIs
+- ✅ Key lifecycle management
+- ✅ Error handling and recovery paths
+- ✅ High-availability behavior
 
-### 加密模式覆盖
-- ✅ SSE-C (Server-Side Encryption with Customer-Provided Keys)
-- ✅ SSE-S3 (Server-Side Encryption with S3-Managed Keys)
-- ✅ SSE-KMS (Server-Side Encryption with KMS-Managed Keys)
+### Encryption modes
+- ✅ SSE-C (customer-provided)
+- ✅ SSE-S3 (S3-managed)
+- ✅ SSE-KMS (KMS-managed)
 
-### S3操作覆盖
-- ✅ 对象上传/下载 (SSE-C模式)
-- 🚧 分片上传 (需要AWS SDK兼容性修复)
-- 🚧 对象复制 (需要AWS SDK兼容性修复)
-- 🚧 Bucket加密配置 (需要AWS SDK兼容性修复)
+### S3 operations
+- ✅ Object upload/download (SSE-C)
+- 🚧 Multipart uploads (pending AWS SDK fixes)
+- 🚧 Object replication (pending AWS SDK fixes)
+- 🚧 Bucket encryption defaults (pending AWS SDK fixes)
 
-### KMS API覆盖
-- ✅ 基础密钥管理 (创建、列表)
-- 🚧 完整密钥生命周期 (需要AWS SDK兼容性修复)
-- 🚧 直接加密/解密操作 (需要AWS SDK兼容性修复)
-- 🚧 数据密钥生成和解密 (需要AWS SDK兼容性修复)
-- ✅ KMS服务管理 (配置、启动、停止、状态)
+### KMS API
+- ✅ Basic key management (create/list)
+- 🚧 Full key lifecycle (pending AWS SDK fixes)
+- 🚧 Direct encrypt/decrypt (pending AWS SDK fixes)
+- 🚧 Data key operations (pending AWS SDK fixes)
+- ✅ Service lifecycle (configure/start/stop/status)
 
-### 认证方式覆盖
-- ✅ Vault Token认证
-- 🚧 Vault AppRole认证
+### Authentication
+- ✅ Vault token auth
+- 🚧 Vault AppRole auth
 
-## 🔄 持续集成
+## 🔄 CI Integration
 
-这些测试设计为可在CI/CD环境中运行：
+Designed to run inside CI/CD pipelines:
 
 ```yaml
-# GitHub Actions 示例
 - name: Run KMS E2E Tests
   run: |
-    # 安装依赖
     sudo apt-get update
     sudo apt-get install -y vault
     pip install awscurl
-    
-    # 构建并测试
+
     cargo build
     cd crates/e2e_test
     cargo test kms -- --nocapture --test-threads=1
 ```
 
-## 📚 相关文档
+## 📚 References
 
-- [KMS 配置文档](../../../../docs/kms/README.md) - KMS功能完整文档
-- [动态配置API](../../../../docs/kms/http-api.md) - REST API接口说明
-- [故障排除指南](../../../../docs/kms/troubleshooting.md) - 常见问题解决
+- [KMS configuration guide](../../../../docs/kms/README.md)
+- [Dynamic configuration API](../../../../docs/kms/http-api.md)
+- [Troubleshooting](../../../../docs/kms/troubleshooting.md)
 
 ---
 
-*这些测试确保KMS功能的稳定性和可靠性，为生产环境部署提供信心。*
+*These suites ensure KMS stability and reliability, building confidence for production deployments.*

crates/e2e_test/src/kms/common.rs

@@ -547,9 +547,9 @@ pub async fn test_multipart_upload_with_config(
 ) -> Result<(), Box<dyn std::error::Error + Send + Sync>> {
     let total_size = config.total_size();
 
-    info!("🧪 开始分片上传测试 - {:?}", config.encryption_type);
+    info!("🧪 Starting multipart upload test - {:?}", config.encryption_type);
     info!(
-        "   对象: {}, 分片: {}个, 每片: {}MB, 总计: {}MB",
+        "   Object: {}, parts: {}, part size: {} MB, total: {} MB",
         config.object_key,
         config.total_parts,
         config.part_size / (1024 * 1024),
@@ -589,7 +589,7 @@ pub async fn test_multipart_upload_with_config(
 
     let create_multipart_output = create_request.send().await?;
     let upload_id = create_multipart_output.upload_id().unwrap();
-    info!("📋 创建分片上传，ID: {}", upload_id);
+    info!("📋 Created multipart upload, ID: {}", upload_id);
 
     // Step 2: Upload parts
     let mut completed_parts = Vec::new();
@@ -598,7 +598,7 @@ pub async fn test_multipart_upload_with_config(
         let end = std::cmp::min(start + config.part_size, total_size);
         let part_data = &test_data[start..end];
 
-        info!("📤 上传分片 {} ({:.2}MB)", part_number, part_data.len() as f64 / (1024.0 * 1024.0));
+        info!("📤 Uploading part {} ({:.2} MB)", part_number, part_data.len() as f64 / (1024.0 * 1024.0));
 
         let mut upload_request = s3_client
             .upload_part()
@@ -625,7 +625,7 @@ pub async fn test_multipart_upload_with_config(
                 .build(),
         );
 
-        debug!("分片 {} 上传完成，ETag: {}", part_number, etag);
+        debug!("Part {} uploaded with ETag {}", part_number, etag);
     }
 
     // Step 3: Complete multipart upload
@@ -633,7 +633,7 @@ pub async fn test_multipart_upload_with_config(
         .set_parts(Some(completed_parts))
         .build();
 
-    info!("🔗 完成分片上传");
+    info!("🔗 Completing multipart upload");
     let complete_output = s3_client
         .complete_multipart_upload()
         .bucket(bucket)
@@ -643,10 +643,10 @@ pub async fn test_multipart_upload_with_config(
         .send()
         .await?;
 
-    debug!("完成分片上传，ETag: {:?}", complete_output.e_tag());
+    debug!("Multipart upload finalized with ETag {:?}", complete_output.e_tag());
 
     // Step 4: Download and verify
-    info!("📥 下载文件并验证");
+    info!("📥 Downloading object for verification");
     let mut get_request = s3_client.get_object().bucket(bucket).key(&config.object_key);
 
     // Add encryption headers for SSE-C GET
@@ -680,7 +680,7 @@ pub async fn test_multipart_upload_with_config(
     assert_eq!(downloaded_data.len(), total_size);
     assert_eq!(&downloaded_data[..], &test_data[..]);
 
-    info!("✅ 分片上传测试通过 - {:?}", config.encryption_type);
+    info!("✅ Multipart upload test passed - {:?}", config.encryption_type);
     Ok(())
 }
 
@@ -700,7 +700,7 @@ pub async fn test_all_multipart_encryption_types(
     bucket: &str,
     base_object_key: &str,
 ) -> Result<(), Box<dyn std::error::Error + Send + Sync>> {
-    info!("🧪 测试所有加密类型的分片上传");
+    info!("🧪 Testing multipart uploads for every encryption type");
 
     let part_size = 5 * 1024 * 1024; // 5MB per part
     let total_parts = 2;
@@ -718,7 +718,7 @@ pub async fn test_all_multipart_encryption_types(
         test_multipart_upload_with_config(s3_client, bucket, &config).await?;
     }
 
-    info!("✅ 所有加密类型的分片上传测试通过");
+    info!("✅ Multipart uploads succeeded for every encryption type");
     Ok(())
 }
 

crates/e2e_test/src/kms/multipart_encryption_test.rs

@@ -201,7 +201,7 @@ async fn test_step3_multipart_upload_with_sse_s3() -> Result<(), Box<dyn std::er
     let total_parts = 2;
     let total_size = part_size * total_parts;
 
-    // 生成测试数据
+    // Generate test data
     let test_data: Vec<u8> = (0..total_size).map(|i| ((i / 1000) % 256) as u8).collect();
 
     info!(
@@ -275,43 +275,43 @@ async fn test_step3_multipart_upload_with_sse_s3() -> Result<(), Box<dyn std::er
         .send()
         .await?;
 
-    debug!("完成加密分片上传，ETag: {:?}", complete_output.e_tag());
+    debug!("Encrypted multipart upload completed with ETag {:?}", complete_output.e_tag());
 
-    // 步骤 4：HEAD 请求检查元数据
-    info!("📋 检查对象元数据");
+    // Step 4: HEAD request to inspect metadata
+    info!("📋 Inspecting object metadata");
     let head_response = s3_client.head_object().bucket(TEST_BUCKET).key(object_key).send().await?;
 
-    debug!("HEAD 响应 SSE: {:?}", head_response.server_side_encryption());
-    debug!("HEAD 响应 元数据：{:?}", head_response.metadata());
+    debug!("HEAD response SSE: {:?}", head_response.server_side_encryption());
+    debug!("HEAD response metadata: {:?}", head_response.metadata());
 
-    // 步骤 5：GET 请求下载并验证
-    info!("📥 下载加密文件并验证");
+    // Step 5: GET request to download and verify
+    info!("📥 Downloading encrypted object for verification");
     let get_response = s3_client.get_object().bucket(TEST_BUCKET).key(object_key).send().await?;
 
-    debug!("GET 响应 SSE: {:?}", get_response.server_side_encryption());
+    debug!("GET response SSE: {:?}", get_response.server_side_encryption());
 
-    // 🎯 关键验证：GET 响应必须包含 SSE-S3 加密头
+    // 🎯 Critical check: GET response must include SSE-S3 headers
     assert_eq!(
         get_response.server_side_encryption(),
         Some(&aws_sdk_s3::types::ServerSideEncryption::Aes256)
     );
 
-    // 验证数据完整性
+    // Verify data integrity
     let downloaded_data = get_response.body.collect().await?.into_bytes();
     assert_eq!(downloaded_data.len(), total_size);
     assert_eq!(&downloaded_data[..], &test_data[..]);
 
     kms_env.base_env.delete_test_bucket(TEST_BUCKET).await?;
-    info!("✅ 步骤 3 通过：分片上传 + SSE-S3 加密功能正常");
+    info!("✅ Step 3 passed: multipart upload with SSE-S3 encryption");
     Ok(())
 }
 
-/// 步骤 4：测试更大的分片上传（测试流式加密）
+/// Step 4: test larger multipart uploads (streaming encryption)
 #[tokio::test]
 #[serial]
 async fn test_step4_large_multipart_upload_with_encryption() -> Result<(), Box<dyn std::error::Error + Send + Sync>> {
     init_logging();
-    info!("🧪 步骤 4：测试大文件分片上传加密");
+    info!("🧪 Step 4: test large-file multipart encryption");
 
     let mut kms_env = LocalKMSTestEnvironment::new().await?;
     let _default_key_id = kms_env.start_rustfs_for_local_kms().await?;
@@ -321,18 +321,18 @@ async fn test_step4_large_multipart_upload_with_encryption() -> Result<(), Box<d
     kms_env.base_env.create_test_bucket(TEST_BUCKET).await?;
 
     let object_key = "test-large-multipart-encrypted";
-    let part_size = 6 * 1024 * 1024; // 6MB per part (大于 1MB 加密块大小)
-    let total_parts = 3; // 总共 18MB
+    let part_size = 6 * 1024 * 1024; // 6 MB per part (greater than the 1 MB encryption chunk)
+    let total_parts = 3; // 18 MB total
     let total_size = part_size * total_parts;
 
     info!(
-        "🗂️ 生成大文件测试数据：{} parts，每个 {}MB，总计 {}MB",
+        "🗂️ Generated large-file test data: {} parts, {} MB each, {} MB total",
         total_parts,
         part_size / (1024 * 1024),
         total_size / (1024 * 1024)
     );
 
-    // 生成大文件测试数据（使用复杂模式便于验证）
+    // Generate large test data (complex pattern for validation)
     let test_data: Vec<u8> = (0..total_size)
         .map(|i| {
             let part_num = i / part_size;
@@ -341,9 +341,9 @@ async fn test_step4_large_multipart_upload_with_encryption() -> Result<(), Box<d
         })
         .collect();
 
-    info!("🔐 开始大文件分片上传（SSE-S3 加密）");
+    info!("🔐 Starting large-file multipart upload (SSE-S3 encryption)");
 
-    // 创建分片上传
+    // Create multipart upload
     let create_multipart_output = s3_client
         .create_multipart_upload()
         .bucket(TEST_BUCKET)
@@ -353,9 +353,9 @@ async fn test_step4_large_multipart_upload_with_encryption() -> Result<(), Box<d
         .await?;
 
     let upload_id = create_multipart_output.upload_id().unwrap();
-    info!("📋 创建大文件加密分片上传，ID: {}", upload_id);
+    info!("📋 Created large encrypted multipart upload, ID: {}", upload_id);
 
-    // 上传各个分片
+    // Upload each part
     let mut completed_parts = Vec::new();
     for part_number in 1..=total_parts {
         let start = (part_number - 1) * part_size;
@@ -363,7 +363,7 @@ async fn test_step4_large_multipart_upload_with_encryption() -> Result<(), Box<d
         let part_data = &test_data[start..end];
 
         info!(
-            "🔐 上传大文件加密分片 {} ({:.2}MB)",
+            "🔐 Uploading encrypted large-file part {} ({:.2} MB)",
             part_number,
             part_data.len() as f64 / (1024.0 * 1024.0)
         );
@@ -386,15 +386,15 @@ async fn test_step4_large_multipart_upload_with_encryption() -> Result<(), Box<d
                 .build(),
         );
 
-        debug!("大文件加密分片 {} 上传完成，ETag: {}", part_number, etag);
+        debug!("Large encrypted part {} uploaded with ETag {}", part_number, etag);
     }
 
-    // 完成分片上传
+    // Complete the multipart upload
     let completed_multipart_upload = aws_sdk_s3::types::CompletedMultipartUpload::builder()
         .set_parts(Some(completed_parts))
         .build();
 
-    info!("🔗 完成大文件加密分片上传");
+    info!("🔗 Completing large encrypted multipart upload");
     let complete_output = s3_client
         .complete_multipart_upload()
         .bucket(TEST_BUCKET)
@@ -404,40 +404,40 @@ async fn test_step4_large_multipart_upload_with_encryption() -> Result<(), Box<d
         .send()
         .await?;
 
-    debug!("完成大文件加密分片上传，ETag: {:?}", complete_output.e_tag());
+    debug!("Large encrypted multipart upload completed with ETag {:?}", complete_output.e_tag());
 
-    // 下载并验证
-    info!("📥 下载大文件并验证");
+    // Download and verify
+    info!("📥 Downloading large object for verification");
     let get_response = s3_client.get_object().bucket(TEST_BUCKET).key(object_key).send().await?;
 
-    // 验证加密头
+    // Verify encryption headers
     assert_eq!(
         get_response.server_side_encryption(),
         Some(&aws_sdk_s3::types::ServerSideEncryption::Aes256)
     );
 
-    // 验证数据完整性
+    // Verify data integrity
     let downloaded_data = get_response.body.collect().await?.into_bytes();
     assert_eq!(downloaded_data.len(), total_size);
 
-    // 逐字节验证数据（对于大文件更严格）
+    // Validate bytes individually (stricter for large files)
     for (i, (&actual, &expected)) in downloaded_data.iter().zip(test_data.iter()).enumerate() {
         if actual != expected {
-            panic!("大文件数据在第{i}字节不匹配：实际={actual}, 期待={expected}");
+            panic!("Large file mismatch at byte {i}: actual={actual}, expected={expected}");
         }
     }
 
     kms_env.base_env.delete_test_bucket(TEST_BUCKET).await?;
-    info!("✅ 步骤 4 通过：大文件分片上传加密功能正常");
+    info!("✅ Step 4 passed: large-file multipart encryption succeeded");
     Ok(())
 }
 
-/// 步骤 5：测试所有加密类型的分片上传
+/// Step 5: test multipart uploads for every encryption mode
 #[tokio::test]
 #[serial]
 async fn test_step5_all_encryption_types_multipart() -> Result<(), Box<dyn std::error::Error + Send + Sync>> {
     init_logging();
-    info!("🧪 步骤 5：测试所有加密类型的分片上传");
+    info!("🧪 Step 5: test multipart uploads for every encryption mode");
 
     let mut kms_env = LocalKMSTestEnvironment::new().await?;
     let _default_key_id = kms_env.start_rustfs_for_local_kms().await?;
@@ -450,8 +450,8 @@ async fn test_step5_all_encryption_types_multipart() -> Result<(), Box<dyn std::
     let total_parts = 2;
     let total_size = part_size * total_parts;
 
-    // 测试 SSE-KMS
-    info!("🔐 测试 SSE-KMS 分片上传");
+    // Test SSE-KMS
+    info!("🔐 Testing SSE-KMS multipart upload");
     test_multipart_encryption_type(
         &s3_client,
         TEST_BUCKET,
@@ -463,8 +463,8 @@ async fn test_step5_all_encryption_types_multipart() -> Result<(), Box<dyn std::
     )
     .await?;
 
-    // 测试 SSE-C
-    info!("🔐 测试 SSE-C 分片上传");
+    // Test SSE-C
+    info!("🔐 Testing SSE-C multipart upload");
     test_multipart_encryption_type(
         &s3_client,
         TEST_BUCKET,
@@ -477,7 +477,7 @@ async fn test_step5_all_encryption_types_multipart() -> Result<(), Box<dyn std::
     .await?;
 
     kms_env.base_env.delete_test_bucket(TEST_BUCKET).await?;
-    info!("✅ 步骤 5 通过：所有加密类型的分片上传功能正常");
+    info!("✅ Step 5 passed: multipart uploads succeeded for every encryption mode");
     Ok(())
 }
 
@@ -487,7 +487,7 @@ enum EncryptionType {
     SSEC,
 }
 
-/// 辅助函数：测试特定加密类型的分片上传
+/// Helper: test multipart uploads for a specific encryption type
 async fn test_multipart_encryption_type(
     s3_client: &aws_sdk_s3::Client,
     bucket: &str,
@@ -497,10 +497,10 @@ async fn test_multipart_encryption_type(
     total_parts: usize,
     encryption_type: EncryptionType,
 ) -> Result<(), Box<dyn std::error::Error + Send + Sync>> {
-    // 生成测试数据
+    // Generate test data
     let test_data: Vec<u8> = (0..total_size).map(|i| ((i * 7) % 256) as u8).collect();
 
-    // 准备 SSE-C 所需的密钥（如果需要）
+    // Prepare SSE-C keys when required
     let (sse_c_key, sse_c_md5) = if matches!(encryption_type, EncryptionType::SSEC) {
         let key = "01234567890123456789012345678901";
         let key_b64 = base64::Engine::encode(&base64::engine::general_purpose::STANDARD, key);
@@ -510,9 +510,9 @@ async fn test_multipart_encryption_type(
         (None, None)
     };
 
-    info!("📋 创建分片上传 - {:?}", encryption_type);
+    info!("📋 Creating multipart upload - {:?}", encryption_type);
 
-    // 创建分片上传
+    // Create multipart upload
     let mut create_request = s3_client.create_multipart_upload().bucket(bucket).key(object_key);
 
     create_request = match encryption_type {
@@ -526,7 +526,7 @@ async fn test_multipart_encryption_type(
     let create_multipart_output = create_request.send().await?;
     let upload_id = create_multipart_output.upload_id().unwrap();
 
-    // 上传分片
+    // Upload parts
     let mut completed_parts = Vec::new();
     for part_number in 1..=total_parts {
         let start = (part_number - 1) * part_size;
@@ -541,7 +541,7 @@ async fn test_multipart_encryption_type(
             .part_number(part_number as i32)
             .body(aws_sdk_s3::primitives::ByteStream::from(part_data.to_vec()));
 
-        // SSE-C 需要在每个 UploadPart 请求中包含密钥
+        // SSE-C requires the key on each UploadPart request
         if matches!(encryption_type, EncryptionType::SSEC) {
             upload_request = upload_request
                 .sse_customer_algorithm("AES256")
@@ -558,10 +558,10 @@ async fn test_multipart_encryption_type(
                 .build(),
         );
 
-        debug!("{:?} 分片 {} 上传完成", encryption_type, part_number);
+        debug!("{:?} part {} uploaded", encryption_type, part_number);
     }
 
-    // 完成分片上传
+    // Complete the multipart upload
     let completed_multipart_upload = aws_sdk_s3::types::CompletedMultipartUpload::builder()
         .set_parts(Some(completed_parts))
         .build();
@@ -575,10 +575,10 @@ async fn test_multipart_encryption_type(
         .send()
         .await?;
 
-    // 下载并验证
+    // Download and verify
     let mut get_request = s3_client.get_object().bucket(bucket).key(object_key);
 
-    // SSE-C 需要在 GET 请求中包含密钥
+    // SSE-C requires the key on GET requests
     if matches!(encryption_type, EncryptionType::SSEC) {
         get_request = get_request
             .sse_customer_algorithm("AES256")
@@ -588,7 +588,7 @@ async fn test_multipart_encryption_type(
 
     let get_response = get_request.send().await?;
 
-    // 验证加密头
+    // Verify encryption headers
     match encryption_type {
         EncryptionType::SSEKMS => {
             assert_eq!(
@@ -601,11 +601,11 @@ async fn test_multipart_encryption_type(
         }
     }
 
-    // 验证数据完整性
+    // Verify data integrity
     let downloaded_data = get_response.body.collect().await?.into_bytes();
     assert_eq!(downloaded_data.len(), total_size);
     assert_eq!(&downloaded_data[..], &test_data[..]);
 
-    info!("✅ {:?} 分片上传测试通过", encryption_type);
+    info!("✅ {:?} multipart upload test passed", encryption_type);
     Ok(())
 }

crates/e2e_test/src/kms/test_runner.rs

@@ -346,7 +346,7 @@ impl KMSTestSuite {
     /// Run the complete test suite
     pub async fn run_test_suite(&self) -> Vec<TestResult> {
         init_logging();
-        info!("🚀 开始KMS统一测试套件");
+        info!("🚀 Starting unified KMS test suite");
 
         let start_time = Instant::now();
         let mut results = Vec::new();
@@ -359,17 +359,17 @@ impl KMSTestSuite {
             .filter(|test| !self.config.include_critical_only || test.is_critical)
             .collect();
 
-        info!("📊 测试计划: {} 个测试将被执行", tests_to_run.len());
+        info!("📊 Test plan: {} test(s) scheduled", tests_to_run.len());
         for (i, test) in tests_to_run.iter().enumerate() {
             info!("  {}. {} ({})", i + 1, test.name, test.category.as_str());
         }
 
         // Execute tests
         for (i, test_def) in tests_to_run.iter().enumerate() {
-            info!("🧪 执行测试 {}/{}: {}", i + 1, tests_to_run.len(), test_def.name);
-            info!("   📝 描述: {}", test_def.description);
-            info!("   🏷️  分类: {}", test_def.category.as_str());
-            info!("   ⏱️  预计时间: {:?}", test_def.estimated_duration);
+            info!("🧪 Running test {}/{}: {}", i + 1, tests_to_run.len(), test_def.name);
+            info!("   📝 Description: {}", test_def.description);
+            info!("   🏷️  Category: {}", test_def.category.as_str());
+            info!("   ⏱️  Estimated duration: {:?}", test_def.estimated_duration);
 
             let test_start = Instant::now();
             let result = self.run_single_test(test_def).await;
@@ -377,11 +377,11 @@ impl KMSTestSuite {
 
             match result {
                 Ok(_) => {
-                    info!("✅ 测试通过: {} ({:.2}s)", test_def.name, test_duration.as_secs_f64());
+                    info!("✅ Test passed: {} ({:.2}s)", test_def.name, test_duration.as_secs_f64());
                     results.push(TestResult::success(test_def.name.clone(), test_def.category.clone(), test_duration));
                 }
                 Err(e) => {
-                    error!("❌ 测试失败: {} ({:.2}s): {}", test_def.name, test_duration.as_secs_f64(), e);
+                    error!("❌ Test failed: {} ({:.2}s): {}", test_def.name, test_duration.as_secs_f64(), e);
                     results.push(TestResult::failure(
                         test_def.name.clone(),
                         test_def.category.clone(),
@@ -393,7 +393,7 @@ impl KMSTestSuite {
 
             // Add delay between tests to avoid resource conflicts
             if i < tests_to_run.len() - 1 {
-                debug!("⏸️  等待2秒后执行下一个测试...");
+                debug!("⏸️  Waiting two seconds before the next test...");
                 sleep(Duration::from_secs(2)).await;
             }
         }
@@ -408,22 +408,22 @@ impl KMSTestSuite {
     async fn run_single_test(&self, test_def: &TestDefinition) -> Result<(), Box<dyn std::error::Error + Send + Sync>> {
         // This is a placeholder for test dispatch logic
         // In a real implementation, this would dispatch to actual test functions
-        warn!("⚠️  测试函数 '{}' 在统一运行器中尚未实现，跳过", test_def.name);
+        warn!("⚠️  Test '{}' is not implemented in the unified runner; skipping", test_def.name);
         Ok(())
     }
 
     /// Print comprehensive test summary
     fn print_test_summary(&self, results: &[TestResult], total_duration: Duration) {
-        info!("📊 KMS测试套件总结");
-        info!("⏱️  总执行时间: {:.2}秒", total_duration.as_secs_f64());
-        info!("📈 总测试数量: {}", results.len());
+        info!("📊 KMS test suite summary");
+        info!("⏱️  Total duration: {:.2} seconds", total_duration.as_secs_f64());
+        info!("📈 Total tests: {}", results.len());
 
         let passed = results.iter().filter(|r| r.success).count();
         let failed = results.iter().filter(|r| !r.success).count();
 
-        info!("✅ 通过: {}", passed);
-        info!("❌ 失败: {}", failed);
-        info!("📊 成功率: {:.1}%", (passed as f64 / results.len() as f64) * 100.0);
+        info!("✅ Passed: {}", passed);
+        info!("❌ Failed: {}", failed);
+        info!("📊 Success rate: {:.1}%", (passed as f64 / results.len() as f64) * 100.0);
 
         // Summary by category
         let mut category_summary: std::collections::HashMap<TestCategory, (usize, usize)> = std::collections::HashMap::new();
@@ -435,7 +435,7 @@ impl KMSTestSuite {
             }
         }
 
-        info!("📊 分类汇总:");
+        info!("📊 Category summary:");
         for (category, (total, passed_count)) in category_summary {
             info!(
                 "  🏷️  {}: {}/{} ({:.1}%)",
@@ -448,7 +448,7 @@ impl KMSTestSuite {
 
         // List failed tests
         if failed > 0 {
-            warn!("❌ 失败的测试:");
+            warn!("❌ Failing tests:");
             for result in results.iter().filter(|r| !r.success) {
                 warn!(
                     "  - {}: {}",
@@ -479,7 +479,7 @@ async fn test_kms_critical_suite() -> Result<(), Box<dyn std::error::Error + Sen
         return Err(format!("Critical test suite failed: {failed_count} tests failed").into());
     }
 
-    info!("✅ 所有关键测试通过");
+    info!("✅ All critical tests passed");
     Ok(())
 }
 
@@ -494,13 +494,13 @@ async fn test_kms_full_suite() -> Result<(), Box<dyn std::error::Error + Send +
     let failed_count = results.iter().filter(|r| !r.success).count();
     let success_rate = ((total_tests - failed_count) as f64 / total_tests as f64) * 100.0;
 
-    info!("📊 完整测试套件结果: {:.1}% 成功率", success_rate);
+    info!("📊 Full suite success rate: {:.1}%", success_rate);
 
     // Allow up to 10% failure rate for non-critical tests
     if success_rate < 90.0 {
         return Err(format!("Test suite success rate too low: {success_rate:.1}%").into());
     }
 
-    info!("✅ 完整测试套件通过");
+    info!("✅ Full test suite succeeded");
     Ok(())
 }

crates/e2e_test/src/reliant/get_deleted_object_test.rs

@@ -0,0 +1,284 @@
+// Copyright 2024 RustFS Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+//! Test for GetObject on deleted objects
+//!
+//! This test reproduces the issue where getting a deleted object returns
+//! a networking error instead of NoSuchKey.
+
+#![cfg(test)]
+
+use aws_config::meta::region::RegionProviderChain;
+use aws_sdk_s3::Client;
+use aws_sdk_s3::config::{Credentials, Region};
+use aws_sdk_s3::error::SdkError;
+use bytes::Bytes;
+use serial_test::serial;
+use std::error::Error;
+use tracing::info;
+
+const ENDPOINT: &str = "http://localhost:9000";
+const ACCESS_KEY: &str = "rustfsadmin";
+const SECRET_KEY: &str = "rustfsadmin";
+const BUCKET: &str = "test-get-deleted-bucket";
+
+async fn create_aws_s3_client() -> Result<Client, Box<dyn Error>> {
+    let region_provider = RegionProviderChain::default_provider().or_else(Region::new("us-east-1"));
+    let shared_config = aws_config::defaults(aws_config::BehaviorVersion::latest())
+        .region(region_provider)
+        .credentials_provider(Credentials::new(ACCESS_KEY, SECRET_KEY, None, None, "static"))
+        .endpoint_url(ENDPOINT)
+        .load()
+        .await;
+
+    let client = Client::from_conf(
+        aws_sdk_s3::Config::from(&shared_config)
+            .to_builder()
+            .force_path_style(true)
+            .build(),
+    );
+    Ok(client)
+}
+
+/// Setup test bucket, creating it if it doesn't exist
+async fn setup_test_bucket(client: &Client) -> Result<(), Box<dyn Error>> {
+    match client.create_bucket().bucket(BUCKET).send().await {
+        Ok(_) => {}
+        Err(SdkError::ServiceError(e)) => {
+            let e = e.into_err();
+            let error_code = e.meta().code().unwrap_or("");
+            if !error_code.eq("BucketAlreadyExists") && !error_code.eq("BucketAlreadyOwnedByYou") {
+                return Err(e.into());
+            }
+        }
+        Err(e) => {
+            return Err(e.into());
+        }
+    }
+    Ok(())
+}
+
+#[tokio::test]
+#[serial]
+#[ignore = "requires running RustFS server at localhost:9000"]
+async fn test_get_deleted_object_returns_nosuchkey() -> Result<(), Box<dyn std::error::Error>> {
+    // Initialize logging
+    let _ = tracing_subscriber::fmt()
+        .with_max_level(tracing::Level::INFO)
+        .with_test_writer()
+        .try_init();
+
+    info!("🧪 Starting test_get_deleted_object_returns_nosuchkey");
+
+    let client = create_aws_s3_client().await?;
+    setup_test_bucket(&client).await?;
+
+    // Upload a test object
+    let key = "test-file-to-delete.txt";
+    let content = b"This will be deleted soon!";
+
+    info!("Uploading object: {}", key);
+    client
+        .put_object()
+        .bucket(BUCKET)
+        .key(key)
+        .body(Bytes::from_static(content).into())
+        .send()
+        .await?;
+
+    // Verify object exists
+    info!("Verifying object exists");
+    let get_result = client.get_object().bucket(BUCKET).key(key).send().await;
+
+    assert!(get_result.is_ok(), "Object should exist after upload");
+
+    // Delete the object
+    info!("Deleting object: {}", key);
+    client.delete_object().bucket(BUCKET).key(key).send().await?;
+
+    // Try to get the deleted object - should return NoSuchKey error
+    info!("Attempting to get deleted object - expecting NoSuchKey error");
+    let get_result = client.get_object().bucket(BUCKET).key(key).send().await;
+
+    // Check that we get an error
+    assert!(get_result.is_err(), "Getting deleted object should return an error");
+
+    // Check that the error is NoSuchKey, not a networking error
+    let err = get_result.unwrap_err();
+
+    // Print the error for debugging
+    info!("Error received: {:?}", err);
+
+    // Check if it's a service error
+    match err {
+        SdkError::ServiceError(service_err) => {
+            let s3_err = service_err.into_err();
+            info!("Service error code: {:?}", s3_err.meta().code());
+
+            // The error should be NoSuchKey
+            assert!(s3_err.is_no_such_key(), "Error should be NoSuchKey, got: {:?}", s3_err);
+
+            info!("✅ Test passed: GetObject on deleted object correctly returns NoSuchKey");
+        }
+        other_err => {
+            panic!("Expected ServiceError with NoSuchKey, but got: {:?}", other_err);
+        }
+    }
+
+    // Cleanup
+    let _ = client.delete_object().bucket(BUCKET).key(key).send().await;
+
+    Ok(())
+}
+
+/// Test that HeadObject on a deleted object also returns NoSuchKey
+#[tokio::test]
+#[serial]
+#[ignore = "requires running RustFS server at localhost:9000"]
+async fn test_head_deleted_object_returns_nosuchkey() -> Result<(), Box<dyn std::error::Error>> {
+    let _ = tracing_subscriber::fmt()
+        .with_max_level(tracing::Level::INFO)
+        .with_test_writer()
+        .try_init();
+
+    info!("🧪 Starting test_head_deleted_object_returns_nosuchkey");
+
+    let client = create_aws_s3_client().await?;
+    setup_test_bucket(&client).await?;
+
+    let key = "test-head-deleted.txt";
+    let content = b"Test content for HeadObject";
+
+    // Upload and verify
+    client
+        .put_object()
+        .bucket(BUCKET)
+        .key(key)
+        .body(Bytes::from_static(content).into())
+        .send()
+        .await?;
+
+    // Delete the object
+    client.delete_object().bucket(BUCKET).key(key).send().await?;
+
+    // Try to head the deleted object
+    let head_result = client.head_object().bucket(BUCKET).key(key).send().await;
+
+    assert!(head_result.is_err(), "HeadObject on deleted object should return an error");
+
+    match head_result.unwrap_err() {
+        SdkError::ServiceError(service_err) => {
+            let s3_err = service_err.into_err();
+            assert!(
+                s3_err.meta().code() == Some("NoSuchKey") || s3_err.meta().code() == Some("NotFound"),
+                "Error should be NoSuchKey or NotFound, got: {:?}",
+                s3_err
+            );
+            info!("✅ HeadObject correctly returns NoSuchKey/NotFound");
+        }
+        other_err => {
+            panic!("Expected ServiceError but got: {:?}", other_err);
+        }
+    }
+
+    Ok(())
+}
+
+/// Test GetObject with non-existent key (never existed)
+#[tokio::test]
+#[serial]
+#[ignore = "requires running RustFS server at localhost:9000"]
+async fn test_get_nonexistent_object_returns_nosuchkey() -> Result<(), Box<dyn std::error::Error>> {
+    let _ = tracing_subscriber::fmt()
+        .with_max_level(tracing::Level::INFO)
+        .with_test_writer()
+        .try_init();
+
+    info!("🧪 Starting test_get_nonexistent_object_returns_nosuchkey");
+
+    let client = create_aws_s3_client().await?;
+    setup_test_bucket(&client).await?;
+
+    // Try to get an object that never existed
+    let key = "this-key-never-existed.txt";
+
+    let get_result = client.get_object().bucket(BUCKET).key(key).send().await;
+
+    assert!(get_result.is_err(), "Getting non-existent object should return an error");
+
+    match get_result.unwrap_err() {
+        SdkError::ServiceError(service_err) => {
+            let s3_err = service_err.into_err();
+            assert!(s3_err.is_no_such_key(), "Error should be NoSuchKey, got: {:?}", s3_err);
+            info!("✅ GetObject correctly returns NoSuchKey for non-existent object");
+        }
+        other_err => {
+            panic!("Expected ServiceError with NoSuchKey, but got: {:?}", other_err);
+        }
+    }
+
+    Ok(())
+}
+
+/// Test multiple consecutive GetObject calls on deleted object
+/// This ensures the fix is stable and doesn't have race conditions
+#[tokio::test]
+#[serial]
+#[ignore = "requires running RustFS server at localhost:9000"]
+async fn test_multiple_gets_deleted_object() -> Result<(), Box<dyn std::error::Error>> {
+    let _ = tracing_subscriber::fmt()
+        .with_max_level(tracing::Level::INFO)
+        .with_test_writer()
+        .try_init();
+
+    info!("🧪 Starting test_multiple_gets_deleted_object");
+
+    let client = create_aws_s3_client().await?;
+    setup_test_bucket(&client).await?;
+
+    let key = "test-multiple-gets.txt";
+    let content = b"Test content";
+
+    // Upload and delete
+    client
+        .put_object()
+        .bucket(BUCKET)
+        .key(key)
+        .body(Bytes::from_static(content).into())
+        .send()
+        .await?;
+
+    client.delete_object().bucket(BUCKET).key(key).send().await?;
+
+    // Try multiple consecutive GetObject calls
+    for i in 1..=5 {
+        info!("Attempt {} to get deleted object", i);
+        let get_result = client.get_object().bucket(BUCKET).key(key).send().await;
+
+        assert!(get_result.is_err(), "Attempt {}: should return error", i);
+
+        match get_result.unwrap_err() {
+            SdkError::ServiceError(service_err) => {
+                let s3_err = service_err.into_err();
+                assert!(s3_err.is_no_such_key(), "Attempt {}: Error should be NoSuchKey, got: {:?}", i, s3_err);
+            }
+            other_err => {
+                panic!("Attempt {}: Expected ServiceError but got: {:?}", i, other_err);
+            }
+        }
+    }
+
+    info!("✅ All 5 attempts correctly returned NoSuchKey");
+    Ok(())
+}

crates/e2e_test/src/reliant/mod.rs

@@ -13,6 +13,7 @@
 // limitations under the License.
 
 mod conditional_writes;
+mod get_deleted_object_test;
 mod lifecycle;
 mod lock;
 mod node_interact_test;

crates/ecstore/Cargo.toml

@@ -103,9 +103,10 @@ parking_lot = { workspace = true }
 moka = { workspace = true }
 base64-simd.workspace = true
 serde_urlencoded.workspace = true
-google-cloud-storage = "1.1.0"
-google-cloud-auth = "1.0.1"
+google-cloud-storage = { workspace = true }
+google-cloud-auth = { workspace = true }
 aws-config = { workspace = true }
+faster-hex = { workspace = true }
 
 [target.'cfg(not(windows))'.dependencies]
 nix = { workspace = true }
@@ -118,7 +119,6 @@ winapi = { workspace = true }
 tokio = { workspace = true, features = ["rt-multi-thread", "macros"] }
 criterion = { workspace = true, features = ["html_reports"] }
 temp-env = { workspace = true }
-heed = "0.22.0"
 
 [build-dependencies]
 shadow-rs = { workspace = true, features = ["build", "metadata"] }

crates/ecstore/run_benchmarks.sh

@@ -14,12 +14,12 @@
 # limitations under the License.
 
 
-# Reed-Solomon SIMD 性能基准测试脚本
-# 使用高性能 SIMD 实现进行纠删码性能测试
+# Reed-Solomon SIMD performance benchmark script
+# Run erasure-coding benchmarks using the high-performance SIMD implementation
 
 set -e
 
-# ANSI 颜色码
+# ANSI color codes
 RED='\033[0;31m'
 GREEN='\033[0;32m'
 YELLOW='\033[1;33m'
@@ -27,7 +27,7 @@ BLUE='\033[0;34m'
 PURPLE='\033[0;35m'
 NC='\033[0m' # No Color
 
-# 打印带颜色的消息
+# Print colored messages
 print_info() {
     echo -e "${BLUE}ℹ️  $1${NC}"
 }
@@ -44,177 +44,177 @@ print_error() {
     echo -e "${RED}❌ $1${NC}"
 }
 
-# 检查系统要求
+# Validate system requirements
 check_requirements() {
-    print_info "检查系统要求..."
+    print_info "Checking system requirements..."
     
-    # 检查 Rust
+    # Check for Rust
     if ! command -v cargo &> /dev/null; then
-        print_error "Cargo 未找到，请确保已安装 Rust"
+        print_error "Cargo not found; install Rust first"
         exit 1
     fi
     
-    # 检查 criterion
+    # Check criterion support
     if ! cargo --list | grep -q "bench"; then
-        print_error "未找到基准测试支持，请确保使用的是支持基准测试的 Rust 版本"
+        print_error "Benchmark support missing; use a Rust toolchain with criterion support"
         exit 1
     fi
     
-    print_success "系统要求检查通过"
+    print_success "System requirements satisfied"
 }
 
-# 清理之前的测试结果
+# Remove previous benchmark artifacts
 cleanup() {
-    print_info "清理之前的测试结果..."
+    print_info "Cleaning previous benchmark artifacts..."
     rm -rf target/criterion
-    print_success "清理完成"
+    print_success "Cleanup complete"
 }
 
-# 运行 SIMD 模式基准测试
+# Run SIMD-only benchmarks
 run_simd_benchmark() {
-    print_info "🎯 开始运行 SIMD 模式基准测试..."
+    print_info "🎯 Starting SIMD-only benchmark run..."
     echo "================================================"
     
     cargo bench --bench comparison_benchmark \
         -- --save-baseline simd_baseline
     
-    print_success "SIMD 模式基准测试完成"
+    print_success "SIMD-only benchmarks completed"
 }
 
-# 运行完整的基准测试套件
+# Run the full benchmark suite
 run_full_benchmark() {
-    print_info "🚀 开始运行完整基准测试套件..."
+    print_info "🚀 Starting full benchmark suite..."
     echo "================================================"
     
-    # 运行详细的基准测试
+    # Execute detailed benchmarks
     cargo bench --bench erasure_benchmark
     
-    print_success "完整基准测试套件完成"
+    print_success "Full benchmark suite finished"
 }
 
-# 运行性能测试
+# Run performance tests
 run_performance_test() {
-    print_info "📊 开始运行性能测试..."
+    print_info "📊 Starting performance tests..."
     echo "================================================"
     
-    print_info "步骤 1: 运行编码基准测试..."
+    print_info "Step 1: running encoding benchmarks..."
     cargo bench --bench comparison_benchmark \
         -- encode --save-baseline encode_baseline
     
-    print_info "步骤 2: 运行解码基准测试..."
+    print_info "Step 2: running decoding benchmarks..."
     cargo bench --bench comparison_benchmark \
         -- decode --save-baseline decode_baseline
     
-    print_success "性能测试完成"
+    print_success "Performance tests completed"
 }
 
-# 运行大数据集测试
+# Run large dataset tests
 run_large_data_test() {
-    print_info "🗂️ 开始运行大数据集测试..."
+    print_info "🗂️ Starting large-dataset tests..."
     echo "================================================"
     
     cargo bench --bench erasure_benchmark \
         -- large_data --save-baseline large_data_baseline
     
-    print_success "大数据集测试完成"
+    print_success "Large-dataset tests completed"
 }
 
-# 生成比较报告
+# Generate comparison report
 generate_comparison_report() {
-    print_info "📊 生成性能报告..."
+    print_info "📊 Generating performance report..."
     
     if [ -d "target/criterion" ]; then
-        print_info "基准测试结果已保存到 target/criterion/ 目录"
-        print_info "你可以打开 target/criterion/report/index.html 查看详细报告"
+        print_info "Benchmark results saved under target/criterion/"
+        print_info "Open target/criterion/report/index.html for the HTML report"
         
-        # 如果有 python 环境，可以启动简单的 HTTP 服务器查看报告
+        # If Python is available, start a simple HTTP server to browse the report
         if command -v python3 &> /dev/null; then
-            print_info "你可以运行以下命令启动本地服务器查看报告:"
+            print_info "Run the following command to serve the report locally:"
             echo "  cd target/criterion && python3 -m http.server 8080"
-            echo "  然后在浏览器中访问 http://localhost:8080/report/index.html"
+            echo "  Then open http://localhost:8080/report/index.html"
         fi
     else
-        print_warning "未找到基准测试结果目录"
+        print_warning "Benchmark result directory not found"
     fi
 }
 
-# 快速测试模式
+# Quick test mode
 run_quick_test() {
-    print_info "🏃 运行快速性能测试..."
+    print_info "🏃 Running quick performance test..."
     
-    print_info "测试 SIMD 编码性能..."
+    print_info "Testing SIMD encoding performance..."
     cargo bench --bench comparison_benchmark \
         -- encode --quick
     
-    print_info "测试 SIMD 解码性能..."
+    print_info "Testing SIMD decoding performance..."
     cargo bench --bench comparison_benchmark \
         -- decode --quick
     
-    print_success "快速测试完成"
+    print_success "Quick test complete"
 }
 
-# 显示帮助信息
+# Display help
 show_help() {
-    echo "Reed-Solomon SIMD 性能基准测试脚本"
+    echo "Reed-Solomon SIMD performance benchmark script"
     echo ""
-    echo "实现模式："
-    echo "  🎯 SIMD 模式 - 高性能 SIMD 优化的 reed-solomon-simd 实现"
+    echo "Modes:"
+    echo "  🎯 simd         High-performance reed-solomon-simd implementation"
     echo ""
-    echo "使用方法:"
+    echo "Usage:"
     echo "  $0 [command]"
     echo ""
-    echo "命令:"
-    echo "  quick        运行快速性能测试"
-    echo "  full         运行完整基准测试套件"
-    echo "  performance  运行详细的性能测试"
-    echo "  simd         运行 SIMD 模式测试"
-    echo "  large        运行大数据集测试"
-    echo "  clean        清理测试结果"
-    echo "  help         显示此帮助信息"
+    echo "Commands:"
+    echo "  quick        Run the quick performance test"
+    echo "  full         Run the full benchmark suite"
+    echo "  performance  Run detailed performance tests"
+    echo "  simd         Run the SIMD-only tests"
+    echo "  large        Run large-dataset tests"
+    echo "  clean        Remove previous results"
+    echo "  help         Show this help message"
     echo ""
-    echo "示例:"
-    echo "  $0 quick              # 快速性能测试"
-    echo "  $0 performance        # 详细性能测试"
-    echo "  $0 full              # 完整测试套件"
-    echo "  $0 simd              # SIMD 模式测试"
-    echo "  $0 large             # 大数据集测试"
+    echo "Examples:"
+    echo "  $0 quick              # Quick performance test"
+    echo "  $0 performance        # Detailed performance test"
+    echo "  $0 full              # Full benchmark suite"
+    echo "  $0 simd              # SIMD-only benchmark"
+    echo "  $0 large             # Large-dataset benchmark"
     echo ""
-    echo "实现特性:"
-    echo "  - 使用 reed-solomon-simd 高性能 SIMD 实现"
-    echo "  - 支持编码器/解码器实例缓存"
-    echo "  - 优化的内存管理和线程安全"
-    echo "  - 跨平台 SIMD 指令支持"
+    echo "Features:"
+    echo "  - Uses the high-performance reed-solomon-simd implementation"
+    echo "  - Caches encoder/decoder instances"
+    echo "  - Optimized memory management and thread safety"
+    echo "  - Cross-platform SIMD instruction support"
 }
 
-# 显示测试配置信息
+# Show benchmark configuration
 show_test_info() {
-    print_info "📋 测试配置信息:"
-    echo "  - 当前目录: $(pwd)"
-    echo "  - Rust 版本: $(rustc --version)"
-    echo "  - Cargo 版本: $(cargo --version)"
-    echo "  - CPU 架构: $(uname -m)"
-    echo "  - 操作系统: $(uname -s)"
+    print_info "📋 Benchmark configuration:"
+    echo "  - Working directory: $(pwd)"
+    echo "  - Rust version: $(rustc --version)"
+    echo "  - Cargo version: $(cargo --version)"
+    echo "  - CPU architecture: $(uname -m)"
+    echo "  - Operating system: $(uname -s)"
     
-    # 检查 CPU 特性
+    # Inspect CPU capabilities
     if [ -f "/proc/cpuinfo" ]; then
-        echo "  - CPU 型号: $(grep 'model name' /proc/cpuinfo | head -1 | cut -d: -f2 | xargs)"
+        echo "  - CPU model: $(grep 'model name' /proc/cpuinfo | head -1 | cut -d: -f2 | xargs)"
         if grep -q "avx2" /proc/cpuinfo; then
-            echo "  - SIMD 支持: AVX2 ✅ (将使用高级 SIMD 优化)"
+            echo "  - SIMD support: AVX2 ✅ (using advanced SIMD optimizations)"
         elif grep -q "sse4" /proc/cpuinfo; then
-            echo "  - SIMD 支持: SSE4 ✅ (将使用 SIMD 优化)"
+            echo "  - SIMD support: SSE4 ✅ (using SIMD optimizations)"
         else
-            echo "  - SIMD 支持: 基础 SIMD 特性"
+            echo "  - SIMD support: baseline features"
         fi
     fi
     
-    echo "  - 实现: reed-solomon-simd (高性能 SIMD 优化)"
-    echo "  - 特性: 实例缓存、线程安全、跨平台 SIMD"
+    echo "  - Implementation: reed-solomon-simd (SIMD-optimized)"
+    echo "  - Highlights: instance caching, thread safety, cross-platform SIMD"
     echo ""
 }
 
-# 主函数
+# Main entry point
 main() {
-    print_info "🧪 Reed-Solomon SIMD 实现性能基准测试"
+    print_info "🧪 Reed-Solomon SIMD benchmark suite"
     echo "================================================"
     
     check_requirements
@@ -252,15 +252,15 @@ main() {
             show_help
             ;;
         *)
-            print_error "未知命令: $1"
+            print_error "Unknown command: $1"
             echo ""
             show_help
             exit 1
             ;;
     esac
     
-    print_success "✨ 基准测试执行完成!"
+    print_success "✨ Benchmark run completed!"
 }
 
-# 启动脚本
+# Launch script
 main "$@" 

crates/ecstore/src/admin_server_info.rs

@@ -34,9 +34,10 @@ use rustfs_protos::{
 };
 use std::{
     collections::{HashMap, HashSet},
-    time::SystemTime,
+    time::{Duration, SystemTime},
 };
 use time::OffsetDateTime;
+use tokio::time::timeout;
 use tonic::Request;
 use tracing::warn;
 
@@ -44,6 +45,8 @@ use shadow_rs::shadow;
 
 shadow!(build);
 
+const SERVER_PING_TIMEOUT: Duration = Duration::from_secs(1);
+
 // pub const ITEM_OFFLINE: &str = "offline";
 // pub const ITEM_INITIALIZING: &str = "initializing";
 // pub const ITEM_ONLINE: &str = "online";
@@ -83,42 +86,45 @@ async fn is_server_resolvable(endpoint: &Endpoint) -> Result<()> {
         endpoint.url.host_str().unwrap(),
         endpoint.url.port().unwrap()
     );
-    let mut fbb = flatbuffers::FlatBufferBuilder::new();
-    let payload = fbb.create_vector(b"hello world");
 
-    let mut builder = PingBodyBuilder::new(&mut fbb);
-    builder.add_payload(payload);
-    let root = builder.finish();
-    fbb.finish(root, None);
+    let ping_task = async {
+        let mut fbb = flatbuffers::FlatBufferBuilder::new();
+        let payload = fbb.create_vector(b"hello world");
 
-    let finished_data = fbb.finished_data();
+        let mut builder = PingBodyBuilder::new(&mut fbb);
+        builder.add_payload(payload);
+        let root = builder.finish();
+        fbb.finish(root, None);
 
-    let decoded_payload = flatbuffers::root::<PingBody>(finished_data);
-    assert!(decoded_payload.is_ok());
+        let finished_data = fbb.finished_data();
 
-    // 创建客户端
-    let mut client = node_service_time_out_client(&addr)
+        let decoded_payload = flatbuffers::root::<PingBody>(finished_data);
+        assert!(decoded_payload.is_ok());
+
+        let mut client = node_service_time_out_client(&addr)
+            .await
+            .map_err(|err| Error::other(err.to_string()))?;
+
+        let request = Request::new(PingRequest {
+            version: 1,
+            body: bytes::Bytes::copy_from_slice(finished_data),
+        });
+
+        let response: PingResponse = client.ping(request).await?.into_inner();
+
+        let ping_response_body = flatbuffers::root::<PingBody>(&response.body);
+        if let Err(e) = ping_response_body {
+            eprintln!("{e}");
+        } else {
+            println!("ping_resp:body(flatbuffer): {ping_response_body:?}");
+        }
+
+        Ok(())
+    };
+
+    timeout(SERVER_PING_TIMEOUT, ping_task)
         .await
-        .map_err(|err| Error::other(err.to_string()))?;
-
-    // 构造 PingRequest
-    let request = Request::new(PingRequest {
-        version: 1,
-        body: bytes::Bytes::copy_from_slice(finished_data),
-    });
-
-    // 发送请求并获取响应
-    let response: PingResponse = client.ping(request).await?.into_inner();
-
-    // 打印响应
-    let ping_response_body = flatbuffers::root::<PingBody>(&response.body);
-    if let Err(e) = ping_response_body {
-        eprintln!("{e}");
-    } else {
-        println!("ping_resp:body(flatbuffer): {ping_response_body:?}");
-    }
-
-    Ok(())
+        .map_err(|_| Error::other("server ping timeout"))?
 }
 
 pub async fn get_local_server_property() -> ServerProperties {

crates/ecstore/src/bucket/bucket_target_sys.rs

@@ -1105,10 +1105,17 @@ impl TargetClient {
             Err(e) => match e {
                 SdkError::ServiceError(oe) => match oe.into_err() {
                     HeadBucketError::NotFound(_) => Ok(false),
-                    other => Err(other.into()),
+                    other => Err(S3ClientError::new(format!(
+                        "failed to check bucket exists for bucket:{bucket} please check the bucket name and credentials, error:{other:?}"
+                    ))),
                 },
+                SdkError::DispatchFailure(e) => Err(S3ClientError::new(format!(
+                    "failed to dispatch bucket exists for bucket:{bucket} error:{e:?}"
+                ))),
 
-                _ => Err(e.into()),
+                _ => Err(S3ClientError::new(format!(
+                    "failed to check bucket exists for bucket:{bucket} error:{e:?}"
+                ))),
             },
         }
     }

crates/ecstore/src/bucket/lifecycle/bucket_lifecycle_ops.rs

@@ -18,7 +18,24 @@
 #![allow(unused_must_use)]
 #![allow(clippy::all)]
 
+use crate::bucket::lifecycle::bucket_lifecycle_audit::{LcAuditEvent, LcEventSrc};
+use crate::bucket::lifecycle::lifecycle::{self, ExpirationOptions, Lifecycle, TransitionOptions};
+use crate::bucket::lifecycle::tier_last_day_stats::{DailyAllTierStats, LastDayTierStats};
+use crate::bucket::lifecycle::tier_sweeper::{Jentry, delete_object_from_remote_tier};
+use crate::bucket::object_lock::objectlock_sys::enforce_retention_for_deletion;
+use crate::bucket::{metadata_sys::get_lifecycle_config, versioning_sys::BucketVersioningSys};
+use crate::client::object_api_utils::new_getobjectreader;
+use crate::error::Error;
 use crate::error::StorageError;
+use crate::error::{error_resp_to_object_err, is_err_object_not_found, is_err_version_not_found, is_network_or_host_down};
+use crate::event::name::EventName;
+use crate::event_notification::{EventArgs, send_event};
+use crate::global::GLOBAL_LocalNodeName;
+use crate::global::{GLOBAL_LifecycleSys, GLOBAL_TierConfigMgr, get_global_deployment_id};
+use crate::store::ECStore;
+use crate::store_api::StorageAPI;
+use crate::store_api::{GetObjectReader, HTTPRangeSpec, ObjectInfo, ObjectOptions, ObjectToDelete};
+use crate::tier::warm_backend::WarmBackendGetOpts;
 use async_channel::{Receiver as A_Receiver, Sender as A_Sender, bounded};
 use bytes::BytesMut;
 use futures::Future;
@@ -27,10 +44,15 @@ use lazy_static::lazy_static;
 use rustfs_common::data_usage::TierStats;
 use rustfs_common::heal_channel::rep_has_active_rules;
 use rustfs_common::metrics::{IlmAction, Metrics};
-use rustfs_filemeta::fileinfo::{NULL_VERSION_ID, RestoreStatusOps, is_restored_object_on_disk};
+use rustfs_filemeta::{NULL_VERSION_ID, RestoreStatusOps, is_restored_object_on_disk};
 use rustfs_utils::path::encode_dir_object;
 use rustfs_utils::string::strings_has_prefix_fold;
 use s3s::Body;
+use s3s::dto::{
+    BucketLifecycleConfiguration, DefaultRetention, ReplicationConfiguration, RestoreRequest, RestoreRequestType, RestoreStatus,
+    ServerSideEncryption, Timestamp,
+};
+use s3s::header::{X_AMZ_RESTORE, X_AMZ_SERVER_SIDE_ENCRYPTION, X_AMZ_STORAGE_CLASS};
 use sha2::{Digest, Sha256};
 use std::any::Any;
 use std::collections::HashMap;
@@ -47,31 +69,6 @@ use tracing::{debug, error, info};
 use uuid::Uuid;
 use xxhash_rust::xxh64;
 
-//use rustfs_notify::{BucketNotificationConfig, Event, EventName, LogLevel, NotificationError, init_logger};
-//use rustfs_notify::{initialize, notification_system};
-use super::bucket_lifecycle_audit::{LcAuditEvent, LcEventSrc};
-use super::lifecycle::{self, ExpirationOptions, Lifecycle, TransitionOptions};
-use super::tier_last_day_stats::{DailyAllTierStats, LastDayTierStats};
-use super::tier_sweeper::{Jentry, delete_object_from_remote_tier};
-use crate::bucket::object_lock::objectlock_sys::enforce_retention_for_deletion;
-use crate::bucket::{metadata_sys::get_lifecycle_config, versioning_sys::BucketVersioningSys};
-use crate::client::object_api_utils::new_getobjectreader;
-use crate::error::Error;
-use crate::error::{error_resp_to_object_err, is_err_object_not_found, is_err_version_not_found, is_network_or_host_down};
-use crate::event::name::EventName;
-use crate::event_notification::{EventArgs, send_event};
-use crate::global::GLOBAL_LocalNodeName;
-use crate::global::{GLOBAL_LifecycleSys, GLOBAL_TierConfigMgr, get_global_deployment_id};
-use crate::store::ECStore;
-use crate::store_api::StorageAPI;
-use crate::store_api::{GetObjectReader, HTTPRangeSpec, ObjectInfo, ObjectOptions, ObjectToDelete};
-use crate::tier::warm_backend::WarmBackendGetOpts;
-use s3s::dto::{
-    BucketLifecycleConfiguration, DefaultRetention, ReplicationConfiguration, RestoreRequest, RestoreRequestType, RestoreStatus,
-    ServerSideEncryption, Timestamp,
-};
-use s3s::header::{X_AMZ_RESTORE, X_AMZ_SERVER_SIDE_ENCRYPTION, X_AMZ_STORAGE_CLASS};
-
 pub type TimeFn = Arc<dyn Fn() -> Pin<Box<dyn Future<Output = ()> + Send>> + Send + Sync + 'static>;
 pub type TraceFn =
     Arc<dyn Fn(String, HashMap<String, String>) -> Pin<Box<dyn Future<Output = ()> + Send>> + Send + Sync + 'static>;
@@ -118,10 +115,9 @@ struct ExpiryTask {
 impl ExpiryOp for ExpiryTask {
     fn op_hash(&self) -> u64 {
         let mut hasher = Sha256::new();
-        let _ = hasher.write(format!("{}", self.obj_info.bucket).as_bytes());
-        let _ = hasher.write(format!("{}", self.obj_info.name).as_bytes());
-        hasher.flush();
-        xxh64::xxh64(hasher.clone().finalize().as_slice(), XXHASH_SEED)
+        hasher.update(format!("{}", self.obj_info.bucket).as_bytes());
+        hasher.update(format!("{}", self.obj_info.name).as_bytes());
+        xxh64::xxh64(hasher.finalize().as_slice(), XXHASH_SEED)
     }
 
     fn as_any(&self) -> &dyn Any {
@@ -174,10 +170,9 @@ struct FreeVersionTask(ObjectInfo);
 impl ExpiryOp for FreeVersionTask {
     fn op_hash(&self) -> u64 {
         let mut hasher = Sha256::new();
-        let _ = hasher.write(format!("{}", self.0.transitioned_object.tier).as_bytes());
-        let _ = hasher.write(format!("{}", self.0.transitioned_object.name).as_bytes());
-        hasher.flush();
-        xxh64::xxh64(hasher.clone().finalize().as_slice(), XXHASH_SEED)
+        hasher.update(format!("{}", self.0.transitioned_object.tier).as_bytes());
+        hasher.update(format!("{}", self.0.transitioned_object.name).as_bytes());
+        xxh64::xxh64(hasher.finalize().as_slice(), XXHASH_SEED)
     }
 
     fn as_any(&self) -> &dyn Any {
@@ -194,10 +189,9 @@ struct NewerNoncurrentTask {
 impl ExpiryOp for NewerNoncurrentTask {
     fn op_hash(&self) -> u64 {
         let mut hasher = Sha256::new();
-        let _ = hasher.write(format!("{}", self.bucket).as_bytes());
-        let _ = hasher.write(format!("{}", self.versions[0].object_name).as_bytes());
-        hasher.flush();
-        xxh64::xxh64(hasher.clone().finalize().as_slice(), XXHASH_SEED)
+        hasher.update(format!("{}", self.bucket).as_bytes());
+        hasher.update(format!("{}", self.versions[0].object_name).as_bytes());
+        xxh64::xxh64(hasher.finalize().as_slice(), XXHASH_SEED)
     }
 
     fn as_any(&self) -> &dyn Any {
@@ -418,10 +412,9 @@ struct TransitionTask {
 impl ExpiryOp for TransitionTask {
     fn op_hash(&self) -> u64 {
         let mut hasher = Sha256::new();
-        let _ = hasher.write(format!("{}", self.obj_info.bucket).as_bytes());
-        //let _ = hasher.write(format!("{}", self.obj_info.versions[0].object_name).as_bytes());
-        hasher.flush();
-        xxh64::xxh64(hasher.clone().finalize().as_slice(), XXHASH_SEED)
+        hasher.update(format!("{}", self.obj_info.bucket).as_bytes());
+        // hasher.update(format!("{}", self.obj_info.versions[0].object_name).as_bytes());
+        xxh64::xxh64(hasher.finalize().as_slice(), XXHASH_SEED)
     }
 
     fn as_any(&self) -> &dyn Any {
@@ -483,7 +476,7 @@ impl TransitionState {
             .and_then(|s| s.parse::<i64>().ok())
             .unwrap_or_else(|| std::cmp::min(num_cpus::get() as i64, 16));
         let mut n = max_workers;
-        let tw = 8; //globalILMConfig.getTransitionWorkers(); 
+        let tw = 8; //globalILMConfig.getTransitionWorkers();
         if tw > 0 {
             n = tw;
         }
@@ -763,9 +756,8 @@ pub async fn expire_transitioned_object(
 pub fn gen_transition_objname(bucket: &str) -> Result<String, Error> {
     let us = Uuid::new_v4().to_string();
     let mut hasher = Sha256::new();
-    let _ = hasher.write(format!("{}/{}", get_global_deployment_id().unwrap_or_default(), bucket).as_bytes());
-    hasher.flush();
-    let hash = rustfs_utils::crypto::hex(hasher.clone().finalize().as_slice());
+    hasher.update(format!("{}/{}", get_global_deployment_id().unwrap_or_default(), bucket).as_bytes());
+    let hash = rustfs_utils::crypto::hex(hasher.finalize().as_slice());
     let obj = format!("{}/{}/{}/{}", &hash[0..16], &us[0..2], &us[2..4], &us);
     Ok(obj)
 }

crates/ecstore/src/bucket/lifecycle/tier_sweeper.rs

@@ -20,7 +20,7 @@
 
 use sha2::{Digest, Sha256};
 use std::any::Any;
-use std::io::{Cursor, Write};
+use std::io::Write;
 use xxhash_rust::xxh64;
 
 use super::bucket_lifecycle_ops::{ExpiryOp, GLOBAL_ExpiryState, TransitionedObject};
@@ -128,10 +128,9 @@ pub struct Jentry {
 impl ExpiryOp for Jentry {
     fn op_hash(&self) -> u64 {
         let mut hasher = Sha256::new();
-        let _ = hasher.write(format!("{}", self.tier_name).as_bytes());
-        let _ = hasher.write(format!("{}", self.obj_name).as_bytes());
-        hasher.flush();
-        xxh64::xxh64(hasher.clone().finalize().as_slice(), XXHASH_SEED)
+        hasher.update(format!("{}", self.tier_name).as_bytes());
+        hasher.update(format!("{}", self.obj_name).as_bytes());
+        xxh64::xxh64(hasher.finalize().as_slice(), XXHASH_SEED)
     }
 
     fn as_any(&self) -> &dyn Any {

crates/ecstore/src/bucket/metadata.rs

@@ -428,8 +428,8 @@ where
     let sec = t.unix_timestamp() - 62135596800;
     let nsec = t.nanosecond();
     buf[0] = 0xc7; // mext8
-    buf[1] = 0x0c; // 长度
-    buf[2] = 0x05; // 时间扩展类型
+    buf[1] = 0x0c; // Length
+    buf[2] = 0x05; // Time extension type
     BigEndian::write_u64(&mut buf[3..], sec as u64);
     BigEndian::write_u32(&mut buf[11..], nsec);
     s.serialize_bytes(&buf)

crates/ecstore/src/bucket/policy_sys.rs

@@ -12,10 +12,10 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use super::{error::BucketMetadataError, metadata_sys::get_bucket_metadata_sys};
-use crate::error::Result;
+use super::metadata_sys::get_bucket_metadata_sys;
+use crate::error::{Result, StorageError};
 use rustfs_policy::policy::{BucketPolicy, BucketPolicyArgs};
-use tracing::warn;
+use tracing::info;
 
 pub struct PolicySys {}
 
@@ -24,9 +24,8 @@ impl PolicySys {
         match Self::get(args.bucket).await {
             Ok(cfg) => return cfg.is_allowed(args),
             Err(err) => {
-                let berr: BucketMetadataError = err.into();
-                if berr != BucketMetadataError::BucketPolicyNotFound {
-                    warn!("config get err {:?}", berr);
+                if err != StorageError::ConfigNotFound {
+                    info!("config get err {:?}", err);
                 }
             }
         }

crates/ecstore/src/bucket/quota/mod.rs

@@ -16,16 +16,16 @@ use crate::error::Result;
 use rmp_serde::Serializer as rmpSerializer;
 use serde::{Deserialize, Serialize};
 
-// 定义 QuotaType 枚举类型
+// Define the QuotaType enum
 #[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
 pub enum QuotaType {
     Hard,
 }
 
-// 定义 BucketQuota 结构体
+// Define the BucketQuota structure
 #[derive(Debug, Deserialize, Serialize, Default, Clone)]
 pub struct BucketQuota {
-    quota: Option<u64>, // 使用 Option 来表示可能不存在的字段
+    quota: Option<u64>, // Use Option to represent optional fields
 
     size: u64,
 

crates/ecstore/src/bucket/replication/config.rs

@@ -46,7 +46,7 @@ pub trait ReplicationConfigurationExt {
 }
 
 impl ReplicationConfigurationExt for ReplicationConfiguration {
-    /// 检查是否有现有对象复制规则
+    /// Check whether any object-replication rules exist
     fn has_existing_object_replication(&self, arn: &str) -> (bool, bool) {
         let mut has_arn = false;
 
@@ -103,6 +103,8 @@ impl ReplicationConfigurationExt for ReplicationConfiguration {
                 if filter.test_tags(&object_tags) {
                     rules.push(rule.clone());
                 }
+            } else {
+                rules.push(rule.clone());
             }
         }
 
@@ -117,7 +119,7 @@ impl ReplicationConfigurationExt for ReplicationConfiguration {
         rules
     }
 
-    /// 获取目标配置
+    /// Retrieve the destination configuration
     fn get_destination(&self) -> Destination {
         if !self.rules.is_empty() {
             self.rules[0].destination.clone()
@@ -134,7 +136,7 @@ impl ReplicationConfigurationExt for ReplicationConfiguration {
         }
     }
 
-    /// 判断对象是否应该被复制
+    /// Determine whether an object should be replicated
     fn replicate(&self, obj: &ObjectOpts) -> bool {
         let rules = self.filter_actionable_rules(obj);
 
@@ -164,16 +166,16 @@ impl ReplicationConfigurationExt for ReplicationConfiguration {
                 }
             }
 
-            // 常规对象/元数据复制
+            // Regular object/metadata replication
             return rule.metadata_replicate(obj);
         }
         false
     }
 
-    /// 检查是否有活跃的规则
-    /// 可选择性地提供前缀
-    /// 如果recursive为true，函数还会在前缀下的任何级别有活跃规则时返回true
-    /// 如果没有指定前缀，recursive实际上为true
+    /// Check for an active rule
+    /// Optionally accept a prefix
+    /// When recursive is true, return true if any level under the prefix has an active rule
+    /// Without a prefix, recursive behaves as true
     fn has_active_rules(&self, prefix: &str, recursive: bool) -> bool {
         if self.rules.is_empty() {
             return false;
@@ -187,13 +189,13 @@ impl ReplicationConfigurationExt for ReplicationConfiguration {
             if let Some(filter) = &rule.filter {
                 if let Some(filter_prefix) = &filter.prefix {
                     if !prefix.is_empty() && !filter_prefix.is_empty() {
-                        // 传入的前缀必须在规则前缀中
+                        // The provided prefix must fall within the rule prefix
                         if !recursive && !prefix.starts_with(filter_prefix) {
                             continue;
                         }
                     }
 
-                    // 如果是递归的，我们可以跳过这个规则，如果它不匹配测试前缀或前缀下的级别不匹配
+                    // When recursive, skip this rule if it does not match the test prefix or hierarchy
                     if recursive && !rule.prefix().starts_with(prefix) && !prefix.starts_with(rule.prefix()) {
                         continue;
                     }
@@ -204,7 +206,7 @@ impl ReplicationConfigurationExt for ReplicationConfiguration {
         false
     }
 
-    /// 过滤目标ARN，返回配置中不同目标ARN的切片
+    /// Filter target ARNs and return a slice of the distinct values in the config
     fn filter_target_arns(&self, obj: &ObjectOpts) -> Vec<String> {
         let mut arns = Vec::new();
         let mut targets_map: HashSet<String> = HashSet::new();
@@ -216,7 +218,7 @@ impl ReplicationConfigurationExt for ReplicationConfiguration {
             }
 
             if !self.role.is_empty() {
-                arns.push(self.role.clone()); // 如果存在，使用传统的RoleArn
+                arns.push(self.role.clone()); // Use the legacy RoleArn when present
                 return arns;
             }