From 6b1b3800d713d73d97da78ae9afe1950850f1c47 Mon Sep 17 00:00:00 2001 From: Zhengchao An Date: Wed, 24 Jun 2026 10:00:36 +0800 Subject: [PATCH] refactor: centralize ecstore owner runtime sources (#3811) --- crates/ecstore/src/notification_sys.rs | 14 +++++------- crates/ecstore/src/pools.rs | 3 +-- crates/ecstore/src/tier/tier.rs | 3 +-- docs/architecture/migration-progress.md | 29 ++++++++++++++++++++----- 4 files changed, 31 insertions(+), 18 deletions(-) diff --git a/crates/ecstore/src/notification_sys.rs b/crates/ecstore/src/notification_sys.rs index e65174300..fe604ca1b 100644 --- a/crates/ecstore/src/notification_sys.rs +++ b/crates/ecstore/src/notification_sys.rs @@ -15,10 +15,10 @@ use crate::admin_server_info::get_commit_id; use crate::endpoints::EndpointServerPools; use crate::error::{Error, Result}; -use crate::global::{GLOBAL_BOOT_TIME, get_global_endpoints, resolve_object_store_handle}; use crate::metrics_realtime::{CollectMetricsOpts, MetricType}; use crate::rebalance::RebalSaveOpt; use crate::rpc::PeerRestClient; +use crate::runtime_sources; use futures::future::join_all; use lazy_static::lazy_static; use rustfs_madmin::health::{Cpus, MemInfo, OsInfo, Partitions, ProcInfo, SysConfig, SysErrors, SysServices}; @@ -289,7 +289,7 @@ impl NotificationSys { S: StorageAdminApi, { let mut futures = Vec::with_capacity(self.peer_clients.len()); - let endpoints = get_global_endpoints(); + let endpoints = runtime_sources::endpoint_pools().unwrap_or_else(|| Vec::new().into()); let peer_timeout = Duration::from_secs(5); for (idx, client) in self.peer_clients.iter().enumerate() { @@ -334,7 +334,7 @@ impl NotificationSys { pub async fn server_info(&self) -> Vec { let mut futures = Vec::with_capacity(self.peer_clients.len()); - let endpoints = get_global_endpoints(); + let endpoints = runtime_sources::endpoint_pools().unwrap_or_else(|| Vec::new().into()); let peer_timeout = Duration::from_secs(5); for (idx, client) in self.peer_clients.iter().enumerate() { @@ -595,7 +595,7 @@ impl NotificationSys { state = "started", "notification peer propagation" ); - let Some(store) = resolve_object_store_handle() else { + let Some(store) = runtime_sources::object_store_handle() else { error!( event = EVENT_NOTIFICATION_PEER_PROPAGATION, component = LOG_COMPONENT_ECSTORE, @@ -1107,11 +1107,7 @@ fn initializing_server_properties(host: &str) -> ServerProperties { fn offline_server_properties(host: &str, endpoints: &EndpointServerPools) -> ServerProperties { ServerProperties { - uptime: GLOBAL_BOOT_TIME - .get() - .and_then(|boot_time| SystemTime::now().duration_since(*boot_time).ok()) - .unwrap_or_default() - .as_secs(), + uptime: runtime_sources::boot_uptime_secs(), version: get_commit_id(), endpoint: host.to_string(), state: ItemState::Offline.to_string().to_owned(), diff --git a/crates/ecstore/src/pools.rs b/crates/ecstore/src/pools.rs index 6ebad8d86..679be7d99 100644 --- a/crates/ecstore/src/pools.rs +++ b/crates/ecstore/src/pools.rs @@ -35,7 +35,6 @@ use crate::error::{Error, Result}; use crate::error::{ StorageError, is_err_bucket_exists, is_err_bucket_not_found, is_err_object_not_found, is_err_version_not_found, }; -use crate::global::resolve_object_store_handle; use crate::object_api::{GetObjectReader, ObjectOptions}; use crate::rebalance::{REBAL_META_NAME, RebalanceMeta, is_rebalance_conflicting_with_decommission}; use crate::runtime_sources; @@ -2420,7 +2419,7 @@ impl ECStore { self.ensure_decommission_rebalance_idle_after_refresh().await?; - let store = require_decommission_store(resolve_object_store_handle(), "start decommission")?; + let store = require_decommission_store(runtime_sources::object_store_handle(), "start decommission")?; let local_indices = local_decommission_queue_prefix(&self.endpoints(), &indices)?; self.start_decommission(indices.clone()).await?; diff --git a/crates/ecstore/src/tier/tier.rs b/crates/ecstore/src/tier/tier.rs index 8ada33bb5..93ba17b33 100644 --- a/crates/ecstore/src/tier/tier.rs +++ b/crates/ecstore/src/tier/tier.rs @@ -38,7 +38,6 @@ use tracing::{debug, error, info, warn}; use crate::client::admin_handler_utils::AdminError; use crate::error::{Error, Result, StorageError}; -use crate::global::resolve_object_store_handle; use crate::tier::{ tier_admin::TierCreds, tier_config::{TierConfig, TierType}, @@ -1026,7 +1025,7 @@ impl TierConfigMgr { #[tracing::instrument(level = "debug", name = "tier_save", skip(self))] pub async fn save(&self) -> std::result::Result<(), std::io::Error> { - let Some(api) = resolve_object_store_handle() else { + let Some(api) = runtime_sources::object_store_handle() else { return Err(tier_config_not_initialized_error("save tiering config")); }; //let (pr, opts) = GLOBAL_TierConfigMgr.write().config_reader()?; diff --git a/docs/architecture/migration-progress.md b/docs/architecture/migration-progress.md index 03e936968..e01027fd1 100644 --- a/docs/architecture/migration-progress.md +++ b/docs/architecture/migration-progress.md @@ -5,9 +5,9 @@ Status values: `[ ]` not started, `[~]` in progress, `[x]` complete, `[!]` block ## Current Context - Issue: [`rustfs/backlog#660`](https://github.com/rustfs/backlog/issues/660) -- Branch: `overtrue/arch-ecstore-remaining-runtime-sources-batch` -- Baseline: completed `C-011/C-012/C-013/API-055/API-059/API-079/API-080/API-081/API-082/API-083/API-084/API-085/API-086/API-087/API-088/API-089/API-090/API-091/API-092/API-093/API-094/API-095/API-096/API-097/API-098/API-099/API-100/API-101/API-102/API-103/API-104/API-105/API-106/API-107/API-108/API-109/API-110/API-111/API-112/API-113/API-114/API-115/API-116/API-117/API-118/API-119/API-120/API-121/API-122/API-123/API-124/API-125/API-126/API-127/API-128/API-129/API-130/API-131/API-132/API-133/API-134/API-135/API-136/API-137/API-138/API-139/API-140/API-141/API-142/API-143/API-144/API-145/API-146/API-147/API-148/API-149/API-150/API-151/API-152/API-153/API-154/API-155/API-156/API-157/API-158/API-159/API-160/API-161/API-162/API-163/API-164/API-165/API-166/API-167/API-168/API-169/API-170/API-171/API-172/API-173/API-174/API-175/API-176/API-177/API-178/API-179/API-180/API-181/API-182/API-183/API-184/API-185/API-186/API-187/API-188/API-189/API-190/API-191/API-192/API-193`. -- Based on: latest default branch after merge 3808. +- Branch: `overtrue/arch-ecstore-owner-runtime-sources-batch` +- Baseline: completed `C-011/C-012/C-013/API-055/API-059/API-079/API-080/API-081/API-082/API-083/API-084/API-085/API-086/API-087/API-088/API-089/API-090/API-091/API-092/API-093/API-094/API-095/API-096/API-097/API-098/API-099/API-100/API-101/API-102/API-103/API-104/API-105/API-106/API-107/API-108/API-109/API-110/API-111/API-112/API-113/API-114/API-115/API-116/API-117/API-118/API-119/API-120/API-121/API-122/API-123/API-124/API-125/API-126/API-127/API-128/API-129/API-130/API-131/API-132/API-133/API-134/API-135/API-136/API-137/API-138/API-139/API-140/API-141/API-142/API-143/API-144/API-145/API-146/API-147/API-148/API-149/API-150/API-151/API-152/API-153/API-154/API-155/API-156/API-157/API-158/API-159/API-160/API-161/API-162/API-163/API-164/API-165/API-166/API-167/API-168/API-169/API-170/API-171/API-172/API-173/API-174/API-175/API-176/API-177/API-178/API-179/API-180/API-181/API-182/API-183/API-184/API-185/API-186/API-187/API-188/API-189/API-190/API-191/API-192/API-193/API-194`. +- Based on: stacked on API-193 PR branch while PR #3810 is pending. - PR type for this branch: `consumer-migration` - Runtime behavior changes: none. - Rust code changes: route replication pool, outbound TLS generation, runtime @@ -25,7 +25,8 @@ Status values: `[ ]` not started, `[~]` in progress, `[x]` complete, `[!]` block id, event-host reads, bucket monitor reads, replication worker pool reads, config/tier first-node checks, rebalance endpoint-locality checks, bucket metadata object-store reads, metadata endpoint/setup reads, lifecycle - object-store reads, and replication object-store readiness checks, + object-store reads, replication object-store readiness checks, and ECStore + pools/tier/notification owner-root runtime source reads, through AppContext-first or owner-crate resolver boundaries. - CI/script changes: lock completed owner and test/fuzz boundaries against bare/glob imports, scattered raw ECStore facade subpaths, and startup @@ -35,7 +36,7 @@ Status values: `[ ]` not started, `[~]` in progress, `[x]` complete, `[!]` block and storage owner thin bridge regressions, plus app context and notify event-bridge thin module regressions; accept the reviewed AppContext resolver reverse dependencies in the layer baseline. -- Docs changes: record the API-136 through API-193 owner facade cleanup. +- Docs changes: record the API-136 through API-194 owner facade cleanup. ## Phase 0 Tasks @@ -4811,6 +4812,21 @@ Status values: `[ ]` not started, `[~]` in progress, `[x]` complete, `[!]` block migration/layer guards, PR-before-push quality gate, and three-expert review. +- [x] `API-194` Centralize ECStore owner-root runtime source reads. + - Do: route ECStore decommission object-store lookups, tier config save + object-store lookups, notification rebalance object-store lookups, + notification peer endpoint snapshots, and offline uptime reads through the + ECStore-owned runtime-source module. + - Acceptance: `pools`, `tier`, and `notification_sys` no longer import these + runtime globals directly outside the runtime-source boundary. + - Must preserve: decommission start validation and rollback behavior, tier + save initialization errors, notification stop-rebalance errors, peer + storage/server fallback behavior, and offline server uptime reporting. + - Verification: ECStore compile coverage, focused owner-root tests, + formatting, diff hygiene, residual runtime-source scan, Rust risk scan, + migration/layer guards, PR-before-push quality gate, and three-expert + review. + ## Next PRs 1. `consumer-migration`: continue reducing direct global reads behind AppContext resolver boundaries. @@ -4819,6 +4835,9 @@ Status values: `[ ]` not started, `[~]` in progress, `[x]` complete, `[!]` block | Expert | Status | Notes | |---|---|---| +| Quality/architecture | pass | API-194 keeps pools, tier, and notification owner-root runtime reads behind the ECStore runtime-source boundary. | +| Migration preservation | pass | Decommission, tier-save, notification rebalance, peer fallback, and offline uptime behavior preserve existing error/fallback semantics. | +| Testing/verification | pass | ECStore compile, focused owner-root tests, formatting, migration/layer guards, residual scan, Rust risk scan, and PR gate are planned for API-194. | | Quality/architecture | pass | API-193 keeps remaining ECStore bucket metadata, lifecycle, and replication runtime reads behind the runtime-source boundary. | | Migration preservation | pass | Metadata fan-out, standalone fallback, lifecycle compensation/expiry, and replication readiness defaults preserve existing behavior. | | Testing/verification | pass | ECStore compile, focused scans, formatting, migration/layer guards, residual scan, Rust risk scan, and PR gate are planned for API-193. |