fix(scanner): preserve unversioned heal retries

* fix(ecstore): fail closed on unresolved decommission entries

* perf(ecstore): avoid successful listing name clone

.github/workflows/ci.yml

@@ -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.

crates/config/src/constants/runtime.rs

@@ -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";

crates/ecstore/src/cluster/rpc/remote_disk.rs

@@ -50,10 +50,10 @@ use rustfs_protos::evict_failed_connection;
 use rustfs_protos::proto_gen::node_service::RenamePartRequest;
 use rustfs_protos::proto_gen::node_service::{
     BatchReadVersionRequest, BatchReadVersionResponse, CheckPartsRequest, DeletePathsRequest, DeleteRequest,
-    DeleteVersionRequest, DeleteVersionsRequest, DeleteVolumeRequest, DiskInfoRequest, ListDirRequest, ListVolumesRequest,
-    MakeVolumeRequest, MakeVolumesRequest, PreparePartTransactionRequest, ReadAllRequest, ReadMetadataRequest,
-    ReadMultipleRequest, ReadMultipleResponse, ReadPartsRequest, ReadVersionRequest, ReadXlRequest, RenameDataRequest,
-    RenameFileRequest, SettlePartTransactionRequest, SnapshotLeaseReleaseRequest, SnapshotLeaseRenewRequest,
+    DeleteVersionRequest, DeleteVersionsRequest, DeleteVersionsResponse, DeleteVolumeRequest, DiskInfoRequest, ListDirRequest,
+    ListVolumesRequest, MakeVolumeRequest, MakeVolumesRequest, PreparePartTransactionRequest, ReadAllRequest,
+    ReadMetadataRequest, ReadMultipleRequest, ReadMultipleResponse, ReadPartsRequest, ReadVersionRequest, ReadXlRequest,
+    RenameDataRequest, RenameFileRequest, SettlePartTransactionRequest, SnapshotLeaseReleaseRequest, SnapshotLeaseRenewRequest,
     SnapshotLeaseRequest, SnapshotLeaseResponse, StatVolumeRequest, UpdateMetadataRequest, VerifyFileRequest, WriteAllRequest,
     WriteMetadataRequest, node_service_client::NodeServiceClient,
 };
@@ -112,6 +112,28 @@ const EVENT_REMOTE_DISK_RPC: &str = "remote_disk_rpc";
 const SNAPSHOT_LEASE_PROTOCOL_VERSION: u32 = 1;
 pub const REMOTE_SNAPSHOT_LEASE_TTL: Duration = Duration::from_secs(60);
 
+fn decode_delete_versions_errors(response: DeleteVersionsResponse, expected_len: usize) -> Vec<Option<Error>> {
+    if !response.item_errors.is_empty() {
+        if response.item_errors.len() != expected_len {
+            return vec![Some(Error::other("malformed delete_versions item errors")); expected_len];
+        }
+        return response
+            .item_errors
+            .into_iter()
+            .map(|error| (error.code != 0).then(|| error.into()))
+            .collect();
+    }
+
+    if response.errors.len() != expected_len {
+        return vec![Some(Error::other("malformed delete_versions errors")); expected_len];
+    }
+    response
+        .errors
+        .into_iter()
+        .map(|error| (!error.is_empty()).then(|| Error::other(error)))
+        .collect()
+}
+
 fn snapshot_lease_token_from_response(response: SnapshotLeaseResponse) -> Result<SnapshotLeaseToken> {
     if !response.success {
         return Err(response.error.unwrap_or_default().into());
@@ -2406,8 +2428,6 @@ impl DiskAPI for RemoteDisk {
             return errors;
         }
 
-        // TODO(backlog): replace string errors with typed `StorageError` variants
-
         let result = self
             .execute_with_timeout(
                 || async {
@@ -2439,17 +2459,7 @@ impl DiskAPI for RemoteDisk {
             }
             return errors;
         }
-        response
-            .errors
-            .iter()
-            .map(|error| {
-                if error.is_empty() {
-                    None
-                } else {
-                    Some(Error::other(error.to_string()))
-                }
-            })
-            .collect()
+        decode_delete_versions_errors(response, versions.len())
     }
 
     #[tracing::instrument(level = "trace", skip_all)]
@@ -3760,6 +3770,63 @@ mod tests {
 
     static INIT: Once = Once::new();
 
+    #[test]
+    fn delete_versions_response_preserves_typed_item_errors() {
+        let errors = decode_delete_versions_errors(
+            DeleteVersionsResponse {
+                success: true,
+                errors: vec!["file not found".to_string(), String::new()],
+                error: None,
+                item_errors: vec![
+                    rustfs_protos::proto_gen::node_service::Error {
+                        code: DiskError::FileNotFound.to_u32(),
+                        error_info: "file not found".to_string(),
+                    },
+                    rustfs_protos::proto_gen::node_service::Error::default(),
+                ],
+            },
+            2,
+        );
+
+        assert!(matches!(errors.as_slice(), [Some(DiskError::FileNotFound), None]));
+    }
+
+    #[test]
+    fn delete_versions_response_accepts_legacy_string_errors() {
+        let errors = decode_delete_versions_errors(
+            DeleteVersionsResponse {
+                success: true,
+                errors: vec!["legacy error".to_string(), String::new()],
+                error: None,
+                item_errors: Vec::new(),
+            },
+            2,
+        );
+
+        assert_eq!(errors.len(), 2);
+        assert_eq!(errors[0].as_ref().map(ToString::to_string).as_deref(), Some("io error legacy error"));
+        assert!(errors[1].is_none());
+    }
+
+    #[test]
+    fn delete_versions_response_rejects_misaligned_item_errors() {
+        let errors = decode_delete_versions_errors(
+            DeleteVersionsResponse {
+                success: true,
+                errors: vec!["file not found".to_string()],
+                error: None,
+                item_errors: vec![rustfs_protos::proto_gen::node_service::Error {
+                    code: DiskError::FileNotFound.to_u32(),
+                    error_info: "file not found".to_string(),
+                }],
+            },
+            2,
+        );
+
+        assert_eq!(errors.len(), 2);
+        assert!(errors.iter().all(Option::is_some));
+    }
+
     #[test]
     fn disk_mutation_digest_marks_rolling_compatibility() {
         let mut request = Request::new(());

crates/ecstore/src/core/pools.rs

@@ -36,7 +36,8 @@ use crate::disk::error::DiskError;
 use crate::disk::{BUCKET_META_PREFIX, RUSTFS_META_BUCKET};
 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,
+    StorageError, is_err_bucket_exists, is_err_bucket_not_found, is_err_object_not_found, is_err_operation_canceled,
+    is_err_version_not_found,
 };
 use crate::layout::endpoints::EndpointServerPools;
 use crate::object_api::{GetObjectReader, ObjectOptions};
@@ -89,6 +90,7 @@ const DECOMMISSION_STAGE_SOURCE_CLEANUP: &str = "source_cleanup";
 const DECOMMISSION_STAGE_ENTRY_FINISHED: &str = "entry_finished";
 const DECOMMISSION_PROGRESS_SAVE_INTERVAL: Duration = Duration::seconds(30);
 const DECOMMISSION_PROGRESS_SAVE_ITEM_THRESHOLD: usize = 1000;
+const DECOMMISSION_PROGRESS_SAVE_RETRY_BACKOFF: Duration = Duration::seconds(1);
 const DECOMMISSION_BUCKET_CONCURRENCY_ENV: &str = "RUSTFS_DECOMMISSION_BUCKET_CONCURRENCY";
 const DECOMMISSION_BUCKET_CONCURRENCY_DEFAULT_CAP: usize = 4;
 const DECOMMISSION_TARGET_CAPACITY_OVERHEAD_PERCENT: usize = 30;
@@ -638,22 +640,6 @@ fn track_decommission_current_object(meta: &mut PoolMeta, idx: usize, bucket: &s
     track_decommission_current_object_stage(meta, idx, bucket, object, "")
 }
 
-fn touch_decommission_progress(meta: &mut PoolMeta, idx: usize) -> Result<()> {
-    let pool_count = meta.pools.len();
-    ensure_valid_decommission_pool_index(pool_count, idx)?;
-
-    let Some(pool) = meta.pools.get_mut(idx) else {
-        return Err(invalid_decommission_pool_index_error(pool_count, idx));
-    };
-    let Some(info) = pool.decommission.as_mut() else {
-        return Err(decommission_metadata_not_initialized_error("touch decommission progress"));
-    };
-
-    pool.last_update = OffsetDateTime::now_utc();
-    info.mark_progress_saved();
-    Ok(())
-}
-
 fn resolve_decommission_update_after_result(result: Result<bool>) -> Result<bool> {
     result.map_err(|err| Error::other(format!("decommission metadata update failed: {err}")))
 }
@@ -773,7 +759,76 @@ async fn load_decommission_entry_exact_versions(
 }
 
 fn resolve_decommission_check_after_list_result(list_result: Result<()>, entry_error: Option<Error>) -> Result<()> {
-    if let Some(err) = entry_error { Err(err) } else { list_result }
+    match list_result {
+        Ok(()) => entry_error.map_or(Ok(()), Err),
+        Err(list_err) => resolve_decommission_listing_error(Some(list_err), entry_error).map_or(Ok(()), Err),
+    }
+}
+
+fn resolve_decommission_listing_error(listing_error: Option<Error>, entry_error: Option<Error>) -> Option<Error> {
+    match (listing_error, entry_error) {
+        (Some(listing_error), Some(entry_error)) if is_err_operation_canceled(&listing_error) => Some(entry_error),
+        (Some(listing_error), Some(entry_error)) if is_err_operation_canceled(&entry_error) => Some(listing_error),
+        (Some(listing_error), _) => Some(listing_error),
+        (None, entry_error) => entry_error,
+    }
+}
+
+fn decommission_unresolved_listing_error(
+    bucket: &str,
+    prefix: &str,
+    candidate: Option<&str>,
+    candidate_count: usize,
+    disk_error_count: usize,
+    pool_index: usize,
+    set_index: usize,
+) -> Error {
+    let location = candidate.unwrap_or(prefix);
+    Error::other(format!(
+        "decommission listing could not resolve metadata for {bucket}/{location} on pool {pool_index} set {set_index} ({candidate_count} candidate(s), {disk_error_count} disk error(s))"
+    ))
+}
+
+fn resolve_decommission_partial_listing_entry(
+    entries: MetaCacheEntries,
+    resolver: MetadataResolutionParams,
+    bucket: &str,
+    prefix: &str,
+    disk_error_count: usize,
+    pool_index: usize,
+    set_index: usize,
+) -> Result<MetaCacheEntry> {
+    let candidate_count = entries.as_ref().iter().flatten().count();
+    if let Some(entry) = entries.resolve(resolver) {
+        return Ok(entry);
+    }
+
+    let candidate = entries.as_ref().iter().flatten().map(|entry| entry.name.as_str()).next();
+    Err(decommission_unresolved_listing_error(
+        bucket,
+        prefix,
+        candidate,
+        candidate_count,
+        disk_error_count,
+        pool_index,
+        set_index,
+    ))
+}
+
+async fn record_decommission_entry_error(
+    entry_error: &Arc<tokio::sync::Mutex<Option<Error>>>,
+    rx: &CancellationToken,
+    err: Error,
+) {
+    if rx.is_cancelled() {
+        return;
+    }
+
+    let mut first_err = entry_error.lock().await;
+    if first_err.is_none() && !rx.is_cancelled() {
+        *first_err = Some(err);
+        rx.cancel();
+    }
 }
 
 fn resolve_decommission_pool_meta_reload_result(result: Result<()>, stage: &str) -> Result<()> {
@@ -1483,6 +1538,7 @@ impl TryFrom<PersistedPoolDecommissionInfo> for PoolDecommissionInfo {
             terminal_reload_attempt_at: value.terminal_reload_attempt_at,
             terminal_reload_failures: value.terminal_reload_failures,
             progress_save_item_baseline: value.items_decommissioned.saturating_add(value.items_decommission_failed),
+            progress_save_retry_after: None,
         })
     }
 }
@@ -1514,6 +1570,7 @@ impl TryFrom<LegacyPoolDecommissionInfo> for PoolDecommissionInfo {
             terminal_reload_attempt_at: None,
             terminal_reload_failures: Vec::new(),
             progress_save_item_baseline: value.items_decommissioned.saturating_add(value.items_decommission_failed),
+            progress_save_retry_after: None,
         })
     }
 }
@@ -1627,6 +1684,82 @@ impl PoolMeta {
         }
     }
 
+    fn decommission_progress_checkpoint(
+        &self,
+        idx: usize,
+        duration: Duration,
+        now: OffsetDateTime,
+    ) -> Result<Option<DecommissionProgressCheckpoint>> {
+        let pool_count = self.pools.len();
+        ensure_valid_decommission_pool_index(pool_count, idx)?;
+
+        let Some(pool) = self.pools.get(idx) else {
+            return Err(invalid_decommission_pool_index_error(pool_count, idx));
+        };
+        let Some(info) = pool.decommission.as_ref() else {
+            return Err(decommission_metadata_not_initialized_error("update decommission metadata timestamp"));
+        };
+
+        if info.progress_save_retry_after.is_some_and(|retry_after| now < retry_after) {
+            return Ok(None);
+        }
+
+        let time_threshold_reached = now.unix_timestamp() - pool.last_update.unix_timestamp() >= duration.whole_seconds();
+        let item_threshold_reached = info.items_since_last_progress_save() >= DECOMMISSION_PROGRESS_SAVE_ITEM_THRESHOLD;
+        if !time_threshold_reached && !item_threshold_reached {
+            return Ok(None);
+        }
+
+        Ok(Some(DecommissionProgressCheckpoint {
+            start_time: info.start_time,
+            queued: info.queued,
+            counted_items: info.counted_items(),
+            checkpoint_at: now,
+        }))
+    }
+
+    fn commit_decommission_progress_checkpoint(&mut self, idx: usize, checkpoint: DecommissionProgressCheckpoint) -> bool {
+        let Some(pool) = self.pools.get_mut(idx) else {
+            return false;
+        };
+        let Some(info) = pool.decommission.as_mut() else {
+            return false;
+        };
+
+        if info.start_time != checkpoint.start_time
+            || info.queued != checkpoint.queued
+            || !is_decommission_active(info.complete, info.failed, info.canceled)
+        {
+            return false;
+        }
+
+        info.progress_save_item_baseline = info.progress_save_item_baseline.max(checkpoint.counted_items);
+        info.progress_save_retry_after = None;
+        pool.last_update = pool.last_update.max(checkpoint.checkpoint_at);
+        true
+    }
+
+    fn defer_decommission_progress_checkpoint(
+        &mut self,
+        idx: usize,
+        checkpoint: DecommissionProgressCheckpoint,
+        retry_after: OffsetDateTime,
+    ) {
+        let Some(pool) = self.pools.get_mut(idx) else {
+            return;
+        };
+        let Some(info) = pool.decommission.as_mut() else {
+            return;
+        };
+
+        if info.start_time == checkpoint.start_time
+            && info.queued == checkpoint.queued
+            && is_decommission_active(info.complete, info.failed, info.canceled)
+        {
+            info.progress_save_retry_after = Some(retry_after);
+        }
+    }
+
     fn load_from_config_data(&mut self, data: Vec<u8>) -> Result<()> {
         if data.is_empty() {
             return Ok(());
@@ -1987,30 +2120,9 @@ impl PoolMeta {
     }
 
     pub fn update_after(&mut self, idx: usize, duration: Duration) -> Result<bool> {
-        let pool_count = self.pools.len();
-        ensure_valid_decommission_pool_index(pool_count, idx)?;
-
-        let (last_update, item_threshold_reached) = match self.pools.get(idx) {
-            Some(pool) if let Some(info) = pool.decommission.as_ref() => (
-                pool.last_update,
-                info.items_since_last_progress_save() >= DECOMMISSION_PROGRESS_SAVE_ITEM_THRESHOLD,
-            ),
-            Some(_) => {
-                return Err(decommission_metadata_not_initialized_error("update decommission metadata timestamp"));
-            }
-            None => return Err(invalid_decommission_pool_index_error(pool_count, idx)),
-        };
-        let now = OffsetDateTime::now_utc();
-
-        if now.unix_timestamp() - last_update.unix_timestamp() >= duration.whole_seconds() || item_threshold_reached {
-            let Some(pool) = self.pools.get_mut(idx) else {
-                return Err(invalid_decommission_pool_index_error(pool_count, idx));
-            };
-            pool.last_update = now;
-            return Ok(true);
-        }
-
-        Ok(false)
+        Ok(self
+            .decommission_progress_checkpoint(idx, duration, OffsetDateTime::now_utc())?
+            .is_some())
     }
 
     pub fn validate(&self, pools: Vec<Arc<Sets>>) -> Result<bool> {
@@ -2151,6 +2263,16 @@ pub struct PoolDecommissionInfo {
     pub terminal_reload_failures: Vec<String>,
     #[serde(skip)]
     pub progress_save_item_baseline: usize,
+    #[serde(skip)]
+    pub progress_save_retry_after: Option<OffsetDateTime>,
+}
+
+#[derive(Debug, Clone, Copy, PartialEq, Eq)]
+struct DecommissionProgressCheckpoint {
+    start_time: Option<OffsetDateTime>,
+    queued: bool,
+    counted_items: usize,
+    checkpoint_at: OffsetDateTime,
 }
 
 impl PoolDecommissionInfo {
@@ -2185,6 +2307,7 @@ impl PoolDecommissionInfo {
 
     fn mark_progress_saved(&mut self) {
         self.progress_save_item_baseline = self.counted_items();
+        self.progress_save_retry_after = None;
     }
 
     pub fn bucket_push(&mut self, bucket: &DecomBucketInfo) {
@@ -2489,6 +2612,40 @@ impl ECStore {
         snapshot.save(self.pools.clone()).await
     }
 
+    async fn save_decommission_progress_checkpoint(&self, idx: usize) -> Result<bool> {
+        // Lock order: save gate, then the short pool metadata read/write sections. Peer
+        // reloads are intentionally performed by the caller after both locks are released.
+        let _save_guard = self.pool_meta_save_gate.lock().await;
+        let (snapshot, checkpoint) = {
+            let pool_meta = self.pool_meta.read().await;
+            let Some(checkpoint) = pool_meta.decommission_progress_checkpoint(
+                idx,
+                DECOMMISSION_PROGRESS_SAVE_INTERVAL,
+                OffsetDateTime::now_utc(),
+            )?
+            else {
+                return Ok(false);
+            };
+
+            let mut snapshot = pool_meta.clone();
+            let Some(pool) = snapshot.pools.get_mut(idx) else {
+                return Err(invalid_decommission_pool_index_error(snapshot.pools.len(), idx));
+            };
+            pool.last_update = checkpoint.checkpoint_at;
+            (snapshot, checkpoint)
+        };
+
+        if let Err(err) = snapshot.save(self.pools.clone()).await {
+            let retry_after = OffsetDateTime::now_utc() + DECOMMISSION_PROGRESS_SAVE_RETRY_BACKOFF;
+            let mut pool_meta = self.pool_meta.write().await;
+            pool_meta.defer_decommission_progress_checkpoint(idx, checkpoint, retry_after);
+            return Err(err);
+        }
+
+        let mut pool_meta = self.pool_meta.write().await;
+        Ok(pool_meta.commit_decommission_progress_checkpoint(idx, checkpoint))
+    }
+
     async fn save_current_pool_meta_for_decommission_start(
         &self,
         indices: &[usize],
@@ -2871,7 +3028,7 @@ impl ECStore {
         Ok(())
     }
 
-    async fn save_decommission_entry_progress_stage(
+    async fn track_decommission_entry_progress_stage(
         &self,
         idx: usize,
         bucket: &str,
@@ -2882,22 +3039,6 @@ impl ECStore {
             let mut pool_meta = self.pool_meta.write().await;
             track_decommission_current_object_stage(&mut pool_meta, idx, bucket, object, stage)
                 .map_err(|err| with_decommission_entry_context(stage, bucket, object, err))?;
-            touch_decommission_progress(&mut pool_meta, idx)
-                .map_err(|err| with_decommission_entry_context(stage, bucket, object, err))?;
-        }
-
-        if let Some(err) = resolve_decommission_progress_save_result(self.save_current_pool_meta().await) {
-            warn!(
-                event = EVENT_DECOMMISSION_ENTRY,
-                component = LOG_COMPONENT_ECSTORE,
-                subsystem = LOG_SUBSYSTEM_POOLS,
-                pool_index = idx,
-                bucket = %bucket,
-                object = %object,
-                stage,
-                error = ?err,
-                "Decommission progress stage save failed"
-            );
         }
 
         Ok(())
@@ -3165,7 +3306,7 @@ impl ECStore {
                 let bucket_name = bucket.clone();
                 let object_name = rd.object_info.name.clone();
 
-                self.save_decommission_entry_progress_stage(
+                self.track_decommission_entry_progress_stage(
                     idx,
                     bucket_name.as_str(),
                     object_name.as_str(),
@@ -3259,7 +3400,7 @@ impl ECStore {
             }
             decommission_cancel_signal_result(rx.is_cancelled())?;
 
-            self.save_decommission_entry_progress_stage(
+            self.track_decommission_entry_progress_stage(
                 idx,
                 bucket.as_str(),
                 entry.name.as_str(),
@@ -3267,7 +3408,7 @@ impl ECStore {
             )
             .await?;
 
-            self.save_decommission_entry_progress_stage(
+            self.track_decommission_entry_progress_stage(
                 idx,
                 bucket.as_str(),
                 entry.name.as_str(),
@@ -3334,34 +3475,42 @@ impl ECStore {
             }
         };
 
-        self.save_decommission_entry_progress_stage(idx, bucket.as_str(), entry.name.as_str(), DECOMMISSION_STAGE_ENTRY_FINISHED)
-            .await?;
+        self.track_decommission_entry_progress_stage(
+            idx,
+            bucket.as_str(),
+            entry.name.as_str(),
+            DECOMMISSION_STAGE_ENTRY_FINISHED,
+        )
+        .await?;
 
         if should_save_progress {
-            let save_result = self.save_current_pool_meta().await;
-            if let Some(err) = resolve_decommission_progress_save_result(save_result) {
-                warn!(
-                    event = EVENT_DECOMMISSION_ENTRY,
-                    component = LOG_COMPONENT_ECSTORE,
-                    subsystem = LOG_SUBSYSTEM_POOLS,
-                    pool_index = idx,
-                    bucket = %bucket,
-                    object = %entry.name,
-                    state = "progress_save_failed",
-                    error = %err,
-                    "Decommission progress save failed; continuing and will retry at the next checkpoint"
-                );
-            } else {
-                let mut pool_meta = self.pool_meta.write().await;
-                pool_meta.mark_decommission_progress_saved();
-                if let Some(notification_sys) = runtime_sources::notification_sys()
-                    && let Err(err) = resolve_decommission_entry_reload_result(
-                        notification_sys.reload_pool_meta().await,
-                        bucket.as_str(),
-                        entry.name.as_str(),
-                    )
-                {
-                    warn!("{err}");
+            match self.save_decommission_progress_checkpoint(idx).await {
+                Ok(true) => {
+                    if let Some(notification_sys) = runtime_sources::notification_sys()
+                        && let Err(err) = resolve_decommission_entry_reload_result(
+                            notification_sys.reload_pool_meta().await,
+                            bucket.as_str(),
+                            entry.name.as_str(),
+                        )
+                    {
+                        warn!("{err}");
+                    }
+                }
+                Ok(false) => {}
+                Err(err) => {
+                    if let Some(err) = resolve_decommission_progress_save_result(Err(err)) {
+                        warn!(
+                            event = EVENT_DECOMMISSION_ENTRY,
+                            component = LOG_COMPONENT_ECSTORE,
+                            subsystem = LOG_SUBSYSTEM_POOLS,
+                            pool_index = idx,
+                            bucket = %bucket,
+                            object = %entry.name,
+                            state = "progress_save_failed",
+                            error = %err,
+                            "Decommission progress save failed; continuing and will retry at the next checkpoint"
+                        );
+                    }
                 }
             }
         }
@@ -3538,6 +3687,7 @@ impl ECStore {
             let rx_clone = rx.clone();
             let bi = bi.clone();
             let set_id = set_idx;
+            let listing_entry_error = entry_error.clone();
             let worker = tokio::spawn(async move {
                 let _listing_permit = listing_permit;
                 run_decommission_listing_with_retry(
@@ -3551,7 +3701,11 @@ impl ECStore {
                         let set = set.clone();
                         let rx = rx_clone.clone();
                         let bucket = bi.clone();
-                        async move { set.list_objects_to_decommission(rx, bucket, callback).await }
+                        let entry_error = listing_entry_error.clone();
+                        async move {
+                            set.list_objects_to_decommission(rx, bucket, callback, entry_error.clone(), idx, set_id)
+                                .await
+                        }
                     },
                 )
                 .await
@@ -3581,11 +3735,7 @@ impl ECStore {
 
         wait_decommission_worker_drain(&workers, worker_limit).await?;
 
-        if let Some(err) = listing_worker_error {
-            return Err(err);
-        }
-
-        if let Some(err) = entry_error.lock().await.clone() {
+        if let Some(err) = resolve_decommission_listing_error(listing_worker_error, entry_error.lock().await.clone()) {
             return Err(err);
         }
 
@@ -4191,7 +4341,7 @@ impl ECStore {
         let buckets = self.get_buckets_to_decommission().await?;
         let pool = self.pools[idx].clone();
 
-        for set in &pool.disk_set {
+        for (set_index, set) in pool.disk_set.iter().enumerate() {
             for bucket_info in &buckets {
                 let mut lifecycle_config = None;
                 let mut object_lock_config = None;
@@ -4286,7 +4436,7 @@ impl ECStore {
                 });
 
                 let list_result = set
-                    .list_objects_to_decommission(callback_rx, bucket_info.clone(), callback)
+                    .list_objects_to_decommission(callback_rx, bucket_info.clone(), callback, entry_error.clone(), idx, set_index)
                     .await;
                 let entry_error = entry_error.lock().await.clone();
                 resolve_decommission_check_after_list_result(list_result, entry_error)?;
@@ -5021,12 +5171,15 @@ mod tests {
 pub type ListCallback = Arc<dyn Fn(MetaCacheEntry) -> BoxFuture<'static, ()> + Send + Sync + 'static>;
 
 impl SetDisks {
-    #[tracing::instrument(skip(self, rx, cb_func))]
+    #[tracing::instrument(skip(self, rx, cb_func, entry_error))]
     async fn list_objects_to_decommission(
         self: &Arc<Self>,
         rx: CancellationToken,
         bucket_info: DecomBucketInfo,
         cb_func: ListCallback,
+        entry_error: Arc<tokio::sync::Mutex<Option<Error>>>,
+        pool_index: usize,
+        set_index: usize,
     ) -> Result<()> {
         let (disks, _) = self.get_online_disks_with_healing(false).await;
         ensure_decommission_listing_disks_available(!disks.is_empty(), &bucket_info.name)?;
@@ -5041,6 +5194,12 @@ impl SetDisks {
         };
 
         let cb1 = cb_func.clone();
+        let unresolved_error = entry_error.clone();
+        let unresolved_rx = rx.clone();
+        let unresolved_bucket = bucket_info.name.clone();
+        let unresolved_prefix = bucket_info.prefix.clone();
+        let unresolved_pool_index = pool_index;
+        let unresolved_set_index = set_index;
 
         list_path_raw(
             rx,
@@ -5053,20 +5212,51 @@ impl SetDisks {
                 skip_walkdir_total_timeout: true,
                 walkdir_stall_timeout: Some(DECOMMISSION_BACKGROUND_WALKDIR_STALL_TIMEOUT),
                 agreed: Some(Box::new(move |entry: MetaCacheEntry| Box::pin(cb1(entry)))),
-                partial: Some(Box::new(move |entries: MetaCacheEntries, _: &[Option<DiskError>]| {
+                partial: Some(Box::new(move |entries: MetaCacheEntries, errs: &[Option<DiskError>]| {
                     let resolver = resolver.clone();
                     let cb_func = cb_func.clone();
-                    match entries.resolve(resolver) {
-                        Some(entry) => {
+                    let bucket = unresolved_bucket.clone();
+                    let prefix = unresolved_prefix.clone();
+                    let unresolved_error = unresolved_error.clone();
+                    let unresolved_rx = unresolved_rx.clone();
+                    let pool_index = unresolved_pool_index;
+                    let set_index = unresolved_set_index;
+                    let disk_error_count = errs.iter().flatten().count();
+                    if unresolved_rx.is_cancelled() {
+                        return Box::pin(async {});
+                    }
+
+                    match resolve_decommission_partial_listing_entry(
+                        entries,
+                        resolver,
+                        &bucket,
+                        &prefix,
+                        disk_error_count,
+                        pool_index,
+                        set_index,
+                    ) {
+                        Ok(entry) => {
                             warn!("decommission_pool: list_objects_to_decommission get {}", &entry.name);
                             Box::pin(async move {
                                 cb_func(entry).await;
                             })
                         }
-                        None => {
-                            warn!("decommission_pool: list_objects_to_decommission get none");
-                            Box::pin(async {})
-                        }
+                        Err(err) => Box::pin(async move {
+                            if unresolved_rx.is_cancelled() {
+                                return;
+                            }
+                            warn!(
+                                event = EVENT_DECOMMISSION_BUCKET,
+                                component = LOG_COMPONENT_ECSTORE,
+                                subsystem = LOG_SUBSYSTEM_POOLS,
+                                bucket = %bucket,
+                                prefix = %prefix,
+                                state = "unresolved_entry",
+                                error = %err,
+                                "Decommission listing failed closed on unresolved metadata"
+                            );
+                            record_decommission_entry_error(&unresolved_error, &unresolved_rx, err).await;
+                        }),
                     }
                 })),
                 ..Default::default()
@@ -5074,6 +5264,10 @@ impl SetDisks {
         )
         .await?;
 
+        if let Some(err) = entry_error.lock().await.clone() {
+            return Err(err);
+        }
+
         Ok(())
     }
 }
@@ -5264,11 +5458,11 @@ pub(crate) fn fallback_free_capacity_dedup(disks: &[rustfs_madmin::Disk]) -> usi
 #[cfg(test)]
 mod pools_tests {
     use super::{
-        DECOMMISSION_PROGRESS_SAVE_INTERVAL, DECOMMISSION_PROGRESS_SAVE_ITEM_THRESHOLD, DecomBucketInfo,
-        DecommissionStartPoolState, DecommissionTerminalState, ListCallback, PoolDecommissionInfo, PoolMeta, PoolSpaceInfo,
-        PoolStatus, apply_decommission_status_space_info, bind_decommission_cancelers, bind_missing_decommission_cancelers,
-        cancel_decommission_canceler, classify_decommission_terminal_state, count_decommission_item,
-        decommission_cancel_signal_result, decommission_item_size, decommission_meta_bucket_options,
+        DECOMMISSION_PROGRESS_SAVE_INTERVAL, DECOMMISSION_PROGRESS_SAVE_ITEM_THRESHOLD, DECOMMISSION_PROGRESS_SAVE_RETRY_BACKOFF,
+        DecomBucketInfo, DecommissionStartPoolState, DecommissionTerminalState, ListCallback, PoolDecommissionInfo, PoolMeta,
+        PoolSpaceInfo, PoolStatus, apply_decommission_status_space_info, bind_decommission_cancelers,
+        bind_missing_decommission_cancelers, cancel_decommission_canceler, classify_decommission_terminal_state,
+        count_decommission_item, decommission_cancel_signal_result, decommission_item_size, decommission_meta_bucket_options,
         decommission_start_pool_state, dedup_indices, default_decommission_bucket_concurrency,
         ensure_decommission_cancel_allowed, ensure_decommission_clear_allowed, ensure_decommission_listing_disks_available,
         ensure_decommission_not_rebalancing, ensure_decommission_start_allowed, ensure_decommission_start_keeps_active_pool,
@@ -5279,11 +5473,12 @@ mod pools_tests {
         has_active_decommission_canceler, is_decommission_active, is_decommission_cancel_requested,
         load_decommission_entry_versions, local_decommission_queue_prefix, mark_decommission_bucket_done,
         merge_pool_status_refresh, missing_decommission_worker_prefix, observe_decommission_terminal_reload_result,
-        pool_meta_has_active_decommission, require_decommission_store, resolve_decommission_bucket_done_save_result,
-        resolve_decommission_bucket_state, resolve_decommission_check_after_list_result,
-        resolve_decommission_entry_cleanup_delete_result, resolve_decommission_entry_exact_versions,
-        resolve_decommission_entry_reload_result, resolve_decommission_listing_worker_result,
-        resolve_decommission_optional_bucket_config_result, resolve_decommission_pool_meta_reload_result,
+        pool_meta_has_active_decommission, record_decommission_entry_error, require_decommission_store,
+        resolve_decommission_bucket_done_save_result, resolve_decommission_bucket_state,
+        resolve_decommission_check_after_list_result, resolve_decommission_entry_cleanup_delete_result,
+        resolve_decommission_entry_exact_versions, resolve_decommission_entry_reload_result, resolve_decommission_listing_error,
+        resolve_decommission_listing_worker_result, resolve_decommission_optional_bucket_config_result,
+        resolve_decommission_partial_listing_entry, resolve_decommission_pool_meta_reload_result,
         resolve_decommission_preflight_heal_result, resolve_decommission_progress_save_result,
         resolve_decommission_spawn_failure_result, resolve_decommission_terminal_mark_after_error_result,
         resolve_decommission_terminal_mark_result, resolve_decommission_update_after_result,
@@ -5293,16 +5488,17 @@ mod pools_tests {
         should_preserve_decommission_canceled_state, should_reject_decommission_cancel_as_terminal,
         should_retry_decommission_cancel_reload, should_retry_decommission_listing, should_skip_canceled_decommission_routine,
         split_decommission_buckets, take_and_cancel_decommission_canceler, take_decommission_canceler,
-        touch_decommission_progress, track_decommission_current_object, track_decommission_current_object_stage,
-        validate_start_decommission_request, wait_decommission_listing_retry, wait_decommission_worker_drain,
-        with_decommission_entry_context,
+        track_decommission_current_object, track_decommission_current_object_stage, validate_start_decommission_request,
+        wait_decommission_listing_retry, wait_decommission_worker_drain, with_decommission_entry_context,
     };
     use crate::data_movement;
     use crate::disk::endpoint::Endpoint;
     use crate::error::{Error, StorageError};
     use crate::layout::endpoints::{EndpointServerPools, Endpoints, PoolEndpoints};
     use crate::services::rebalance::{RebalStatus, RebalanceInfo, RebalanceMeta, RebalanceStats};
-    use rustfs_filemeta::{FileInfo, FileInfoVersions, MetaCacheEntry, ObjectPartInfo};
+    use rustfs_filemeta::{
+        FileInfo, FileInfoVersions, MetaCacheEntries, MetaCacheEntry, MetadataResolutionParams, ObjectPartInfo,
+    };
     use rustfs_rio::Index;
     use std::sync::{
         Arc,
@@ -6321,6 +6517,65 @@ mod pools_tests {
         assert!(matches!(err, Error::SlowDown));
     }
 
+    #[test]
+    fn test_resolve_decommission_partial_listing_entry_rejects_unresolved_metadata() {
+        let err = resolve_decommission_partial_listing_entry(
+            MetaCacheEntries(vec![None]),
+            MetadataResolutionParams {
+                dir_quorum: 2,
+                obj_quorum: 2,
+                bucket: "bucket-a".to_string(),
+                ..Default::default()
+            },
+            "bucket-a",
+            "prefix/",
+            1,
+            2,
+            3,
+        )
+        .expect_err("unresolved partial listing must fail closed");
+
+        let message = err.to_string();
+        assert!(message.contains("decommission listing could not resolve metadata"));
+        assert!(message.contains("bucket-a/prefix/"));
+        assert!(message.contains("pool 2 set 3"));
+        assert!(message.contains("1 disk error(s)"));
+    }
+
+    #[tokio::test]
+    async fn test_record_decommission_entry_error_cancels_listing_and_preserves_first_error() {
+        let entry_error = Arc::new(tokio::sync::Mutex::new(None));
+        let rx = CancellationToken::new();
+
+        record_decommission_entry_error(&entry_error, &rx, Error::SlowDown).await;
+        record_decommission_entry_error(&entry_error, &rx, Error::OperationCanceled).await;
+
+        assert!(rx.is_cancelled());
+        assert!(matches!(*entry_error.lock().await, Some(Error::SlowDown)));
+    }
+
+    #[tokio::test]
+    async fn test_record_decommission_entry_error_ignores_already_canceled_listing() {
+        let entry_error = Arc::new(tokio::sync::Mutex::new(None));
+        let rx = CancellationToken::new();
+        rx.cancel();
+
+        record_decommission_entry_error(&entry_error, &rx, Error::SlowDown).await;
+
+        assert!(entry_error.lock().await.is_none());
+    }
+
+    #[test]
+    fn test_resolve_decommission_listing_error_preserves_real_listing_failure() {
+        let err = resolve_decommission_listing_error(Some(Error::SlowDown), Some(Error::OperationCanceled))
+            .expect("listing failure should be returned");
+        assert!(matches!(err, Error::SlowDown));
+
+        let err = resolve_decommission_listing_error(Some(Error::OperationCanceled), Some(Error::SlowDown))
+            .expect("entry failure should be returned");
+        assert!(matches!(err, Error::SlowDown));
+    }
+
     #[test]
     fn test_resolve_decommission_check_after_list_result_returns_list_result_without_entry_error() {
         let err = resolve_decommission_check_after_list_result(Err(Error::OperationCanceled), None)
@@ -6538,7 +6793,7 @@ mod pools_tests {
     }
 
     #[test]
-    fn test_touch_decommission_progress_updates_last_update_and_save_baseline() {
+    fn test_track_decommission_stage_does_not_advance_checkpoint_state() {
         let mut meta = PoolMeta {
             pools: vec![PoolStatus {
                 id: 0,
@@ -6553,11 +6808,13 @@ mod pools_tests {
             ..Default::default()
         };
 
-        touch_decommission_progress(&mut meta, 0).expect("valid decommission progress should be touched");
+        track_decommission_current_object_stage(&mut meta, 0, "bucket", "object", "migrate_object")
+            .expect("valid decommission progress should be tracked");
 
-        assert!(meta.pools[0].last_update > OffsetDateTime::UNIX_EPOCH);
+        assert_eq!(meta.pools[0].last_update, OffsetDateTime::UNIX_EPOCH);
         let info = meta.pools[0].decommission.as_ref().expect("decommission info should exist");
-        assert_eq!(info.items_since_last_progress_save(), 0);
+        assert_eq!(info.items_since_last_progress_save(), 5);
+        assert_eq!(info.stage, "migrate_object");
     }
 
     #[test]
@@ -6632,6 +6889,134 @@ mod pools_tests {
         assert_eq!(info.items_since_last_progress_save(), 1);
     }
 
+    #[test]
+    fn test_pool_meta_update_after_does_not_advance_last_update_before_save() {
+        let last_update = OffsetDateTime::UNIX_EPOCH;
+        let mut meta = PoolMeta {
+            pools: vec![PoolStatus {
+                id: 0,
+                cmd_line: "pool-0".to_string(),
+                last_update,
+                decommission: Some(PoolDecommissionInfo {
+                    start_time: Some(last_update),
+                    items_decommissioned: DECOMMISSION_PROGRESS_SAVE_ITEM_THRESHOLD,
+                    ..Default::default()
+                }),
+            }],
+            ..Default::default()
+        };
+
+        assert!(
+            meta.update_after(0, DECOMMISSION_PROGRESS_SAVE_INTERVAL)
+                .expect("item threshold should request a checkpoint")
+        );
+        assert_eq!(meta.pools[0].last_update, last_update);
+    }
+
+    #[test]
+    fn test_decommission_progress_checkpoint_commits_exact_snapshot_watermark() {
+        let start_time = OffsetDateTime::UNIX_EPOCH;
+        let checkpoint_at = start_time + Duration::seconds(30);
+        let mut meta = PoolMeta {
+            pools: vec![PoolStatus {
+                id: 0,
+                cmd_line: "pool-0".to_string(),
+                last_update: start_time,
+                decommission: Some(PoolDecommissionInfo {
+                    start_time: Some(start_time),
+                    items_decommissioned: DECOMMISSION_PROGRESS_SAVE_ITEM_THRESHOLD,
+                    ..Default::default()
+                }),
+            }],
+            ..Default::default()
+        };
+
+        let checkpoint = meta
+            .decommission_progress_checkpoint(0, DECOMMISSION_PROGRESS_SAVE_INTERVAL, checkpoint_at)
+            .expect("valid decommission state should produce a checkpoint")
+            .expect("item threshold should produce a checkpoint");
+        meta.count_item(0, 1, false);
+
+        assert!(meta.commit_decommission_progress_checkpoint(0, checkpoint));
+        let info = meta.pools[0].decommission.as_ref().expect("decommission info should exist");
+        assert_eq!(info.progress_save_item_baseline, checkpoint.counted_items);
+        assert_eq!(info.items_since_last_progress_save(), 1);
+        assert_eq!(meta.pools[0].last_update, checkpoint_at);
+    }
+
+    #[test]
+    fn test_decommission_progress_checkpoint_backoff_does_not_advance_baseline() {
+        let start_time = OffsetDateTime::UNIX_EPOCH;
+        let checkpoint_at = start_time + Duration::seconds(30);
+        let retry_after = checkpoint_at + DECOMMISSION_PROGRESS_SAVE_RETRY_BACKOFF;
+        let mut meta = PoolMeta {
+            pools: vec![PoolStatus {
+                id: 0,
+                cmd_line: "pool-0".to_string(),
+                last_update: start_time,
+                decommission: Some(PoolDecommissionInfo {
+                    start_time: Some(start_time),
+                    items_decommissioned: DECOMMISSION_PROGRESS_SAVE_ITEM_THRESHOLD,
+                    ..Default::default()
+                }),
+            }],
+            ..Default::default()
+        };
+
+        let checkpoint = meta
+            .decommission_progress_checkpoint(0, DECOMMISSION_PROGRESS_SAVE_INTERVAL, checkpoint_at)
+            .expect("valid decommission state should produce a checkpoint")
+            .expect("item threshold should produce a checkpoint");
+        meta.defer_decommission_progress_checkpoint(0, checkpoint, retry_after);
+
+        assert!(
+            meta.decommission_progress_checkpoint(0, DECOMMISSION_PROGRESS_SAVE_INTERVAL, checkpoint_at)
+                .expect("retry backoff check should succeed")
+                .is_none()
+        );
+        assert_eq!(meta.pools[0].last_update, start_time);
+        assert_eq!(
+            meta.pools[0]
+                .decommission
+                .as_ref()
+                .expect("decommission info should exist")
+                .progress_save_item_baseline,
+            0
+        );
+    }
+
+    #[test]
+    fn test_decommission_progress_checkpoint_count_scales_with_threshold() {
+        let start_time = OffsetDateTime::UNIX_EPOCH;
+        let checkpoint_at = start_time;
+        let mut meta = PoolMeta {
+            pools: vec![PoolStatus {
+                id: 0,
+                cmd_line: "pool-0".to_string(),
+                last_update: start_time,
+                decommission: Some(PoolDecommissionInfo {
+                    start_time: Some(start_time),
+                    ..Default::default()
+                }),
+            }],
+            ..Default::default()
+        };
+        let mut checkpoint_count = 0;
+
+        for _ in 0..(DECOMMISSION_PROGRESS_SAVE_ITEM_THRESHOLD * 10) {
+            meta.count_item(0, 1, false);
+            if let Some(checkpoint) = meta
+                .decommission_progress_checkpoint(0, DECOMMISSION_PROGRESS_SAVE_INTERVAL, checkpoint_at)
+                .expect("valid decommission state should produce a checkpoint")
+            {
+                checkpoint_count += 1;
+                assert!(meta.commit_decommission_progress_checkpoint(0, checkpoint));
+            }
+        }
+
+        assert_eq!(checkpoint_count, 10);
+    }
+
     #[test]
     fn test_ensure_decommission_not_rebalancing_rejects_running_rebalance() {
         let err = ensure_decommission_not_rebalancing(true).expect_err("rebalance running should be rejected");

crates/ecstore/src/disk/os.rs

@@ -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()

crates/filemeta/src/metacache.rs

@@ -21,6 +21,7 @@ use arc_swap::ArcSwapOption;
 use rmp::Marker;
 use serde::{Deserialize, Serialize};
 use std::cmp::Ordering;
+use std::collections::{HashMap, HashSet};
 use std::str::from_utf8;
 use std::{
     fmt::Debug,
@@ -37,8 +38,10 @@ use tokio::io::{AsyncRead, AsyncReadExt, AsyncWrite, AsyncWriteExt};
 use tokio::spawn;
 use tokio::sync::Mutex;
 use tracing::{debug, warn};
+use uuid::Uuid;
 
 const SLASH_SEPARATOR: &str = "/";
+pub const MAX_META_CACHE_HEAL_CANDIDATES: usize = 1024;
 
 #[derive(Clone, Debug, Default)]
 pub struct MetadataResolutionParams {
@@ -66,6 +69,41 @@ pub struct MetaCacheEntry {
     pub reusable: bool,
 }
 
+#[derive(Clone, Debug, PartialEq, Eq, Hash)]
+pub enum MetaCacheHealCandidateKind {
+    Object,
+    DeleteMarker,
+    UnversionedObject,
+}
+
+#[derive(Clone, Debug, PartialEq, Eq, Hash)]
+pub struct MetaCacheHealCandidate {
+    pub object: String,
+    pub version_id: Option<Uuid>,
+    pub kind: MetaCacheHealCandidateKind,
+    /// Number of raw disk entries that carried this validated version.
+    pub replica_count: usize,
+}
+
+impl MetaCacheHealCandidate {
+    pub fn validated_version(&self) -> Option<Uuid> {
+        match self.kind {
+            MetaCacheHealCandidateKind::Object | MetaCacheHealCandidateKind::DeleteMarker => self.version_id,
+            MetaCacheHealCandidateKind::UnversionedObject => None,
+        }
+    }
+
+    pub fn is_unversioned(&self) -> bool {
+        self.kind == MetaCacheHealCandidateKind::UnversionedObject
+    }
+}
+
+#[derive(Clone, Debug, Default, PartialEq, Eq)]
+pub struct MetaCacheHealDiscovery {
+    pub candidates: Vec<MetaCacheHealCandidate>,
+    pub unverified_count: usize,
+}
+
 impl MetaCacheEntry {
     pub fn marshal_msg(&self) -> Result<Vec<u8>> {
         let mut wr = Vec::new();
@@ -370,6 +408,176 @@ impl MetaCacheEntries {
         })
     }
 
+    /// Discover validated object/delete-marker versions and safe unversioned
+    /// inspection candidates in the raw entries without applying read quorum.
+    /// This is intentionally separate from [`Self::resolve`]: a sub-quorum
+    /// version is a valid heal target even though it must not participate in
+    /// normal reads or writes.
+    ///
+    /// The validated list is bounded and deduplicated by object, version id,
+    /// and metadata kind; each candidate retains the number of raw disk
+    /// entries that carried it so callers can classify sub-quorum versions.
+    /// Entries whose xl.meta cannot be decoded are counted separately for
+    /// discovery accounting; they never become versionless destructive heal
+    /// requests and do not consume the validated quota. An
+    /// [`MetaCacheHealCandidateKind::UnversionedObject`] is always consumed by
+    /// a non-destructive scanner request.
+    pub fn discover_heal_candidates(&self, bucket: &str, max_candidates: usize) -> MetaCacheHealDiscovery {
+        let limit = max_candidates.min(MAX_META_CACHE_HEAL_CANDIDATES);
+        if limit == 0 || bucket.is_empty() {
+            return MetaCacheHealDiscovery::default();
+        }
+
+        let mut discovery = MetaCacheHealDiscovery {
+            candidates: Vec::<MetaCacheHealCandidate>::with_capacity(limit.min(self.0.len())),
+            unverified_count: 0,
+        };
+        let mut seen: HashMap<(String, Option<Uuid>, MetaCacheHealCandidateKind), usize> =
+            HashMap::with_capacity(limit.min(self.0.len()));
+
+        for entry in self.0.iter().flatten() {
+            if !valid_heal_candidate_name(bucket, entry) {
+                continue;
+            }
+
+            let meta = match FileMeta::load(&entry.metadata) {
+                Ok(meta) => meta,
+                Err(_) => {
+                    discovery.unverified_count = discovery.unverified_count.saturating_add(1);
+                    continue;
+                }
+            };
+            let mut entry_seen = HashSet::new();
+
+            for shallow in meta.versions {
+                let version = match shallow.parse_version_meta() {
+                    Ok(version) if version.valid() => version,
+                    Ok(_) | Err(_) => {
+                        discovery.unverified_count = discovery.unverified_count.saturating_add(1);
+                        continue;
+                    }
+                };
+                if version.free_version() {
+                    continue;
+                }
+
+                let payload_header = version.header();
+                if normalize_version_id(shallow.header.version_id) != normalize_version_id(payload_header.version_id)
+                    || shallow.header.version_type != payload_header.version_type
+                {
+                    discovery.unverified_count = discovery.unverified_count.saturating_add(1);
+                    continue;
+                }
+
+                let (kind, version_id) = match version.version_type {
+                    VersionType::Object
+                        if version.object.is_some() && version.delete_marker.is_none() && version.legacy_object.is_none() =>
+                    {
+                        match version.object.as_ref().and_then(|object| object.version_id) {
+                            Some(id) if !id.is_nil() => (MetaCacheHealCandidateKind::Object, Some(id)),
+                            Some(_) | None => (MetaCacheHealCandidateKind::UnversionedObject, None),
+                        }
+                    }
+                    VersionType::Delete
+                        if version.delete_marker.is_some() && version.object.is_none() && version.legacy_object.is_none() =>
+                    {
+                        let Some(id) = version.delete_marker.as_ref().and_then(|marker| marker.version_id) else {
+                            discovery.unverified_count = discovery.unverified_count.saturating_add(1);
+                            continue;
+                        };
+                        if id.is_nil() {
+                            discovery.unverified_count = discovery.unverified_count.saturating_add(1);
+                            continue;
+                        }
+                        (MetaCacheHealCandidateKind::DeleteMarker, Some(id))
+                    }
+                    VersionType::Legacy
+                        if version.legacy_object.is_some() && version.object.is_none() && version.delete_marker.is_none() =>
+                    {
+                        let Some(legacy) = version.legacy_object.as_ref() else {
+                            continue;
+                        };
+                        if legacy.version_id.is_empty() {
+                            (MetaCacheHealCandidateKind::UnversionedObject, None)
+                        } else {
+                            let Ok(id) = Uuid::parse_str(&legacy.version_id) else {
+                                discovery.unverified_count = discovery.unverified_count.saturating_add(1);
+                                continue;
+                            };
+                            if id.is_nil() {
+                                discovery.unverified_count = discovery.unverified_count.saturating_add(1);
+                                continue;
+                            }
+                            (MetaCacheHealCandidateKind::Object, Some(id))
+                        }
+                    }
+                    _ => {
+                        discovery.unverified_count = discovery.unverified_count.saturating_add(1);
+                        continue;
+                    }
+                };
+
+                if normalize_version_id(payload_header.version_id) != version_id {
+                    discovery.unverified_count = discovery.unverified_count.saturating_add(1);
+                    continue;
+                }
+
+                // `all_parts=true` is the trust-boundary check for versioned
+                // candidates. A null/legacy object may still need the old
+                // non-destructive inspection fallback when its part arrays
+                // are parseable but incomplete; never use that fallback for
+                // a candidate carrying a real version id.
+                let file_info = match version.clone().into_fileinfo(bucket, &entry.name, true) {
+                    Ok(file_info) => file_info,
+                    Err(_) if version_id.is_none() && matches!(kind, MetaCacheHealCandidateKind::UnversionedObject) => {
+                        discovery.unverified_count = discovery.unverified_count.saturating_add(1);
+                        match version.into_fileinfo(bucket, &entry.name, false) {
+                            Ok(file_info) => file_info,
+                            Err(_) => continue,
+                        }
+                    }
+                    Err(_) => {
+                        discovery.unverified_count = discovery.unverified_count.saturating_add(1);
+                        continue;
+                    }
+                };
+                if file_info.volume != bucket || file_info.name != entry.name {
+                    discovery.unverified_count = discovery.unverified_count.saturating_add(1);
+                    continue;
+                }
+
+                let candidate = MetaCacheHealCandidate {
+                    object: entry.name.clone(),
+                    version_id,
+                    kind,
+                    replica_count: 1,
+                };
+                let key = (candidate.object.clone(), candidate.version_id, candidate.kind.clone());
+                if !entry_seen.contains(&key) {
+                    // Keep per-entry dedupe bounded as well as the global
+                    // candidate union. Once the cap is reached, only keys
+                    // already present in the global map may update replica
+                    // counts; novel versions are accounting-only.
+                    if entry_seen.len() >= limit && !seen.contains_key(&key) {
+                        continue;
+                    }
+                    entry_seen.insert(key.clone());
+                }
+                if let Some(index) = seen.get(&key).copied() {
+                    discovery.candidates[index].replica_count = discovery.candidates[index].replica_count.saturating_add(1);
+                } else {
+                    seen.insert(key, discovery.candidates.len());
+                    discovery.candidates.push(candidate);
+                    if discovery.candidates.len() >= limit {
+                        return discovery;
+                    }
+                }
+            }
+        }
+
+        discovery
+    }
+
     fn resolve_inner(&self, mut params: MetadataResolutionParams, enforce_write_quorum: bool) -> Option<MetaCacheEntry> {
         if self.0.is_empty() {
             debug!(
@@ -546,6 +754,28 @@ impl MetaCacheEntries {
     }
 }
 
+fn valid_heal_candidate_name(bucket: &str, entry: &MetaCacheEntry) -> bool {
+    if bucket.is_empty() || entry.name.is_empty() || entry.is_dir() || entry.name.contains('\0') {
+        return false;
+    }
+
+    // Validate raw key components without normalizing them. A dot component
+    // could otherwise escape the bucket when the key is later mapped back to
+    // a disk path; a final empty component is retained for valid keys ending
+    // in '/'.
+    let mut components = entry.name.split('/').peekable();
+    while let Some(component) = components.next() {
+        if component == "." || component == ".." || (component.is_empty() && components.peek().is_some()) {
+            return false;
+        }
+    }
+    true
+}
+
+fn normalize_version_id(version_id: Option<Uuid>) -> Option<Uuid> {
+    version_id.filter(|id| !id.is_nil())
+}
+
 #[derive(Debug, Default)]
 pub struct MetaCacheEntriesSortedResult {
     pub entries: Option<MetaCacheEntriesSorted>,
@@ -991,7 +1221,7 @@ impl<T: Clone + Debug + Send + Sync + 'static> Cache<T> {
 mod tests {
     use super::*;
     use crate::test_data::create_real_xlmeta;
-    use crate::{FileMetaVersion, MetaDeleteMarker, TRANSITION_COMPLETE};
+    use crate::{FileMetaVersion, MetaDeleteMarker, MetaObjectV1, MetaObjectV1Erasure, MetaObjectV1Stat, TRANSITION_COMPLETE};
     use std::collections::HashMap;
     use std::io::Cursor;
     use std::sync::{
@@ -1592,6 +1822,277 @@ mod tests {
         );
     }
 
+    #[test]
+    fn discover_heal_candidates_keeps_sub_quorum_versions_and_deduplicates() {
+        let now = OffsetDateTime::from_unix_timestamp(1_705_312_300).expect("valid timestamp");
+        let entries = MetaCacheEntries(vec![
+            Some(metacache_entry_single_version(1, now, "one")),
+            Some(metacache_entry_single_version(2, now, "two")),
+            Some(metacache_entry_single_version(2, now, "two")),
+            Some(metacache_entry_single_version(3, now, "three")),
+        ]);
+
+        let discovery = entries.discover_heal_candidates("bucket", 16);
+        let ids: std::collections::HashSet<Uuid> = discovery
+            .candidates
+            .iter()
+            .filter_map(|candidate| candidate.version_id)
+            .collect();
+        assert_eq!(
+            ids,
+            [Uuid::from_u128(1), Uuid::from_u128(2), Uuid::from_u128(3)]
+                .into_iter()
+                .collect()
+        );
+        assert_eq!(discovery.candidates.len(), 3, "duplicate tied versions must be emitted once");
+        assert_eq!(
+            discovery
+                .candidates
+                .iter()
+                .find(|candidate| candidate.version_id == Some(Uuid::from_u128(2)))
+                .expect("duplicate version should be discovered")
+                .replica_count,
+            2
+        );
+    }
+
+    #[test]
+    fn discover_heal_candidates_covers_divergent_quorum_boundaries_n2_n4_n6() {
+        let now = OffsetDateTime::from_unix_timestamp(1_705_312_300).expect("valid timestamp");
+
+        for (disk_count, quorum) in [(2usize, 1usize), (4, 2), (6, 3)] {
+            let target_id = Uuid::from_u128(0x1000 + disk_count as u128);
+            for target_replicas in [quorum.saturating_sub(1), quorum, quorum + 1] {
+                let entries = (0..disk_count)
+                    .map(|disk| {
+                        let version_id = if disk < target_replicas {
+                            target_id
+                        } else {
+                            Uuid::from_u128(0x2000 + disk as u128)
+                        };
+                        Some(metacache_entry_single_version(version_id.as_u128(), now, "divergent"))
+                    })
+                    .collect();
+                let discovery = MetaCacheEntries(entries).discover_heal_candidates("bucket", 32);
+                let target = discovery
+                    .candidates
+                    .iter()
+                    .find(|candidate| candidate.version_id == Some(target_id));
+                assert_eq!(target.is_some(), target_replicas > 0, "N={disk_count}, replicas={target_replicas}");
+                if let Some(target) = target {
+                    assert_eq!(target.replica_count, target_replicas);
+                }
+            }
+        }
+    }
+
+    #[test]
+    fn discover_heal_candidates_separates_delete_markers_and_preserves_unversioned_objects() {
+        let mut marker_meta = FileMeta::new();
+        marker_meta
+            .add_version(FileInfo {
+                volume: "bucket".to_string(),
+                name: "object".to_string(),
+                version_id: Some(Uuid::from_u128(99)),
+                deleted: true,
+                mod_time: Some(OffsetDateTime::from_unix_timestamp(1_705_312_300).expect("valid timestamp")),
+                ..Default::default()
+            })
+            .expect("delete marker should be added");
+        let marker = MetaCacheEntry {
+            name: "object".to_string(),
+            metadata: marker_meta.marshal_msg().expect("delete marker metadata should marshal"),
+            cached: Some(marker_meta),
+            reusable: false,
+        };
+
+        let unversioned_entry = metacache_entry_with_mod_time(
+            OffsetDateTime::from_unix_timestamp(1_705_312_300).expect("valid timestamp"),
+            "unversioned",
+        );
+        let discovery = MetaCacheEntries(vec![Some(marker), Some(unversioned_entry)]).discover_heal_candidates("bucket", 16);
+        assert!(discovery.candidates.iter().any(|candidate| {
+            candidate.kind == MetaCacheHealCandidateKind::DeleteMarker && candidate.version_id == Some(Uuid::from_u128(99))
+        }));
+        assert!(discovery.candidates.iter().any(|candidate| {
+            candidate.kind == MetaCacheHealCandidateKind::UnversionedObject && candidate.version_id.is_none()
+        }));
+    }
+
+    #[test]
+    fn discover_heal_candidates_skips_free_versions() {
+        let object_id = Uuid::from_u128(100);
+        let free_id = Uuid::from_u128(101);
+        let mut meta = FileMeta::new();
+        meta.add_version(FileInfo {
+            volume: "bucket".to_string(),
+            name: "object".to_string(),
+            version_id: Some(object_id),
+            transition_status: TRANSITION_COMPLETE.to_string(),
+            transitioned_objname: "remote/object".to_string(),
+            transition_version_id: Some(Uuid::from_u128(102)),
+            transition_tier: "WARM".to_string(),
+            mod_time: Some(OffsetDateTime::now_utc()),
+            ..Default::default()
+        })
+        .expect("transitioned object should be added");
+        let mut delete = FileInfo {
+            volume: "bucket".to_string(),
+            name: "object".to_string(),
+            version_id: Some(object_id),
+            mod_time: Some(OffsetDateTime::now_utc()),
+            ..Default::default()
+        };
+        delete.set_tier_free_version_id(&free_id.to_string());
+        meta.delete_version(&delete).expect("free version should be persisted");
+
+        let discovery = MetaCacheEntries(vec![Some(MetaCacheEntry {
+            name: "object".to_string(),
+            metadata: meta.marshal_msg().expect("free version metadata should marshal"),
+            cached: Some(meta),
+            reusable: false,
+        })])
+        .discover_heal_candidates("bucket", 16);
+        assert!(discovery.candidates.is_empty());
+    }
+
+    #[test]
+    fn discover_heal_candidates_preserves_unversioned_legacy_object() {
+        let legacy = MetaObjectV1 {
+            version: "1.0.1".to_string(),
+            format: "xl".to_string(),
+            stat: MetaObjectV1Stat {
+                size: 1,
+                mod_time: Some(OffsetDateTime::from_unix_timestamp(1_705_312_300).expect("valid timestamp")),
+                name: "object".to_string(),
+                ..Default::default()
+            },
+            erasure: MetaObjectV1Erasure {
+                data_blocks: 4,
+                parity_blocks: 2,
+                index: 1,
+                distribution: vec![1, 2, 3, 4, 5, 6],
+                ..Default::default()
+            },
+            ..Default::default()
+        };
+        let version = FileMetaVersion {
+            version_type: VersionType::Legacy,
+            legacy_object: Some(legacy),
+            ..Default::default()
+        };
+        let mut meta = FileMeta::new();
+        meta.versions
+            .push(FileMetaShallowVersion::try_from(version).expect("legacy metadata should marshal"));
+        let discovery = MetaCacheEntries(vec![Some(MetaCacheEntry {
+            name: "object".to_string(),
+            metadata: meta.marshal_msg().expect("legacy metadata should marshal"),
+            cached: Some(meta),
+            reusable: false,
+        })])
+        .discover_heal_candidates("bucket", 16);
+        assert!(discovery.candidates.iter().any(|candidate| {
+            candidate.kind == MetaCacheHealCandidateKind::UnversionedObject && candidate.version_id.is_none()
+        }));
+    }
+
+    #[test]
+    fn discover_heal_candidates_rejects_nil_and_malformed_metadata_and_is_bounded() {
+        let now = OffsetDateTime::from_unix_timestamp(1_705_312_300).expect("valid timestamp");
+        let mut nil = metacache_entry_single_version(1, now, "nil");
+        let mut nil_meta = FileMeta::load(&nil.metadata).expect("nil fixture should decode");
+        let mut nil_version = nil_meta.versions[0]
+            .parse_version_meta()
+            .expect("nil fixture version should decode");
+        nil_version.object.as_mut().expect("object fixture").version_id = Some(Uuid::nil());
+        nil_meta.versions[0] = FileMetaShallowVersion::try_from(nil_version).expect("nil fixture should marshal");
+        nil.metadata = nil_meta.marshal_msg().expect("nil fixture metadata should marshal");
+
+        let mut mismatched = metacache_entry_single_version(2, now, "mismatched");
+        let mut mismatched_meta = FileMeta::load(&mismatched.metadata).expect("mismatched fixture should decode");
+        mismatched_meta.versions[0].header.version_id = Some(Uuid::from_u128(200));
+        mismatched.metadata = mismatched_meta.marshal_msg().expect("mismatched metadata should marshal");
+
+        let mut short_parts = metacache_entry_single_version(3, now, "short-parts");
+        let mut short_parts_meta = FileMeta::load(&short_parts.metadata).expect("short-parts fixture should decode");
+        let mut short_parts_version = short_parts_meta.versions[0]
+            .parse_version_meta()
+            .expect("short-parts fixture version should decode");
+        let object = short_parts_version.object.as_mut().expect("object fixture");
+        object.part_numbers = vec![1];
+        object.part_actual_sizes = vec![1];
+        object.part_sizes.clear();
+        short_parts_meta.versions[0] = FileMetaShallowVersion::try_from(short_parts_version).expect("short-parts should marshal");
+        short_parts.metadata = short_parts_meta.marshal_msg().expect("short-parts metadata should marshal");
+
+        let mut short_unversioned = metacache_entry_with_mod_time(now, "short-unversioned");
+        let mut short_unversioned_meta =
+            FileMeta::load(&short_unversioned.metadata).expect("short-unversioned fixture should decode");
+        let mut short_unversioned_version = short_unversioned_meta.versions[0]
+            .parse_version_meta()
+            .expect("short-unversioned version should decode");
+        let unversioned_object = short_unversioned_version.object.as_mut().expect("unversioned object fixture");
+        unversioned_object.part_numbers = vec![1];
+        unversioned_object.part_actual_sizes = vec![1];
+        unversioned_object.part_sizes.clear();
+        short_unversioned_meta.versions[0] =
+            FileMetaShallowVersion::try_from(short_unversioned_version).expect("short-unversioned should marshal");
+        short_unversioned.metadata = short_unversioned_meta
+            .marshal_msg()
+            .expect("short-unversioned metadata should marshal");
+
+        let mut malformed = nil.clone();
+        malformed.name = "malformed".to_string();
+        malformed.metadata = vec![1, 2, 3];
+
+        let entries = MetaCacheEntries(
+            std::iter::once(Some(nil))
+                .chain(std::iter::once(Some(mismatched)))
+                .chain(std::iter::once(Some(short_parts)))
+                .chain(std::iter::once(Some(short_unversioned)))
+                .chain(std::iter::once(Some(malformed)))
+                .chain((0..32).map(|id| Some(metacache_entry_single_version(id + 10, now, "bounded"))))
+                .collect(),
+        );
+        let discovery = entries.discover_heal_candidates("bucket", 5);
+        assert!(discovery.candidates.len() <= 5);
+        assert!(
+            !discovery
+                .candidates
+                .iter()
+                .any(|candidate| candidate.version_id == Some(Uuid::nil()))
+        );
+        assert!(
+            !discovery
+                .candidates
+                .iter()
+                .any(|candidate| candidate.version_id == Some(Uuid::from_u128(2)))
+        );
+        assert!(
+            !discovery
+                .candidates
+                .iter()
+                .any(|candidate| candidate.version_id == Some(Uuid::from_u128(3)))
+        );
+        assert!(discovery.candidates.iter().any(|candidate| {
+            candidate.kind == MetaCacheHealCandidateKind::UnversionedObject && candidate.version_id.is_none()
+        }));
+        assert!(
+            discovery.unverified_count >= 1,
+            "malformed and rejected metadata must remain observable during discovery"
+        );
+
+        for invalid_name in ["../object", "object//", "object\0name"] {
+            let mut entry = metacache_entry_single_version(400, now, invalid_name);
+            entry.name = invalid_name.to_string();
+            let discovery = MetaCacheEntries(vec![Some(entry)]).discover_heal_candidates("bucket", 5);
+            assert!(
+                discovery.candidates.is_empty(),
+                "invalid key should not become a heal candidate: {invalid_name:?}"
+            );
+        }
+    }
+
     #[test]
     fn resolve_rejects_partial_latest_and_returns_committed_previous_metadata() {
         let old_mod_time = OffsetDateTime::from_unix_timestamp(1_705_312_300).expect("valid timestamp");

crates/heal/src/heal/channel.rs

@@ -1640,6 +1640,60 @@ mod tests {
         assert_eq!(payload["items"].as_array().expect("items should be an array").len(), 0);
     }
 
+    #[tokio::test]
+    async fn test_process_query_request_reports_displaced_terminal_detail() {
+        let heal_manager = Arc::new(HealManager::new(
+            Arc::new(MockStorage),
+            Some(HealConfig {
+                queue_size: 1,
+                ..HealConfig::default()
+            }),
+        ));
+        let mut displaced = HealRequest::new(
+            HealType::Bucket {
+                bucket: "displaced-channel".to_string(),
+            },
+            HealOptions::default(),
+            HealPriority::Low,
+        );
+        displaced.id = "displaced-channel-task".to_string();
+        let displaced_id = displaced.id.clone();
+        heal_manager
+            .submit_heal_request(displaced)
+            .await
+            .expect("initial channel task should queue");
+        heal_manager
+            .submit_heal_request(HealRequest::new(
+                HealType::Bucket {
+                    bucket: "successor-channel".to_string(),
+                },
+                HealOptions::default(),
+                HealPriority::High,
+            ))
+            .await
+            .expect("successor channel task should displace the initial task");
+
+        let processor = HealChannelProcessor::new(heal_manager);
+        let (tx, rx) = oneshot::channel();
+        processor
+            .process_query_request("displaced-channel".to_string(), displaced_id, None, tx)
+            .await
+            .expect("displaced query should process");
+        let response = rx
+            .await
+            .expect("query response should be returned")
+            .expect("displaced query should remain successful");
+        let payload: serde_json::Value = serde_json::from_slice(response.data.as_deref().expect("status payload should exist"))
+            .expect("status payload should be json");
+        assert_eq!(payload["summary"], "stopped");
+        assert!(
+            response
+                .error
+                .as_deref()
+                .is_some_and(|detail| detail.contains("reason=displaced"))
+        );
+    }
+
     #[tokio::test]
     async fn test_process_query_request_reports_running_for_queued_task() {
         let heal_manager = create_test_heal_manager();

crates/heal/src/heal/manager.rs

@@ -40,6 +40,7 @@ use tracing::{debug, error, info, warn};
 use super::{DiskError, Endpoint, HealDiskExt as _, local_disk_map_read};
 
 const KEEP_HEAL_TASK_STATUS_DURATION: Duration = Duration::from_secs(10 * 60);
+const DISPLACED_HEAL_REASON: &str = "reason=displaced; retry_hint=submit_again";
 const LOG_COMPONENT_HEAL: &str = "heal";
 const LOG_SUBSYSTEM_DISK_SCANNER: &str = "disk_scanner";
 const LOG_SUBSYSTEM_MANAGER: &str = "manager";
@@ -120,26 +121,30 @@ struct MrfRepairNoticeTarget {
     version_id: Option<[u8; 16]>,
 }
 
-#[derive(Debug, Clone, PartialEq, Eq)]
+#[derive(Debug, Clone)]
 struct HealAdmissionDecision {
     result: HealAdmissionResult,
-    displaced_task_id: Option<String>,
+    displaced_request: Option<HealRequest>,
 }
 
 impl HealAdmissionDecision {
     const fn new(result: HealAdmissionResult) -> Self {
         Self {
             result,
-            displaced_task_id: None,
+            displaced_request: None,
         }
     }
 
-    fn accepted_with_displacement(displaced_task_id: String) -> Self {
+    fn accepted_with_displacement(displaced_request: HealRequest) -> Self {
         Self {
             result: HealAdmissionResult::Accepted,
-            displaced_task_id: Some(displaced_task_id),
+            displaced_request: Some(displaced_request),
         }
     }
+
+    fn displaced_task_id(&self) -> Option<&str> {
+        self.displaced_request.as_ref().map(|request| request.id.as_str())
+    }
 }
 
 fn lock_mrf_repair_notice_targets(
@@ -151,6 +156,55 @@ fn lock_mrf_repair_notice_targets(
     }
 }
 
+fn lock_displaced_terminals(
+    registry: &StdMutex<HashMap<String, Arc<CompletedHealStatus>>>,
+) -> StdMutexGuard<'_, HashMap<String, Arc<CompletedHealStatus>>> {
+    match registry.lock() {
+        Ok(guard) => guard,
+        Err(poisoned) => poisoned.into_inner(),
+    }
+}
+
+fn record_displaced_terminal(
+    registry: &StdMutex<HashMap<String, Arc<CompletedHealStatus>>>,
+    request: &HealRequest,
+) -> Arc<CompletedHealStatus> {
+    let terminal = Arc::new(CompletedHealStatus {
+        heal_type: request.heal_type.clone(),
+        status: HealTaskStatus::Failed {
+            error: format!("heal task displaced by a higher-priority request ({DISPLACED_HEAL_REASON})"),
+        },
+        result_items_truncated: false,
+        completed_at: SystemTime::now(),
+        seqed_items: Vec::new(),
+        next_seq: 0,
+        min_seq: 0,
+    });
+    let mut terminals = lock_displaced_terminals(registry);
+    prune_completed_heal_statuses(&mut terminals);
+    terminals.insert(request.id.clone(), Arc::clone(&terminal));
+    terminal
+}
+
+async fn remove_displaced_task_aliases(
+    aliases: &Arc<Mutex<HashMap<String, HealTaskAlias>>>,
+    terminals: &StdMutex<HashMap<String, Arc<CompletedHealStatus>>>,
+    task_id: &str,
+    terminal: &Arc<CompletedHealStatus>,
+) {
+    let mut aliases = aliases.lock().await;
+    let alias_ids = aliases
+        .iter()
+        .filter_map(|(alias_id, alias)| (alias.task_id == task_id).then_some(alias_id.clone()))
+        .collect::<Vec<_>>();
+    let mut displaced_terminals = lock_displaced_terminals(terminals);
+    prune_completed_heal_statuses(&mut displaced_terminals);
+    for alias_id in alias_ids {
+        displaced_terminals.insert(alias_id, Arc::clone(terminal));
+    }
+    aliases.retain(|alias_id, alias| alias_id != task_id && alias.task_id != task_id);
+}
+
 async fn remove_task_aliases_for_task(registry: &Arc<Mutex<HashMap<String, HealTaskAlias>>>, task_id: &str) {
     registry
         .lock()
@@ -618,6 +672,14 @@ pub struct HealManager {
     /// are shared so the lookup helper can hand a completed entry to a
     /// caller without cloning the retained result window.
     completed_heals: Arc<Mutex<HashMap<String, Arc<CompletedHealStatus>>>>,
+    /// Terminals for requests removed by priority displacement. An Accepted
+    /// task ID remains queryable for the same process lifetime and the normal
+    /// ten-minute status TTL; clients should treat `reason=displaced` as a
+    /// terminal result and submit a fresh request. This sidecar is synchronous
+    /// so admission can publish the terminal while the queue transition is
+    /// still under its lock, without awaiting another tokio lock. Queue state
+    /// is process-local, so this guarantee does not extend across restart.
+    displaced_terminals: Arc<StdMutex<HashMap<String, Arc<CompletedHealStatus>>>>,
     /// Client tokens merged into an existing task id.
     task_aliases: Arc<Mutex<HashMap<String, HealTaskAlias>>>,
     /// Heal tasks waiting for a retry backoff to expire.
@@ -659,6 +721,7 @@ struct HealQueueContext<'a> {
     heal_queue: &'a Arc<Mutex<PriorityHealQueue>>,
     active_heals: &'a Arc<Mutex<HashMap<String, Arc<HealTask>>>>,
     completed_heals: &'a Arc<Mutex<HashMap<String, Arc<CompletedHealStatus>>>>,
+    displaced_terminals: &'a Arc<StdMutex<HashMap<String, Arc<CompletedHealStatus>>>>,
     task_aliases: &'a Arc<Mutex<HashMap<String, HealTaskAlias>>>,
     retrying_heals: &'a Arc<Mutex<HashMap<String, RetryingHeal>>>,
     mrf_repair_notice_targets: &'a Arc<StdMutex<HashMap<String, Vec<MrfRepairNoticeTarget>>>>,
@@ -874,7 +937,7 @@ impl HealManager {
                         result = "accepted_by_displacement",
                         "Heal queue request accepted by displacement"
                     });
-                    return HealAdmissionDecision::accepted_with_displacement(displaced.id);
+                    return HealAdmissionDecision::accepted_with_displacement(displaced);
                 }
 
                 demote_to_debug_when!(per_object_request, warn, target: "rustfs::heal::manager", {
@@ -1105,6 +1168,7 @@ impl HealManager {
             active_heals: Arc::new(Mutex::new(HashMap::new())),
             heal_queue: Arc::new(Mutex::new(PriorityHealQueue::new())),
             completed_heals: Arc::new(Mutex::new(HashMap::new())),
+            displaced_terminals: Arc::new(StdMutex::new(HashMap::new())),
             task_aliases: Arc::new(Mutex::new(HashMap::new())),
             retrying_heals: Arc::new(Mutex::new(HashMap::new())),
             mrf_repair_notice_targets: Arc::new(StdMutex::new(HashMap::new())),
@@ -1209,6 +1273,10 @@ impl HealManager {
         active_heals.clear();
         publish_active_heal_count(&active_heals);
         self.completed_heals.lock().await.clear();
+        // Do not let the synchronous guard live across the following async lock.
+        {
+            lock_displaced_terminals(&self.displaced_terminals).clear();
+        }
         self.task_aliases.lock().await.clear();
         self.retrying_heals.lock().await.clear();
         lock_mrf_repair_notice_targets(&self.mrf_repair_notice_targets).clear();
@@ -1459,7 +1527,11 @@ impl HealManager {
             task_id = queued_id.to_owned();
         }
         let should_notify = matches!(admission, HealAdmissionResult::Accepted) && config.event_driven_scheduler_enable;
-        let displaced_task_id = admission_decision.displaced_task_id;
+        let displaced_task_id = admission_decision.displaced_task_id().map(ToOwned::to_owned);
+        let displaced_terminal = admission_decision
+            .displaced_request
+            .as_ref()
+            .map(|request| record_displaced_terminal(&self.displaced_terminals, request));
         if matches!(admission, HealAdmissionResult::Accepted | HealAdmissionResult::Merged)
             && let Some(target) = mrf_notice_target
         {
@@ -1473,8 +1545,12 @@ impl HealManager {
         drop(queue);
         drop(active_heals);
 
-        if let Some(displaced_task_id) = displaced_task_id {
-            self.remove_aliases_for_task(&displaced_task_id).await;
+        if let (Some(displaced_task_id), Some(displaced_terminal)) = (displaced_task_id, displaced_terminal) {
+            // The queue has already removed the displaced request, so the
+            // synchronous terminal sidecar was published before aliases and
+            // MRF ownership are cleaned up.
+            remove_displaced_task_aliases(&self.task_aliases, &self.displaced_terminals, &displaced_task_id, &displaced_terminal)
+                .await;
         }
 
         if should_notify {
@@ -1549,6 +1625,15 @@ impl HealManager {
             }
         }
 
+        if terminal_completed.is_none() {
+            let mut displaced_terminals = lock_displaced_terminals(&self.displaced_terminals);
+            prune_completed_heal_statuses(&mut displaced_terminals);
+            terminal_completed = displaced_terminals
+                .get(canonical_task_id)
+                .filter(|terminal| matches_path(&terminal.heal_type))
+                .cloned();
+        }
+
         match terminal_completed {
             Some(completed) => TaskStateLookup::Completed(completed),
             None => TaskStateLookup::NotFound,
@@ -1669,9 +1754,19 @@ impl HealManager {
 
         let mut completed_heals = self.completed_heals.lock().await;
         prune_completed_heal_statuses(&mut completed_heals);
-        completed_heals
+        if completed_heals
             .values()
             .any(|completed| heal_type_matches_path(&completed.heal_type, heal_path))
+        {
+            return true;
+        }
+        drop(completed_heals);
+
+        let mut displaced_terminals = lock_displaced_terminals(&self.displaced_terminals);
+        prune_completed_heal_statuses(&mut displaced_terminals);
+        displaced_terminals
+            .values()
+            .any(|terminal| heal_type_matches_path(&terminal.heal_type, heal_path))
     }
 
     /// Get task progress

crates/heal/src/heal/manager/auto_scan.rs

@@ -21,6 +21,7 @@ impl HealManager {
         let heal_queue = self.heal_queue.clone();
         let active_heals = self.active_heals.clone();
         let task_aliases = self.task_aliases.clone();
+        let displaced_terminals = self.displaced_terminals.clone();
         let mrf_repair_notice_targets = self.mrf_repair_notice_targets.clone();
         let storage = self.storage.clone();
         let replacement_recovery_anchors = self.replacement_recovery_anchors.clone();
@@ -481,6 +482,10 @@ impl HealManager {
                             let admission = admission_decision.result;
                             let should_notify =
                                 matches!(admission, HealAdmissionResult::Accepted) && config.event_driven_scheduler_enable;
+                            let displaced_terminal = admission_decision
+                                .displaced_request
+                                .as_ref()
+                                .map(|request| record_displaced_terminal(&displaced_terminals, request));
                             if matches!(admission, HealAdmissionResult::Accepted)
                                 && let Some(anchor) = recovery_anchor
                             {
@@ -491,8 +496,16 @@ impl HealManager {
                             }
                             drop(queue);
                             drop(config);
-                            if let Some(displaced_task_id) = admission_decision.displaced_task_id {
-                                remove_task_aliases_for_task(&task_aliases, &displaced_task_id).await;
+                            if let (Some(displaced_task_id), Some(displaced_terminal)) =
+                                (admission_decision.displaced_task_id().map(ToOwned::to_owned), displaced_terminal)
+                            {
+                                remove_displaced_task_aliases(
+                                    &task_aliases,
+                                    &displaced_terminals,
+                                    &displaced_task_id,
+                                    &displaced_terminal,
+                                )
+                                .await;
                                 lock_mrf_repair_notice_targets(&mrf_repair_notice_targets).remove(&displaced_task_id);
                             }
                             if matches!(admission, HealAdmissionResult::Accepted) {

crates/heal/src/heal/manager/scheduler.rs

@@ -21,6 +21,7 @@ impl HealManager {
         let heal_queue = self.heal_queue.clone();
         let active_heals = self.active_heals.clone();
         let completed_heals = self.completed_heals.clone();
+        let displaced_terminals = self.displaced_terminals.clone();
         let task_aliases = self.task_aliases.clone();
         let retrying_heals = self.retrying_heals.clone();
         let mrf_repair_notice_targets = self.mrf_repair_notice_targets.clone();
@@ -53,6 +54,7 @@ impl HealManager {
                             heal_queue: &heal_queue,
                             active_heals: &active_heals,
                             completed_heals: &completed_heals,
+                            displaced_terminals: &displaced_terminals,
                             task_aliases: &task_aliases,
                             retrying_heals: &retrying_heals,
                             mrf_repair_notice_targets: &mrf_repair_notice_targets,
@@ -71,6 +73,7 @@ impl HealManager {
                             heal_queue: &heal_queue,
                             active_heals: &active_heals,
                             completed_heals: &completed_heals,
+                            displaced_terminals: &displaced_terminals,
                             task_aliases: &task_aliases,
                             retrying_heals: &retrying_heals,
                             mrf_repair_notice_targets: &mrf_repair_notice_targets,
@@ -98,6 +101,7 @@ impl HealManager {
             heal_queue,
             active_heals,
             completed_heals,
+            displaced_terminals,
             task_aliases,
             retrying_heals,
             mrf_repair_notice_targets,
@@ -183,6 +187,7 @@ impl HealManager {
                 let active_heals_clone = active_heals.clone();
                 let heal_queue_clone = heal_queue.clone();
                 let completed_heals_clone = completed_heals.clone();
+                let displaced_terminals_clone = displaced_terminals.clone();
                 let task_aliases_clone = task_aliases.clone();
                 let retrying_heals_clone = retrying_heals.clone();
                 let mrf_repair_notice_targets_clone = mrf_repair_notice_targets.clone();
@@ -363,6 +368,7 @@ impl HealManager {
                         let retry_heal_queue = heal_queue_clone.clone();
                         let retrying_heals_for_spawn = retrying_heals_clone.clone();
                         let retry_task_aliases = task_aliases_clone.clone();
+                        let retry_displaced_terminals = displaced_terminals_clone.clone();
                         let retry_mrf_repair_notice_targets = mrf_repair_notice_targets_clone.clone();
                         let retry_completed_heals = completed_heals_clone.clone();
                         let retry_notify = notify_clone.clone();
@@ -430,6 +436,14 @@ impl HealManager {
                                 let admission = admission_decision.result;
                                 let should_notify = matches!(admission, HealAdmissionResult::Accepted)
                                     && retry_config.event_driven_scheduler_enable;
+                                // Publish the terminal synchronously while the
+                                // queue transition is protected. The subsequent
+                                // queue -> retrying handoff retains the lock order
+                                // used by operations_snapshot.
+                                let displaced_terminal = admission_decision
+                                    .displaced_request
+                                    .as_ref()
+                                    .map(|request| record_displaced_terminal(&retry_displaced_terminals, request));
                                 match admission {
                                     HealAdmissionResult::Accepted => {
                                         // Transfer ownership while holding queue -> retrying,
@@ -437,10 +451,18 @@ impl HealManager {
                                         #[cfg(test)]
                                         pause_retry_ownership_transition(&retry_request_id, true).await;
                                         retrying_heals_for_spawn.lock().await.remove(&retry_request_id);
-                                        let displaced_task_id = admission_decision.displaced_task_id;
+                                        let displaced_task_id = admission_decision.displaced_task_id().map(ToOwned::to_owned);
                                         drop(queue);
-                                        if let Some(displaced_task_id) = displaced_task_id {
-                                            remove_task_aliases_for_task(&retry_task_aliases, &displaced_task_id).await;
+                                        if let (Some(displaced_task_id), Some(displaced_terminal)) =
+                                            (displaced_task_id, displaced_terminal)
+                                        {
+                                            remove_displaced_task_aliases(
+                                                &retry_task_aliases,
+                                                &retry_displaced_terminals,
+                                                &displaced_task_id,
+                                                &displaced_terminal,
+                                            )
+                                            .await;
                                             remove_mrf_repair_notice_targets(
                                                 &retry_mrf_repair_notice_targets,
                                                 &displaced_task_id,

crates/heal/src/heal/manager/tests.rs

@@ -84,6 +84,7 @@ async fn process_manager_queue_once(manager: &HealManager) {
         heal_queue: &manager.heal_queue,
         active_heals: &manager.active_heals,
         completed_heals: &manager.completed_heals,
+        displaced_terminals: &manager.displaced_terminals,
         task_aliases: &manager.task_aliases,
         retrying_heals: &manager.retrying_heals,
         mrf_repair_notice_targets: &manager.mrf_repair_notice_targets,
@@ -2778,7 +2779,10 @@ async fn test_high_priority_request_displaces_lower_priority_when_queue_full() {
         HealAdmissionResult::Accepted
     );
     assert_eq!(manager.get_queue_length().await, 1);
-    assert!(matches!(manager.get_task_status(&low_id).await, Err(Error::TaskNotFound { .. })));
+    assert!(matches!(
+        manager.get_task_status(&low_id).await,
+        Ok(HealTaskStatus::Failed { error }) if error.contains("reason=displaced")
+    ));
     assert_eq!(
         manager
             .get_task_status(&high_id)
@@ -2788,6 +2792,263 @@ async fn test_high_priority_request_displaces_lower_priority_when_queue_full() {
     );
 }
 
+#[tokio::test]
+async fn displaced_task_remains_queryable() {
+    let manager = HealManager::new(
+        Arc::new(MockStorage),
+        Some(HealConfig {
+            queue_size: 1,
+            ..HealConfig::default()
+        }),
+    );
+    let mut displaced = HealRequest::new(
+        HealType::Bucket {
+            bucket: "displaced-bucket".to_string(),
+        },
+        HealOptions::default(),
+        HealPriority::Low,
+    );
+    displaced.id = "displaced-task".to_string();
+    let displaced_id = displaced.id.clone();
+    manager
+        .submit_heal_request(displaced)
+        .await
+        .expect("displaced request should queue");
+
+    let successor = HealRequest::new(
+        HealType::Bucket {
+            bucket: "successor-bucket".to_string(),
+        },
+        HealOptions::default(),
+        HealPriority::High,
+    );
+    manager
+        .submit_heal_request(successor)
+        .await
+        .expect("successor should displace low work");
+
+    let report = manager
+        .get_task_report(&displaced_id)
+        .await
+        .expect("displaced report should remain queryable");
+    assert!(matches!(report.status, HealTaskStatus::Failed { ref error } if error.contains("reason=displaced")));
+}
+
+#[tokio::test]
+async fn displaced_archive_failure_keeps_queryable_terminal() {
+    let manager = HealManager::new(Arc::new(MockStorage), None);
+    let mut request = HealRequest::new(
+        HealType::Bucket {
+            bucket: "archive-failure".to_string(),
+        },
+        HealOptions::default(),
+        HealPriority::Low,
+    );
+    request.id = "archive-failure-task".to_string();
+    let request_id = request.id.clone();
+    // The synchronous sidecar is the authoritative fallback when the normal
+    // completed-task archive has no entry (the failure window that must not
+    // turn an Accepted ID into NotFound).
+    record_displaced_terminal(&manager.displaced_terminals, &request);
+    assert!(manager.completed_heals.lock().await.is_empty());
+    assert!(matches!(
+        manager.get_task_status(&request_id).await,
+        Ok(HealTaskStatus::Failed { error }) if error.contains("reason=displaced")
+    ));
+}
+
+#[tokio::test]
+async fn scheduler_retry_displacement_keeps_evicted_task_queryable() {
+    let manager = Arc::new(HealManager::new(
+        Arc::new(MockStorage),
+        Some(HealConfig {
+            queue_size: 1,
+            event_driven_scheduler_enable: false,
+            ..HealConfig::default()
+        }),
+    ));
+    let mut retry_request = HealRequest::object("retry-transition".to_string(), "object".to_string(), None);
+    retry_request.priority = HealPriority::High;
+    let retry_id = retry_request.id.clone();
+    manager
+        .submit_heal_request(retry_request)
+        .await
+        .expect("retry request should queue");
+
+    // Process exactly one queue cycle so the retry task is spawned without a
+    // background scheduler consuming the filler request before the retry wakes.
+    process_manager_queue_once(&manager).await;
+    tokio::time::timeout(Duration::from_secs(1), async {
+        loop {
+            if manager.retrying_heals.lock().await.contains_key(&retry_id) {
+                break;
+            }
+            tokio::task::yield_now().await;
+        }
+    })
+    .await
+    .expect("retry request should enter backoff");
+
+    let filler = HealRequest::new(
+        HealType::Bucket {
+            bucket: "retry-displaced-filler".to_string(),
+        },
+        HealOptions::default(),
+        HealPriority::Low,
+    );
+    let filler_id = filler.id.clone();
+    manager
+        .submit_heal_request(filler)
+        .await
+        .expect("filler request should occupy the queue");
+
+    tokio::time::timeout(Duration::from_secs(5), async {
+        loop {
+            if matches!(
+                manager.get_task_status(&filler_id).await,
+                Ok(HealTaskStatus::Failed { ref error }) if error.contains("reason=displaced")
+            ) {
+                break;
+            }
+            tokio::time::sleep(Duration::from_millis(10)).await;
+        }
+    })
+    .await
+    .expect("retry admission should displace the filler request");
+    assert_eq!(manager.get_queue_length().await, 1);
+    assert_eq!(
+        manager.get_task_status(&retry_id).await.expect("retry should be queued"),
+        HealTaskStatus::Pending
+    );
+}
+
+#[tokio::test]
+async fn concurrent_displacers_produce_one_terminal_generation() {
+    let manager = Arc::new(HealManager::new(
+        Arc::new(MockStorage),
+        Some(HealConfig {
+            queue_size: 1,
+            ..HealConfig::default()
+        }),
+    ));
+    let mut displaced = HealRequest::new(
+        HealType::Bucket {
+            bucket: "concurrent-displaced".to_string(),
+        },
+        HealOptions::default(),
+        HealPriority::Low,
+    );
+    displaced.id = "concurrent-displaced-task".to_string();
+    let displaced_id = displaced.id.clone();
+    manager
+        .submit_heal_request(displaced)
+        .await
+        .expect("initial request should queue");
+
+    let first = HealRequest::new(
+        HealType::Bucket {
+            bucket: "concurrent-successor-a".to_string(),
+        },
+        HealOptions::default(),
+        HealPriority::High,
+    );
+    let second = HealRequest::new(
+        HealType::Bucket {
+            bucket: "concurrent-successor-b".to_string(),
+        },
+        HealOptions::default(),
+        HealPriority::High,
+    );
+    let (first_result, second_result) = tokio::join!(manager.submit_heal_request(first), manager.submit_heal_request(second));
+    let accepted = [&first_result, &second_result]
+        .into_iter()
+        .filter(|result| matches!(result, Ok(HealAdmissionResult::Accepted)))
+        .count();
+    assert_eq!(accepted, 1, "exactly one concurrent displacer should win the full queue");
+    assert!(
+        first_result.is_ok() && second_result.is_ok(),
+        "the losing request should receive a typed Full result"
+    );
+    let terminals = lock_displaced_terminals(&manager.displaced_terminals);
+    assert_eq!(terminals.len(), 1);
+    assert!(terminals.contains_key(&displaced_id));
+}
+
+#[tokio::test]
+async fn successor_chain_is_bounded_and_authorized() {
+    let manager = HealManager::new(
+        Arc::new(MockStorage),
+        Some(HealConfig {
+            queue_size: 1,
+            ..HealConfig::default()
+        }),
+    );
+    let mut original = HealRequest::new(
+        HealType::Bucket {
+            bucket: "authorized-original".to_string(),
+        },
+        HealOptions::default(),
+        HealPriority::Low,
+    );
+    original.id = "authorized-original-task".to_string();
+    let original_id = original.id.clone();
+    manager.submit_heal_request(original).await.expect("original should queue");
+    let mut duplicate = HealRequest::new(
+        HealType::Bucket {
+            bucket: "authorized-original".to_string(),
+        },
+        HealOptions::default(),
+        HealPriority::Low,
+    );
+    duplicate.id = "authorized-duplicate-task".to_string();
+    let duplicate_id = duplicate.id.clone();
+    manager
+        .submit_heal_request(duplicate)
+        .await
+        .expect("same-target duplicate should merge");
+    let successor = HealRequest::new(
+        HealType::Bucket {
+            bucket: "authorized-successor".to_string(),
+        },
+        HealOptions::default(),
+        HealPriority::High,
+    );
+    let successor_id = successor.id.clone();
+    manager.submit_heal_request(successor).await.expect("successor should queue");
+    assert!(manager.task_aliases.lock().await.is_empty());
+    assert!(matches!(manager.get_task_status(&original_id).await, Ok(HealTaskStatus::Failed { .. })));
+    assert!(matches!(manager.get_task_status(&duplicate_id).await, Ok(HealTaskStatus::Failed { .. })));
+    assert_eq!(
+        manager
+            .get_task_status(&successor_id)
+            .await
+            .expect("successor should remain queued"),
+        HealTaskStatus::Pending
+    );
+}
+
+#[tokio::test]
+async fn displaced_terminal_expires_after_bounded_ttl() {
+    let manager = HealManager::new(Arc::new(MockStorage), None);
+    let mut request = HealRequest::new(
+        HealType::Bucket {
+            bucket: "expires".to_string(),
+        },
+        HealOptions::default(),
+        HealPriority::Low,
+    );
+    request.id = "expires-task".to_string();
+    let request_id = request.id.clone();
+    record_displaced_terminal(&manager.displaced_terminals, &request);
+    {
+        let mut terminals = lock_displaced_terminals(&manager.displaced_terminals);
+        let entry =
+            Arc::get_mut(terminals.get_mut(&request_id).expect("terminal should be retained")).expect("test owns terminal entry");
+        entry.completed_at = SystemTime::now() - KEEP_HEAL_TASK_STATUS_DURATION - Duration::from_secs(1);
+    }
+    assert!(matches!(manager.get_task_status(&request_id).await, Err(Error::TaskNotFound { .. })));
+}
+
 #[tokio::test]
 async fn test_displacing_registered_mrf_task_drops_notice_ownership() {
     let storage: Arc<dyn HealStorageAPI> = Arc::new(MockStorage);

crates/protos/src/generated/proto_gen/node_service.rs

@@ -722,6 +722,10 @@ pub struct DeleteVersionsResponse {
     pub errors: ::prost::alloc::vec::Vec<::prost::alloc::string::String>,
     #[prost(message, optional, tag = "3")]
     pub error: ::core::option::Option<Error>,
+    /// Senders dual-write the legacy strings and typed entries. Receivers prefer typed entries
+    /// when present and fall back to strings for peers that predate this field. Code zero means success.
+    #[prost(message, repeated, tag = "4")]
+    pub item_errors: ::prost::alloc::vec::Vec<Error>,
 }
 #[derive(Clone, PartialEq, Eq, Hash, ::prost::Message)]
 pub struct ReadMultipleRequest {

crates/protos/src/lib.rs

@@ -2106,6 +2106,9 @@ pub enum ChannelClass {
     Bulk,
 }
 
+// Keep multiplexed unary RPCs below h2's per-connection small-frame budget.
+const INTERNODE_RPC_CONCURRENCY_LIMIT: usize = 64;
+
 /// Whether control/bulk channel isolation is enabled (env-gated, default off for safe rollout).
 fn channel_isolation_enabled() -> bool {
     rustfs_utils::get_env_bool(
@@ -2188,6 +2191,7 @@ async fn build_channel(dial_addr: &str, cache_key: &str) -> Result<Channel, Box<
     let mut connector = Endpoint::from_shared(dial_addr.to_string())?
         // Fast connection timeout for dead peer detection
         .connect_timeout(connect_timeout)
+        .concurrency_limit(INTERNODE_RPC_CONCURRENCY_LIMIT)
         // TCP-level keepalive - OS will probe connection
         .tcp_keepalive(Some(tcp_keepalive))
         // Disable Nagle so latency-sensitive control-plane RPCs (locks/health) are not batched

crates/protos/src/node.proto

@@ -493,6 +493,9 @@ message DeleteVersionsResponse {
   bool success = 1;
   repeated string errors = 2;
   optional Error error = 3;
+  // Senders dual-write the legacy strings and typed entries. Receivers prefer typed entries
+  // when present and fall back to strings for peers that predate this field. Code zero means success.
+  repeated Error item_errors = 4;
 }
 
 message ReadMultipleRequest {

crates/scanner/src/scanner_folder.rs

@@ -43,7 +43,7 @@ use rustfs_common::metrics::{
     UpdateCurrentPathFn, current_path_updater, global_metrics,
 };
 use rustfs_common::trace_bus::{TraceEvent, TraceFunc, TraceKind, trace_emit, trace_subscriber_count};
-use rustfs_filemeta::{MetaCacheEntries, MetaCacheEntry, MetadataResolutionParams};
+use rustfs_filemeta::{MAX_META_CACHE_HEAL_CANDIDATES, MetaCacheEntries, MetaCacheEntry, MetaCacheHealCandidateKind};
 use rustfs_utils::path::{SLASH_SEPARATOR, path_join_buf};
 use s3s::dto::{BucketLifecycleConfiguration, ObjectLockConfiguration, VersioningConfiguration};
 use time::OffsetDateTime;
@@ -96,6 +96,10 @@ const METRIC_SCANNER_EXCESS_OBJECT_VERSION_SIZE_TOTAL: &str = "rustfs_scanner_ex
 const METRIC_SCANNER_EXCESS_FOLDERS_TOTAL: &str = "rustfs_scanner_excess_folders_total";
 const METRIC_SCANNER_PENDING_HEAL_PRUNE_TOTAL: &str = "rustfs_scanner_pending_heal_prune_total";
 const METRIC_SCANNER_PENDING_HEAL_MALFORMED_TOTAL: &str = "rustfs_scanner_pending_heal_malformed_total";
+const METRIC_SCANNER_HEAL_DISCOVERY_CANDIDATES_TOTAL: &str = "rustfs_scanner_heal_discovery_candidates_total";
+const METRIC_SCANNER_HEAL_DISCOVERY_SUB_QUORUM_TOTAL: &str = "rustfs_scanner_heal_discovery_sub_quorum_total";
+const METRIC_SCANNER_HEAL_DISCOVERY_UNVERIFIED_TOTAL: &str = "rustfs_scanner_heal_discovery_unverified_total";
+const METRIC_SCANNER_HEAL_DISCOVERY_QUEUED_TOTAL: &str = "rustfs_scanner_heal_discovery_queued_total";
 const MAX_PENDING_SCANNER_HEAL_RETRIES_PER_BUCKET: usize = 128;
 
 // --- scanner excess alerts as S3 notification events (rustfs/backlog#1868) --
@@ -883,7 +887,7 @@ impl FolderScanner {
         object: Option<String>,
         version_id: Option<String>,
         request: HealChannelRequest,
-    ) -> Result<(), ScannerError> {
+    ) -> Result<HealAdmissionResult, ScannerError> {
         let candidate_type = pending_scanner_heal_candidate_type(kind);
         let priority = request.priority;
         let scan_mode = request.scan_mode.unwrap_or(self.scan_mode);
@@ -911,7 +915,7 @@ impl FolderScanner {
                     error = %err,
                     "Scanner deferred heal request after channel error"
                 );
-                return Ok(());
+                return Ok(HealAdmissionResult::Full);
             }
         };
         self.update_pending_scanner_heal_after_admission(
@@ -923,7 +927,7 @@ impl FolderScanner {
             result,
         );
         if result.is_admitted() {
-            return Ok(());
+            return Ok(result);
         }
 
         record_high_priority_heal_escalation(candidate_type, priority, result);
@@ -944,7 +948,7 @@ impl FolderScanner {
             state = "high_priority_not_admitted",
             "Scanner high-priority heal admission failed"
         );
-        Ok(())
+        Ok(result)
     }
 
     pub fn set_heal_object_select(&mut self, prob: u32) {
@@ -1736,14 +1740,7 @@ impl FolderScanner {
                 break;
             }
 
-            let mut resolver = MetadataResolutionParams {
-                dir_quorum: self.disks_quorum,
-                obj_quorum: self.disks_quorum,
-                bucket: "".to_string(),
-                strict: false,
-                ..Default::default()
-            };
-
+            let mut previous_bucket = String::new();
             for name in abandoned_children {
                 if !self.should_heal().await {
                     break;
@@ -1751,7 +1748,7 @@ impl FolderScanner {
 
                 let (bucket, prefix) = path2_bucket_object(name.as_str());
 
-                if bucket != resolver.bucket {
+                if bucket != previous_bucket {
                     self.send_required_scanner_heal_request(
                         PendingScannerHealKind::Bucket,
                         bucket.clone(),
@@ -1760,10 +1757,9 @@ impl FolderScanner {
                         build_bucket_heal_request(bucket.clone(), HealChannelPriority::High),
                     )
                     .await?;
+                    previous_bucket = bucket.clone();
                 }
 
-                resolver.bucket = bucket.clone();
-
                 let child_ctx = ctx.child_token();
 
                 let (agreed_tx, mut agreed_rx) = mpsc::channel::<String>(1);
@@ -1880,6 +1876,7 @@ impl FolderScanner {
                 let mut agreed_closed = false;
                 let mut partial_closed = false;
                 let mut finished_closed = false;
+                let mut seen_heal_candidates: HashSet<(String, Option<String>, MetaCacheHealCandidateKind)> = HashSet::new();
 
                 loop {
                     if agreed_closed && partial_closed && finished_closed {
@@ -1904,65 +1901,56 @@ impl FolderScanner {
                                 break;
                             }
 
-                            let Some(entry) = resolve_object_heal_entry(&entries, resolver.clone()) else {
-                                continue;
-                            };
+                            let discovery = entries.discover_heal_candidates(&bucket, MAX_META_CACHE_HEAL_CANDIDATES);
+                            counter!(METRIC_SCANNER_HEAL_DISCOVERY_CANDIDATES_TOTAL)
+                                .increment(u64::try_from(discovery.candidates.len()).unwrap_or(u64::MAX));
+                            counter!(METRIC_SCANNER_HEAL_DISCOVERY_SUB_QUORUM_TOTAL).increment(
+                                u64::try_from(
+                                    discovery
+                                        .candidates
+                                        .iter()
+                                        .filter(|candidate| candidate.replica_count < disks_quorum)
+                                        .count(),
+                                )
+                                .unwrap_or(u64::MAX),
+                            );
+                            counter!(METRIC_SCANNER_HEAL_DISCOVERY_UNVERIFIED_TOTAL).increment(
+                                u64::try_from(discovery.unverified_count).unwrap_or(u64::MAX),
+                            );
 
-                            (self.update_current_path)(&entry.name).await;
-
-                            if entry.is_dir() {
-                                continue;
-                            }
-
-                            let fivs = match entry.file_info_versions(&bucket) {
-                                Ok(fivs) => fivs,
-                                Err(e) => {
-                                    error!(
-                                        target: "rustfs::scanner::folder",
-                                        event = EVENT_SCANNER_FOLDER_STATE,
-                                        component = LOG_COMPONENT_SCANNER,
-                                        subsystem = LOG_SUBSYSTEM_FOLDER,
-                                        bucket = %bucket,
-                                        entry = %entry.name,
-                                        state = "file_info_versions_failed",
-                                        error = %e,
-                                        "Scanner list_path_raw failed to resolve file versions"
-                                    );
-                                    self.send_required_scanner_heal_request(
-                                        PendingScannerHealKind::Object,
-                                        bucket.clone(),
-                                        Some(entry.name.clone()),
-                                        None,
-                                        build_object_heal_request(
-                                            bucket.clone(),
-                                            entry.name.clone(),
-                                            None,
-                                            self.scan_mode,
-                                            HealChannelPriority::High,
-                                        ),
-                                    )
-                                    .await?;
-                                    found_objects = true;
+                            for candidate in discovery.candidates {
+                                let version_id = candidate.validated_version().map(|id| id.to_string());
+                                let identity = (candidate.object.clone(), version_id.clone(), candidate.kind.clone());
+                                if seen_heal_candidates.len() >= MAX_META_CACHE_HEAL_CANDIDATES
+                                    && !seen_heal_candidates.contains(&identity)
+                                {
                                     continue;
                                 }
-                            };
-
-                            for fiv in fivs.versions {
-                                let version_id = fiv.version_id.and_then(|v| if v.is_nil() { None } else { Some(v.to_string()) });
-                                self.send_required_scanner_heal_request(
+                                if !seen_heal_candidates.insert(identity) {
+                                    continue;
+                                }
+                                let mut request = build_object_heal_request(
+                                    bucket.clone(),
+                                    candidate.object.clone(),
+                                    version_id.clone(),
+                                    self.scan_mode,
+                                    HealChannelPriority::High,
+                                );
+                                if candidate.is_unversioned() {
+                                    request.remove_corrupted = Some(false);
+                                }
+                                (self.update_current_path)(&candidate.object).await;
+                                let admission = self.send_required_scanner_heal_request(
                                     PendingScannerHealKind::Object,
                                     bucket.clone(),
-                                    Some(entry.name.clone()),
-                                    version_id.clone(),
-                                    build_object_heal_request(
-                                        bucket.clone(),
-                                        entry.name.clone(),
-                                        version_id,
-                                        self.scan_mode,
-                                        HealChannelPriority::High,
-                                    ),
+                                    Some(candidate.object.clone()),
+                                    version_id,
+                                    request,
                                 )
                                 .await?;
+                                if admission.is_admitted() {
+                                    counter!(METRIC_SCANNER_HEAL_DISCOVERY_QUEUED_TOTAL).increment(1);
+                                }
                                 found_objects = true;
                             }
 

crates/scanner/src/scanner_folder/item_actions.rs

@@ -13,6 +13,8 @@
 // limitations under the License.
 /// Per-object scan actions: ScannerItem, the get-size failure policy, and the heal/ILM admission helpers.
 use super::*;
+#[cfg(test)]
+use rustfs_filemeta::MetadataResolutionParams;
 
 /// Cached folder information for scanning
 #[derive(Clone, Debug)]
@@ -88,6 +90,7 @@ pub(super) fn build_object_heal_request(
     }
 }
 
+#[cfg(test)]
 pub(super) fn resolve_object_heal_entry(
     entries: &MetaCacheEntries,
     resolver: MetadataResolutionParams,

crates/scanner/src/scanner_folder/ledger.rs

@@ -305,13 +305,17 @@ pub(super) fn build_pending_scanner_heal_request(entry: &PendingScannerHeal) ->
     match entry.kind {
         PendingScannerHealKind::Bucket => Some(build_bucket_heal_request(entry.bucket.clone(), HealChannelPriority::High)),
         PendingScannerHealKind::Object => entry.object.as_ref().map(|object| {
-            build_object_heal_request(
+            let mut request = build_object_heal_request(
                 entry.bucket.clone(),
                 object.clone(),
                 entry.version_id.clone(),
                 entry.scan_mode,
                 HealChannelPriority::High,
-            )
+            );
+            if entry.version_id.is_none() {
+                request.remove_corrupted = Some(false);
+            }
+            request
         }),
     }
 }

crates/scanner/src/scanner_folder/tests.rs

@@ -17,7 +17,7 @@ use crate::SCANNER_SLEEPER;
 use super::*;
 use crate::storage_api::VersionPurgeStatusType;
 use crate::{DiskOption, Endpoint, STORAGE_FORMAT_FILE, TierStats, new_disk, storageclass};
-use rustfs_filemeta::{FileInfo, FileMeta};
+use rustfs_filemeta::{FileInfo, FileMeta, MetadataResolutionParams};
 use std::io::Write;
 #[cfg(unix)]
 use std::os::unix::fs::{PermissionsExt, symlink};
@@ -1121,6 +1121,17 @@ fn test_pending_heal_reconstructs_object_request_with_version() {
     assert_eq!(request.source, HealRequestSource::Scanner);
 }
 
+#[test]
+fn test_pending_heal_reconstructs_unversioned_request_without_removal() {
+    let pending = pending_heal(PendingScannerHealKind::Object, "bucket", Some("object"), None, 1, 1);
+
+    let request = build_pending_scanner_heal_request(&pending).expect("unversioned object request should rebuild");
+
+    assert!(request.object_version_id.is_none());
+    assert_eq!(request.remove_corrupted, Some(false));
+    assert_eq!(request.recreate_missing, Some(false));
+}
+
 #[test]
 fn test_pending_heal_retry_candidates_respect_cap_and_order() {
     let pending: Vec<PendingScannerHeal> = (0..(MAX_PENDING_SCANNER_HEAL_RETRIES_PER_BUCKET + 2))
@@ -1323,6 +1334,20 @@ fn metadata_for_object(bucket: &str, object: &str) -> Vec<u8> {
     meta.marshal_msg().expect("test metadata should marshal")
 }
 
+fn metadata_for_object_version(bucket: &str, object: &str, version_id: Option<Uuid>) -> Vec<u8> {
+    let mut file_info = FileInfo::new(object, 4, 2);
+    file_info.volume = bucket.to_string();
+    file_info.name = object.to_string();
+    file_info.version_id = version_id;
+    file_info.versioned = version_id.is_some();
+    file_info.mod_time = Some(OffsetDateTime::now_utc());
+    file_info.size = 1;
+
+    let mut meta = FileMeta::new();
+    meta.add_version(file_info).expect("test metadata version should be accepted");
+    meta.marshal_msg().expect("test metadata should marshal")
+}
+
 async fn write_test_object_metadata(root: &std::path::Path, bucket: &str, object: &str) {
     write_test_object_metadata_bytes(root, bucket, object, &metadata_for_object(bucket, object)).await;
 }
@@ -1726,12 +1751,21 @@ async fn test_scan_folder_exits_when_abandoned_child_listing_finishes() {
     let _guard = TestGuard::new(60, 100, &mut scanner, temp_dir.clone());
     let heal_starts = Arc::new(AtomicUsize::new(0));
     let heal_starts_clone = heal_starts.clone();
+    let healed_versions = Arc::new(Mutex::new(Vec::<Option<String>>::new()));
+    let healed_versions_clone = healed_versions.clone();
     let mut heal_rx =
         rustfs_common::heal_channel::init_heal_channel().expect("heal channel should initialize once for scanner tests");
     let _heal_responder = tokio::spawn(async move {
         while let Some(command) = heal_rx.recv().await {
-            if let rustfs_common::heal_channel::HealChannelCommand::Start { response_tx, .. } = command {
+            if let rustfs_common::heal_channel::HealChannelCommand::Start {
+                request, response_tx, ..
+            } = command
+            {
                 heal_starts_clone.fetch_add(1, Ordering::Relaxed);
+                healed_versions_clone
+                    .lock()
+                    .expect("heal version capture lock should not be poisoned")
+                    .push(request.object_version_id);
                 let _ = response_tx.send(Ok(HealAdmissionResult::Accepted));
             }
         }
@@ -1739,13 +1773,18 @@ async fn test_scan_folder_exits_when_abandoned_child_listing_finishes() {
 
     let bucket = "src-archive";
     let object = "snapshots/37b3f20d941e2f5e6d99114d9bb2f3e67a8a2e5c9c4c5a1b0d6e7f8091a2b3c4";
-    let metadata = metadata_for_object(bucket, object);
-    write_test_object_metadata_bytes(&temp_dir, bucket, object, &metadata).await;
+    let orphan_version = Uuid::from_u128(0x1934);
+    let shared_version = Uuid::from_u128(0x1935);
+    let orphan_metadata = metadata_for_object_version(bucket, object, Some(orphan_version));
+    let shared_metadata = metadata_for_object_version(bucket, object, Some(shared_version));
+    write_test_object_metadata_bytes(&temp_dir, bucket, object, &orphan_metadata).await;
+    let mut expected_metadata = vec![(temp_dir.join(bucket).join(object).join("xl.meta"), orphan_metadata.clone())];
 
     let mut disks = vec![scanner.local_disk.clone()];
     for disk_name in ["disk2", "disk3", "disk4"] {
         let disk_root = temp_dir.join(disk_name);
-        write_test_object_metadata_bytes(&disk_root, bucket, object, &metadata).await;
+        write_test_object_metadata_bytes(&disk_root, bucket, object, &shared_metadata).await;
+        expected_metadata.push((disk_root.join(bucket).join(object).join("xl.meta"), shared_metadata.clone()));
         let endpoint = Endpoint::try_from(disk_root.to_string_lossy().as_ref()).expect("failed to create extra disk endpoint");
         let disk = new_disk(
             &endpoint,
@@ -1794,8 +1833,29 @@ async fn test_scan_folder_exits_when_abandoned_child_listing_finishes() {
         .new_cache
         .checked_flatten(bucket)
         .expect("healed cache must contain canonical child links");
-    assert_eq!(root.objects, 1);
+    // The fixture intentionally exposes two divergent version histories, so
+    // the scanner keeps both logical versions visible while discovering heals.
+    assert_eq!(root.objects, 2);
     assert!(heal_starts.load(Ordering::Relaxed) > 0, "test must execute the heal child-link path");
+    let orphan_version_text = orphan_version.to_string();
+    assert!(
+        healed_versions
+            .lock()
+            .expect("heal version capture lock should not be poisoned")
+            .iter()
+            .any(|version| version.as_deref() == Some(orphan_version_text.as_str())),
+        "sub-quorum orphan version must be submitted as an exact heal candidate"
+    );
+    for (path, expected) in expected_metadata {
+        assert_eq!(
+            tokio::fs::read(&path)
+                .await
+                .expect("scanner discovery must not delete metadata"),
+            expected,
+            "scanner discovery must not modify candidate metadata: {}",
+            path.display()
+        );
+    }
 }
 
 #[tokio::test]

rustfs/src/storage/rpc/node_service/disk.rs

@@ -146,6 +146,29 @@ fn encode_file_info_msgpack(value: &FileInfo) -> std::result::Result<Vec<u8>, Di
     encode_msgpack_with_capacity(value, "FileInfo", FILE_INFO_MSGPACK_ENCODE_CAPACITY_HINT)
 }
 
+fn encode_delete_versions_errors(disk_errors: Vec<Option<DiskError>>) -> (Vec<String>, Vec<Error>) {
+    let mut errors = Vec::with_capacity(disk_errors.len());
+    let mut item_errors = Vec::with_capacity(disk_errors.len());
+    for error in disk_errors {
+        match error {
+            Some(error) => {
+                let code = match &error {
+                    DiskError::Io(source) if source.kind() == std::io::ErrorKind::NotFound => DiskError::FileNotFound.to_u32(),
+                    _ => error.to_u32(),
+                };
+                let error_info = error.to_string();
+                errors.push(error_info.clone());
+                item_errors.push(Error { code, error_info });
+            }
+            None => {
+                errors.push(String::new());
+                item_errors.push(Error::default());
+            }
+        }
+    }
+    (errors, item_errors)
+}
+
 fn encode_msgpack_named<T: serde::Serialize>(value: &T, value_name: &str) -> std::result::Result<Vec<u8>, DiskError> {
     let mut serializer = rmp_serde::Serializer::new(Vec::with_capacity(MSGPACK_ENCODE_CAPACITY_HINT)).with_struct_map();
     value
@@ -552,6 +575,7 @@ impl NodeService {
                             success: false,
                             errors: Vec::new(),
                             error: Some(DiskError::other(format!("decode FileInfoVersions failed: {err}")).into()),
+                            item_errors: Vec::new(),
                         }));
                     }
                 };
@@ -563,30 +587,26 @@ impl NodeService {
                         success: false,
                         errors: Vec::new(),
                         error: Some(DiskError::other(format!("decode DeleteOptions failed: {err}")).into()),
+                        item_errors: Vec::new(),
                     }));
                 }
             };
 
-            let errors = disk
-                .delete_versions(&request.volume, versions, opts)
-                .await
-                .into_iter()
-                .map(|error| match error {
-                    Some(e) => e.to_string(),
-                    None => "".to_string(),
-                })
-                .collect();
+            let (errors, item_errors) =
+                encode_delete_versions_errors(disk.delete_versions(&request.volume, versions, opts).await);
 
             Ok(Response::new(DeleteVersionsResponse {
                 success: true,
                 errors,
                 error: None,
+                item_errors,
             }))
         } else {
             Ok(Response::new(DeleteVersionsResponse {
                 success: false,
                 errors: Vec::new(),
                 error: Some(DiskError::other("cannot find disk".to_string()).into()),
+                item_errors: Vec::new(),
             }))
         }
     }
@@ -1612,8 +1632,8 @@ impl NodeService {
 mod tests {
     use super::{
         compat_response_json, decode_msgpack_or_json, decode_rename_data_request_file_info,
-        encode_batch_read_version_response_payloads, encode_file_info_msgpack, encode_msgpack, encode_msgpack_named,
-        encode_read_multiple_response_payloads, encode_rename_data_response_payloads,
+        encode_batch_read_version_response_payloads, encode_delete_versions_errors, encode_file_info_msgpack, encode_msgpack,
+        encode_msgpack_named, encode_read_multiple_response_payloads, encode_rename_data_response_payloads,
     };
     use crate::storage::rpc::node_service::make_server;
     use crate::storage::storage_api::ReadMultipleResp;
@@ -1632,6 +1652,18 @@ mod tests {
         count: u32,
     }
 
+    #[test]
+    fn delete_versions_response_dual_writes_typed_item_errors() {
+        let raw_not_found = super::DiskError::Io(std::io::Error::from(std::io::ErrorKind::NotFound));
+        let (errors, item_errors) = encode_delete_versions_errors(vec![Some(raw_not_found), None]);
+
+        assert!(errors[0].starts_with("io error "));
+        assert!(errors[1].is_empty());
+        assert_eq!(item_errors[0].code, super::DiskError::FileNotFound.to_u32());
+        assert_eq!(item_errors[0].error_info, errors[0]);
+        assert_eq!(item_errors[1].code, 0);
+    }
+
     #[tokio::test]
     #[serial]
     async fn handle_read_version_records_attribution_for_missing_disk() {

scripts/check_test_wiring.py

@@ -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
 
 

scripts/s3-tests/run.sh

@@ -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" \