test(e2e): activate conditional write regressions

.config/e2e-full-selection.txt

@@ -1,2 +1,2 @@
-sha256-darwin=9f767b37ed8b1c82da62ea441462d75487785c8086e56f08fb6f6cd89c6e2e52
-sha256-linux=fbdaf42b220958d4b1e8880e0f8b5a7992d38e21051bb60596dd4538424757d6
+sha256-darwin=03bdfb6a9d6e25d744c385f1461e651f05e4da78e5b0ead2adb3e8b2463e3834
+sha256-linux=78c46adad135231017fb8679fb91877b5ae3ec6ce1f75d3d045d2c1076c12a49

.config/nextest.toml

@@ -78,12 +78,6 @@ test-group = 'embedded-test-ports'
 filter = 'package(rustfs-ecstore) & test(manual_transition_page_checkpoint_persists_durable_job_progress)'
 test-group = 'ecstore-serial-flaky'
 
-# The durable ILM decommission regressions build isolated multi-pool stores and
-# deliberately take source or target disks offline while checking fencing.
-[[profile.default.overrides]]
-filter = 'package(rustfs-ecstore) & (test(decommission_migrates_and_verifies_registered_durable_ilm_records) | test(decommission_durable_ilm_target_read_error_is_not_masked_by_peer_success) | test(decommission_durable_ilm_terminal_receipt_recovers_failed_source_cleanup) | test(decommission_durable_ilm_receipt_pagination_fails_closed_on_second_page) | test(decommission_durable_ilm_recovery_keeps_multiple_active_sources))'
-test-group = 'ecstore-serial-flaky'
-
 # Serialize the bucket-incarnation / lifecycle-fence tests. They drive
 # init_bucket_metadata_sys and bucket_metadata_sys_of, i.e. process-global
 # OnceLock state that serial_test's #[serial] cannot protect across nextest's
@@ -196,10 +190,6 @@ test-group = 'embedded-test-ports'
 filter = 'package(rustfs-ecstore) & test(manual_transition_page_checkpoint_persists_durable_job_progress)'
 test-group = 'ecstore-serial-flaky'
 
-[[profile.ci.overrides]]
-filter = 'package(rustfs-ecstore) & (test(decommission_migrates_and_verifies_registered_durable_ilm_records) | test(decommission_durable_ilm_target_read_error_is_not_masked_by_peer_success) | test(decommission_durable_ilm_terminal_receipt_recovers_failed_source_cleanup) | test(decommission_durable_ilm_receipt_pagination_fails_closed_on_second_page) | test(decommission_durable_ilm_recovery_keeps_multiple_active_sources))'
-test-group = 'ecstore-serial-flaky'
-
 # Serialize the bucket-incarnation / lifecycle-fence tests under the ci profile
 # too (see the matching default-profile override near the top). No retries.
 [[profile.ci.overrides]]

CLAUDE.md

@@ -30,8 +30,7 @@ make build-docker BUILD_OS=ubuntu22.04
 - Crate membership: `Cargo.toml` `[workspace].members`
 - Architecture, layering, crate map: [ARCHITECTURE.md](ARCHITECTURE.md)
 - Migration guardrails & readiness contracts: [docs/architecture/](docs/architecture/README.md)
-- CI workflow steps: `.github/workflows/`; event, timeout, and required-status
-  matrix: [docs/testing/ci-gates.md](docs/testing/ci-gates.md)
+- CI gates: `.github/workflows/ci.yml` (source of truth; never copy its steps into docs)
 - Test-layer taxonomy, per-layer entry commands, serial/nextest rules, flake
   policy: [docs/testing/README.md](docs/testing/README.md)
 - Tier/ILM transition debugging (xl.meta inspection, versionId tracing):

CONTRIBUTING.md

@@ -70,8 +70,6 @@ make pre-pr
 
 > For the full test-layer taxonomy (unit / ecstore black-box / e2e / s3s-e2e / S3 compatibility / chaos / fuzz / bench), each layer's entry command, the naming conventions the migration gate depends on, and the serial/nextest rules, see [docs/testing/README.md](docs/testing/README.md).
 
-> For the event, timeout, required-status, and local reproduction matrix, see [docs/testing/ci-gates.md](docs/testing/ci-gates.md).
-
 ### 🔒 Automated Pre-commit Hooks
 #### What `make pre-commit` and `make pre-pr` actually run
 

Cargo.lock

@@ -1858,9 +1858,9 @@ dependencies = [
 
 [[package]]
 name = "cc"
-version = "1.4.4"
+version = "1.4.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0ad534f4357a5264cce5019c989cf66a4f0dc4e0d1b1d15f8aacec0ff7360273"
+checksum = "509591b7bcd67f4ef775afad7662703b4935daaa6ec0e5605cfb1090b32a2b6d"
 dependencies = [
  "find-msvc-tools",
  "jobserver",
@@ -2522,6 +2522,12 @@ dependencies = [
  "subtle",
 ]
 
+[[package]]
+name = "cty"
+version = "0.2.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b365fabc795046672053e29c954733ec3b05e4be654ab130fe8f1f94d7051f35"
+
 [[package]]
 name = "curve25519-dalek"
 version = "4.1.3"
@@ -5982,6 +5988,15 @@ version = "0.2.16"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "b6d2cec3eae94f9f509c767b45932f1ada8350c4bdb85af2fcab4a3c14807981"
 
+[[package]]
+name = "libmimalloc-sys"
+version = "0.1.49"
+source = "git+https://github.com/xonatius/mimalloc_rust.git?rev=6d4c41bb10c6d9da1d1b6f07b38c4cc051667f11#6d4c41bb10c6d9da1d1b6f07b38c4cc051667f11"
+dependencies = [
+ "cc",
+ "cty",
+]
+
 [[package]]
 name = "libredox"
 version = "0.1.20"
@@ -6382,6 +6397,14 @@ dependencies = [
  "synstructure 0.13.2",
 ]
 
+[[package]]
+name = "mimalloc"
+version = "0.1.52"
+source = "git+https://github.com/xonatius/mimalloc_rust.git?rev=6d4c41bb10c6d9da1d1b6f07b38c4cc051667f11#6d4c41bb10c6d9da1d1b6f07b38c4cc051667f11"
+dependencies = [
+ "libmimalloc-sys",
+]
+
 [[package]]
 name = "mime"
 version = "0.3.17"
@@ -9139,11 +9162,13 @@ dependencies = [
  "insta",
  "jiff",
  "libc",
+ "libmimalloc-sys",
  "libsystemd",
  "matchit 0.9.2",
  "md-5 0.11.0",
  "metrics",
  "metrics-util",
+ "mimalloc",
  "mime_guess",
  "opentelemetry",
  "opentelemetry_sdk",
@@ -9179,8 +9204,6 @@ dependencies = [
  "rustfs-lock",
  "rustfs-log-analyzer",
  "rustfs-madmin",
- "rustfs-mimalloc",
- "rustfs-mimalloc-sys",
  "rustfs-notify",
  "rustfs-object-capacity",
  "rustfs-object-data-cache",
@@ -9852,24 +9875,6 @@ dependencies = [
  "tokio",
 ]
 
-[[package]]
-name = "rustfs-mimalloc"
-version = "0.5.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a406f4aa07084301d485beec873af6dccc8e3f8762da244743df92038b1db1a6"
-dependencies = [
- "rustfs-mimalloc-sys",
-]
-
-[[package]]
-name = "rustfs-mimalloc-sys"
-version = "0.5.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c3051b819175f58445d4c369a72f0ab88149f3885ba8bea2aff3be01f53fe7cd"
-dependencies = [
- "cc",
-]
-
 [[package]]
 name = "rustfs-notify"
 version = "1.0.0-rc.3"

Cargo.toml

@@ -350,8 +350,8 @@ russh-sftp = "2.4.0"
 dav-server = "0.11.0"
 
 # Performance Analysis and Memory Profiling
-rustfs-mimalloc = { version = "0.5.0" }
-rustfs-mimalloc-sys = { version = "0.5.0" }
+mimalloc = { version = "0.1.52", git = "https://github.com/xonatius/mimalloc_rust.git", rev = "6d4c41bb10c6d9da1d1b6f07b38c4cc051667f11" }
+libmimalloc-sys = { version = "0.1.49", git = "https://github.com/xonatius/mimalloc_rust.git", rev = "6d4c41bb10c6d9da1d1b6f07b38c4cc051667f11", features = ["extended"] }
 hotpath = { version = "0.23.3", default-features = false }
 # Snapshot testing for output format regression detection
 insta = { version = "1.48" }

crates/e2e_test/src/reliant/conditional_writes.rs

@@ -1,52 +1,26 @@
 #![cfg(test)]
 
-use aws_config::meta::region::RegionProviderChain;
+use crate::common::{RustFSTestEnvironment, TEST_BUCKET, init_logging};
 use aws_sdk_s3::Client;
-use aws_sdk_s3::config::{Credentials, Region};
-use aws_sdk_s3::error::SdkError;
+use aws_sdk_s3::error::{ProvideErrorMetadata, SdkError};
 use aws_sdk_s3::types::{CompletedMultipartUpload, CompletedPart};
 use bytes::Bytes;
 use std::error::Error;
+use std::fmt::Debug;
 
-const ENDPOINT: &str = "http://localhost:9000";
-const ACCESS_KEY: &str = "rustfsadmin";
-const SECRET_KEY: &str = "rustfsadmin";
-const BUCKET: &str = "api-test";
+type TestResult = Result<(), Box<dyn Error + Send + Sync>>;
 
-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(),
+fn assert_s3_error_code<T, E>(result: Result<T, SdkError<E>>, expected: &str)
+where
+    T: Debug,
+    E: ProvideErrorMetadata + Debug,
+{
+    let error = result.expect_err("conditional request must fail");
+    assert_eq!(
+        error.as_service_error().and_then(ProvideErrorMetadata::code),
+        Some(expected),
+        "unexpected conditional request error: {error:?}"
     );
-    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") {
-                return Err(e.into());
-            }
-        }
-        Err(e) => {
-            return Err(e.into());
-        }
-    }
-    Ok(())
 }
 
 /// Generate test data of specified size
@@ -60,7 +34,12 @@ fn generate_test_data(size: usize) -> Vec<u8> {
 }
 
 /// Upload an object and return its ETag
-async fn upload_object_with_metadata(client: &Client, bucket: &str, key: &str, data: &[u8]) -> Result<String, Box<dyn Error>> {
+async fn upload_object_with_metadata(
+    client: &Client,
+    bucket: &str,
+    key: &str,
+    data: &[u8],
+) -> Result<String, Box<dyn Error + Send + Sync>> {
     let response = client
         .put_object()
         .bucket(bucket)
@@ -69,188 +48,164 @@ async fn upload_object_with_metadata(client: &Client, bucket: &str, key: &str, d
         .send()
         .await?;
 
-    let etag = response.e_tag().unwrap_or("").to_string();
-    Ok(etag)
+    response
+        .e_tag()
+        .map(str::to_owned)
+        .ok_or_else(|| std::io::Error::other("put object response did not include an ETag").into())
 }
 
-/// Cleanup test objects from bucket
-async fn cleanup_objects(client: &Client, bucket: &str, keys: &[&str]) {
-    for key in keys {
-        let _ = client.delete_object().bucket(bucket).key(*key).send().await;
-    }
-}
-
-/// Generate unique test object key
-fn generate_test_key(prefix: &str) -> String {
-    use std::time::{SystemTime, UNIX_EPOCH};
-    let timestamp = SystemTime::now().duration_since(UNIX_EPOCH).unwrap().as_nanos();
-    format!("{prefix}-{timestamp}")
+async fn object_body(client: &Client, key: &str) -> Result<Bytes, Box<dyn Error + Send + Sync>> {
+    let response = client.get_object().bucket(TEST_BUCKET).key(key).send().await?;
+    Ok(response.body.collect().await?.into_bytes())
 }
 
 #[tokio::test]
-#[ignore = "requires running RustFS server at localhost:9000"]
-async fn test_conditional_put_okay() -> Result<(), Box<dyn std::error::Error>> {
-    let client = create_aws_s3_client().await?;
-    setup_test_bucket(&client).await?;
+async fn test_conditional_put_okay() -> TestResult {
+    init_logging();
+    let mut env = RustFSTestEnvironment::new().await?;
+    env.start_rustfs_server(vec![]).await?;
+    env.create_test_bucket(TEST_BUCKET).await?;
+    let client = env.create_s3_client();
 
-    let test_key = generate_test_key("conditional-put-ok");
+    let test_key = "conditional-put-ok";
     let initial_data = generate_test_data(1024); // 1KB test data
-    let updated_data = generate_test_data(2048); // 2KB updated data
+    let matching_data = generate_test_data(2048); // 2KB updated data
+    let non_matching_data = generate_test_data(3072); // 3KB updated data
 
     // Upload initial object and get its ETag
-    let initial_etag = upload_object_with_metadata(&client, BUCKET, &test_key, &initial_data).await?;
+    let initial_etag = upload_object_with_metadata(&client, TEST_BUCKET, test_key, &initial_data).await?;
 
     // Test 1: PUT with matching If-Match condition (should succeed)
-    let response1 = client
+    client
         .put_object()
-        .bucket(BUCKET)
-        .key(&test_key)
-        .body(Bytes::from(updated_data.clone()).into())
+        .bucket(TEST_BUCKET)
+        .key(test_key)
+        .body(Bytes::from(matching_data.clone()).into())
         .if_match(&initial_etag)
         .send()
-        .await;
-    assert!(response1.is_ok(), "PUT with matching If-Match should succeed");
+        .await?;
+    assert_eq!(object_body(&client, test_key).await?.as_ref(), matching_data);
 
     // Test 2: PUT with non-matching If-None-Match condition (should succeed)
     let fake_etag = "\"fake-etag-12345\"";
-    let response2 = client
+    client
         .put_object()
-        .bucket(BUCKET)
-        .key(&test_key)
-        .body(Bytes::from(updated_data.clone()).into())
+        .bucket(TEST_BUCKET)
+        .key(test_key)
+        .body(Bytes::from(non_matching_data.clone()).into())
         .if_none_match(fake_etag)
         .send()
-        .await;
-    assert!(response2.is_ok(), "PUT with non-matching If-None-Match should succeed");
-
-    // Cleanup
-    cleanup_objects(&client, BUCKET, &[&test_key]).await;
+        .await?;
+    assert_eq!(object_body(&client, test_key).await?.as_ref(), non_matching_data);
 
     Ok(())
 }
 
 #[tokio::test]
-#[ignore = "requires running RustFS server at localhost:9000"]
-async fn test_conditional_put_failed() -> Result<(), Box<dyn std::error::Error>> {
-    let client = create_aws_s3_client().await?;
-    setup_test_bucket(&client).await?;
+async fn test_conditional_put_failed() -> TestResult {
+    init_logging();
+    let mut env = RustFSTestEnvironment::new().await?;
+    env.start_rustfs_server(vec![]).await?;
+    env.create_test_bucket(TEST_BUCKET).await?;
+    let client = env.create_s3_client();
 
-    let test_key = generate_test_key("conditional-put-failed");
+    let test_key = "conditional-put-failed";
     let initial_data = generate_test_data(1024);
     let updated_data = generate_test_data(2048);
 
     // Upload initial object and get its ETag
-    let initial_etag = upload_object_with_metadata(&client, BUCKET, &test_key, &initial_data).await?;
+    let initial_etag = upload_object_with_metadata(&client, TEST_BUCKET, test_key, &initial_data).await?;
 
     // Test 1: PUT with non-matching If-Match condition (should fail with 412)
     let fake_etag = "\"fake-etag-should-not-match\"";
     let response1 = client
         .put_object()
-        .bucket(BUCKET)
-        .key(&test_key)
+        .bucket(TEST_BUCKET)
+        .key(test_key)
         .body(Bytes::from(updated_data.clone()).into())
         .if_match(fake_etag)
         .send()
         .await;
 
-    assert!(response1.is_err(), "PUT with non-matching If-Match should fail");
-    if let Err(e) = response1 {
-        if let SdkError::ServiceError(e) = e {
-            let e = e.into_err();
-            let error_code = e.meta().code().unwrap_or("");
-            assert_eq!("PreconditionFailed", error_code);
-        } else {
-            panic!("Unexpected error: {e:?}");
-        }
-    }
+    assert_s3_error_code(response1, "PreconditionFailed");
+    assert_eq!(object_body(&client, test_key).await?.as_ref(), initial_data);
 
     // Test 2: PUT with matching If-None-Match condition (should fail with 412)
     let response2 = client
         .put_object()
-        .bucket(BUCKET)
-        .key(&test_key)
+        .bucket(TEST_BUCKET)
+        .key(test_key)
         .body(Bytes::from(updated_data.clone()).into())
         .if_none_match(&initial_etag)
         .send()
         .await;
 
-    assert!(response2.is_err(), "PUT with matching If-None-Match should fail");
-    if let Err(e) = response2 {
-        if let SdkError::ServiceError(e) = e {
-            let e = e.into_err();
-            let error_code = e.meta().code().unwrap_or("");
-            assert_eq!("PreconditionFailed", error_code);
-        } else {
-            panic!("Unexpected error: {e:?}");
-        }
-    }
-
-    // Cleanup - only need to clean up the initial object since failed PUTs shouldn't create objects
-    cleanup_objects(&client, BUCKET, &[&test_key]).await;
+    assert_s3_error_code(response2, "PreconditionFailed");
+    assert_eq!(object_body(&client, test_key).await?.as_ref(), initial_data);
 
     Ok(())
 }
 
 #[tokio::test]
-#[ignore = "requires running RustFS server at localhost:9000"]
-async fn test_conditional_put_when_object_does_not_exist() -> Result<(), Box<dyn std::error::Error>> {
-    let client = create_aws_s3_client().await?;
-    setup_test_bucket(&client).await?;
+async fn test_conditional_put_when_object_does_not_exist() -> TestResult {
+    init_logging();
+    let mut env = RustFSTestEnvironment::new().await?;
+    env.start_rustfs_server(vec![]).await?;
+    env.create_test_bucket(TEST_BUCKET).await?;
+    let client = env.create_s3_client();
 
-    let key = "some_key";
-    cleanup_objects(&client, BUCKET, &[key]).await;
+    let key = "conditional-put-missing";
 
     // When the object does not exist, the If-Match condition should always fail
     let response1 = client
         .put_object()
-        .bucket(BUCKET)
+        .bucket(TEST_BUCKET)
         .key(key)
         .body(Bytes::from(generate_test_data(1024)).into())
         .if_match("*")
         .send()
         .await;
-    assert!(response1.is_err());
-    if let Err(e) = response1 {
-        if let SdkError::ServiceError(e) = e {
-            let e = e.into_err();
-            let error_code = e.meta().code().unwrap_or("");
-            assert_eq!("NoSuchKey", error_code);
-        } else {
-            panic!("Unexpected error: {e:?}");
-        }
-    }
+    assert_s3_error_code(response1, "NoSuchKey");
 
     // When the object does not exist, the If-None-Match condition should be able to succeed
-    let response2 = client
+    let created_data = generate_test_data(1024);
+    client
         .put_object()
-        .bucket(BUCKET)
+        .bucket(TEST_BUCKET)
         .key(key)
-        .body(Bytes::from(generate_test_data(1024)).into())
+        .body(Bytes::from(created_data.clone()).into())
         .if_none_match("*")
         .send()
-        .await;
-    assert!(response2.is_ok());
+        .await?;
+    assert_eq!(object_body(&client, key).await?.as_ref(), created_data);
 
-    cleanup_objects(&client, BUCKET, &[key]).await;
     Ok(())
 }
 
 #[tokio::test]
-#[ignore = "requires running RustFS server at localhost:9000"]
-async fn test_conditional_multi_part_upload() -> Result<(), Box<dyn std::error::Error>> {
-    let client = create_aws_s3_client().await?;
-    setup_test_bucket(&client).await?;
+async fn test_conditional_multi_part_upload() -> TestResult {
+    init_logging();
+    let mut env = RustFSTestEnvironment::new().await?;
+    env.start_rustfs_server(vec![]).await?;
+    env.create_test_bucket(TEST_BUCKET).await?;
+    let client = env.create_s3_client();
 
-    let test_key = generate_test_key("multipart-upload-ok");
+    let test_key = "conditional-multipart-upload";
     let test_data = generate_test_data(1024);
-    let initial_etag = upload_object_with_metadata(&client, BUCKET, &test_key, &test_data).await?;
+    let initial_etag = upload_object_with_metadata(&client, TEST_BUCKET, test_key, &test_data).await?;
 
     let part_size = 5 * 1024 * 1024; // 5MB per part (minimum for multipart)
     let num_parts = 3;
     let mut parts = Vec::new();
+    let mut expected_data = Vec::with_capacity(part_size * usize::try_from(num_parts)?);
 
     // Initiate multipart upload
-    let initiate_response = client.create_multipart_upload().bucket(BUCKET).key(&test_key).send().await?;
+    let initiate_response = client
+        .create_multipart_upload()
+        .bucket(TEST_BUCKET)
+        .key(test_key)
+        .send()
+        .await?;
 
     let upload_id = initiate_response
         .upload_id()
@@ -258,12 +213,13 @@ async fn test_conditional_multi_part_upload() -> Result<(), Box<dyn std::error::
 
     // Upload parts
     for part_number in 1..=num_parts {
-        let part_data = generate_test_data(part_size);
+        let part_data = vec![u8::try_from(part_number)?; part_size];
+        expected_data.extend_from_slice(&part_data);
 
         let upload_part_response = client
             .upload_part()
-            .bucket(BUCKET)
-            .key(&test_key)
+            .bucket(TEST_BUCKET)
+            .key(test_key)
             .upload_id(upload_id)
             .part_number(part_number)
             .body(Bytes::from(part_data).into())
@@ -286,57 +242,62 @@ async fn test_conditional_multi_part_upload() -> Result<(), Box<dyn std::error::
     // Test 1: Multipart upload with wildcard If-None-Match, should fail
     let complete_response = client
         .complete_multipart_upload()
-        .bucket(BUCKET)
-        .key(&test_key)
+        .bucket(TEST_BUCKET)
+        .key(test_key)
         .upload_id(upload_id)
         .multipart_upload(completed_upload.clone())
         .if_none_match("*")
         .send()
         .await;
 
-    assert!(complete_response.is_err());
+    assert_s3_error_code(complete_response, "PreconditionFailed");
 
     // Test 2: Multipart upload with matching If-None-Match, should fail
     let complete_response = client
         .complete_multipart_upload()
-        .bucket(BUCKET)
-        .key(&test_key)
+        .bucket(TEST_BUCKET)
+        .key(test_key)
         .upload_id(upload_id)
         .multipart_upload(completed_upload.clone())
         .if_none_match(initial_etag.clone())
         .send()
         .await;
 
-    assert!(complete_response.is_err());
+    assert_s3_error_code(complete_response, "PreconditionFailed");
 
     // Test 3: Multipart upload with unmatching If-Match, should fail
     let complete_response = client
         .complete_multipart_upload()
-        .bucket(BUCKET)
-        .key(&test_key)
+        .bucket(TEST_BUCKET)
+        .key(test_key)
         .upload_id(upload_id)
         .multipart_upload(completed_upload.clone())
         .if_match("\"abcdef\"")
         .send()
         .await;
 
-    assert!(complete_response.is_err());
+    assert_s3_error_code(complete_response, "PreconditionFailed");
+
+    let staged_parts = client
+        .list_parts()
+        .bucket(TEST_BUCKET)
+        .key(test_key)
+        .upload_id(upload_id)
+        .send()
+        .await?;
+    assert_eq!(staged_parts.parts().len(), usize::try_from(num_parts)?);
 
     // Test 4: Multipart upload with matching If-Match, should succeed
-    let complete_response = client
+    client
         .complete_multipart_upload()
-        .bucket(BUCKET)
-        .key(&test_key)
+        .bucket(TEST_BUCKET)
+        .key(test_key)
         .upload_id(upload_id)
-        .multipart_upload(completed_upload.clone())
+        .multipart_upload(completed_upload)
         .if_match(initial_etag)
         .send()
-        .await;
-
-    assert!(complete_response.is_ok());
-
-    // Cleanup
-    cleanup_objects(&client, BUCKET, &[&test_key]).await;
+        .await?;
+    assert_eq!(object_body(&client, test_key).await?.as_ref(), expected_data);
 
     Ok(())
 }

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

@@ -9136,7 +9136,6 @@ mod tests {
         assert_eq!(loaded.report.scanned, 37);
         assert_eq!(loaded.report.eligible, 11);
         assert_eq!(loaded.report.enqueued, 5);
-        assert_eq!(loaded.cursor_revision, Some(1));
         assert!(loaded.lease_expires_at_unix_nanos > 0);
         let token = loaded
             .report
@@ -9155,30 +9154,6 @@ mod tests {
         assert_eq!(admission.lease_id, loaded.lease_id);
         assert_eq!(admission.lease_expires_at_unix_nanos, loaded.lease_expires_at_unix_nanos);
 
-        let mut same_marker_report = report.clone();
-        same_marker_report.scanned += 1;
-        persist_manual_transition_page_checkpoint(
-            &checkpoint_options,
-            &same_marker_report,
-            Some("logs/page-end".to_string()),
-            Some("opaque-next-version".to_string()),
-        )
-        .await
-        .expect("same-marker version checkpoint should persist through the durable progress sink");
-        let same_marker_checkpointed = load_manual_transition_job_record(ecstore.clone(), job_id)
-            .await
-            .expect("same-marker version checkpoint should reload");
-        assert_eq!(same_marker_checkpointed.cursor_revision, Some(2));
-        let (_, version_marker) = decode_manual_transition_continuation_token(
-            same_marker_checkpointed
-                .report
-                .continuation_token
-                .as_deref()
-                .expect("same-marker version checkpoint should persist a cursor"),
-        )
-        .expect("same-marker version cursor should decode");
-        assert_eq!(version_marker.as_deref(), Some("opaque-next-version"));
-
         create_test_bucket(&ecstore, &bucket).await;
         let lifecycle_xml = format!(
             r#"<?xml version="1.0" encoding="UTF-8"?>
@@ -9235,7 +9210,6 @@ mod tests {
         assert_eq!(checkpointed.report.scanned, 1000);
         assert_eq!(checkpointed.report.eligible, 1000);
         assert_eq!(checkpointed.report.dry_run_eligible, 1000);
-        assert_eq!(checkpointed.cursor_revision, Some(3));
         let token = checkpointed
             .report
             .continuation_token

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

@@ -24,10 +24,6 @@ use crate::bucket::lifecycle::bucket_lifecycle_ops::{
     ManualTransitionQueueSnapshot, ManualTransitionRunOptions, ManualTransitionRunReport,
 };
 use crate::bucket::lifecycle::config_boundary;
-use crate::bucket::lifecycle::durable_namespace::{
-    MANUAL_TRANSITION_JOB_NAMESPACE, MANUAL_TRANSITION_SCOPE_NAMESPACE, MANUAL_TRANSITION_TASK_NAMESPACE,
-    MANUAL_TRANSITION_WORKER_RESULT_NAMESPACE,
-};
 use crate::disk::RUSTFS_META_BUCKET;
 use crate::error::{Error, Result as EcstoreResult};
 use crate::object_api::ObjectOptions;
@@ -38,10 +34,10 @@ use crate::store::ECStore;
 pub const MANUAL_TRANSITION_JOB_SCHEMA: &str = "rustfs-manual-transition-job-v1";
 pub const MANUAL_TRANSITION_TASK_SCHEMA: &str = "rustfs-manual-transition-task-v1";
 pub const MANUAL_TRANSITION_WORKER_RESULT_SCHEMA: &str = "rustfs-manual-transition-worker-result-v1";
-pub const MANUAL_TRANSITION_JOB_RECORD_PREFIX: &str = MANUAL_TRANSITION_JOB_NAMESPACE.prefix;
-pub const MANUAL_TRANSITION_SCOPE_RECORD_PREFIX: &str = MANUAL_TRANSITION_SCOPE_NAMESPACE.prefix;
-pub const MANUAL_TRANSITION_TASK_PREFIX: &str = MANUAL_TRANSITION_TASK_NAMESPACE.prefix;
-pub const MANUAL_TRANSITION_WORKER_RESULT_PREFIX: &str = MANUAL_TRANSITION_WORKER_RESULT_NAMESPACE.prefix;
+pub const MANUAL_TRANSITION_JOB_RECORD_PREFIX: &str = "ilm/manual-transition/jobs";
+pub const MANUAL_TRANSITION_SCOPE_RECORD_PREFIX: &str = "ilm/manual-transition/scopes";
+pub const MANUAL_TRANSITION_TASK_PREFIX: &str = "ilm/manual-transition/tasks";
+pub const MANUAL_TRANSITION_WORKER_RESULT_PREFIX: &str = "ilm/manual-transition/results";
 pub const MAX_MANUAL_TRANSITION_JOB_RECORD_SIZE: usize = 64 * 1024;
 pub const MAX_MANUAL_TRANSITION_TASK_RECORD_SIZE: usize = 16 * 1024;
 pub const MAX_MANUAL_TRANSITION_WORKER_RESULT_RECORD_SIZE: usize = 8 * 1024;
@@ -199,8 +195,6 @@ pub struct ManualTransitionJobRecord {
     pub updated_at_unix_nanos: i128,
     #[serde(default, skip_serializing_if = "Option::is_none")]
     pub completed_at_unix_nanos: Option<i128>,
-    #[serde(default, skip_serializing_if = "Option::is_none")]
-    pub cursor_revision: Option<u64>,
     pub report: ManualTransitionRunReport,
     pub queue_snapshot: ManualTransitionQueueSnapshot,
     #[serde(default, skip_serializing_if = "Option::is_none")]
@@ -230,7 +224,6 @@ impl ManualTransitionJobRecord {
             created_at_unix_nanos: now,
             updated_at_unix_nanos: now,
             completed_at_unix_nanos: None,
-            cursor_revision: Some(0),
             report: ManualTransitionRunReport {
                 bucket: bucket.to_string(),
                 prefix: options.prefix.clone(),
@@ -245,7 +238,7 @@ impl ManualTransitionJobRecord {
 
     pub fn complete(&mut self, report: ManualTransitionRunReport, queue_snapshot: ManualTransitionQueueSnapshot) {
         self.scan_completed = true;
-        self.merge_scan_report(&report);
+        self.report.merge_scan_report_preserving_worker(&report);
         self.queue_snapshot = queue_snapshot;
         self.error = None;
         self.mark_terminal_if_worker_drained();
@@ -323,7 +316,7 @@ impl ManualTransitionJobRecord {
             }
         }
         self.queue_snapshot = queue_snapshot;
-        self.advance_updated_at();
+        self.updated_at_unix_nanos = OffsetDateTime::now_utc().unix_timestamp_nanos();
         self.mark_terminal_if_worker_drained();
     }
 
@@ -366,14 +359,14 @@ impl ManualTransitionJobRecord {
         self.report.tier_failure = scan_tier_failure.saturating_add(transition_failed);
         self.report.tier_failure_by_reason = scan_tier_failure_by_reason;
         self.queue_snapshot = queue_snapshot;
-        self.advance_updated_at();
+        self.updated_at_unix_nanos = OffsetDateTime::now_utc().unix_timestamp_nanos();
         self.mark_terminal_if_worker_drained();
         true
     }
 
     pub fn mark_cancel_requested(&mut self) {
         self.cancel_requested = true;
-        self.advance_updated_at();
+        self.updated_at_unix_nanos = OffsetDateTime::now_utc().unix_timestamp_nanos();
     }
 
     pub fn claim_recovery_lease(&mut self, owner_id: impl Into<String>, queue_snapshot: ManualTransitionQueueSnapshot) {
@@ -387,7 +380,7 @@ impl ManualTransitionJobRecord {
     pub fn abandon_recovery_lease(&mut self, lease_id: Uuid) {
         if self.state == ManualTransitionJobState::Running && self.lease_id == lease_id {
             self.lease_expires_at_unix_nanos = 0;
-            self.advance_updated_at();
+            self.updated_at_unix_nanos = OffsetDateTime::now_utc().unix_timestamp_nanos();
         }
     }
 
@@ -405,7 +398,7 @@ impl ManualTransitionJobRecord {
 
     pub fn renew_lease(&mut self, queue_snapshot: ManualTransitionQueueSnapshot) {
         let now = OffsetDateTime::now_utc().unix_timestamp_nanos();
-        self.updated_at_unix_nanos = self.updated_at_unix_nanos.saturating_add(1).max(now);
+        self.updated_at_unix_nanos = now;
         self.lease_expires_at_unix_nanos = manual_transition_job_lease_expires_at(now);
         self.queue_snapshot = queue_snapshot;
     }
@@ -445,18 +438,11 @@ impl ManualTransitionJobRecord {
 
     pub fn update_running_progress(&mut self, report: ManualTransitionRunReport, queue_snapshot: ManualTransitionQueueSnapshot) {
         if self.state == ManualTransitionJobState::Running {
-            self.merge_scan_report(&report);
+            self.report.merge_scan_report_preserving_worker(&report);
             self.renew_lease(queue_snapshot);
         }
     }
 
-    fn merge_scan_report(&mut self, report: &ManualTransitionRunReport) {
-        if self.report.continuation_token != report.continuation_token {
-            self.cursor_revision = Some(self.cursor_revision.unwrap_or(0).saturating_add(1));
-        }
-        self.report.merge_scan_report_preserving_worker(report);
-    }
-
     pub fn mark_unknown_if_unowned(&mut self) {
         if self.state == ManualTransitionJobState::Running {
             self.state = ManualTransitionJobState::Unknown;
@@ -477,13 +463,9 @@ impl ManualTransitionJobRecord {
     }
 
     fn mark_updated_terminal(&mut self) {
-        self.advance_updated_at();
-        self.completed_at_unix_nanos = Some(self.updated_at_unix_nanos);
-    }
-
-    fn advance_updated_at(&mut self) {
         let now = OffsetDateTime::now_utc().unix_timestamp_nanos();
-        self.updated_at_unix_nanos = self.updated_at_unix_nanos.saturating_add(1).max(now);
+        self.updated_at_unix_nanos = now;
+        self.completed_at_unix_nanos = Some(now);
     }
 
     fn mark_terminal_if_worker_drained(&mut self) {
@@ -1127,8 +1109,7 @@ pub fn manual_transition_scope_record_object_name(scope_key: &str) -> Result<Str
 pub async fn save_manual_transition_job_record(api: Arc<ECStore>, job: &ManualTransitionJobRecord) -> EcstoreResult<()> {
     let object = manual_transition_job_record_object_name(job.job_id).map_err(manual_transition_job_store_error)?;
     let data = job.encode().map_err(manual_transition_job_store_error)?;
-    config_boundary::save_config(api.clone(), &object, data.clone()).await?;
-    api.record_durable_ilm_decommission_progress(&object, &data).await
+    config_boundary::save_config(api, &object, data).await
 }
 
 pub async fn load_manual_transition_job_record(api: Arc<ECStore>, job_id: Uuid) -> EcstoreResult<ManualTransitionJobRecord> {
@@ -1161,9 +1142,9 @@ pub async fn save_manual_transition_job_record_if_current(
     let object = manual_transition_job_record_object_name(job.job_id).map_err(manual_transition_job_store_error)?;
     let data = job.encode().map_err(manual_transition_job_store_error)?;
     config_boundary::save_config_with_opts_quiet(
-        api.clone(),
+        api,
         &object,
-        data.clone(),
+        data,
         &ObjectOptions {
             max_parity: true,
             http_preconditions: Some(HTTPPreconditions {
@@ -1173,8 +1154,7 @@ pub async fn save_manual_transition_job_record_if_current(
             ..Default::default()
         },
     )
-    .await?;
-    api.record_durable_ilm_decommission_progress(&object, &data).await
+    .await
 }
 
 /// Applies a job-record mutation with optimistic concurrency control.
@@ -1612,9 +1592,9 @@ pub async fn save_manual_transition_scope_admission_if_absent(
     let object = manual_transition_scope_record_object_name(&admission.scope_key).map_err(manual_transition_job_store_error)?;
     let data = serde_json::to_vec(admission).map_err(Error::other)?;
     config_boundary::save_config_with_opts(
-        api.clone(),
+        api,
         &object,
-        data.clone(),
+        data,
         &ObjectOptions {
             max_parity: true,
             http_preconditions: Some(HTTPPreconditions {
@@ -1624,8 +1604,7 @@ pub async fn save_manual_transition_scope_admission_if_absent(
             ..Default::default()
         },
     )
-    .await?;
-    api.record_durable_ilm_decommission_progress(&object, &data).await
+    .await
 }
 
 pub async fn load_manual_transition_scope_admission(
@@ -1663,9 +1642,9 @@ pub async fn save_manual_transition_scope_admission_if_current(
     let object = manual_transition_scope_record_object_name(&admission.scope_key).map_err(manual_transition_job_store_error)?;
     let data = serde_json::to_vec(admission).map_err(Error::other)?;
     match config_boundary::save_config_with_opts(
-        api.clone(),
+        api,
         &object,
-        data.clone(),
+        data,
         &ObjectOptions {
             max_parity: true,
             http_preconditions: Some(HTTPPreconditions {
@@ -1681,8 +1660,7 @@ pub async fn save_manual_transition_scope_admission_if_current(
             Err(Error::PreconditionFailed)
         }
         result => result,
-    }?;
-    api.record_durable_ilm_decommission_progress(&object, &data).await
+    }
 }
 
 pub async fn claim_manual_transition_scope_admission(
@@ -1975,15 +1953,13 @@ pub async fn delete_manual_transition_scope_admission_if_current(
     job_id: Uuid,
     lease_id: Uuid,
 ) -> EcstoreResult<bool> {
-    let (admission, etag) = match load_manual_transition_scope_admission_with_etag(api.clone(), scope_key).await {
-        Ok((admission, etag)) if admission.job_id == job_id && admission.lease_id == lease_id => (admission, etag),
+    let etag = match load_manual_transition_scope_admission_with_etag(api.clone(), scope_key).await {
+        Ok((admission, etag)) if admission.job_id == job_id && admission.lease_id == lease_id => etag,
         Ok(_) => return Ok(false),
         Err(Error::ConfigNotFound) => return Ok(true),
         Err(err) => return Err(err),
     };
     let object = manual_transition_scope_record_object_name(scope_key).map_err(manual_transition_job_store_error)?;
-    let data = serde_json::to_vec(&admission).map_err(Error::other)?;
-    api.record_durable_ilm_decommission_terminal(&object, &data).await?;
     match config_boundary::delete_config_if_match(api, &object, &etag).await {
         Ok(()) | Err(Error::ConfigNotFound) => Ok(true),
         Err(Error::PreconditionFailed) => Ok(false),
@@ -2601,25 +2577,6 @@ mod tests {
         assert!(decoded.report.tier_failure_by_reason.is_empty());
     }
 
-    #[test]
-    fn manual_transition_job_record_decodes_legacy_cursor_without_revision() {
-        let options = ManualTransitionRunOptions::default();
-        let record = ManualTransitionJobRecord::new(Uuid::new_v4(), "bucket", &options, TEST_OWNER);
-        let encoded = record.encode().expect("job record should encode");
-        let mut value: serde_json::Value = serde_json::from_slice(&encoded).expect("encoded job should be json");
-        value["job"]
-            .as_object_mut()
-            .expect("job should be object")
-            .remove("cursor_revision");
-        let record_bytes = serde_json::to_vec(&value["job"]).expect("legacy job should encode");
-        value["content_sha256"] = serde_json::Value::String(hex_sha256(&record_bytes, ToOwned::to_owned));
-        let legacy = serde_json::to_vec(&value).expect("legacy envelope should encode");
-
-        let decoded = ManualTransitionJobRecord::decode(record.job_id, &legacy).expect("legacy job should decode");
-
-        assert_eq!(decoded.cursor_revision, None);
-    }
-
     #[test]
     fn manual_transition_job_record_rejects_unknown_report_fields() {
         let options = ManualTransitionRunOptions::default();

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

@@ -16,7 +16,6 @@ pub mod bucket_lifecycle_audit;
 pub mod bucket_lifecycle_ops;
 mod config_boundary;
 pub mod core;
-mod durable_namespace;
 pub mod evaluator;
 pub mod manual_transition_job;
 mod metadata_boundary;
@@ -32,8 +31,3 @@ pub mod tier_free_version_recovery;
 pub mod tier_last_day_stats;
 pub mod tier_sweeper;
 pub mod transition_transaction;
-
-pub(crate) use durable_namespace::{
-    DurableIlmRecordCheckpoint, ILM_META_PREFIX, ValidatedDurableIlmRecord, classify_durable_ilm_record,
-    validate_durable_ilm_record,
-};

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

@@ -20,7 +20,6 @@ use tokio_util::sync::CancellationToken;
 use tracing::{debug, warn};
 
 use crate::bucket::lifecycle::config_boundary;
-use crate::bucket::lifecycle::durable_namespace::TIER_DELETE_JOURNAL_NAMESPACE;
 use crate::bucket::lifecycle::runtime_boundary;
 use crate::bucket::lifecycle::tier_sweeper::{
     Jentry, TierDeleteJournalState, TierDeleteSourceIdentity,
@@ -50,7 +49,7 @@ const TIER_DELETE_JOURNAL_VERSION: u8 = 2;
 const TIER_DELETE_JOURNAL_EXACT_VERSION: u8 = 3;
 const TIER_DELETE_JOURNAL_STATE_VERSION: u8 = 4;
 const TIER_DELETE_JOURNAL_TRANSACTION_VERSION: u8 = 5;
-pub(crate) const TIER_DELETE_JOURNAL_PREFIX: &str = TIER_DELETE_JOURNAL_NAMESPACE.prefix;
+pub(crate) const TIER_DELETE_JOURNAL_PREFIX: &str = "ilm/tier-delete-journal/";
 
 #[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Eq)]
 #[serde(deny_unknown_fields)]
@@ -433,21 +432,6 @@ async fn process_committed_tier_delete_journal_entry(api: Arc<ECStore>, je: &Jen
         )
         .await?;
     }
-    let path = tier_delete_journal_object_name(je);
-    let data = encode_tier_delete_journal_entry(je).map_err(std::io::Error::other)?;
-    let target_pool_indices = api
-        .record_durable_ilm_decommission_terminal_target_pools(&path, &data)
-        .await
-        .map_err(std::io::Error::other)?;
-    if let Some(target_pool_indices) = target_pool_indices {
-        for target_pool_idx in target_pool_indices {
-            match config_boundary::delete_config(api.pools[target_pool_idx].clone(), &path).await {
-                Ok(()) | Err(Error::ConfigNotFound) => {}
-                Err(err) => return Err(std::io::Error::other(err)),
-            }
-        }
-        return Ok(());
-    }
     remove_tier_delete_journal_entry(api, je).await
 }
 

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

@@ -21,7 +21,6 @@ use tracing::{debug, warn};
 use uuid::Uuid;
 
 use crate::bucket::lifecycle::config_boundary;
-use crate::bucket::lifecycle::durable_namespace::TRANSITION_TRANSACTION_NAMESPACE;
 use crate::bucket::lifecycle::lifecycle::TRANSITION_COMPLETE;
 use crate::bucket::lifecycle::tier_sweeper::{
     delete_confirmed_transition_candidate_exact_with_lease_idempotent,
@@ -43,7 +42,7 @@ const TRANSITION_TRANSACTION_RECOVERY_INTERVAL: Duration = Duration::from_secs(6
 const TRANSITION_TRANSACTION_RECOVERY_TIMEOUT: Duration = Duration::from_secs(300);
 pub const TRANSITION_TRANSACTION_SCHEMA: &str = "rustfs-transition-transaction-v1";
 pub const TRANSITION_TRANSACTION_PREFIX: &str = "ilm/transition-transactions";
-pub const TRANSITION_TRANSACTION_RECORD_PREFIX: &str = TRANSITION_TRANSACTION_NAMESPACE.prefix;
+pub const TRANSITION_TRANSACTION_RECORD_PREFIX: &str = "ilm/transition-transactions/records";
 pub const MAX_TRANSITION_TRANSACTION_SIZE: usize = 64 * 1024;
 
 pub type Result<T> = std::result::Result<T, TransitionTransactionError>;
@@ -585,8 +584,7 @@ pub(crate) async fn save_transition_transaction_record(
     let object =
         transition_transaction_record_object_name(transaction.transaction_id).map_err(transition_transaction_store_error)?;
     let data = transaction.encode().map_err(transition_transaction_store_error)?;
-    config_boundary::save_config(api.clone(), &object, data.clone()).await?;
-    api.record_durable_ilm_decommission_progress(&object, &data).await
+    config_boundary::save_config(api, &object, data).await
 }
 
 pub(crate) async fn load_transition_transaction_record(
@@ -598,14 +596,8 @@ pub(crate) async fn load_transition_transaction_record(
     TransitionTransaction::decode(transaction_id, &data).map_err(transition_transaction_store_error)
 }
 
-pub(crate) async fn delete_transition_transaction_record(
-    api: Arc<ECStore>,
-    transaction: &TransitionTransaction,
-) -> EcstoreResult<()> {
-    let object =
-        transition_transaction_record_object_name(transaction.transaction_id).map_err(transition_transaction_store_error)?;
-    let data = transaction.encode().map_err(transition_transaction_store_error)?;
-    api.record_durable_ilm_decommission_terminal(&object, &data).await?;
+pub(crate) async fn delete_transition_transaction_record(api: Arc<ECStore>, transaction_id: Uuid) -> EcstoreResult<()> {
+    let object = transition_transaction_record_object_name(transaction_id).map_err(transition_transaction_store_error)?;
     match config_boundary::delete_config(api, &object).await {
         Ok(()) | Err(Error::ConfigNotFound) => Ok(()),
         Err(err) => Err(err),
@@ -821,7 +813,7 @@ pub async fn finalize_missing_transition_transaction_for_operator(
     if probe != TransitionOperatorProbe::Missing {
         return Err(TransitionOperatorError::CandidateNotMissing(probe));
     }
-    delete_transition_transaction_record(api, &transaction)
+    delete_transition_transaction_record(api, transaction_id)
         .await
         .map_err(TransitionOperatorError::Store)
 }
@@ -857,22 +849,22 @@ pub async fn process_transition_transaction_record(
     match transaction.state {
         TransitionTransactionState::Uploaded => {
             delete_transition_remote_candidate(api.clone(), transaction).await?;
-            delete_transition_transaction_record(api, transaction).await?;
+            delete_transition_transaction_record(api, transaction.transaction_id).await?;
             Ok(TransitionTransactionRecoveryOutcome::RemoteCandidateDeleted)
         }
         TransitionTransactionState::CleanupPending => match local_commit_matches_transaction(api.clone(), transaction).await {
             Ok(true) => {
-                delete_transition_transaction_record(api, transaction).await?;
+                delete_transition_transaction_record(api, transaction.transaction_id).await?;
                 Ok(TransitionTransactionRecoveryOutcome::RecordDeleted)
             }
             Ok(false) => {
                 delete_transition_remote_candidate(api.clone(), transaction).await?;
-                delete_transition_transaction_record(api, transaction).await?;
+                delete_transition_transaction_record(api, transaction.transaction_id).await?;
                 Ok(TransitionTransactionRecoveryOutcome::RemoteCandidateDeleted)
             }
             Err(err) if transition_source_is_missing(&err) => {
                 delete_transition_remote_candidate(api.clone(), transaction).await?;
-                delete_transition_transaction_record(api, transaction).await?;
+                delete_transition_transaction_record(api, transaction.transaction_id).await?;
                 Ok(TransitionTransactionRecoveryOutcome::RemoteCandidateDeleted)
             }
             Err(err) => Err(err),
@@ -880,7 +872,7 @@ pub async fn process_transition_transaction_record(
         TransitionTransactionState::LocalCommitStarted => {
             match local_commit_matches_transaction(api.clone(), transaction).await {
                 Ok(true) => {
-                    delete_transition_transaction_record(api, transaction).await?;
+                    delete_transition_transaction_record(api, transaction.transaction_id).await?;
                     Ok(TransitionTransactionRecoveryOutcome::RecordDeleted)
                 }
                 Ok(false) => Ok(TransitionTransactionRecoveryOutcome::Retained),
@@ -889,7 +881,7 @@ pub async fn process_transition_transaction_record(
             }
         }
         TransitionTransactionState::AbortedNoRemote | TransitionTransactionState::Committed => {
-            delete_transition_transaction_record(api, transaction).await?;
+            delete_transition_transaction_record(api, transaction.transaction_id).await?;
             Ok(TransitionTransactionRecoveryOutcome::RecordDeleted)
         }
         TransitionTransactionState::UploadOutcomeUnknown => recover_unknown_upload_outcome(api, transaction).await,
@@ -915,7 +907,7 @@ async fn recover_unknown_upload_outcome(
         .map_err(Error::other)?
     {
         TransitionCandidateProbe::Missing => {
-            delete_transition_transaction_record(api, transaction).await?;
+            delete_transition_transaction_record(api, transaction.transaction_id).await?;
             Ok(TransitionTransactionRecoveryOutcome::RecordDeleted)
         }
         TransitionCandidateProbe::UnversionedPresent => {
@@ -933,7 +925,7 @@ async fn recover_unknown_upload_outcome(
             )
             .await
             .map_err(Error::other)?;
-            delete_transition_transaction_record(api, transaction).await?;
+            delete_transition_transaction_record(api, transaction.transaction_id).await?;
             Ok(TransitionTransactionRecoveryOutcome::RemoteCandidateDeleted)
         }
         TransitionCandidateProbe::VersionedPresent(version_id) => {
@@ -966,7 +958,7 @@ async fn cleanup_recovered_unknown_upload_candidate(
         .map_err(transition_transaction_store_error)?;
     save_transition_transaction_record(api.clone(), &cleanup).await?;
     delete_transition_remote_candidate(api.clone(), &cleanup).await?;
-    delete_transition_transaction_record(api, &cleanup).await?;
+    delete_transition_transaction_record(api, cleanup.transaction_id).await?;
     Ok(TransitionTransactionRecoveryOutcome::RemoteCandidateDeleted)
 }
 

crates/ecstore/src/config/com.rs

@@ -406,25 +406,6 @@ where
     Ok(data)
 }
 
-pub(crate) async fn read_config_limited_preserve_empty<S>(api: Arc<S>, file: &str, max_bytes: usize) -> Result<Vec<u8>>
-where
-    S: EcstoreObjectIO,
-{
-    let (data, _obj) = read_config_limited_preserve_empty_with_metadata(api, file, max_bytes).await?;
-    Ok(data)
-}
-
-pub(crate) async fn read_config_limited_preserve_empty_with_metadata<S>(
-    api: Arc<S>,
-    file: &str,
-    max_bytes: usize,
-) -> Result<(Vec<u8>, ObjectInfo)>
-where
-    S: EcstoreObjectIO,
-{
-    read_config_with_metadata_inner(api, file, &ObjectOptions::default(), true, Some(max_bytes)).await
-}
-
 /// Read an existing config object without treating an empty payload as absent.
 /// Callers that validate their own payload format need to distinguish corruption
 /// from `ConfigNotFound`.

crates/ecstore/src/set_disk/ops/object.rs

@@ -37,7 +37,7 @@ use crate::bucket::lifecycle::{
     transition_transaction::{
         TransitionRemoteVersion, TransitionSourceIdentity, TransitionSourceVersionMode, TransitionTransaction,
         TransitionTransactionInit, TransitionTransactionState, delete_transition_transaction_record,
-        load_transition_transaction_record, save_transition_transaction_record,
+        save_transition_transaction_record,
     },
 };
 use crate::bucket::quota::reservation;
@@ -2124,13 +2124,26 @@ impl SetDisks {
 
             let put_object_size = known_put_object_storage_size(data.size());
             let shard_file_size_raw = erasure.shard_file_size(put_object_size);
-            let is_inline_buffer = storage_class_config.should_inline(shard_file_size_raw, erasure.data_shards, opts.versioned);
+            let is_inline_buffer =
+                storage_class_config.should_inline(shard_file_size_raw, erasure.data_shards, opts.versioned);
 
             let collect_stage_timing = rustfs_io_metrics::put_stage_metrics_enabled() || issue3031_diag_enabled();
             let shard_file_size = shard_file_size_raw;
             let shard_size = erasure.shard_size();
             let write_path = classify_put_write_path(is_inline_buffer, put_object_size, fi.erasure.block_size);
             let direct_inline_commit = matches!(write_path, SmallWritePath::Inline);
+            {
+                use std::io::Write;
+                let msg = format!(
+                    "INLINE_DEBUG: bucket={} obj={} size={} shard_fs={} ds={} bs={} inline={} direct={} path={} iblock={} ver={}\n",
+                    bucket, object, put_object_size, shard_file_size_raw, erasure.data_shards, fi.erasure.block_size,
+                    is_inline_buffer, direct_inline_commit, write_path.metric_label(), storage_class_config.inline_block(), opts.versioned
+                );
+                if let Ok(mut f) = std::fs::OpenOptions::new().create(true).append(true).open("/tmp/rustfs_inline_debug.log") {
+                    let _ = f.write_all(msg.as_bytes());
+                }
+                let _ = std::io::stderr().write_all(msg.as_bytes());
+            }
             rustfs_io_metrics::record_put_object_path(write_path.metric_label());
             let writer_setup_stage_start = collect_stage_timing.then(Instant::now);
             let (mut writers, errors) = if direct_inline_commit {
@@ -4248,12 +4261,7 @@ fn record_transition_uploaded_save_attempt(transaction: &TransitionTransaction,
 
 async fn delete_transition_transaction_if_available(api: Option<&Arc<ECStore>>, transaction_id: Uuid) -> Result<()> {
     if let Some(api) = api {
-        let transaction = match load_transition_transaction_record(api.clone(), transaction_id).await {
-            Ok(transaction) => transaction,
-            Err(Error::ConfigNotFound) => return Ok(()),
-            Err(err) => return Err(err),
-        };
-        return delete_transition_transaction_record(api.clone(), &transaction).await;
+        return delete_transition_transaction_record(api.clone(), transaction_id).await;
     }
     Ok(())
 }

crates/ecstore/src/store/object.rs

@@ -3194,7 +3194,7 @@ impl ECStore {
 
         // Default return value
         let mut del_objects = vec![DeletedObject::default(); objects.len()];
-        let accounting = vec![None; objects.len()];
+        let mut accounting = vec![None; objects.len()];
 
         let mut del_errs = Vec::with_capacity(objects.len());
         for _ in 0..objects.len() {

crates/ecstore/src/store/rebalance/support.rs

@@ -271,7 +271,7 @@ pub(super) fn resolve_latest_object_info_candidates(
             .filter(|candidate| latest_candidate_mod_time(candidate) == Some(latest_mod_time))
             .collect::<Vec<_>>();
 
-        latest_candidates.sort_by_key(|candidate| std::cmp::Reverse(candidate.idx));
+        latest_candidates.sort_by(|left, right| right.idx.cmp(&left.idx));
 
         let Some(winner) = latest_candidates.first() else {
             return Err(Error::ErasureReadQuorum);

deny.toml

@@ -43,6 +43,9 @@ allow-git = [
     # RustFS fork carrying presigned expiry and constant-time authentication fixes.
     # owner: rustfs-maintainers  review: 2026-10
     "https://github.com/rustfs/s3s.git",
+    # MiMalloc fork pinned for hotpath allocation counting support.
+    # owner: houseme  review: 2026-10
+    "https://github.com/xonatius/mimalloc_rust.git",
 ]
 
 [bans]

docs/testing/ci-gates.md

@@ -1,149 +0,0 @@
-# CI gate matrix
-
-This file is the source of truth for which validation runs on each event, its
-configured wall-clock budget, and whether it can block a merge. Test taxonomy,
-naming, and nextest serialization rules remain in [README.md](README.md); e2e
-membership and counts remain in
-[e2e-suite-inventory.md](e2e-suite-inventory.md).
-
-The distinction between **required** and **report-only** is load-bearing:
-a failing job blocks a merge only when its exact check name is present in the
-live `main` ruleset. A workflow name, a `merge_group` trigger, or a red PR check
-does not make a job required by itself.
-
-## Required merge checks
-
-The live `main` ruleset (`6436880`) currently requires exactly these contexts:
-
-| Required context | Producer | Validation |
-|---|---|---|
-| `CLA Check` | `.github/workflows/cla.yml` | Contributor agreement |
-| `Quick Checks` | `.github/workflows/ci.yml` | Formatting and repository guard scripts |
-| `Test and Lint` | `.github/workflows/ci.yml` | Clippy, workspace nextest excluding `e2e_test`, doctests, and migration proofs |
-
-For pull requests limited to the paths excluded by the main CI workflow,
-`.github/workflows/ci-docs-only.yml` reports `Quick Checks` and
-`Test and Lint` under the same names. It runs the real quick checks and the
-planning-document guard; it does not claim that Rust compilation or runtime
-tests ran. Despite the workflow name, these paths also include selected deploy,
-workflow, and lock files.
-
-Verify the live rule rather than trusting this snapshot before changing merge
-policy:
-
-```bash
-gh api repos/rustfs/rustfs/rulesets/6436880 \
-  --jq '.rules[] | select(.type == "required_status_checks") | .parameters'
-```
-
-The ruleset currently has `strict_required_status_checks_policy=false`.
-`Continuous Integration` accepts `merge_group` events and runs `e2e-full` for
-them, but `End-to-End Tests (full merge gate)` is not currently a required
-context. Therefore the repository is prepared to test a merge-queue SHA, but
-the workflow alone does not prove that every merge passed that lane.
-
-## Pull request and merge matrix
-
-Budgets below are job `timeout-minutes`, not typical runtimes. “Report-only”
-means the result is visible and actionable but is not in the live required
-context list.
-
-| Event | Validation | Budget | Merge status | Reproduction |
-|---|---|---:|---|---|
-| PR, non-doc change | `Quick Checks` | 10 min | Required | `make pre-commit` (broader local umbrella) |
-| PR, non-doc change | `Test and Lint` | 90 min | Required | `cargo nextest run --profile ci --all --exclude e2e_test` |
-| PR, non-doc change | `Typos` | 10 min | Report-only | `typos` |
-| PR, non-doc change | `ILM Integration (serial)` | 90 min | Report-only | Use the exact command in `.github/workflows/ci.yml` |
-| PR, non-doc change | rio-v2 / swift / sftp test-and-lint variants | 90 min each | Report-only | `cargo nextest run` with the workflow's feature set |
-| PR, non-doc change | `Build RustFS Debug Binary` | 30 min | Report-only; prerequisite for black-box lanes | `cargo build -p rustfs --bins` |
-| PR, non-doc change | `io_uring Integration (real)` | 30 min | Report-only | `cargo test -p rustfs-ecstore --lib uring_ -- --test-threads=1 --nocapture` |
-| PR, non-doc change | `End-to-End Tests` (`e2e-smoke` plus `s3s-e2e`) | 30 min | Report-only | `cargo nextest run --profile e2e-smoke -p e2e_test`; then `./scripts/e2e-run.sh ./target/debug/rustfs <data-dir>` |
-| PR, non-doc change | `S3 Implemented Tests` | 60 min | Report-only | Build `rustfs`, then run `scripts/s3-tests/run.sh` with `DEPLOY_MODE=binary`, `TEST_MODE=single`, and `MAXFAIL=0` |
-| PR, non-doc change | `S3 Lifecycle Behavior Tests` | 30 min | Report-only | Use the accelerated scanner environment in `.github/workflows/ci.yml` with `scripts/s3-tests/run.sh` |
-| PR touching dependency or workflow inputs | Cargo Deny / Workflow Pin Report / Dependency Review | 20 / 5 / 30 min | Report-only | `cargo deny check`; `scripts/security/check_workflow_pins.sh` |
-| PR touching architecture rules or architecture docs | `Architecture Migration Rules` | 10 min | Report-only | `scripts/check_architecture_migration_rules.sh` |
-| PR touching Nix or workspace manifests | `Nix Build & Check` | 60 min | Report-only | `nix flake check` |
-| PR limited to main-CI-excluded paths | companion `Quick Checks` and `Test and Lint` | 10 min each | Required | `git diff --check`; `make doc-paths-check` when documentation paths changed |
-| `merge_group` | Standard CI plus `e2e-full` | 55 min for `e2e-full` | Standard required contexts only; `e2e-full` report-only | `cargo nextest run --profile e2e-full -p e2e_test` |
-| Push to `main` | Standard CI plus `e2e-full` | 55 min for `e2e-full` | Post-merge detection | Same as `merge_group` |
-| PR touching fuzz inputs or harness paths | Build plus five 60-second fuzz smoke targets | 60 min build; 30 min per target | Report-only | `MAX_TOTAL_TIME=60 ./scripts/fuzz/run.sh` |
-| PR touching selected ecstore disk/format paths | `Rename Safety` on Windows | 60 min | Report-only | Run the four `cargo test -p rustfs-ecstore --lib <filter>` commands in `windows-filesystem.yml` on Windows |
-
-The authoritative e2e filters live in `.config/nextest.toml`; extend a profile
-instead of adding a second ad-hoc selector. Before a profile runs,
-`scripts/check_test_wiring.py` compares its exact membership to the committed
-digest so a silent test drop fails closed.
-
-## Scheduled and manual validation
-
-Scheduled lanes are independent fault domains. They do not block a pull
-request, but their workflow-local gate can fail the run and scheduled failures
-are routed to the shared failure-issue action. The scheduled-validation
-watchdog and freshness workflow separately detect incomplete runs and missing
-schedules.
-
-| Cadence (UTC unless noted) | Workflow / validation | Budget | Verdict and artifacts | Reproduction |
-|---|---|---:|---|---|
-| Daily 02:17 | Fuzz: five nightly corpus targets | 60 min build; 60 min per target | Gate; corpus/crash artifacts, scheduled failure alert | `MAX_TOTAL_TIME=<seconds> ./scripts/fuzz/run.sh` |
-| Daily 03:17 | MinIO interop (EC + SSE read parity) | 40 min | Gate; scheduled failure alert | Dispatch `minio-interop.yml` or follow its pinned Docker fixture steps |
-| Daily 04:29 | Replication / cluster-fault / protocol e2e | 45 / 90 / 90 min | Three independent gates; JUnit, membership, and server logs | `cargo nextest run --profile e2e-repl-nightly -p e2e_test`; `--profile e2e-nightly`; `-j 1 --profile e2e-protocols` |
-| Daily 06:31 | Warp performance A/B | 180 min | Regression budget gate; A/B summaries and server logs | `bash scripts/run_hotpath_warp_abba.sh --help` |
-| Daily 00:07 Asia/Shanghai (16:07 UTC previous day) | Nightly GNU build and Vault lanes | 150 / 90 / 60 min | Build, live Vault, and HA failover gates | Use the commands and pinned Vault images in `nightly-gnu.yml` |
-| Daily 03:23 | Security Audit | 20 / 5 min, plus 30 min on PR dependency review | Cargo Deny and workflow-pin gates; scheduled failure alert | `cargo deny check`; `scripts/security/check_workflow_pins.sh` |
-| Daily 23:47 | Scheduled Validation Freshness | 10 min | Fails when a critical schedule was never created or is stale | Dispatch `scheduled-validation-freshness.yml` |
-| Sunday 00:11 | Full `Continuous Integration` matrix | Per-job budgets above | Weekly variant coverage, including dormant rio-v2 binary/e2e lanes | Dispatch `ci.yml` |
-| Sunday 01:13 | Seven-platform build matrix | 150 min per platform | Build/package integrity; scheduled failure alert | Dispatch `build.yml` with an exact platform set |
-| Sunday 02:19 | Ceph s3-tests full sweep: single and real four-node, four shards each | 180 min per shard | Compatibility gate; report, JUnit, exact node IDs, and server logs | `scripts/s3-tests/run.sh` against an existing single or distributed target |
-| Sunday 06:41 | Mint | 120 min | **Report-only by design**; per-suite PASS/FAIL/NA and raw `log.json` | Reproduce the pinned Docker sequence in `mint.yml` or dispatch it |
-| Sunday 07:43 | Workspace line coverage | 120 min | Report-only trend; lcov and JSON retained 90 days | `make coverage` |
-| Monthly, day 1 06:37 | Runner Hygiene | 15 min | Validates runner ephemerality; scheduled failure alert | Dispatch `runner-hygiene.yml` |
-
-Manual `workflow_dispatch` exists for the scheduled workflows above. Manual
-runs are debugging evidence and intentionally do not open scheduled-failure
-issues. A manual performance run may explicitly allow a known regression; that
-override must not be treated as an ordinary passing baseline.
-
-## Release validation
-
-Release validation is post-merge and tag-driven; it does not substitute for a
-pull-request gate.
-
-| Event | Validation | Budget | Result |
-|---|---|---:|---|
-| Push to `main` or weekly schedule | `Build and Release` platform matrix | 150 min per platform | Build artifacts for all selected targets; no release publication on a main push |
-| Valid release or preview tag | `Build and Release` plus asset checks | 150 min per platform | Draft release, checksummed assets, and publish step |
-| Successful non-preview release-tag build | Docker image build and image scan | 60 min build; 30 min scan | Multi-architecture images plus vulnerability report |
-| Successful release-tag build | DEB/RPM packaging | 30 min per architecture | Packages and checksum files uploaded to the release |
-| Successful non-preview release-tag build | Helm template test and package | 30 min build; 30 min publish | Versioned chart and repository index |
-
-Use an exact preview tag for end-to-end release rehearsal. Manual dispatches
-are backfill/debug paths and do not prove the automatic `workflow_run` chain.
-
-## Evidence requirements
-
-A green check is useful only when it proves the intended behavior ran:
-
-- Record the exact commit SHA and run URL.
-- Separate product failure from runner prerequisites, service readiness, and
-  cancellation. Repair the precondition, then rerun the exact workload.
-- Preserve membership manifests, JUnit, raw compatibility logs, seeds, and
-  server logs where the workflow provides them.
-- For a bug fix or a new fault checker, provide sensitivity evidence: the old
-  behavior or an intentional mutation must fail the new oracle, and the fixed
-  behavior must pass it.
-- Never promote a report-only lane to required from one green run. Require at
-  least 14 days and 30 representative pull requests with at least 99% complete
-  execution, then update the ruleset and this table together.
-
-## Change checklist
-
-Update this file in the same pull request when any of these change:
-
-- workflow triggers, job names, timeouts, or nextest profile ownership;
-- required status contexts or strict/merge-queue policy;
-- scheduled cadence, alert routing, artifact contract, or local reproduction;
-- report-only versus gating semantics.
-
-Do not copy per-module test counts here. Update
-[e2e-suite-inventory.md](e2e-suite-inventory.md) and its enforced membership
-digest instead.

docs/testing/e2e-suite-inventory.md

@@ -84,7 +84,7 @@
 | protocols | 16 | 🌙 |
 | quota_test | 14 |  |
 | reliability_disk_fault_test | 4 |  |
-| reliant | 25 | 19 ✅ |
+| reliant | 29 | 19 ✅ |
 | replication_extension_test | 75 | 20 ✅ +55 🌙 |
 | security_boundary_test | 4 |  |
 | server_startup_failfast_test | 1 |  |

rustfs/Cargo.toml

@@ -336,13 +336,13 @@ opentelemetry = { workspace = true }
 tracing-opentelemetry = { workspace = true }
 # Data structures
 hashbrown = { workspace = true, features = ["serde", "rayon"] }
-rustfs-mimalloc = { workspace = true }
+mimalloc = { workspace = true }
 
 [target.'cfg(target_os = "linux")'.dependencies]
 libsystemd.workspace = true
 
 [target.'cfg(not(target_os = "windows"))'.dependencies]
-rustfs-mimalloc-sys.workspace = true
+libmimalloc-sys.workspace = true
 
 [dev-dependencies]
 uuid = { workspace = true, features = ["v4", "v5", "fast-rng", "macro-diagnostics"] }

rustfs/src/allocator_reclaim.rs

@@ -369,8 +369,14 @@ pub fn allocator_reclaim_controller_snapshot(ctx: &CancellationToken) -> Allocat
 }
 
 #[cfg(not(target_os = "windows"))]
+#[allow(unsafe_code)]
 fn collect_allocator_memory(force: bool) -> Result<(), String> {
-    rustfs_mimalloc::MiMalloc::collect(force);
+    // SAFETY: `mi_collect` is provided by the active global allocator backend
+    // on this target family. It is explicitly intended to reclaim retained
+    // pages/segments and does not require additional invariants from the caller.
+    unsafe {
+        libmimalloc_sys::mi_collect(force);
+    }
     Ok(())
 }
 

rustfs/src/main.rs

@@ -26,22 +26,22 @@ struct MiMallocAllocator;
 unsafe impl GlobalAlloc for MiMallocAllocator {
     unsafe fn alloc(&self, layout: Layout) -> *mut u8 {
         // SAFETY: the caller upholds GlobalAlloc's contract for layout.
-        unsafe { rustfs_mimalloc::MiMalloc.alloc(layout) }
+        unsafe { mimalloc::MiMalloc.alloc(layout) }
     }
 
     unsafe fn alloc_zeroed(&self, layout: Layout) -> *mut u8 {
         // SAFETY: the caller upholds GlobalAlloc's contract for layout.
-        unsafe { rustfs_mimalloc::MiMalloc.alloc_zeroed(layout) }
+        unsafe { mimalloc::MiMalloc.alloc_zeroed(layout) }
     }
 
     unsafe fn dealloc(&self, ptr: *mut u8, layout: Layout) {
         // SAFETY: ptr and layout came from this allocator and are forwarded unchanged.
-        unsafe { rustfs_mimalloc::MiMalloc.dealloc(ptr, layout) }
+        unsafe { mimalloc::MiMalloc.dealloc(ptr, layout) }
     }
 
     unsafe fn realloc(&self, ptr: *mut u8, layout: Layout, new_size: usize) -> *mut u8 {
         // SAFETY: ptr and layout came from this allocator and are forwarded unchanged.
-        unsafe { rustfs_mimalloc::MiMalloc.realloc(ptr, layout, new_size) }
+        unsafe { mimalloc::MiMalloc.realloc(ptr, layout, new_size) }
     }
 }
 
@@ -51,7 +51,7 @@ static GLOBAL: hotpath::CountingAllocator<MiMallocAllocator> = hotpath::Counting
 
 #[cfg(not(all(feature = "hotpath", feature = "hotpath-alloc")))]
 #[global_allocator]
-static GLOBAL: rustfs_mimalloc::MiMalloc = rustfs_mimalloc::MiMalloc;
+static GLOBAL: mimalloc::MiMalloc = mimalloc::MiMalloc;
 
 fn main() {
     let _hotpath_guard = hotpath::HotpathGuardBuilder::new("main").build();
@@ -71,9 +71,8 @@ mod tests {
         allocation.extend_from_slice(&[7_u8; 64]);
 
         assert_eq!(allocation.len(), 64);
-        let heap = rustfs_mimalloc::heap::Heap::main();
         // SAFETY: the live Vec pointer is valid to inspect for heap ownership.
-        assert!(unsafe { heap.contains(allocation.as_ptr()) });
+        assert!(unsafe { libmimalloc_sys::mi_is_in_heap_region(allocation.as_ptr().cast()) });
     }
 
     #[test]
@@ -86,13 +85,12 @@ mod tests {
         let layout = Layout::from_size_align(32, 8).expect("valid test allocation layout");
         let grown_layout = Layout::from_size_align(64, 8).expect("valid grown test allocation layout");
         let allocator = super::MiMallocAllocator;
-        let heap = rustfs_mimalloc::heap::Heap::main();
 
         // SAFETY: The pointer is checked for null before use and later released
         // through the same allocator with the corresponding layout.
         let ptr = unsafe { allocator.alloc_zeroed(layout) };
         assert!(!ptr.is_null());
-        assert!(unsafe { heap.contains(ptr) });
+        assert!(unsafe { libmimalloc_sys::mi_is_in_heap_region(ptr.cast()) });
         assert!(unsafe { std::slice::from_raw_parts(ptr, 32).iter().all(|byte| *byte == 0) });
 
         // SAFETY: `ptr` was allocated by `allocator` with `layout`; on failure
@@ -104,7 +102,7 @@ mod tests {
             panic!("mimalloc realloc failed in allocator smoke test");
         }
 
-        assert!(unsafe { heap.contains(grown_ptr) });
+        assert!(unsafe { libmimalloc_sys::mi_is_in_heap_region(grown_ptr.cast()) });
         // SAFETY: `grown_ptr` was reallocated by `allocator` and is released
         // with the matching grown layout.
         unsafe { allocator.dealloc(grown_ptr, grown_layout) };

rustfs/src/memory_observability.rs

@@ -17,7 +17,10 @@ use rustfs_io_metrics::{
     record_cpu_usage, record_memory_usage, record_process_memory_split,
 };
 use serde::Serialize;
+#[cfg(any(test, not(target_os = "windows")))]
 use serde_json::Value;
+#[cfg(not(target_os = "windows"))]
+use std::ffi::CStr;
 use std::path::Path;
 use std::sync::{Arc, Mutex, OnceLock};
 use std::time::Duration;
@@ -228,18 +231,7 @@ fn read_cgroup_memory_snapshot() -> Option<CgroupMemorySnapshot> {
     read_cgroup_v2().or_else(read_cgroup_v1)
 }
 
-fn read_allocator_memory_snapshot() -> Option<AllocatorMemorySnapshot> {
-    let json = rustfs_mimalloc::MiMalloc::stats_json();
-    if json.is_empty() {
-        return None;
-    }
-    let observation = parse_mimalloc_stats_json(&json)?;
-    Some(AllocatorMemorySnapshot {
-        backend: crate::allocator_reclaim::allocator_backend(),
-        observation,
-    })
-}
-
+#[cfg(any(test, not(target_os = "windows")))]
 fn numeric_json_value(value: &Value) -> Option<u64> {
     match value {
         Value::Number(number) => number
@@ -250,6 +242,7 @@ fn numeric_json_value(value: &Value) -> Option<u64> {
     }
 }
 
+#[cfg(any(test, not(target_os = "windows")))]
 fn numeric_json_field(value: &Value, field: &str) -> Option<u64> {
     match value {
         Value::Object(fields) => fields
@@ -261,6 +254,7 @@ fn numeric_json_field(value: &Value, field: &str) -> Option<u64> {
     }
 }
 
+#[cfg(any(test, not(target_os = "windows")))]
 fn mimalloc_stat_field(value: &Value, metric: &str, field: &str) -> Option<u64> {
     match value {
         Value::Object(fields) => {
@@ -277,10 +271,12 @@ fn mimalloc_stat_field(value: &Value, metric: &str, field: &str) -> Option<u64>
     }
 }
 
+#[cfg(any(test, not(target_os = "windows")))]
 fn mimalloc_stat_current(value: &Value, metric: &str) -> Option<u64> {
     mimalloc_stat_field(value, metric, "current")
 }
 
+#[cfg(any(test, not(target_os = "windows")))]
 fn mimalloc_stat_sum(value: &Value, metrics: &[&str], field: &str) -> Option<u64> {
     metrics
         .iter()
@@ -289,6 +285,7 @@ fn mimalloc_stat_sum(value: &Value, metrics: &[&str], field: &str) -> Option<u64
         .filter(|value| *value > 0)
 }
 
+#[cfg(any(test, not(target_os = "windows")))]
 fn parse_mimalloc_stats_json(stats_json: &str) -> Option<AllocatorMemoryObservation> {
     let value = serde_json::from_str::<Value>(stats_json).ok()?;
     let malloc_metrics = ["malloc_normal", "malloc_huge"];
@@ -315,6 +312,33 @@ fn parse_mimalloc_stats_json(stats_json: &str) -> Option<AllocatorMemoryObservat
     }
 }
 
+#[cfg(not(target_os = "windows"))]
+#[allow(unsafe_code)]
+fn read_allocator_memory_snapshot() -> Option<AllocatorMemorySnapshot> {
+    // SAFETY: `mi_stats_get_json` returns a null-terminated JSON buffer owned by
+    // mimalloc when called with a null input buffer. The mimalloc API requires
+    // freeing that buffer with `mi_free`; parsing finishes before the buffer is freed.
+    let observation = unsafe {
+        let stats_ptr = libmimalloc_sys::mi_stats_get_json(0, std::ptr::null_mut());
+        if stats_ptr.is_null() {
+            return None;
+        }
+
+        let observation = CStr::from_ptr(stats_ptr).to_str().ok().and_then(parse_mimalloc_stats_json);
+        libmimalloc_sys::mi_free(stats_ptr.cast());
+        observation?
+    };
+    Some(AllocatorMemorySnapshot {
+        backend: crate::allocator_reclaim::allocator_backend(),
+        observation,
+    })
+}
+
+#[cfg(target_os = "windows")]
+fn read_allocator_memory_snapshot() -> Option<AllocatorMemorySnapshot> {
+    None
+}
+
 fn configured_memory_observability_interval_secs() -> u64 {
     rustfs_utils::get_env_u64(ENV_MEMORY_OBSERVABILITY_INTERVAL_SECS, DEFAULT_MEMORY_OBSERVABILITY_INTERVAL_SECS).max(1)
 }
@@ -542,13 +566,6 @@ mod tests {
         assert_eq!(parse_mimalloc_stats_json(r#"{ "allocator": "unknown" }"#), None);
     }
 
-    #[test]
-    fn read_allocator_memory_snapshot_uses_mimalloc_stats_json() {
-        let snapshot = super::read_allocator_memory_snapshot();
-        #[cfg(not(target_os = "windows"))]
-        assert!(snapshot.is_some(), "allocator snapshot should be available on non-Windows");
-    }
-
     #[test]
     fn memory_observability_snapshot_reports_disabled_when_metrics_are_disabled() {
         let snapshot = build_memory_observability_status_snapshot(false, 15, false);