Files
rustfs/crates/ecstore/src/disk/disk_store.rs
T

805 lines
27 KiB
Rust

// Copyright 2024 RustFS Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
use crate::disk::{
CheckPartsResp, DeleteOptions, DiskAPI, DiskError, DiskInfo, DiskInfoOptions, DiskLocation, Endpoint, Error,
FileInfoVersions, ReadMultipleReq, ReadMultipleResp, ReadOptions, RenameDataResp, Result, UpdateMetadataOpts, VolumeInfo,
WalkDirOptions,
local::{LocalDisk, ScanGuard},
};
use crate::global::GLOBAL_LOCAL_DISK_ID_MAP;
use bytes::Bytes;
use rustfs_filemeta::{FileInfo, ObjectPartInfo, RawFileInfo};
use std::{
path::PathBuf,
sync::{
Arc,
atomic::{AtomicI64, AtomicU32, Ordering},
},
time::Duration,
};
use tokio::{sync::RwLock, time};
use tokio_util::sync::CancellationToken;
use tracing::{info, warn};
use uuid::Uuid;
/// Disk health status constants
const DISK_HEALTH_OK: u32 = 0;
const DISK_HEALTH_FAULTY: u32 = 1;
pub const ENV_RUSTFS_DRIVE_ACTIVE_MONITORING: &str = "RUSTFS_DRIVE_ACTIVE_MONITORING";
pub const DEFAULT_RUSTFS_DRIVE_ACTIVE_MONITORING: bool = true;
pub const ENV_RUSTFS_DRIVE_MAX_TIMEOUT_DURATION: &str = "RUSTFS_DRIVE_MAX_TIMEOUT_DURATION";
pub const CHECK_EVERY: Duration = Duration::from_secs(15);
pub const SKIP_IF_SUCCESS_BEFORE: Duration = Duration::from_secs(5);
pub const CHECK_TIMEOUT_DURATION: Duration = Duration::from_secs(5);
lazy_static::lazy_static! {
static ref TEST_DATA: Bytes = Bytes::from(vec![42u8; 2048]);
static ref TEST_BUCKET: String = ".rustfs.sys/tmp".to_string();
}
pub fn get_max_timeout_duration() -> Duration {
std::env::var(ENV_RUSTFS_DRIVE_MAX_TIMEOUT_DURATION)
.map(|v| Duration::from_secs(v.parse::<u64>().unwrap_or(30)))
.unwrap_or(Duration::from_secs(30))
}
/// DiskHealthTracker tracks the health status of a disk.
/// Similar to Go's diskHealthTracker.
#[derive(Debug)]
pub struct DiskHealthTracker {
/// Atomic timestamp of last successful operation
pub last_success: AtomicI64,
/// Atomic timestamp of last operation start
pub last_started: AtomicI64,
/// Atomic disk status (OK or Faulty)
pub status: AtomicU32,
/// Atomic number of waiting operations
pub waiting: AtomicU32,
}
impl DiskHealthTracker {
/// Create a new disk health tracker
pub fn new() -> Self {
let now = std::time::SystemTime::now()
.duration_since(std::time::UNIX_EPOCH)
.unwrap()
.as_nanos() as i64;
Self {
last_success: AtomicI64::new(now),
last_started: AtomicI64::new(now),
status: AtomicU32::new(DISK_HEALTH_OK),
waiting: AtomicU32::new(0),
}
}
/// Log a successful operation
pub fn log_success(&self) {
let now = std::time::SystemTime::now()
.duration_since(std::time::UNIX_EPOCH)
.unwrap()
.as_nanos() as i64;
self.last_success.store(now, Ordering::Relaxed);
}
/// Check if disk is faulty
pub fn is_faulty(&self) -> bool {
self.status.load(Ordering::Acquire) == DISK_HEALTH_FAULTY
}
/// Set disk as faulty
pub fn set_faulty(&self) {
self.status.store(DISK_HEALTH_FAULTY, Ordering::Release);
}
/// Set disk as OK
pub fn set_ok(&self) {
self.status.store(DISK_HEALTH_OK, Ordering::Release);
}
pub fn swap_ok_to_faulty(&self) -> bool {
self.status
.compare_exchange(DISK_HEALTH_OK, DISK_HEALTH_FAULTY, Ordering::AcqRel, Ordering::Relaxed)
.is_ok()
}
/// Increment waiting operations counter
pub fn increment_waiting(&self) {
self.waiting.fetch_add(1, Ordering::Relaxed);
}
/// Decrement waiting operations counter
pub fn decrement_waiting(&self) {
self.waiting.fetch_sub(1, Ordering::Relaxed);
}
/// Get waiting operations count
pub fn waiting_count(&self) -> u32 {
self.waiting.load(Ordering::Relaxed)
}
/// Get last success timestamp
pub fn last_success(&self) -> i64 {
self.last_success.load(Ordering::Acquire)
}
}
impl Default for DiskHealthTracker {
fn default() -> Self {
Self::new()
}
}
/// Health check context key for tracking disk operations
#[derive(Debug, Clone)]
struct HealthDiskCtxKey;
#[derive(Debug)]
struct HealthDiskCtxValue {
last_success: Arc<AtomicI64>,
}
impl HealthDiskCtxValue {
fn log_success(&self) {
let now = std::time::SystemTime::now()
.duration_since(std::time::UNIX_EPOCH)
.unwrap()
.as_nanos() as i64;
self.last_success.store(now, Ordering::Relaxed);
}
}
/// LocalDiskWrapper wraps a DiskStore with health tracking capabilities.
/// This is similar to Go's xlStorageDiskIDCheck.
#[derive(Debug, Clone)]
pub struct LocalDiskWrapper {
/// The underlying disk store
disk: Arc<LocalDisk>,
/// Health tracker
health: Arc<DiskHealthTracker>,
/// Whether health checking is enabled
health_check: bool,
/// Cancellation token for monitoring tasks
cancel_token: CancellationToken,
/// Disk ID for stale checking
disk_id: Arc<RwLock<Option<Uuid>>>,
}
impl LocalDiskWrapper {
/// Create a new LocalDiskWrapper
pub fn new(disk: Arc<LocalDisk>, health_check: bool) -> Self {
// Check environment variable for health check override.
// Only enable if both param and env are true.
let env_health_check =
rustfs_utils::get_env_bool(ENV_RUSTFS_DRIVE_ACTIVE_MONITORING, DEFAULT_RUSTFS_DRIVE_ACTIVE_MONITORING);
Self {
disk,
health: Arc::new(DiskHealthTracker::new()),
health_check: health_check && env_health_check,
cancel_token: CancellationToken::new(),
disk_id: Arc::new(RwLock::new(None)),
}
}
pub fn get_disk(&self) -> Arc<LocalDisk> {
self.disk.clone()
}
/// Enable health monitoring after disk creation.
/// Used to defer health checks until after startup format loading completes.
pub fn enable_health_check(&self) {
if !self.health_check {
return;
}
let health = Arc::clone(&self.health);
let cancel_token = self.cancel_token.clone();
let disk = Arc::clone(&self.disk);
tokio::spawn(async move {
Self::monitor_disk_writable(disk, health, cancel_token).await;
});
}
/// Stop the disk monitoring
pub async fn stop_monitoring(&self) {
self.cancel_token.cancel();
}
/// Monitor disk writability periodically
async fn monitor_disk_writable(disk: Arc<LocalDisk>, health: Arc<DiskHealthTracker>, cancel_token: CancellationToken) {
// TODO: config interval
let mut interval = time::interval(CHECK_EVERY);
loop {
tokio::select! {
_ = cancel_token.cancelled() => {
return;
}
_ = interval.tick() => {
if cancel_token.is_cancelled() {
return;
}
if health.status.load(Ordering::Relaxed) != DISK_HEALTH_OK {
continue;
}
let last_success_nanos = health.last_success.load(Ordering::Relaxed);
let elapsed = Duration::from_nanos(
(std::time::SystemTime::now()
.duration_since(std::time::UNIX_EPOCH)
.unwrap()
.as_nanos() as i64 - last_success_nanos) as u64
);
if elapsed < SKIP_IF_SUCCESS_BEFORE {
continue;
}
tokio::time::sleep(Duration::from_secs(1)).await;
let test_obj = format!("health-check-{}", Uuid::new_v4());
if Self::perform_health_check(disk.clone(), &TEST_BUCKET, &test_obj, &TEST_DATA, true, CHECK_TIMEOUT_DURATION).await.is_err() && health.swap_ok_to_faulty() {
// Health check failed, disk is considered faulty
warn!("health check: failed, disk is considered faulty");
health.increment_waiting(); // Balance the increment from failed operation
let health_clone = Arc::clone(&health);
let disk_clone = disk.clone();
let cancel_clone = cancel_token.clone();
tokio::spawn(async move {
Self::monitor_disk_status(disk_clone, health_clone, cancel_clone).await;
});
}
}
}
}
}
/// Perform a health check by writing and reading a test file
async fn perform_health_check(
disk: Arc<LocalDisk>,
test_bucket: &str,
test_filename: &str,
test_data: &Bytes,
check_faulty_only: bool,
timeout_duration: Duration,
) -> Result<()> {
// Perform health check with timeout
let health_check_result = tokio::time::timeout(timeout_duration, async {
// Try to write test data
disk.write_all(test_bucket, test_filename, test_data.clone()).await?;
// Try to read back the data
let read_data = disk.read_all(test_bucket, test_filename).await?;
// Verify data integrity
if read_data.len() != test_data.len() {
warn!(
"health check: test file data length mismatch: expected {} bytes, got {}",
test_data.len(),
read_data.len()
);
if check_faulty_only {
return Ok(());
}
return Err(DiskError::FaultyDisk);
}
// Clean up
disk.delete(
test_bucket,
test_filename,
DeleteOptions {
recursive: false,
immediate: false,
undo_write: false,
old_data_dir: None,
},
)
.await?;
Ok(())
})
.await;
match health_check_result {
Ok(result) => match result {
Ok(()) => Ok(()),
Err(e) => {
warn!("health check: failed: {:?}", e);
if e == DiskError::FaultyDisk {
return Err(e);
}
if check_faulty_only { Ok(()) } else { Err(e) }
}
},
Err(_) => {
// Timeout occurred
warn!("health check: timeout after {:?}", timeout_duration);
Err(DiskError::FaultyDisk)
}
}
}
/// Monitor disk status and try to bring it back online
async fn monitor_disk_status(disk: Arc<LocalDisk>, health: Arc<DiskHealthTracker>, cancel_token: CancellationToken) {
const CHECK_EVERY: Duration = Duration::from_secs(5);
let mut interval = time::interval(CHECK_EVERY);
loop {
tokio::select! {
_ = cancel_token.cancelled() => {
return;
}
_ = interval.tick() => {
if cancel_token.is_cancelled() {
return;
}
let test_obj = format!("health-check-{}", Uuid::new_v4());
match Self::perform_health_check(disk.clone(), &TEST_BUCKET, &test_obj, &TEST_DATA, false, CHECK_TIMEOUT_DURATION).await {
Ok(_) => {
info!("Disk {} is back online", disk.to_string());
health.set_ok();
health.decrement_waiting();
return;
}
Err(e) => {
warn!("Disk {} still faulty: {:?}", disk.to_string(), e);
}
}
}
}
}
}
async fn check_id(&self, want_id: Option<Uuid>) -> Result<()> {
if want_id.is_none() {
return Ok(());
}
let stored_disk_id = self.disk.get_disk_id().await?;
if stored_disk_id != want_id {
return Err(Error::other(format!("Disk ID mismatch wanted {want_id:?}, got {stored_disk_id:?}")));
}
Ok(())
}
/// Check if disk ID is stale
async fn check_disk_stale(&self) -> Result<()> {
let Some(current_disk_id) = *self.disk_id.read().await else {
return Ok(());
};
let stored_disk_id = match self.disk.get_disk_id().await? {
Some(id) => id,
None => return Ok(()), // Empty disk ID is allowed during initialization
};
if current_disk_id != stored_disk_id {
return Err(DiskError::DiskNotFound);
}
Ok(())
}
/// Set the disk ID
pub async fn set_disk_id_internal(&self, id: Option<Uuid>) -> Result<()> {
let mut disk_id = self.disk_id.write().await;
let previous = *disk_id;
*disk_id = id;
drop(disk_id);
if self.disk.is_local() {
let mut disk_id_map = GLOBAL_LOCAL_DISK_ID_MAP.write().await;
if let Some(previous_id) = previous {
disk_id_map.remove(&previous_id);
}
if let Some(current_id) = id {
disk_id_map.insert(current_id, self.disk.endpoint().to_string());
}
}
Ok(())
}
/// Get the current disk ID
pub async fn get_current_disk_id(&self) -> Option<Uuid> {
*self.disk_id.read().await
}
/// Track disk health for an operation.
/// This method should wrap disk operations to ensure health checking.
pub async fn track_disk_health<T, F, Fut>(&self, operation: F, timeout_duration: Duration) -> Result<T>
where
F: FnOnce() -> Fut,
Fut: std::future::Future<Output = Result<T>>,
{
// Check if disk is faulty
if self.health.is_faulty() {
warn!("local disk {} health is faulty, returning error", self.to_string());
return Err(DiskError::FaultyDisk);
}
// Check if disk is stale
self.check_disk_stale().await?;
// Record operation start
let now = std::time::SystemTime::now()
.duration_since(std::time::UNIX_EPOCH)
.unwrap()
.as_nanos() as i64;
self.health.last_started.store(now, Ordering::Relaxed);
self.health.increment_waiting();
if timeout_duration == Duration::ZERO {
let result = operation().await;
self.health.decrement_waiting();
if result.is_ok() {
self.health.log_success();
}
return result;
}
// Execute the operation with timeout
let result = tokio::time::timeout(timeout_duration, operation()).await;
match result {
Ok(operation_result) => {
// Log success and decrement waiting counter
if operation_result.is_ok() {
self.health.log_success();
}
self.health.decrement_waiting();
operation_result
}
Err(_) => {
// Timeout occurred, mark disk as potentially faulty and decrement waiting counter
self.health.decrement_waiting();
warn!("disk operation timeout after {:?}", timeout_duration);
Err(DiskError::other(format!("disk operation timeout after {timeout_duration:?}")))
}
}
}
}
#[async_trait::async_trait]
impl DiskAPI for LocalDiskWrapper {
async fn read_metadata(&self, volume: &str, path: &str) -> Result<Bytes> {
self.track_disk_health(|| async { self.disk.read_metadata(volume, path).await }, Duration::ZERO)
.await
}
fn start_scan(&self) -> ScanGuard {
self.disk.start_scan()
}
fn to_string(&self) -> String {
self.disk.to_string()
}
async fn is_online(&self) -> bool {
let Ok(Some(disk_id)) = self.disk.get_disk_id().await else {
return false;
};
// if disk_id is not set use the current disk_id
if let Some(current_disk_id) = *self.disk_id.read().await {
return current_disk_id == disk_id;
} else {
// if disk_id is not set, update the disk_id
let _ = self.set_disk_id_internal(Some(disk_id)).await;
}
return true;
}
fn is_local(&self) -> bool {
self.disk.is_local()
}
fn host_name(&self) -> String {
self.disk.host_name()
}
fn endpoint(&self) -> Endpoint {
self.disk.endpoint()
}
async fn close(&self) -> Result<()> {
self.stop_monitoring().await;
self.disk.close().await
}
async fn get_disk_id(&self) -> Result<Option<Uuid>> {
self.disk.get_disk_id().await
}
async fn set_disk_id(&self, id: Option<Uuid>) -> Result<()> {
self.set_disk_id_internal(id).await
}
fn path(&self) -> PathBuf {
self.disk.path()
}
fn get_disk_location(&self) -> DiskLocation {
self.disk.get_disk_location()
}
async fn disk_info(&self, opts: &DiskInfoOptions) -> Result<DiskInfo> {
if opts.noop && opts.metrics {
let mut info = DiskInfo::default();
// Add health metrics
info.metrics.total_waiting = self.health.waiting_count();
if self.health.is_faulty() {
return Err(DiskError::FaultyDisk);
}
return Ok(info);
}
if self.health.is_faulty() {
return Err(DiskError::FaultyDisk);
}
let result = self.disk.disk_info(opts).await?;
if let Some(current_disk_id) = *self.disk_id.read().await
&& Some(current_disk_id) != result.id
{
return Err(DiskError::DiskNotFound);
};
Ok(result)
}
async fn make_volume(&self, volume: &str) -> Result<()> {
self.track_disk_health(|| async { self.disk.make_volume(volume).await }, get_max_timeout_duration())
.await
}
async fn make_volumes(&self, volumes: Vec<&str>) -> Result<()> {
self.track_disk_health(|| async { self.disk.make_volumes(volumes).await }, get_max_timeout_duration())
.await
}
async fn list_volumes(&self) -> Result<Vec<VolumeInfo>> {
self.track_disk_health(|| async { self.disk.list_volumes().await }, Duration::ZERO)
.await
}
async fn stat_volume(&self, volume: &str) -> Result<VolumeInfo> {
self.track_disk_health(|| async { self.disk.stat_volume(volume).await }, get_max_timeout_duration())
.await
}
async fn delete_volume(&self, volume: &str) -> Result<()> {
self.track_disk_health(|| async { self.disk.delete_volume(volume).await }, Duration::ZERO)
.await
}
async fn walk_dir<W: tokio::io::AsyncWrite + Unpin + Send>(&self, opts: WalkDirOptions, wr: &mut W) -> Result<()> {
self.track_disk_health(|| async { self.disk.walk_dir(opts, wr).await }, Duration::ZERO)
.await
}
async fn delete_version(
&self,
volume: &str,
path: &str,
fi: FileInfo,
force_del_marker: bool,
opts: DeleteOptions,
) -> Result<()> {
self.track_disk_health(
|| async { self.disk.delete_version(volume, path, fi, force_del_marker, opts).await },
get_max_timeout_duration(),
)
.await
}
async fn delete_versions(&self, volume: &str, versions: Vec<FileInfoVersions>, opts: DeleteOptions) -> Vec<Option<Error>> {
// Check if disk is faulty before proceeding
if self.health.is_faulty() {
return vec![Some(DiskError::FaultyDisk); versions.len()];
}
// Check if disk is stale
if let Err(e) = self.check_disk_stale().await {
return vec![Some(e); versions.len()];
}
// Record operation start
let now = std::time::SystemTime::now()
.duration_since(std::time::UNIX_EPOCH)
.unwrap()
.as_nanos() as i64;
self.health.last_started.store(now, Ordering::Relaxed);
self.health.increment_waiting();
// Execute the operation
let result = self.disk.delete_versions(volume, versions, opts).await;
self.health.decrement_waiting();
let has_err = result.iter().any(|e| e.is_some());
if !has_err {
// Log success and decrement waiting counter
self.health.log_success();
}
result
}
async fn delete_paths(&self, volume: &str, paths: &[String]) -> Result<()> {
self.track_disk_health(|| async { self.disk.delete_paths(volume, paths).await }, get_max_timeout_duration())
.await
}
async fn write_metadata(&self, org_volume: &str, volume: &str, path: &str, fi: FileInfo) -> Result<()> {
self.track_disk_health(
|| async { self.disk.write_metadata(org_volume, volume, path, fi).await },
get_max_timeout_duration(),
)
.await
}
async fn update_metadata(&self, volume: &str, path: &str, fi: FileInfo, opts: &UpdateMetadataOpts) -> Result<()> {
self.track_disk_health(
|| async { self.disk.update_metadata(volume, path, fi, opts).await },
get_max_timeout_duration(),
)
.await
}
async fn read_version(
&self,
org_volume: &str,
volume: &str,
path: &str,
version_id: &str,
opts: &ReadOptions,
) -> Result<FileInfo> {
self.track_disk_health(
|| async { self.disk.read_version(org_volume, volume, path, version_id, opts).await },
get_max_timeout_duration(),
)
.await
}
async fn read_xl(&self, volume: &str, path: &str, read_data: bool) -> Result<RawFileInfo> {
self.track_disk_health(|| async { self.disk.read_xl(volume, path, read_data).await }, get_max_timeout_duration())
.await
}
async fn rename_data(
&self,
src_volume: &str,
src_path: &str,
fi: FileInfo,
dst_volume: &str,
dst_path: &str,
) -> Result<RenameDataResp> {
self.track_disk_health(
|| async { self.disk.rename_data(src_volume, src_path, fi, dst_volume, dst_path).await },
get_max_timeout_duration(),
)
.await
}
async fn list_dir(&self, origvolume: &str, volume: &str, dir_path: &str, count: i32) -> Result<Vec<String>> {
self.track_disk_health(
|| async { self.disk.list_dir(origvolume, volume, dir_path, count).await },
get_max_timeout_duration(),
)
.await
}
async fn read_file(&self, volume: &str, path: &str) -> Result<crate::disk::FileReader> {
self.track_disk_health(|| async { self.disk.read_file(volume, path).await }, get_max_timeout_duration())
.await
}
async fn read_file_stream(&self, volume: &str, path: &str, offset: usize, length: usize) -> Result<crate::disk::FileReader> {
self.track_disk_health(
|| async { self.disk.read_file_stream(volume, path, offset, length).await },
get_max_timeout_duration(),
)
.await
}
async fn read_file_zero_copy(&self, volume: &str, path: &str, offset: usize, length: usize) -> Result<bytes::Bytes> {
self.track_disk_health(
|| async { self.disk.read_file_zero_copy(volume, path, offset, length).await },
get_max_timeout_duration(),
)
.await
}
async fn append_file(&self, volume: &str, path: &str) -> Result<crate::disk::FileWriter> {
self.track_disk_health(|| async { self.disk.append_file(volume, path).await }, Duration::ZERO)
.await
}
async fn create_file(&self, origvolume: &str, volume: &str, path: &str, file_size: i64) -> Result<crate::disk::FileWriter> {
self.track_disk_health(
|| async { self.disk.create_file(origvolume, volume, path, file_size).await },
Duration::ZERO,
)
.await
}
async fn rename_file(&self, src_volume: &str, src_path: &str, dst_volume: &str, dst_path: &str) -> Result<()> {
self.track_disk_health(
|| async { self.disk.rename_file(src_volume, src_path, dst_volume, dst_path).await },
get_max_timeout_duration(),
)
.await
}
async fn rename_part(&self, src_volume: &str, src_path: &str, dst_volume: &str, dst_path: &str, meta: Bytes) -> Result<()> {
self.track_disk_health(
|| async { self.disk.rename_part(src_volume, src_path, dst_volume, dst_path, meta).await },
get_max_timeout_duration(),
)
.await
}
async fn delete(&self, volume: &str, path: &str, opt: DeleteOptions) -> Result<()> {
self.track_disk_health(|| async { self.disk.delete(volume, path, opt).await }, get_max_timeout_duration())
.await
}
async fn verify_file(&self, volume: &str, path: &str, fi: &FileInfo) -> Result<CheckPartsResp> {
self.track_disk_health(|| async { self.disk.verify_file(volume, path, fi).await }, Duration::ZERO)
.await
}
async fn check_parts(&self, volume: &str, path: &str, fi: &FileInfo) -> Result<CheckPartsResp> {
self.track_disk_health(|| async { self.disk.check_parts(volume, path, fi).await }, Duration::ZERO)
.await
}
async fn read_parts(&self, bucket: &str, paths: &[String]) -> Result<Vec<ObjectPartInfo>> {
self.track_disk_health(|| async { self.disk.read_parts(bucket, paths).await }, Duration::ZERO)
.await
}
async fn read_multiple(&self, req: ReadMultipleReq) -> Result<Vec<ReadMultipleResp>> {
self.track_disk_health(|| async { self.disk.read_multiple(req).await }, Duration::ZERO)
.await
}
async fn write_all(&self, volume: &str, path: &str, data: Bytes) -> Result<()> {
self.track_disk_health(|| async { self.disk.write_all(volume, path, data).await }, get_max_timeout_duration())
.await
}
async fn read_all(&self, volume: &str, path: &str) -> Result<Bytes> {
self.track_disk_health(|| async { self.disk.read_all(volume, path).await }, get_max_timeout_duration())
.await
}
}