Compare commits

...

20 Commits

Author SHA1 Message Date
overtrue 2f048466b4 fix(ecstore): distinguish active ILM target cleanup 2026-08-22 12:10:41 +08:00
overtrue 2a4021c3c5 fix(ecstore): preserve active ILM source journals 2026-08-22 12:10:41 +08:00
overtrue c6ea2dd9b3 test(ecstore): isolate durable ILM scenario stack 2026-08-22 12:10:41 +08:00
overtrue 83b87d2a3e test(ecstore): compile multi-source ILM recovery 2026-08-22 12:10:41 +08:00
overtrue a61a95ae48 test(ecstore): serialize multi-source ILM recovery 2026-08-22 12:10:41 +08:00
overtrue fd3b47f92b test(ecstore): cover durable ILM recovery boundaries 2026-08-22 12:10:41 +08:00
overtrue fe1d5c7b5d fix(ecstore): repair durable ILM receipt recovery 2026-08-22 12:10:41 +08:00
overtrue 9fa3e9c41c fix(ecstore): harden durable ILM cursor receipts 2026-08-22 12:10:41 +08:00
overtrue 1fec344f43 fix(ecstore): avoid terminal receipt shadowing 2026-08-22 12:10:41 +08:00
overtrue acffa7511f fix(ecstore): re-export durable ILM checkpoint 2026-08-22 12:10:41 +08:00
overtrue 9a0cd48f9e fix(ecstore): anchor decommission ILM receipts 2026-08-22 12:10:41 +08:00
overtrue 15b8e8860f fix(ecstore): close ILM receipt recovery gaps 2026-08-22 12:10:41 +08:00
overtrue 812cf9662a fix(ecstore): track ILM recovery across decommission 2026-08-22 12:10:41 +08:00
overtrue 50c208f716 fix(ecstore): verify ILM metadata before decommission 2026-08-22 12:10:41 +08:00
overtrue 44351323e5 fix(ecstore): migrate ILM metadata during decommission 2026-08-22 12:10:41 +08:00
Zhengchao An 2c3e68ad89 ci: let feature validation jobs finish (#6364) 2026-08-22 04:09:21 +00:00
houseme 2f0918f60b feat(disk): fsync dedicated blocking pool (default-off) (#6366) 2026-08-22 11:24:24 +08:00
Zhengchao An 5b951de2b7 test(ci): bound s3-tests failure logs (#6361) 2026-08-22 02:58:01 +00:00
Zhengchao An 98c4675617 refactor(e2e): consolidate duplicated helper functions into common.rs (#6355) 2026-08-22 01:10:14 +00:00
Zhengchao An eec34331de chore(madmin): remove dead trace structs, keep TraceType only (#6343)
chore(madmin): remove dead trace structs, keep TraceType bitflag helper only

TraceInfo, TraceInfoLegacy, TraceHTTPStats, TraceCallStats, TraceRequestInfo,
TraceResponseInfo, StorageStats, and OSStats are unreferenced outside trace.rs.
Trim to TraceType + its bitflag operations which are actively used by
service_commands.rs and profile_admin.rs.

-139 lines (215 -> 76 lines)
2026-08-22 01:01:24 +00:00
21 changed files with 4275 additions and 421 deletions
+10
View File
@@ -78,6 +78,12 @@ 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
@@ -190,6 +196,10 @@ 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]]
+3 -3
View File
@@ -400,7 +400,7 @@ jobs:
if: github.event_name != 'pull_request' || github.event.action != 'closed'
needs: [ quick-checks ]
runs-on: sm-standard-4
timeout-minutes: 45
timeout-minutes: 90
env:
FORCE_JAVASCRIPT_ACTIONS_TO_NODE24: "true"
steps:
@@ -440,7 +440,7 @@ jobs:
if: github.event_name != 'pull_request' || github.event.action != 'closed'
needs: [ quick-checks ]
runs-on: sm-standard-4
timeout-minutes: 60
timeout-minutes: 90
env:
FORCE_JAVASCRIPT_ACTIONS_TO_NODE24: "true"
steps:
@@ -470,7 +470,7 @@ jobs:
if: github.event_name != 'pull_request' || github.event.action != 'closed'
needs: [ quick-checks ]
runs-on: sm-standard-4
timeout-minutes: 60
timeout-minutes: 90
strategy:
# On a PR, one failing protocol leg is enough to know the PR is not ready,
# so stop the sibling leg instead of paying another ~40 minutes for it.
+7
View File
@@ -57,6 +57,13 @@ pub const DEFAULT_MAX_IO_EVENTS_PER_TICK: usize = 1024;
pub const DEFAULT_EVENT_INTERVAL: u32 = 61;
pub const DEFAULT_RNG_SEED: Option<u64> = None; // None means random
/// Dedicated blocking thread pool for fsync/fdatasync operations.
/// When > 1, fsync operations are isolated from the main blocking pool to
/// prevent device-bound fsync from starving read operations (pread/stat/open).
/// Default 0 means auto (no isolation, use main runtime).
pub const ENV_FSYNC_BLOCKING_THREADS: &str = "RUSTFS_RUNTIME_FSYNC_BLOCKING_THREADS";
pub const DEFAULT_FSYNC_BLOCKING_THREADS: usize = 0;
// Dial9 Tokio Telemetry Default values
pub const DEFAULT_RUNTIME_DIAL9_ENABLED: bool = false; // Disabled by default
pub const DEFAULT_RUNTIME_DIAL9_OUTPUT_DIR: &str = "/var/log/rustfs/telemetry";
+151
View File
@@ -30,6 +30,7 @@ use reqwest::StatusCode;
use rustfs_signer::constants::UNSIGNED_PAYLOAD;
use rustfs_signer::sign_v4;
use s3s::Body;
use serde_json;
use std::ffi::OsStr;
use std::fs as stdfs;
use std::io::ErrorKind;
@@ -1583,6 +1584,156 @@ impl Drop for RustFSTestClusterEnvironment {
}
}
/// Send a SigV4-signed HTTP request and return the raw `reqwest::Response`.
///
/// Unlike [`signed_s3_request`], this variant accepts `body: Option<Vec<u8>>`
/// (binary-safe) and reorders parameters so that `access_key`/`secret_key`
/// appear before the body — matching the convention used by the replication
/// extension and object-lambda e2e suites.
pub(crate) async fn signed_request(
method: http::Method,
url: &str,
access_key: &str,
secret_key: &str,
body: Option<Vec<u8>>,
content_type: Option<&str>,
) -> Result<reqwest::Response, Box<dyn std::error::Error + Send + Sync>> {
let uri = url.parse::<http::Uri>()?;
let authority = uri.authority().ok_or("request URL missing authority")?.to_string();
let mut request = http::Request::builder().method(method.clone()).uri(uri);
request = request.header(HOST, authority);
request = request.header("x-amz-content-sha256", UNSIGNED_PAYLOAD);
if let Some(content_type) = content_type {
request = request.header(CONTENT_TYPE, content_type);
}
let content_len = body.as_ref().map(|body| body.len() as i64).unwrap_or_default();
let signed = sign_v4(request.body(Body::empty())?, content_len, access_key, secret_key, "", "us-east-1");
let reqwest_method = reqwest::Method::from_bytes(method.as_str().as_bytes())?;
let client = local_http_client();
let mut request_builder = client.request(reqwest_method, url);
for (name, value) in signed.headers() {
request_builder = request_builder.header(name, value);
}
if let Some(body) = body {
request_builder = request_builder.body(body);
}
Ok(request_builder.send().await?)
}
/// Like [`signed_request`], but uses a caller-supplied `reqwest::Client`
/// instead of the shared [`local_http_client`].
pub(crate) async fn signed_request_with_client(
client: &reqwest::Client,
method: http::Method,
url: &str,
access_key: &str,
secret_key: &str,
body: Option<Vec<u8>>,
content_type: Option<&str>,
) -> Result<reqwest::Response, Box<dyn std::error::Error + Send + Sync>> {
let uri = url.parse::<http::Uri>()?;
let authority = uri.authority().ok_or("request URL missing authority")?.to_string();
let mut request = http::Request::builder().method(method.clone()).uri(uri);
request = request.header(HOST, authority);
request = request.header("x-amz-content-sha256", UNSIGNED_PAYLOAD);
if let Some(content_type) = content_type {
request = request.header(CONTENT_TYPE, content_type);
}
let content_len = body.as_ref().map(|body| body.len() as i64).unwrap_or_default();
let signed = sign_v4(request.body(Body::empty())?, content_len, access_key, secret_key, "", "us-east-1");
let reqwest_method = reqwest::Method::from_bytes(method.as_str().as_bytes())?;
let mut request_builder = client.request(reqwest_method, url);
for (name, value) in signed.headers() {
request_builder = request_builder.header(name, value);
}
if let Some(body) = body {
request_builder = request_builder.body(body);
}
Ok(request_builder.send().await?)
}
/// Like [`signed_request`], but includes a `session_token` in the
/// `x-amz-security-token` header and passes it to the SigV4 signer.
pub(crate) async fn signed_request_with_session_token(
method: http::Method,
url: &str,
access_key: &str,
secret_key: &str,
session_token: &str,
body: Option<Vec<u8>>,
content_type: Option<&str>,
) -> Result<reqwest::Response, Box<dyn std::error::Error + Send + Sync>> {
let uri = url.parse::<http::Uri>()?;
let authority = uri.authority().ok_or("request URL missing authority")?.to_string();
let mut request = http::Request::builder().method(method.clone()).uri(uri);
request = request.header(HOST, authority);
request = request.header("x-amz-content-sha256", UNSIGNED_PAYLOAD);
if !session_token.is_empty() {
request = request.header("x-amz-security-token", session_token);
}
if let Some(content_type) = content_type {
request = request.header(CONTENT_TYPE, content_type);
}
let content_len = body.as_ref().map(|body| body.len() as i64).unwrap_or_default();
let signed = sign_v4(
request.body(Body::empty())?,
content_len,
access_key,
secret_key,
session_token,
"us-east-1",
);
let reqwest_method = reqwest::Method::from_bytes(method.as_str().as_bytes())?;
let client = local_http_client();
let mut request_builder = client.request(reqwest_method, url);
for (name, value) in signed.headers() {
request_builder = request_builder.header(name, value);
}
if let Some(body) = body {
request_builder = request_builder.body(body);
}
Ok(request_builder.send().await?)
}
/// Create a new user via the admin API.
pub(crate) async fn admin_create_user(
env: &RustFSTestEnvironment,
username: &str,
secret_key: &str,
) -> Result<(), Box<dyn std::error::Error + Send + Sync>> {
let url = format!("{}/rustfs/admin/v3/add-user?accessKey={}", env.url, username);
let body = serde_json::json!({
"secretKey": secret_key,
"status": "enabled"
});
let response = signed_request(
http::Method::PUT,
&url,
&env.access_key,
&env.secret_key,
Some(body.to_string().into_bytes()),
Some("application/json"),
)
.await?;
if response.status() != reqwest::StatusCode::OK {
let status = response.status();
let body = response.text().await.unwrap_or_default();
return Err(format!("create user failed: {status} {body}").into());
}
Ok(())
}
#[cfg(test)]
mod tests {
use super::*;
+51 -1
View File
@@ -40,7 +40,7 @@ use std::time::Duration;
use tokio::fs;
use tokio::net::TcpStream;
use tokio::time::sleep;
use tracing::{debug, error, info};
use tracing::{debug, error, info, warn};
// KMS-specific constants
pub const TEST_BUCKET: &str = "kms-test-bucket";
@@ -177,6 +177,49 @@ pub async fn get_kms_status(
Ok(status)
}
/// Poll the KMS status endpoint until the backend reports ready or the timeout
/// expires. Replaces hard-coded `sleep(Duration::from_secs(3))` startup waits
/// with an active readiness probe so tests start as soon as KMS is usable
/// (typically < 1 s) instead of always waiting the full 3 s.
///
/// Uses exponential back-off starting at 200 ms (doubling each attempt, capped
/// at 1 s) up to a total wall-clock budget of 5 s.
pub async fn wait_for_kms_ready(
base_url: &str,
access_key: &str,
secret_key: &str,
) -> Result<(), Box<dyn std::error::Error + Send + Sync>> {
let total_deadline = Duration::from_secs(5);
let start = tokio::time::Instant::now();
let mut backoff = Duration::from_millis(200);
let max_backoff = Duration::from_secs(1);
let mut first_attempt = true;
loop {
if !first_attempt {
if start.elapsed() >= total_deadline {
return Err("KMS failed to become ready within 5 seconds".into());
}
sleep(backoff).await;
backoff = (backoff * 2).min(max_backoff);
}
first_attempt = false;
match get_kms_status(base_url, access_key, secret_key).await {
Ok(status) => {
info!("KMS is ready (status: {})", status);
return Ok(());
}
Err(e) => {
if start.elapsed() >= total_deadline {
return Err(format!("KMS did not become ready within 5 s: last error: {e}").into());
}
warn!(error = %e, elapsed_ms = start.elapsed().as_millis() as u64, "KMS not ready yet, retrying…");
}
}
}
}
/// Create a default KMS key for testing and return the created key ID
pub async fn create_default_key(
base_url: &str,
@@ -861,6 +904,13 @@ impl LocalKMSTestEnvironment {
Ok(default_key_id.to_string())
}
/// Poll the KMS status endpoint until the backend reports ready.
///
/// Prefer this over a fixed `sleep` after calling `start_rustfs_for_local_kms`.
pub async fn wait_for_kms_ready(&self) -> Result<(), Box<dyn std::error::Error + Send + Sync>> {
wait_for_kms_ready(&self.base_env.url, &self.base_env.access_key, &self.base_env.secret_key).await
}
/// Configure Local KMS backend with a predefined default key
pub async fn configure_local_kms(&self) -> Result<String, Box<dyn std::error::Error + Send + Sync>> {
// Use a fixed, predictable default key ID
+2 -36
View File
@@ -12,12 +12,11 @@
// See the License for the specific language governing permissions and
// limitations under the License.
use crate::common::{RustFSTestClusterEnvironment, RustFSTestEnvironment, init_logging, local_http_client};
use crate::common::{RustFSTestClusterEnvironment, RustFSTestEnvironment, init_logging, local_http_client, signed_request};
use aws_sdk_s3::primitives::ByteStream;
use http::header::{CONTENT_TYPE, HOST};
use reqwest::StatusCode;
use rustfs_signer::constants::UNSIGNED_PAYLOAD;
use rustfs_signer::{pre_sign_v4, sign_v4};
use rustfs_signer::pre_sign_v4;
use rustfs_utils::egress::ENV_OUTBOUND_ALLOW_ORIGINS;
use s3s::Body;
use std::collections::HashMap;
@@ -227,39 +226,6 @@ async fn presigned_get_request(
Ok(local_http_client().get(signed.uri().to_string()).send().await?)
}
async fn signed_request(
method: http::Method,
url: &str,
access_key: &str,
secret_key: &str,
body: Option<Vec<u8>>,
content_type: Option<&str>,
) -> Result<reqwest::Response, Box<dyn Error + Send + Sync>> {
let uri = url.parse::<http::Uri>()?;
let authority = uri.authority().ok_or("request URL missing authority")?.to_string();
let mut request = http::Request::builder().method(method.clone()).uri(uri);
request = request.header(HOST, authority);
request = request.header("x-amz-content-sha256", UNSIGNED_PAYLOAD);
if let Some(content_type) = content_type {
request = request.header(CONTENT_TYPE, content_type);
}
let content_len = body.as_ref().map(|body| body.len() as i64).unwrap_or_default();
let signed = sign_v4(request.body(Body::empty())?, content_len, access_key, secret_key, "", "us-east-1");
let reqwest_method = reqwest::Method::from_bytes(method.as_str().as_bytes())?;
let client = local_http_client();
let mut request_builder = client.request(reqwest_method, url);
for (name, value) in signed.headers() {
request_builder = request_builder.header(name, value);
}
if let Some(body) = body {
request_builder = request_builder.body(body);
}
Ok(request_builder.send().await?)
}
async fn configure_webhook_target(
env: &RustFSTestEnvironment,
target_name: &str,
@@ -13,8 +13,9 @@
// limitations under the License.
use crate::common::{
RustFSTestEnvironment, awscurl_available, awscurl_post_sts_form_urlencoded, init_logging, local_http_client,
replication_fast_env, rustfs_binary_path,
RustFSTestEnvironment, admin_create_user, awscurl_available, awscurl_post_sts_form_urlencoded, init_logging,
local_http_client, replication_fast_env, rustfs_binary_path, signed_request, signed_request_with_client,
signed_request_with_session_token,
};
use crate::fake_s3_target::{
FAKE_ACCESS_KEY, FAKE_SECRET_KEY, FakeS3Target, FaultAction as FakeTargetFault, Operation as FakeTargetOperation,
@@ -35,7 +36,7 @@ use base64::{Engine, engine::general_purpose::STANDARD as BASE64_STANDARD};
use bytes::Bytes;
use flate2::read::GzDecoder;
use futures::{Stream, StreamExt};
use http::header::{CONTENT_ENCODING, CONTENT_TYPE, HOST};
use http::header::CONTENT_ENCODING;
use http_body_util::{BodyExt, Full};
use hyper::body::Incoming;
use hyper::server::conn::http1;
@@ -56,9 +57,6 @@ use rustfs_madmin::{
AddServiceAccountReq, ListServiceAccountsResp, PeerInfo, PeerSite, ReplicateAddStatus, ReplicateEditStatus,
ReplicateRemoveStatus, SRRemoveReq, SRResyncOpStatus, SRStatusInfo, SiteReplicationInfo, SyncStatus,
};
use rustfs_signer::constants::UNSIGNED_PAYLOAD;
use rustfs_signer::sign_v4;
use s3s::Body;
use s3s::header::X_AMZ_REPLICATION_STATUS;
use sha2::{Digest, Sha256};
use std::collections::BTreeMap;
@@ -387,116 +385,6 @@ struct ReplicationResetStatusTarget {
object: String,
}
async fn signed_request(
method: http::Method,
url: &str,
access_key: &str,
secret_key: &str,
body: Option<Vec<u8>>,
content_type: Option<&str>,
) -> Result<reqwest::Response, Box<dyn Error + Send + Sync>> {
let uri = url.parse::<http::Uri>()?;
let authority = uri.authority().ok_or("request URL missing authority")?.to_string();
let mut request = http::Request::builder().method(method.clone()).uri(uri);
request = request.header(HOST, authority);
request = request.header("x-amz-content-sha256", UNSIGNED_PAYLOAD);
if let Some(content_type) = content_type {
request = request.header(CONTENT_TYPE, content_type);
}
let content_len = body.as_ref().map(|body| body.len() as i64).unwrap_or_default();
let signed = sign_v4(request.body(Body::empty())?, content_len, access_key, secret_key, "", "us-east-1");
let reqwest_method = reqwest::Method::from_bytes(method.as_str().as_bytes())?;
let client = local_http_client();
let mut request_builder = client.request(reqwest_method, url);
for (name, value) in signed.headers() {
request_builder = request_builder.header(name, value);
}
if let Some(body) = body {
request_builder = request_builder.body(body);
}
Ok(request_builder.send().await?)
}
async fn signed_request_with_client(
client: &reqwest::Client,
method: http::Method,
url: &str,
access_key: &str,
secret_key: &str,
body: Option<Vec<u8>>,
content_type: Option<&str>,
) -> Result<reqwest::Response, Box<dyn Error + Send + Sync>> {
let uri = url.parse::<http::Uri>()?;
let authority = uri.authority().ok_or("request URL missing authority")?.to_string();
let mut request = http::Request::builder().method(method.clone()).uri(uri);
request = request.header(HOST, authority);
request = request.header("x-amz-content-sha256", UNSIGNED_PAYLOAD);
if let Some(content_type) = content_type {
request = request.header(CONTENT_TYPE, content_type);
}
let content_len = body.as_ref().map(|body| body.len() as i64).unwrap_or_default();
let signed = sign_v4(request.body(Body::empty())?, content_len, access_key, secret_key, "", "us-east-1");
let reqwest_method = reqwest::Method::from_bytes(method.as_str().as_bytes())?;
let mut request_builder = client.request(reqwest_method, url);
for (name, value) in signed.headers() {
request_builder = request_builder.header(name, value);
}
if let Some(body) = body {
request_builder = request_builder.body(body);
}
Ok(request_builder.send().await?)
}
async fn signed_request_with_session_token(
method: http::Method,
url: &str,
access_key: &str,
secret_key: &str,
session_token: &str,
body: Option<Vec<u8>>,
content_type: Option<&str>,
) -> Result<reqwest::Response, Box<dyn Error + Send + Sync>> {
let uri = url.parse::<http::Uri>()?;
let authority = uri.authority().ok_or("request URL missing authority")?.to_string();
let mut request = http::Request::builder().method(method.clone()).uri(uri);
request = request.header(HOST, authority);
request = request.header("x-amz-content-sha256", UNSIGNED_PAYLOAD);
if !session_token.is_empty() {
request = request.header("x-amz-security-token", session_token);
}
if let Some(content_type) = content_type {
request = request.header(CONTENT_TYPE, content_type);
}
let content_len = body.as_ref().map(|body| body.len() as i64).unwrap_or_default();
let signed = sign_v4(
request.body(Body::empty())?,
content_len,
access_key,
secret_key,
session_token,
"us-east-1",
);
let reqwest_method = reqwest::Method::from_bytes(method.as_str().as_bytes())?;
let client = local_http_client();
let mut request_builder = client.request(reqwest_method, url);
for (name, value) in signed.headers() {
request_builder = request_builder.header(name, value);
}
if let Some(body) = body {
request_builder = request_builder.body(body);
}
Ok(request_builder.send().await?)
}
fn extract_xml_tag(xml: &str, tag: &str) -> Option<String> {
let open = format!("<{tag}>");
let close = format!("</{tag}>");
@@ -1016,35 +904,6 @@ fn create_user_s3_client(env: &RustFSTestEnvironment, access_key: &str, secret_k
Client::from_conf(config)
}
async fn admin_create_user(
env: &RustFSTestEnvironment,
username: &str,
secret_key: &str,
) -> Result<(), Box<dyn Error + Send + Sync>> {
let url = format!("{}/rustfs/admin/v3/add-user?accessKey={}", env.url, username);
let body = serde_json::json!({
"secretKey": secret_key,
"status": "enabled"
});
let response = signed_request(
http::Method::PUT,
&url,
&env.access_key,
&env.secret_key,
Some(body.to_string().into_bytes()),
Some("application/json"),
)
.await?;
if response.status() != StatusCode::OK {
let status = response.status();
let body = response.text().await.unwrap_or_default();
return Err(format!("create user failed: {status} {body}").into());
}
Ok(())
}
async fn admin_add_canned_policy(
env: &RustFSTestEnvironment,
policy_name: &str,
@@ -9136,6 +9136,7 @@ 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
@@ -9154,6 +9155,30 @@ 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"?>
@@ -9210,6 +9235,7 @@ 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
File diff suppressed because it is too large Load Diff
@@ -24,6 +24,10 @@ 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;
@@ -34,10 +38,10 @@ 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 = "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 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 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;
@@ -195,6 +199,8 @@ 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")]
@@ -224,6 +230,7 @@ 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(),
@@ -238,7 +245,7 @@ impl ManualTransitionJobRecord {
pub fn complete(&mut self, report: ManualTransitionRunReport, queue_snapshot: ManualTransitionQueueSnapshot) {
self.scan_completed = true;
self.report.merge_scan_report_preserving_worker(&report);
self.merge_scan_report(&report);
self.queue_snapshot = queue_snapshot;
self.error = None;
self.mark_terminal_if_worker_drained();
@@ -316,7 +323,7 @@ impl ManualTransitionJobRecord {
}
}
self.queue_snapshot = queue_snapshot;
self.updated_at_unix_nanos = OffsetDateTime::now_utc().unix_timestamp_nanos();
self.advance_updated_at();
self.mark_terminal_if_worker_drained();
}
@@ -359,14 +366,14 @@ 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.updated_at_unix_nanos = OffsetDateTime::now_utc().unix_timestamp_nanos();
self.advance_updated_at();
self.mark_terminal_if_worker_drained();
true
}
pub fn mark_cancel_requested(&mut self) {
self.cancel_requested = true;
self.updated_at_unix_nanos = OffsetDateTime::now_utc().unix_timestamp_nanos();
self.advance_updated_at();
}
pub fn claim_recovery_lease(&mut self, owner_id: impl Into<String>, queue_snapshot: ManualTransitionQueueSnapshot) {
@@ -380,7 +387,7 @@ 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.updated_at_unix_nanos = OffsetDateTime::now_utc().unix_timestamp_nanos();
self.advance_updated_at();
}
}
@@ -398,7 +405,7 @@ impl ManualTransitionJobRecord {
pub fn renew_lease(&mut self, queue_snapshot: ManualTransitionQueueSnapshot) {
let now = OffsetDateTime::now_utc().unix_timestamp_nanos();
self.updated_at_unix_nanos = now;
self.updated_at_unix_nanos = self.updated_at_unix_nanos.saturating_add(1).max(now);
self.lease_expires_at_unix_nanos = manual_transition_job_lease_expires_at(now);
self.queue_snapshot = queue_snapshot;
}
@@ -438,11 +445,18 @@ impl ManualTransitionJobRecord {
pub fn update_running_progress(&mut self, report: ManualTransitionRunReport, queue_snapshot: ManualTransitionQueueSnapshot) {
if self.state == ManualTransitionJobState::Running {
self.report.merge_scan_report_preserving_worker(&report);
self.merge_scan_report(&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;
@@ -463,9 +477,13 @@ 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 = now;
self.completed_at_unix_nanos = Some(now);
self.updated_at_unix_nanos = self.updated_at_unix_nanos.saturating_add(1).max(now);
}
fn mark_terminal_if_worker_drained(&mut self) {
@@ -1109,7 +1127,8 @@ 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, &object, data).await
config_boundary::save_config(api.clone(), &object, data.clone()).await?;
api.record_durable_ilm_decommission_progress(&object, &data).await
}
pub async fn load_manual_transition_job_record(api: Arc<ECStore>, job_id: Uuid) -> EcstoreResult<ManualTransitionJobRecord> {
@@ -1142,9 +1161,9 @@ 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,
api.clone(),
&object,
data,
data.clone(),
&ObjectOptions {
max_parity: true,
http_preconditions: Some(HTTPPreconditions {
@@ -1154,7 +1173,8 @@ pub async fn save_manual_transition_job_record_if_current(
..Default::default()
},
)
.await
.await?;
api.record_durable_ilm_decommission_progress(&object, &data).await
}
/// Applies a job-record mutation with optimistic concurrency control.
@@ -1592,9 +1612,9 @@ 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,
api.clone(),
&object,
data,
data.clone(),
&ObjectOptions {
max_parity: true,
http_preconditions: Some(HTTPPreconditions {
@@ -1604,7 +1624,8 @@ pub async fn save_manual_transition_scope_admission_if_absent(
..Default::default()
},
)
.await
.await?;
api.record_durable_ilm_decommission_progress(&object, &data).await
}
pub async fn load_manual_transition_scope_admission(
@@ -1642,9 +1663,9 @@ 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,
api.clone(),
&object,
data,
data.clone(),
&ObjectOptions {
max_parity: true,
http_preconditions: Some(HTTPPreconditions {
@@ -1660,7 +1681,8 @@ 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(
@@ -1953,13 +1975,15 @@ pub async fn delete_manual_transition_scope_admission_if_current(
job_id: Uuid,
lease_id: Uuid,
) -> EcstoreResult<bool> {
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,
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),
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),
@@ -2577,6 +2601,25 @@ 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();
@@ -16,6 +16,7 @@ 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;
@@ -31,3 +32,8 @@ 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,
};
@@ -20,6 +20,7 @@ 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,
@@ -49,7 +50,7 @@ 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 = "ilm/tier-delete-journal/";
pub(crate) const TIER_DELETE_JOURNAL_PREFIX: &str = TIER_DELETE_JOURNAL_NAMESPACE.prefix;
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Eq)]
#[serde(deny_unknown_fields)]
@@ -432,6 +433,21 @@ 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
}
@@ -21,6 +21,7 @@ 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,
@@ -42,7 +43,7 @@ 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 = "ilm/transition-transactions/records";
pub const TRANSITION_TRANSACTION_RECORD_PREFIX: &str = TRANSITION_TRANSACTION_NAMESPACE.prefix;
pub const MAX_TRANSITION_TRANSACTION_SIZE: usize = 64 * 1024;
pub type Result<T> = std::result::Result<T, TransitionTransactionError>;
@@ -584,7 +585,8 @@ 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, &object, data).await
config_boundary::save_config(api.clone(), &object, data.clone()).await?;
api.record_durable_ilm_decommission_progress(&object, &data).await
}
pub(crate) async fn load_transition_transaction_record(
@@ -596,8 +598,14 @@ 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_id: Uuid) -> EcstoreResult<()> {
let object = transition_transaction_record_object_name(transaction_id).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?;
match config_boundary::delete_config(api, &object).await {
Ok(()) | Err(Error::ConfigNotFound) => Ok(()),
Err(err) => Err(err),
@@ -813,7 +821,7 @@ pub async fn finalize_missing_transition_transaction_for_operator(
if probe != TransitionOperatorProbe::Missing {
return Err(TransitionOperatorError::CandidateNotMissing(probe));
}
delete_transition_transaction_record(api, transaction_id)
delete_transition_transaction_record(api, &transaction)
.await
.map_err(TransitionOperatorError::Store)
}
@@ -849,22 +857,22 @@ 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.transaction_id).await?;
delete_transition_transaction_record(api, transaction).await?;
Ok(TransitionTransactionRecoveryOutcome::RemoteCandidateDeleted)
}
TransitionTransactionState::CleanupPending => match local_commit_matches_transaction(api.clone(), transaction).await {
Ok(true) => {
delete_transition_transaction_record(api, transaction.transaction_id).await?;
delete_transition_transaction_record(api, transaction).await?;
Ok(TransitionTransactionRecoveryOutcome::RecordDeleted)
}
Ok(false) => {
delete_transition_remote_candidate(api.clone(), transaction).await?;
delete_transition_transaction_record(api, transaction.transaction_id).await?;
delete_transition_transaction_record(api, transaction).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.transaction_id).await?;
delete_transition_transaction_record(api, transaction).await?;
Ok(TransitionTransactionRecoveryOutcome::RemoteCandidateDeleted)
}
Err(err) => Err(err),
@@ -872,7 +880,7 @@ 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.transaction_id).await?;
delete_transition_transaction_record(api, transaction).await?;
Ok(TransitionTransactionRecoveryOutcome::RecordDeleted)
}
Ok(false) => Ok(TransitionTransactionRecoveryOutcome::Retained),
@@ -881,7 +889,7 @@ pub async fn process_transition_transaction_record(
}
}
TransitionTransactionState::AbortedNoRemote | TransitionTransactionState::Committed => {
delete_transition_transaction_record(api, transaction.transaction_id).await?;
delete_transition_transaction_record(api, transaction).await?;
Ok(TransitionTransactionRecoveryOutcome::RecordDeleted)
}
TransitionTransactionState::UploadOutcomeUnknown => recover_unknown_upload_outcome(api, transaction).await,
@@ -907,7 +915,7 @@ async fn recover_unknown_upload_outcome(
.map_err(Error::other)?
{
TransitionCandidateProbe::Missing => {
delete_transition_transaction_record(api, transaction.transaction_id).await?;
delete_transition_transaction_record(api, transaction).await?;
Ok(TransitionTransactionRecoveryOutcome::RecordDeleted)
}
TransitionCandidateProbe::UnversionedPresent => {
@@ -925,7 +933,7 @@ async fn recover_unknown_upload_outcome(
)
.await
.map_err(Error::other)?;
delete_transition_transaction_record(api, transaction.transaction_id).await?;
delete_transition_transaction_record(api, transaction).await?;
Ok(TransitionTransactionRecoveryOutcome::RemoteCandidateDeleted)
}
TransitionCandidateProbe::VersionedPresent(version_id) => {
@@ -958,7 +966,7 @@ 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.transaction_id).await?;
delete_transition_transaction_record(api, &cleanup).await?;
Ok(TransitionTransactionRecoveryOutcome::RemoteCandidateDeleted)
}
+19
View File
@@ -406,6 +406,25 @@ 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`.
File diff suppressed because it is too large Load Diff
+42 -4
View File
@@ -315,7 +315,7 @@ pub async fn fsync_dir(dir: impl AsRef<Path>) -> io::Result<()> {
#[cfg(unix)]
{
let dir = dir.as_ref().to_path_buf();
tokio::task::spawn_blocking(move || fsync_dir_std(dir)).await?
fsync_spawn_blocking(move || fsync_dir_std(dir)).await?
}
#[cfg(not(unix))]
@@ -683,7 +683,7 @@ async fn fsync_open_dst_dir_group(group: &DstDirFsyncGroup) -> io::Result<()> {
#[cfg(test)]
let dir = group.dir.clone();
let dir_file = group.dir_file.clone();
tokio::task::spawn_blocking(move || {
fsync_spawn_blocking(move || {
#[cfg(test)]
{
if let Some(kind) = fsync_dir_recorder::take_grouped_failure(&dir) {
@@ -1080,6 +1080,44 @@ const TEST_GLOBAL_FILE_SYNCS: usize = 64;
static FILE_SYNC_PERMITS: LazyLock<Semaphore> = LazyLock::new(|| Semaphore::new(global_file_sync_limit()));
static DISK_FILE_SYNC_LIMITERS: LazyLock<Mutex<HashMap<PathBuf, Weak<Semaphore>>>> = LazyLock::new(|| Mutex::new(HashMap::new()));
/// Dedicated tokio runtime for fsync/fdatasync blocking operations. When
/// configured with >1 threads, isolates device-bound fsync from the main
/// blocking pool so reads (pread/stat/open) are not starved. `None` means
/// fall back to the main runtime (zero behavior change).
static FSYNC_RUNTIME: LazyLock<Option<tokio::runtime::Runtime>> = LazyLock::new(|| {
let threads =
rustfs_utils::get_env_usize(rustfs_config::ENV_FSYNC_BLOCKING_THREADS, rustfs_config::DEFAULT_FSYNC_BLOCKING_THREADS);
if threads <= 1 {
return None;
}
let mut builder = tokio::runtime::Builder::new_multi_thread();
builder
.worker_threads(num_cpus::get().min(8))
.max_blocking_threads(threads)
.thread_name("rustfs-fsync")
.thread_stack_size(512 * 1024)
.enable_all();
match builder.build() {
Ok(rt) => {
tracing::info!(threads, "fsync dedicated blocking pool enabled");
Some(rt)
}
Err(err) => {
tracing::warn!(%err, "failed to build fsync runtime, falling back to main pool");
None
}
}
});
/// Spawn a blocking task on the fsync-dedicated runtime if configured,
/// otherwise fall back to the main tokio blocking pool.
fn fsync_spawn_blocking<T: Send + 'static>(f: impl FnOnce() -> T + Send + 'static) -> tokio::task::JoinHandle<T> {
match FSYNC_RUNTIME.as_ref() {
Some(rt) => rt.spawn_blocking(f),
None => tokio::task::spawn_blocking(f),
}
}
static DISK_VOLUME_MUTATION_LOCKS: LazyLock<Mutex<HashMap<PathBuf, Weak<RwLock<()>>>>> =
LazyLock::new(|| Mutex::new(HashMap::new()));
type NamespaceMutationLock = AsyncMutex<()>;
@@ -1217,7 +1255,7 @@ where
F: FnOnce() -> io::Result<T> + Send + 'static,
{
let (disk_permit, global_permit) = acquire_file_sync_permits(disk_permits).await?;
let result = tokio::task::spawn_blocking(move || {
let result = fsync_spawn_blocking(move || {
let _disk_permit = disk_permit;
work()
})
@@ -2146,7 +2184,7 @@ async fn run_blocking_namespace_file_sync_operation_with_global<T: Send + 'stati
wait_started,
);
let disk_permit = admission.disk_permit.clone();
let result = tokio::task::spawn_blocking(move || {
let result = fsync_spawn_blocking(move || {
let _lease = lease;
let _disk_permit = disk_permit;
operation()
+7 -2
View File
@@ -37,7 +37,7 @@ use crate::bucket::lifecycle::{
transition_transaction::{
TransitionRemoteVersion, TransitionSourceIdentity, TransitionSourceVersionMode, TransitionTransaction,
TransitionTransactionInit, TransitionTransactionState, delete_transition_transaction_record,
save_transition_transaction_record,
load_transition_transaction_record, save_transition_transaction_record,
},
};
use crate::bucket::quota::reservation;
@@ -4246,7 +4246,12 @@ 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 {
return delete_transition_transaction_record(api.clone(), transaction_id).await;
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;
}
Ok(())
}
File diff suppressed because it is too large Load Diff
+32 -138
View File
@@ -12,18 +12,16 @@
// See the License for the specific language governing permissions and
// limitations under the License.
use std::{collections::HashMap, time::Duration};
use jiff::Timestamp;
use serde::{Deserialize, Serialize};
use crate::heal_commands::HealResultItem;
/// Bitflag helper for service trace categories.
///
/// Each variant occupies a single bit so that a `TraceType` value can represent
/// an arbitrary combination of categories via bitwise OR.
#[derive(Debug, Clone, Copy, Serialize, Deserialize, Default)]
pub struct TraceType(u64);
impl TraceType {
// Define some constants
pub const OS: TraceType = TraceType(1 << 0);
pub const STORAGE: TraceType = TraceType(1 << 1);
pub const S3: TraceType = TraceType(1 << 2);
@@ -40,15 +38,13 @@ impl TraceType {
pub const FTP: TraceType = TraceType(1 << 13);
pub const ILM: TraceType = TraceType(1 << 14);
// MetricsAll must be last.
/// All trace categories combined. Must be updated when adding new variants.
pub const ALL: TraceType = TraceType((1 << 15) - 1);
pub fn new(t: u64) -> Self {
Self(t)
}
}
impl TraceType {
pub fn contains(&self, x: &TraceType) -> bool {
(self.0 & x.0) == x.0
}
@@ -76,140 +72,38 @@ impl TraceType {
}
}
#[derive(Debug, Default, Clone, Serialize, Deserialize)]
pub struct TraceInfo {
#[serde(rename = "type")]
trace_type: u64,
#[serde(rename = "nodename")]
node_name: String,
#[serde(rename = "funcname")]
func_name: String,
#[serde(rename = "time")]
time: Timestamp,
#[serde(rename = "path")]
path: String,
#[serde(rename = "dur")]
duration: Duration,
#[serde(rename = "bytes", skip_serializing_if = "Option::is_none")]
bytes: Option<i64>,
#[serde(rename = "msg", skip_serializing_if = "Option::is_none")]
message: Option<String>,
#[serde(rename = "error", skip_serializing_if = "Option::is_none")]
error: Option<String>,
#[serde(rename = "custom", skip_serializing_if = "Option::is_none")]
custom: Option<HashMap<String, String>>,
#[serde(rename = "http", skip_serializing_if = "Option::is_none")]
http: Option<TraceHTTPStats>,
#[serde(rename = "healResult", skip_serializing_if = "Option::is_none")]
heal_result: Option<HealResultItem>,
}
impl TraceInfo {
pub fn mask(&self) -> u64 {
TraceType::new(self.trace_type).mask()
}
}
#[derive(Debug, Default, Clone, Serialize, Deserialize)]
pub struct TraceInfoLegacy {
trace_info: TraceInfo,
#[serde(rename = "request")]
req_info: Option<TraceRequestInfo>,
#[serde(rename = "response")]
resp_info: Option<TraceResponseInfo>,
#[serde(rename = "stats")]
call_stats: Option<TraceCallStats>,
#[serde(rename = "storageStats")]
storage_stats: Option<StorageStats>,
#[serde(rename = "osStats")]
os_stats: Option<OSStats>,
}
#[derive(Debug, Default, Clone, Serialize, Deserialize)]
pub struct StorageStats {
path: String,
duration: Duration,
}
#[derive(Debug, Default, Clone, Serialize, Deserialize)]
pub struct OSStats {
path: String,
duration: Duration,
}
#[derive(Debug, Default, Clone, Serialize, Deserialize)]
pub struct TraceHTTPStats {
req_info: TraceRequestInfo,
resp_info: TraceResponseInfo,
call_stats: TraceCallStats,
}
#[derive(Debug, Default, Clone, Serialize, Deserialize)]
pub struct TraceCallStats {
input_bytes: i32,
output_bytes: i32,
latency: Duration,
time_to_first_byte: Duration,
}
#[derive(Debug, Default, Clone, Serialize, Deserialize)]
pub struct TraceRequestInfo {
time: Timestamp,
proto: String,
method: String,
#[serde(skip_serializing_if = "Option::is_none")]
path: Option<String>,
#[serde(skip_serializing_if = "Option::is_none")]
raw_query: Option<String>,
#[serde(skip_serializing_if = "Option::is_none")]
headers: Option<HashMap<String, String>>,
#[serde(skip_serializing_if = "Option::is_none")]
body: Option<Vec<u8>>,
client: String,
}
#[derive(Debug, Default, Clone, Serialize, Deserialize)]
pub struct TraceResponseInfo {
time: Timestamp,
#[serde(skip_serializing_if = "Option::is_none")]
headers: Option<HashMap<String, String>>,
#[serde(skip_serializing_if = "Option::is_none")]
body: Option<Vec<u8>>,
#[serde(skip_serializing_if = "Option::is_none")]
status_code: Option<i32>,
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn trace_timestamps_serialize_as_rfc3339_utc() {
let timestamp = Timestamp::constant(1_700_000_000, 123_456_000);
let trace = TraceInfo {
time: timestamp,
http: Some(TraceHTTPStats {
req_info: TraceRequestInfo {
time: timestamp,
..Default::default()
},
resp_info: TraceResponseInfo {
time: timestamp,
..Default::default()
},
..Default::default()
}),
..Default::default()
};
fn trace_type_contains_and_overlaps() {
let mut combined = TraceType::default();
combined.merge(&TraceType::S3);
combined.merge(&TraceType::HEALING);
let value = serde_json::to_value(trace).expect("trace should serialize");
assert_eq!(value["time"], "2023-11-14T22:13:20.123456Z");
assert_eq!(value["http"]["req_info"]["time"], "2023-11-14T22:13:20.123456Z");
assert_eq!(value["http"]["resp_info"]["time"], "2023-11-14T22:13:20.123456Z");
let trace: TraceInfo = serde_json::from_value(value).expect("trace should deserialize");
assert_eq!(trace.time, timestamp);
let http = trace.http.expect("http trace should deserialize");
assert_eq!(http.req_info.time, timestamp);
assert_eq!(http.resp_info.time, timestamp);
assert!(combined.contains(&TraceType::S3));
assert!(combined.contains(&TraceType::HEALING));
assert!(!combined.contains(&TraceType::SCANNER));
assert!(combined.overlaps(&TraceType::S3));
assert!(combined.overlaps(&TraceType::HEALING));
assert!(!combined.overlaps(&TraceType::SCANNER));
}
#[test]
fn trace_type_set_if() {
let mut tt = TraceType::default();
tt.set_if(true, &TraceType::OS);
tt.set_if(false, &TraceType::S3);
assert!(tt.contains(&TraceType::OS));
assert!(!tt.contains(&TraceType::S3));
}
#[test]
fn trace_type_single_type() {
assert!(TraceType::S3.single_type());
let mut combined = TraceType::S3;
combined.merge(&TraceType::HEALING);
assert!(!combined.single_type());
}
}
+26 -1
View File
@@ -206,6 +206,13 @@ def check_runner_selection(root: Path) -> list[str]:
return errors
def check_s3_tests_runner(root: Path) -> list[str]:
runner = (root / "scripts/s3-tests/run.sh").read_text()
if "--showlocals" in runner:
return ["scripts/s3-tests/run.sh: pytest failure diagnostics must not dump local values"]
return []
def profile_selection(root: Path, profile: str) -> str:
if not re.fullmatch(r"e2e-[a-z0-9-]+", profile):
raise ValueError(f"invalid e2e profile name: {profile}")
@@ -272,6 +279,7 @@ def validate(root: Path) -> list[str]:
errors.extend(check_e2e_modules(root))
errors.extend(check_fuzz_targets(root))
errors.extend(check_runner_selection(root))
errors.extend(check_s3_tests_runner(root))
errors.extend(check_profile_definitions(root))
return errors
@@ -341,6 +349,23 @@ class SelfTests(unittest.TestCase):
)
self.assertEqual(len(check_fuzz_targets(root)), 1)
def test_s3_runner_rejects_unbounded_failure_locals(self) -> None:
with tempfile.TemporaryDirectory() as tmp:
root = Path(tmp)
runner = root / "scripts/s3-tests/run.sh"
runner.parent.mkdir(parents=True)
runner.write_text("tox -- -vv -ra --tb=long\n")
self.assertEqual(check_s3_tests_runner(root), [])
runner.write_text("tox -- -vv -ra --showlocals --tb=long\n")
self.assertEqual(len(check_s3_tests_runner(root)), 1)
with (
mock.patch(__name__ + ".check_e2e_modules", return_value=[]),
mock.patch(__name__ + ".check_fuzz_targets", return_value=[]),
mock.patch(__name__ + ".check_runner_selection", return_value=[]),
mock.patch(__name__ + ".check_profile_definitions", return_value=[]),
):
self.assertEqual(len(validate(root)), 1)
def test_profile_listing_enforces_selection(self) -> None:
with tempfile.TemporaryDirectory() as tmp:
root = Path(tmp)
@@ -411,7 +436,7 @@ def main() -> int:
for error in errors:
print(f"ERROR: {error}", file=sys.stderr)
return 1
print("OK: e2e modules, runner selection, fuzz matrices, and profile guards are wired")
print("OK: e2e modules, runner selection, fuzz matrices, profiles, and bounded diagnostics are wired")
return 0
+2 -1
View File
@@ -1028,10 +1028,11 @@ else
fi
# Run tests from s3tests/functional
# Failure locals can contain multi-MiB request bodies; keep tracebacks without expanding local values.
set +e
S3TEST_CONF="${CONF_OUTPUT_PATH}" \
tox -- \
-vv -ra --showlocals --tb=long \
-vv -ra --tb=long \
--maxfail="${MAXFAIL}" \
--timeout="${TEST_TIMEOUT}" \
--junitxml="${ARTIFACTS_DIR}/junit.xml" \