Merge the hotpath-rs wall-time instrumentation from the backlog#936 analysis worktree behind an opt-in 'hotpath' cargo feature, keeping the default build at zero overhead and zero dependency.
- hotpath is an optional dependency everywhere (dep:hotpath feature syntax); the default dependency tree contains no hotpath crate at all
- 40+ measurement points across S3 handlers, ECStore/SetDisks object and multipart ops, erasure encode/decode, bitrot, LocalDisk I/O, FileMeta codec, and HashReader
- attribute sites use #[cfg_attr(feature = "hotpath", hotpath::measure)]; async_trait bodies use per-crate hp_guard! macros (ecstore + rustfs bin); rio gates measure_block! behind hp_measure_block!
- feature chain: rustfs -> rustfs-ecstore -> rustfs-rio / rustfs-filemeta, each crate owning its own gate
- hotpath-alloc is intentionally not wired up (hotpath 0.21.x TLS panic on cross-thread guard drop under tokio, see backlog#935); mimalloc stays the unconditional global allocator
- docs/development/hotpath-profiling.md documents building, HOTPATH_* env vars, SIGTERM report flow, and how to reproduce the backlog#936 timing reports
Refs: https://github.com/rustfs/backlog/issues/935 (HP-14, item 2), https://github.com/rustfs/backlog/issues/936
Co-authored-by: heihutu <heihutu@gmail.com>
* fix(object-capacity): stop cancelled/remote-disk refreshes from corrupting capacity
Two independent capacity-refresh bugs (backlog rustfs/backlog#805):
- refresh_or_join set the singleflight `running` flag then awaited the
refresh future with no drop guard. When the admin request that became
leader was cancelled (client disconnect) mid-await, `running` stayed true
forever: joiners blocked indefinitely and the 120s scheduled refresh could
never start again. catch_unwind covered panics but not cancellation. A
RefreshLeaderGuard now resets the state and publishes an error on drop.
- capacity_disk_refs mapped the cluster-wide storage_info disk list without
filtering non-local disks, so admin-triggered refreshes ran a local WalkDir
over remote disks' drive_path. On multi-node clusters this double-counted
local bytes (shared mount layout) or hit NotFound (per-node layouts), and
poisoned the per-disk cache the scheduled local-only refresh depends on,
making the cached total oscillate. Filter to local disks.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* fix(rio): harden internode HTTP client build, cache, and PUT body integrity
Follow-ups from the internode HTTP review (backlog rustfs/backlog#805):
- handle_put_file accepted a truncated body as success: a HttpWriter dropped
mid-stream closes the chunked body cleanly, indistinguishable from EOF, and
the server never compared bytes copied against the declared size. Reject
size mismatches on the create path (append/unknown-size writes send size<=0
and are exempt).
- build_http_client used `.expect()` on ClientBuilder::build(), which runs
lazily on the first request and on every TLS generation bump (cert
rotation), so a build failure panicked a serving task. It now returns an
io::Error; get_http_client falls back to the previous TLS generation when a
rebuild fails instead of failing the request.
- CLIENT_CACHE was a tokio::Mutex taken on every stream open (data_shards
times per GET). Replaced with arc_swap::ArcSwapOption for lock-free reads on
the hot path; the generation-monotonic replacement guard is preserved via rcu.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
* chore(deps): refresh workspace deps and linux fs_type gating
- refresh workspace dependency pins and lockfile updates
- remove now-unused crate dependency entries in multiple Cargo.toml files
- enable profiling export defaults in config and scripts/run.sh
- gate os::fs_type module/function/tests to Linux to avoid non-Linux dead_code warnings
* fix(utils): simplify fs_type linux gating
- keep fs_type module-level linux cfg in os::mod
- remove redundant linux cfg on get_fs_type and test module
* chore(deps): bump s3s git revision
- update workspace s3s dependency to rev 507e1312b211c3ddc214b03875d6fabd15d22ed5
- refresh Cargo.lock source entry for s3s
* chore(dev): allow mysql_async git source and env overrides
- allow mysql_async git source in deny.toml allow-git list
- make scripts/run.sh core env vars overrideable via existing shell env
* fix(utils): import get_fs_type in fs_type tests
- add explicit super::get_fs_type import in fs_type test module
- fix Linux E0425 unresolved function errors in unit tests
* chore(dev): tune run script observability defaults
- make profiling export env overrideable in scripts/run.sh
- set RUSTFS_OBS_SAMPLE_RATIO default from 2.0 to 1.0
- update allow-git review window comments in deny.toml
* test(obs): stabilize profiling env alias tests
* fix
* chore: upgrade cryptography libraries to RC versions
- Upgrade aes-gcm to 0.11.0-rc.2 with rand_core support
- Upgrade chacha20poly1305 to 0.11.0-rc.2
- Upgrade argon2 to 0.6.0-rc.2 with std features
- Upgrade hmac to 0.13.0-rc.3
- Upgrade pbkdf2 to 0.13.0-rc.2
- Upgrade rsa to 0.10.0-rc.10
- Upgrade sha1 and sha2 to 0.11.0-rc.3
- Upgrade md-5 to 0.11.0-rc.3
These upgrades provide enhanced security features and performance
improvements while maintaining backward compatibility with existing
encryption workflows.
* add
* improve code
* fix
* feat(kms): implement key management service with local and vault backends
Signed-off-by: junxiang Mu <1948535941@qq.com>
* feat(kms): enhance security with zeroize for sensitive data and improve key management
Signed-off-by: junxiang Mu <1948535941@qq.com>
* remove Hashi word
Signed-off-by: junxiang Mu <1948535941@qq.com>
* refactor: remove unused request structs from kms handlers
Signed-off-by: junxiang Mu <1948535941@qq.com>
---------
Signed-off-by: junxiang Mu <1948535941@qq.com>
* init audit logger module
* add audit webhook default config kvs
* feat: Add comprehensive tests for authentication module (#309)
* feat: add comprehensive tests for authentication module
- Add 33 unit tests covering all public functions in auth.rs
- Test IAMAuth struct creation and secret key validation
- Test check_claims_from_token with various credential types and scenarios
- Test session token extraction from headers and query parameters
- Test condition values generation for different user types
- Test query parameter parsing with edge cases
- Test Credentials helper methods (is_expired, is_temp, is_service_account)
- Ensure tests handle global state dependencies gracefully
- All tests pass successfully with 100% coverage of testable functions
* style: fix code formatting issues
* Add verification script for checking PR branch statuses and tests
Co-authored-by: anzhengchao <anzhengchao@gmail.com>
* fix: resolve clippy uninlined format args warning
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
* feat: add basic tests for core storage module (#313)
* feat: add basic tests for core storage module
- Add 6 unit tests for FS struct and basic functionality
- Test FS creation, Debug and Clone trait implementations
- Test RUSTFS_OWNER constant definition and values
- Test S3 error code creation and handling
- Test compression format detection for common file types
- Include comprehensive documentation about integration test needs
Note: Full S3 API testing requires complex setup with storage backend,
global configuration, and network infrastructure - better suited for
integration tests rather than unit tests.
* style: fix code formatting issues
* fix: resolve clippy warnings in storage tests
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
* feat: add tests for admin handlers module (#314)
* feat: add tests for admin handlers module
- Add 5 new unit tests for admin handler functionality
- Test AccountInfo struct creation, serialization and default values
- Test creation of all admin handler structs (13 handlers)
- Test HealOpts JSON serialization and deserialization
- Test HealOpts URL encoding/decoding with proper field types
- Maintain existing test while adding comprehensive coverage
- Include documentation about integration test requirements
All tests pass successfully with proper error handling for complex dependencies.
* style: fix code formatting issues
* fix: resolve clippy warnings in admin handlers tests
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
* build(deps): bump the dependencies group with 3 updates (#326)
* perf: avoid transmitting parity shards when the object is good (#322)
* upgrade version
* Fix: fix data integrity check
Signed-off-by: junxiang Mu <1948535941@qq.com>
* Fix: Separate Clippy's fix and check commands into two commands.
Signed-off-by: junxiang Mu <1948535941@qq.com>
* fix: miss inline metadata (#345)
* Update dependabot.yml
* fix: Fixed an issue where the list_objects_v2 API did not return dire… (#352)
* fix: Fixed an issue where the list_objects_v2 API did not return directory names when they conflicted with file names in the same bucket (e.g., test/ vs. test.txt, aaa/ vs. aaa.csv) (#335)
* fix: adjusted the order of directory listings
* init
* fix
* fix
* feat: add docker usage for rustfs mcp (#365)
* feat: enhance metadata extraction with object name for MIME type detection
Signed-off-by: junxiang Mu <1948535941@qq.com>
* Feature: lock support auto release
Signed-off-by: junxiang Mu <1948535941@qq.com>
* improve lock
Signed-off-by: junxiang Mu <1948535941@qq.com>
* Fix: fix scanner detect
Signed-off-by: junxiang Mu <1948535941@qq.com>
* Fix: clippy && fmt
Signed-off-by: junxiang Mu <1948535941@qq.com>
* refactor(ecstore): Optimize memory usage for object integrity verification
Change the object integrity verification from reading all data to streaming processing to avoid memory overflow caused by large objects.
Modify the TLS key log check to use environment variables directly instead of configuration constants.
Add memory limits for object data reading in the AHM module.
Signed-off-by: junxiang Mu <1948535941@qq.com>
* Chore: reduce PR template checklist
Signed-off-by: junxiang Mu <1948535941@qq.com>
* Chore: remove comment code (#376)
Signed-off-by: junxiang Mu <1948535941@qq.com>
* chore: upgrade actions/checkout from v4 to v5 (#381)
* chore: upgrade actions/checkout from v4 to v5
- Update GitHub Actions checkout action version
- Ensure compatibility with latest workflow features
- Maintain existing checkout behavior and configuration
* upgrade version
* fix
* add and improve code for notify
* feat: extend rustfs mcp with bucket creation and deletion (#416)
* feat: extend rustfs mcp with bucket creation and deletion
* update file to fix pipeline error
* change variable name to fix pipeline error
* fix(ecstore): add async-recursion to resolve nightly trait solver reg… (#415)
* fix(ecstore): add async-recursion to resolve nightly trait solver regression
The newest nightly compiler switched to the new trait solver, which
currently rejects async recursive functions that were previously accepted.
This causes the following compilation failures:
- `LocalDisk::delete_file()`
- `LocalDisk::scan_dir()`
Add `async-recursion` as a workspace dependency and annotate both functions with `#[async_recursion]` so that the crate compiles cleanly with the latest nightly and will continue to build once the new solver lands in stable.
Signed-off-by: reigadegr <2722688642@qq.com>
* fix: resolve duplicate bound error in scan_dir function
Replaced inline trait bounds with where clause to avoid duplication caused by macro expansion.
Signed-off-by: reigadegr <2722688642@qq.com>
---------
Signed-off-by: reigadegr <2722688642@qq.com>
Co-authored-by: 安正超 <anzhengchao@gmail.com>
* fix:make bucket exists (#428)
* feat: include user-defined metadata in S3 response (#431)
* fix: simplify Docker entrypoint following efficient user switching pattern (#421)
* fix: simplify Docker entrypoint following efficient user switching pattern
- Remove ALL file permission modifications (no chown at all)
- Use chroot --userspec or gosu to switch user context
- Extremely simple and fast implementation
- Zero filesystem modifications for permissions
Fixes#388
* Update entrypoint.sh
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Update entrypoint.sh
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Update entrypoint.sh
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* wip
* wip
* wip
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* docs: update doc/docker-data-dir README.md (#432)
* add targets crates
* feat(targets): extract targets module into a standalone crate
- Move all target-related code (MQTT, Webhook, etc.) into a new `targets` crate
- Update imports and dependencies to reference the new crate
- Refactor interfaces to ensure compatibility with the new crate structure
- Adjust Cargo.toml and workspace configuration accordingly
* fix
* fix
---------
Signed-off-by: junxiang Mu <1948535941@qq.com>
Signed-off-by: reigadegr <2722688642@qq.com>
Co-authored-by: 安正超 <anzhengchao@gmail.com>
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: zzhpro <56196563+zzhpro@users.noreply.github.com>
Co-authored-by: junxiang Mu <1948535941@qq.com>
Co-authored-by: weisd <im@weisd.in>
Co-authored-by: shiro.lee <69624924+shiroleeee@users.noreply.github.com>
Co-authored-by: majinghe <42570491+majinghe@users.noreply.github.com>
Co-authored-by: guojidan <63799833+guojidan@users.noreply.github.com>
Co-authored-by: reigadegr <103645642+reigadegr@users.noreply.github.com>
Co-authored-by: 0xdx2 <xuedamon2@gmail.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* fix: add tokio-test
* fix: "called `unwrap` on `v` after checking its variant with `is_some`"
= help: try using `if let` or `match`
= help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#unnecessary_unwrap
= note: `-D clippy::unnecessary-unwrap` implied by `-D warnings`
= help: to override `-D warnings` add `#[allow(clippy::unnecessary_unwrap)]`
* fmt
* set toolchain 1.88.0
* fmt
* fix: cliip
Move the construction of the hybrid service stack, including all middleware and the RPC service, from the main `run` function into the `process_connection` function.
This change ensures that each incoming connection gets its own isolated service instance. This improves modularity by making the connection handling logic more self-contained and simplifies the main server loop.
Key changes:
- The `hybrid_service` and `rpc_service` are now created inside `process_connection`.
- The `run` function's responsibility is reduced to accepting TCP connections and spawning tasks for `process_connection`.
This commit introduces a significant reorganization of the project structure to improve maintainability and clarity.
Key changes include:
- Adjusted the directory layout for a more logical module organization.
- Removed unused crate dependencies, reducing the overall project size and potentially speeding up build times.
- Updated import paths and configuration files to reflect the structural changes.
* chore: Add copyright and license headers
This commit adds the Apache 2.0 license and a copyright notice to the header of all source files. This ensures that the licensing and copyright information is clearly stated within the codebase.
* cargo fmt
* fix
* fmt
* fix clippy
houseme