BucketMetadataSys::get_config lazily fabricated a default BucketMetadata
(object-lock off) for any bucket whose .metadata.bin was ConfigNotFound and
cached it in the map that the map-only, fail-closed metadata_sys::get()
serves. The object-lock batch-delete gate (object_lock_delete_check_required,
backlog#929 / #4297) treats that map as authoritative, so a metadata miss
became a cached "no lock" answer: a versioning peek could poison the cache
and let delete_objects skip the per-object retention/legal-hold stat. The
same fabrication raced make_bucket (lost update overwriting freshly
persisted lock-enabled metadata) and let the 15-minute refresh loop replace
good cached metadata on a transient quorum dip.
Production changes:
- get_config caches only metadata actually read from disk; misses are
recorded in a bounded negative cache (30s TTL, 10k entries, invalidated by
set()) so repeated lookups for metadata-less names cost no extra
namespace-lock + erasure-set fanout (reachable pre-auth via CORS
preflight and per-key in DeleteObjects).
- concurrent_load never lets a fabricated default REPLACE an existing map
entry; startup insert-if-vacant behavior for legacy buckets is preserved.
- delete_objects and new_ns_lock resolve dist-erasure, versioning, and the
object-lock gate from the set's own instance context (backlog#1052)
instead of the ambient facade, so a second in-process instance (or, in
tests, another test's transient DistErasure window) cannot reroute
locking onto an empty dist locker list or answer with the wrong
instance's bucket state.
Test-isolation changes (the bug that surfaced all of the above: the
delete_objects lock-gating test failed deterministically when sharing a
process with the lifecycle env tests):
- The MinIO-migration test builds on an isolated InstanceContext instead of
registering soon-deleted disks in the shared bootstrap registry.
- The cached lifecycle env re-registers its disks on every use, surviving
other serial tests' reset_local_disk_test_state.
- Hermetic SetDisks helpers gain isolated-context variants pinned to plain
erasure; tier-free non-serial test modules use them, guard-based
SetupTypeGuard tests stay on the bootstrap context.
- Three deterministic pin tests (nextest-safe) cover the caching contract,
the delete gate resolution source, and the ns-lock resolution source.
Verification:
- cargo test -p rustfs-ecstore --lib -- --exact <4-test combo from the
report> (previously failing, now green)
- cargo test -p rustfs-ecstore --lib: 3169 passed / 0 failed across
repeated runs; cargo fmt --check and cargo clippy --lib --tests clean
- Adversarial validation (high-risk tier, all seven roles) run per
AGENTS.md; all findings fixed or rebutted with evidence
fix(ilm): fail closed on unsafe manual job recovery
Mark expired manual transition jobs Unknown when recovery cannot prove queued worker outcomes are durable. Enforce ETag-guarded admission release with exact delete preconditions so stale releasers cannot remove a replacement admission, while still allowing drained cancelled jobs to terminate.
Co-authored-by: heihutu <heihutu@gmail.com>
* fix(tier): gate exact remote version consumption
Reject non-empty remote tier versions before transitioned GET and remote delete backend I/O when the tier backend does not support exact version operations.
Co-Authored-By: heihutu <heihutu@gmail.com>
* test(tier): split mock remote version validation fault
Separate one-shot mock remote version validation failures from persistent unsupported-backend behavior so cleanup durability tests can still verify exact-version recovery while #1358 fail-closed gate tests keep asserting no backend I/O.
Co-Authored-By: heihutu <heihutu@gmail.com>
---------
Co-authored-by: heihutu <heihutu@gmail.com>
Advance transition transactions to UploadOutcomeUnknown before remote tier PUT so response-loss windows can be recovered through provider-authoritative probing.
Co-authored-by: heihutu <heihutu@gmail.com>
* fix(ilm): accept null version transition sources
Treat nil/null source version IDs as null-version transition sources when building transition transaction source identity, while preserving fail-closed validation for truly missing IDs on versioned sources.
Co-Authored-By: heihutu <heihutu@gmail.com>
* fix(ilm): preserve multipart source versions
Propagate bucket versioning into CompleteMultipartUpload, assign a concrete version ID for versioned multipart completions, and classify unversioned FileInfo sources as null-version transition sources.
Co-Authored-By: heihutu <heihutu@gmail.com>
* fix(ilm): normalize multipart completion versions
Treat nil staged version IDs as missing for versioned multipart completion and clear staged version IDs when completion publishes a null-version object under suspended or unversioned semantics.
Co-Authored-By: heihutu <heihutu@gmail.com>
---------
Co-authored-by: heihutu <heihutu@gmail.com>
* fix(ilm): recover uploaded transition transactions
Persist transition transaction records through production transition uploads, use transaction-scoped remote object names, and start a conservative recovery loop for uploaded candidates.
Co-Authored-By: heihutu <heihutu@gmail.com>
* fix(ilm): reconcile committed transition records
Drop LocalCommitStarted transaction records only after object metadata confirms that the local transition commit already points at the same tier object and version.
Co-Authored-By: heihutu <heihutu@gmail.com>
---------
Co-authored-by: heihutu <heihutu@gmail.com>
* fix(tiering): make rejected upload cleanup durable
* fix(tiering): close transition upload cancellation gap
* test(tiering): cover failed upload without candidate
* test(tiering): synchronize cancelled cleanup recovery
* test(tiering): stabilize cancelled cleanup recovery
Prefer cancellation when the tier delete journal recovery worker is racing an immediate tick, and build the cancelled-cleanup regression store with an already-cancelled token so production recovery cannot consume the test journal.
Co-Authored-By: heihutu <heihutu@gmail.com>
---------
Co-authored-by: houseme <housemecn@gmail.com>
Co-authored-by: heihutu <heihutu@gmail.com>
* fix(filemeta): add state-aware file info validation
* fix(filemeta): validate shard arithmetic and delete paths
* fix(ecstore): add fallible erasure construction
* fix(ecstore): resolve storage parity per pool
* fix(storage): report heterogeneous erasure layouts
* fix(admin): publish prepared storage config atomically
* fix(storage): harden per-pool parity boundaries
* fix(storage): address pre-PR validation findings
* test(ci): fix strict-topology validation fixtures
* fix(heal): preserve delete markers during repair
* refactor(filemeta): drop unused ValidatedFileInfo witness
ValidatedFileInfo wrapped an unread `_file_info` reference alongside an `Option<ValidatedErasureLayout>`, but only the layout was ever consumed. Return the layout directly from `FileInfo::validate` so the sole production consumer (`LocalDisk::check_parts`) and the two unit tests read it without the extra witness type and lifetime.
No behavior change.
* fix(filemeta): keep compressed and MinIO-migrated tiered objects readable
The new decode-path validation rejected several legitimate on-disk shapes that older RustFS and MinIO-migrated data carry, turning readable objects into FileCorrupt:
- Compressed objects written with an unknown upload size persist a negative per-part actual_size (the documented "unknown size" sentinel that ObjectInfo::get_actual_size already tolerates). validate_collection_contents rejected it via usize::try_from; now a negative actual_size skips shard validation and only real, non-negative sizes are checked.
- MinIO-migrated objects transitioned to a versioned remote tier store the tier version id as a UUID string, not 16 raw bytes. MetaObject::into_fileinfo returned FileCorrupt (main tolerated it as None), making all versions of the object unreadable; MetaDeleteMarker free-version records took a Some(nil) sentinel path with the same effect, which also breaks free-version expiry (remote-tier leak). Both now decode through a shared transitioned_version_id_from_meta_sys helper: 16 raw bytes or a UUID string are accepted, anything else is tolerated as None instead of failing the read.
Regression tests updated to assert the readable/compat behavior, with new tests covering MinIO string-form recovery.
* fix(scanner): build the delete-marker test fixture without erasure geometry
get_size_counts_delete_markers_separately_from_versions built its delete marker with `FileInfo::new(object, 1, 1)`, which attaches erasure geometry (data=1/parity=1/distribution). This PR classifies versions by shape via `is_storage_delete_marker()` (no geometry) rather than the raw `deleted` flag, so a geometry-bearing "delete marker" is correctly serialized as a purge-pending payload Object and counted as a version — CI saw summary.versions=3, expected 2.
Real delete markers carry no erasure geometry (delete paths build them as `FileInfo { deleted: true, ..Default::default() }`), so construct the fixture the same way. It then classifies as a storage delete marker and the counts (versions=2, delete_markers=1) hold. This keeps the PR's more-correct classification, which prevents a purge-pending object's geometry from being dropped when serialized as a bare delete marker.
* docs(changelog): note per-pool parity fix and storage-class startup upgrade caveat
Records the #4801 per-pool erasure parity fix under Fixed, and documents the upgrade behavior where a persisted storage class that a small or heterogeneous pool cannot satisfy now fails startup — with the RUSTFS_STORAGE_CLASS_STANDARD recovery steps. Docs-only; covers R4 from the on-disk compatibility audit.
* fix(heal): report parity from erasure geometry, not is_valid()
heal_object set HealResultItem.parity_blocks via `if lfi.is_valid()`, which was missed by the migration of the other quorum/metadata predicates. With the new `is_valid()` semantics (full payload validation; delete markers now return false), a delete marker or a geometry-bearing version with a benign collection quirk would misreport parity as the pool default instead of its own. Use `has_valid_erasure_geometry()` — the narrow "does this carry erasure geometry" predicate the rest of the migration uses — so reporting matches the object's actual layout. Reporting-only; no data-path change.
* fix(filemeta): do not silently serialize a non-canonical deleted FileInfo as an Object
`From<FileInfo> for FileMetaVersion` classifies by `is_storage_delete_marker()` (shape), which correctly routes canonical delete markers to Delete and purge-pending payloads (deleted=true with real erasure geometry) to Object. But a `deleted` FileInfo that is neither a canonical marker nor a valid erasure payload would silently serialize as a zero-geometry MetaObject that later fails `validate_for_metadata_read`. Write paths validate first (`validate_for_erasure_write` / `validate_for_metadata_read`), so this is a caller bug; `From` is infallible, so surface it with a structured `warn!` on the malformed branch instead of writing corrupt metadata silently. Legitimate purge-pending objects (valid geometry) are unaffected — the guard only fires for `deleted && !has_valid_erasure_geometry()`.
* test(filemeta): assert real historical xl.meta versions pass metadata-read validation
Empirical companion to the code-reasoned decode-tolerance invariants (docs/architecture/erasure-coding.md §11) and the rolling-upgrade / MinIO-migration compatibility concern: the tightened `validate_for_metadata_read` runs on every local disk read and peer-RPC-decoded FileInfo, so it must accept every version of real historically-written xl.meta, never reject it as FileCorrupt.
Loads five real fixtures — MinIO small-inline, MinIO versioned (two object versions + a delete marker), MinIO large multipart, a legacy V1 (xl.json-derived) object, and a legacy meta_ver 2 object — decodes every version with parts materialized, and asserts validate_for_metadata_read() is Ok for each. Reverting the tolerant handling (delete-marker shape, legacy per-part checksums, string/short transitioned-versionID, negative actual_size) turns this red.
* fix(ci): remove duplicate storage test re-exports
---------
Co-authored-by: overtrue <anzhengchao@gmail.com>
* fix(ecstore): add fallible erasure construction
(cherry picked from commit bd148b20f7)
* fix(ecstore): resolve storage parity per pool
(cherry picked from commit c05c2cb24b)
* fix(ecstore): keep carved per-pool parity core self-contained on main
Fixups so the cherry-picked fallible-erasure + per-pool-parity core builds standalone on current main without the excluded scope-creep commits:
- runtime/sources.rs: re-add backend_storage_class_parities (removed by the per-pool commit; its rebalance caller was updated in an unrelated reporting commit that was left out). Reimplemented over the snapshot API, behavior-identical.
- config/mod.rs: rename the storage-class publish test module (main independently added a mod tests, so the cherry-pick collided).
- rustfs storage_api.rs + startup_storage.rs: route the storage-class ENV consts through the startup storage facade and use a local const for the erasure-set-drive-count env name, satisfying the layer/facade guardrail (main's guardrail is stricter than when the core was authored).
* fix(ecstore): use struct-init in erasure test helper to satisfy clippy field_reassign_with_default
The cherry-picked fallible-erasure commit's `erasure_with_invalid_dimensions` test helper built `Erasure` via `default()` then reassigned fields, which trips `clippy::field_reassign_with_default` under `-D warnings` (only surfaced by `--all-targets`, which lints test code). #4977 fixed this in a later commit that was not part of the carved core. Use struct-init with `..Default::default()`, matching #4977's final form.
* fix(rustfs): gate the test-only storage-class ENV facade re-export behind cfg(test)
The ENV constants (INLINE_BLOCK_ENV/OPTIMIZE_ENV/RRS_ENV/STANDARD_ENV) re-exported through the startup storage facade are only consumed by a #[cfg(test)] test in startup_storage.rs, so in a non-test lib build the re-export is unused and trips -D unused-imports under clippy --all-targets. Gate it with #[cfg(test)], matching #4977's final form.
---------
Co-authored-by: cxymds <cxymds@gmail.com>
A duplicate transition task admitted after the winner released its
in-flight claim (#4839) re-reads the version before uploading, but on
unversioned buckets that read could hit a stale pre-transition entry in
the 2s-TTL GET metadata cache: transition_object never invalidated the
cache after delete_object_version persisted transition_status=complete
and freed the local data. The stale hit defeated the TRANSITION_COMPLETE
early-return, so the duplicate streamed the already-deleted local data
to the remote tier (NotFound reader errors + rejected duplicate tier
PUT with UnexpectedContent).
Invalidate the cache right after the transitioned metadata is persisted,
matching the other metadata-mutating paths, and add a regression test
that runs a duplicate transition against an already-transitioned version
and asserts no second tier upload and unchanged remote object metadata.
Fixes#4827
DeleteRestoredAction is supposed to demote a restored object back to its
pure transitioned state: remove only the local restored copy, strip the
x-amz-restore headers, and leave the version (and the remote tier data)
untouched. expire_transitioned_object set
opts.transition.expire_restored accordingly, but no delete path ever
read the flag, so delete_object ran an ordinary delete: on unversioned
buckets the whole object vanished and the free-version record scheduled
remote tier cleanup (tier data loss); on versioned buckets the latest
version got a spurious delete marker that replication propagated.
Route expire_restored explicitly in SetDisks::delete_object before
delete-marker resolution and replication dispatch: target the found
version with FileInfo.expire_restored=true and return early. The
FileMeta::delete_version layer already implements the semantics (strip
restore headers, keep the version, hand back the local data dir); this
wires it up.
Also fix the action matching in expire_transitioned_object (extracted
into transitioned_object_delete_opts): DeleteRestoredVersionAction
previously fell through to the full transitioned-object delete, which
removed the remote tier data of a noncurrent restored version. It now
routes through the same restored-copy cleanup with the exact version id,
matching MinIO's Action.DeleteVersioned()/DeleteRestored() dispatch.
Re-enable test_restore_chain_local_read_expiry_keeps_remote_and_allows_
re_restore in the ILM Integration (serial) lane; add unit tests pinning
the event->options routing and the filemeta expire_restored branch.
Closesrustfs/backlog#1302
perf(capacity): remove per-PUT global lock and per-disk allocation from write dirty-scope
Every successful write recorded its capacity dirty scope by allocating an endpoint/path String per online disk, deduplicating through a HashSet, entering the global dirty-scope Mutex, and — in the app response path — taking a global async RwLock to record the write frequency. Under small-object high concurrency this created a global serialization point and O(disks) allocation on the hot path (https://github.com/rustfs/backlog/issues/1315).
This change makes the steady-state write path allocation-free and lock-free without altering capacity accounting semantics:
- Memoize the per-set dirty scope. Each set resolves its disks' immutable endpoint/path identity lazily into a slot-indexed cache and reuses a shared `Arc<CapacityScope>`; steady-state writes clone the Arc under a read lock instead of rebuilding String/HashSet. The heal path keeps an ad-hoc scope builder because it passes disks in erasure-distribution order rather than physical-slot order.
- Add a monotonic generation to the global dirty-scope registry, advanced only when a non-empty drain removes disks. A set upgrades the global registry mutex only on the first write of each generation and then skips it while the generation is unchanged; the observed generation is read under the registry lock so a concurrent drain forces a re-mark, preventing lost updates. The write commits its bytes before recording the scope, so any drain that could remove the mark is ordered after the commit and the following refresh reads the committed bytes.
- Replace the write-frequency `RwLock<WriteRecord>` with lock-free atomics: per-second CAS buckets, an atomic last-write timestamp, and an atomic total counter. The frequency window and debounce semantics the refresh scheduler relies on are unchanged.
Capacity marking remains a conservative superset of the disks actually written, so admin/scan totals are byte-for-byte identical: extra dirty marks only trigger a re-read of a disk whose usage is unchanged. White-box tests assert the memoized scope equals the previous ad-hoc construction, that the global registry is upgraded exactly once per generation and re-marked after a drain, and that the lock-free write record is exact under concurrent contention.
Ref: https://github.com/rustfs/backlog/issues/1315
* test(ilm): fix restore test object key to match transition filter
restore_object_usecase_reports_ongoing_conflict_and_completion used the
object key "restore/api-object.bin", but the shared
set_bucket_lifecycle_transition_with_tier helper only transitions objects
under the "test/" prefix. enqueue_transition_for_existing_objects therefore
matched nothing and wait_for_transition timed out at 15s, failing the test
deterministically.
The test was added in #4860 but its ILM Integration (serial) lane is
skipped on regular PRs, so it merged red and has failed on every main run
since. Move the object under the test/ prefix like every passing sibling
test in this file.
* ci(ilm): exclude broken RestoreObject API test from serial lane
restore_object_usecase_reports_ongoing_conflict_and_completion exposes a
real regression, not a test bug: the RestoreObject copy-back
(handle_restore_transitioned_object) now holds the object write lock added
in #4877 across the entire tier read-back, so it never releases in time and
the test's concurrent get_object_info times out with Lock(Timeout, 5s). The
failure is deterministic and independent of the mock tier's injected latency.
This is the same class of known-broken restore/transition failure already
tracked under backlog#1148 (three sibling scanner tests are excluded here by
name for the same reason), so exclude this one the same way until the restore
copy-back path is fixed or the #4877 lock scope is revisited. The prior commit
keeps its correct fix (the object key must live under the test/ transition
prefix); that was masking this deeper issue by never letting the object
transition in the first place.
Restore copy-back deadlock/hang under the #4877 lock is escalated separately
for a product-level decision (fix the copy-back vs. narrow/revert #4877).
* test(ilm): fix scanner restore test object keys to match transition filter
test_restore_chain_local_read_expiry_keeps_remote_and_allows_re_restore and
test_multipart_restore_preserves_parts_and_etag (both added in #4860) keyed
their objects under restore/ instead of the test/ prefix that
set_bucket_lifecycle_transition_with_tier filters on, so the objects never
transitioned and wait_for_transition timed out at 15s.
These surfaced only after the prior commit excluded the rustfs-side restore
API test: nextest runs -j1 fail-fast, so that earlier failure stopped the run
before these scanner tests executed. Unlike the excluded API test, both call
restore_transitioned_object().await sequentially and only read afterwards, so
they don't hit the concurrent-read-vs-#4877-write-lock timeout; the key
prefix was their only problem.
* ci(ilm): exclude the two remaining #4877-broken restore tests
test_multipart_restore_preserves_parts_and_etag and
test_restore_chain_local_read_expiry_keeps_remote_and_allows_re_restore both
call restore_transitioned_object().await, which since #4877 acquires the
object write lock and deterministically times out (Lock Timeout, 5s) against
an already-held lock, so restore never completes. They surfaced one at a time
because nextest runs -j1 fail-fast. The earlier prefix fix was necessary but
only advanced them from the transition wait to this restore-lock timeout.
Exclude both by name alongside their already-excluded sibling
test_transition_and_restore_flows (same root cause, tracked under
backlog#1148) so the ILM Integration (serial) lane goes green. The #4877 lock
scope still needs a product fix before any of these re-enable.
* docs(ilm): describe the excluded restore tests' symptom as a lock timeout, not a deadlock
The #4877 write lock is held across the tier read-back and outlives the 5s
lock timeout; nothing proves a true deadlock. Wording flagged by Copilot
review.
* fix(ecstore): stop restore copy-back self-deadlocking on the #4877 write lock
#4877 made handle_restore_transitioned_object hold the object write lock for
the whole restore and forward no_lock=true so the set layer would not
reacquire it. But the set-level copy-back rebuilds its own options
(put_restore_opts -> ropts, and the complete_multipart_upload opts) that
default no_lock=false, so the inner put_object / new_multipart_upload /
complete_multipart_upload each re-acquire this object's write lock in their
commit phase and block on the lock the restore already holds -> Lock(Timeout,
5s), and restore never completes.
Confirmed via RUSTFS_OBJECT_LOCK_DIAG_ENABLE: restore_transitioned_object
acquires the write lock, then holds it ~10.5s across two nested 5s acquire
timeouts before failing. This is a real product deadlock: a RestoreObject on
any transitioned object (multipart especially) hangs, not just the tests.
Propagate no_lock into the copy-back options so the inner writes inherit the
already-held lock. Use opts.no_lock (not a hardcoded true) so a caller that
restores without the outer lock still locks correctly. put_object_part is left
as-is: it locks the multipart upload-id resource, not the object key, so it
does not conflict. Verified test_multipart_restore_preserves_parts_and_etag
now passes (3.6s, was a 15s+ hang).
* ci(ilm): re-enable multipart restore test; scope remaining exclusions
The prior commit fixes the #4877 restore self-deadlock, so
test_multipart_restore_preserves_parts_and_etag passes again - drop it from
the serial-lane exclusion list and remove its 'currently excluded' note.
The other restore/transition tests still fail, but each on a DIFFERENT,
independent issue unrelated to the (now-fixed) lock, verified locally:
- test_restore_chain_...: DeleteRestoredAction sets expire_restored but no
delete path reads it, so cleanup deletes the whole object (unimplemented
semantics), not the local restored copy only.
- test_transition_and_restore_flows: transition xl.meta missing on one drive
(EC metadata distribution), not restore.
- restore_object_usecase_reports_ongoing_conflict_and_completion: asserts a
concurrent mid-restore ongoing=true read that #4877's read-vs-restore
serialization rules out (backlog#1148 ilm-8 criterion 1, an API-semantics
decision).
Comments and #[ignore] reasons updated to reflect each real cause. All remain
tracked under backlog#1148.
* perf(metrics): drop needless per-emission work on hot metric paths
Audit of the metrics hot paths surfaced four low-risk wins where emission did
work it did not need to:
- `record_file_cache_reclaim_success/error` (disk/local.rs) called `.to_string()`
on `kind` (already `&'static str`) and on the `"ok"`/`"err"` literals, heap-
allocating up to four `String`s per page-cache reclaim window — which runs per
read-stream reclaim. The `metrics` macros accept `&'static str` label values
directly, so pass them as-is.
- `record_read_repair_dedup` (set_disk/core/io_primitives.rs) likewise
`.to_string()`-ed an already-`&'static str` `reason`.
- `SetDisks::get_object_reader` (set_disk/ops/object.rs) captured `Instant::now()`
and emitted the `rustfs.lock.acquire.*` counter and histogram unconditionally
on every GET, right beside an already-gated stage timer. Gate them behind
`get_stage_metrics_enabled()` too, so an inactive observability config pays no
per-GET clock read or recorder lookups.
- The per-response-body-chunk counter in server/http.rs re-ran the `counter!`
registry lookup on every chunk (a streamed GET emits many). Resolve the
label-less handle once into a `LazyLock<metrics::Counter>`; the global recorder
is installed at startup before any response streams, so the cached handle binds
to the final recorder.
No metric names or label values change. The only behavior change is that the
`rustfs.lock.acquire.*` GET-path metrics now follow the GET stage-metrics flag,
consistent with the neighbouring stage timings.
Co-Authored-By: heihutu <heihutu@gmail.com>
* perf(metrics): gate page-cache reclaim metrics behind metrics_enabled()
`record_file_cache_reclaim_success/error` run per read-stream reclaim window on
large-object reads and emitted unconditionally. When general metrics are
disabled the `counter!`/`histogram!` macros still construct three metric keys
per call for nothing. Skip the emission behind `rustfs_io_metrics::metrics_enabled()`,
matching how the io-metrics free functions self-gate. The serial reclaim-metrics
test now enables the flag (save/restore) alongside the existing stage gate.
Left ungated deliberately: `record_read_repair_dedup` (rare read-repair path,
and its non-serial test would need a global-flag toggle), and the HTTP body-chunk
counter (its cached handle already makes the disabled case a no-op increment).
Co-Authored-By: heihutu <heihutu@gmail.com>
---------
Co-authored-by: heihutu <heihutu@gmail.com>
`full_object_plaintext_len` decides whether a body-cache hook hit may serve
bytes in place of the erasure read. It is a fail-closed allow-list: it excludes
every read whose `ReadPlan::build` applies some other transform (ranged/part,
raw/data-movement, restore, encrypted, remote) with an early `return None`,
then returns a `Some(..)` length only for the whole-plaintext cases. A newly
added `ReadPlan` branch that nobody teaches this gate about falls through to
`None` and safely bypasses the cache. Flip it to a deny-list and the same new
branch silently serves bytes in the wrong representation — the backlog#1108 /
#1109 / #1146 class of bug.
The existing unit and e2e tests only cover the branches that exist today. This
adds `scripts/check_body_cache_whitelist.sh`, a structural guard wired into
pre-commit / pre-pr / dev-check and CI, that asserts every exclusion predicate
and a `return None` still precede the first `Some(..)`. Reordering a predicate,
dropping one, moving the positive return ahead of the gate, or renaming/removing
the function all fail; wording, formatting, and adding a new exclusion in the
same gate do not. Mutation-tested against all four regression shapes.
This machine-enforces the structural invariant that backlog#1146 was kept open
to guard by hand.
Co-authored-by: heihutu <heihutu@gmail.com>
fix(object-data-cache): make GET body cache key write-unique and dedup lookups
Address four object-data-cache GET-path findings (backlog#1107 batch):
ODC-06 (backlog#1111): the cache key was content-unique, not write-unique.
Extend ObjectDataCacheKey with the resolved version's modification time
(i128 unix nanoseconds, None -> 0), derived once in the shared planner so the
ecstore hook and the usecase layer produce an identical key. An unversioned
overwrite advances mod_time, so a stale node can no longer serve old bytes for
up to the TTL under an MD5 collision; etag + size stay as belt-and-braces.
ODC-16 (backlog#1121): every cacheable GET planned and looked up twice (once in
the ecstore hook, once in the usecase layer), double-counting hits, hit_bytes
and lookups. GetObjectReader now carries a GetObjectBodySource marker
(Unprobed / HookMissed / HookServed); the hook stamps it, and
build_get_object_body_with_cache serves a hook-served body directly and skips
its lookup whenever the hook already probed. One hook-served GET now records
exactly one lookup.
ODC-19 (backlog#1124): ENABLE=true with no explicit mode defaulted to HitOnly,
which never fills and keeps a permanent 0% hit rate. Default to
FillBufferedOnly, log the resolved mode at startup, and warn when HitOnly is
selected explicitly.
ODC-24 (backlog#1129): max_entry_bytes above the in-memory GET fill limits was
silently inert. Clamp the planner's size eligibility to
min(max_entry_bytes, seek-support threshold, 64 MiB buffer cap) so ineligible
sizes plan SkipTooLarge instead of being reported eligible, and warn at startup
when the excess is inert.
Co-authored-by: heihutu <heihutu@gmail.com>
fix(ecstore): roll back delete on disks that staged then errored (backlog#1158)
#4300 rolls back a failed delete only on disks that returned Ok, skipping any
disk that staged its rollback backup, applied the delete, and then errored --
leaving that disk deleted while its peers are restored. On rollback, fan the
undo out to every online disk instead; the disk-side restore_delete_rollback
is already idempotent (Ok no-op when nothing was staged), so unstaged disks
are unaffected. The err.is_some() skip now applies only to the success/cleanup
path. Covers both single-object and batch delete.
Refs backlog#1158.
fix(ecstore): make body-cache hook re-registrable + add e2e regressions
ODC-21 (backlog#1126): the GET body-cache hook lived in a first-wins
OnceLock. When AppContext is rebuilt (config reload, test re-init) a fresh
ObjectDataCacheAdapter is constructed and re-registered, but the OnceLock
kept ecstore's GET probe pointed at adapter #1 while every usecase-layer
fill and invalidation targeted adapter #2 — silently degrading the feature
to a 0% hit rate with no error, log, or metric, and stranding entries in the
unreachable cache until their TTL.
Replace the slot with RwLock<Option<Arc<dyn GetObjectBodyCacheHook>>> so
re-registration atomically swaps to the newest adapter, and log at WARN when
a swap replaces a *different* instance (Arc::ptr_eq). RwLock over
ArcSwapOption because arc-swap's RefCnt is impl<T> (Sized, thin *mut T) and
cannot hold an Arc<dyn Trait> without a sized newtype wrapper; the probe
reads the slot once per full-object GET but only clones an Arc, negligible
next to the metadata quorum fan-out already done before the probe. Add a
test-only clear_get_object_body_cache_hook so tests register/unregister
deterministically.
With the hook now re-registrable, add true end-to-end regressions that drive
get_object_reader (not the full_object_plaintext_len predicate) against a
real erasure-coded, genuinely-compressed object via the blackbox
make_local_set_disks harness, with a stand-in hook playing the app-layer
cache (the injection point production uses; the adapter itself lives above
ecstore). These close the gap the predicate-only tests left — a caller that
opens a new shortcut serving the cached body directly, the original form of
both P0s:
- backlog#1108: a raw_data_movement_read must yield the STORED (compressed)
bytes, never the cached plaintext.
- backlog#1109: a compressed cache hit must publish the DECOMPRESSED length
as object_info.size (the UploadPartCopy invariant), with the streamed
length matching.
- backlog#1146: a restore read (restore_request.days) must serve STORED
bytes, not the cache.
Mutation-verified each e2e test bites: dropping the raw_data_movement_read
gate serves plaintext (fails #1108); removing the hit-site size republication
publishes 2972 vs 660000 (fails #1109); dropping the restore gate serves
plaintext (fails #1146).
Co-authored-by: heihutu <heihutu@gmail.com>
* fix(cache): gate GET body-cache hook to preserve ReadPlan output
The ecstore GET body-cache hook serves cached full-object plaintext
directly, bypassing ReadPlan/ReadTransform. That is only sound when the
normal read path returns that same plaintext byte-for-byte. Two probe
conditions were missing, plus two usecase-layer planner gaps.
ODC-01 (backlog#1108): raw/data-movement reads. ReadPlan::build returns
the STORED representation for raw_data_movement_read (e.g. compressed
bytes, length = oi.size), but the cache holds the post-decompression
body. Decommission (raw_data_movement_read: true) would receive
decompressed plaintext where raw compressed bytes are required, silently
corrupting the destination pool.
ODC-02 (backlog#1109): compressed objects. ReadTransform::Compressed
rewrites object_info.size to the decompressed length; on a hook hit
object_info is returned unchanged, so object_info.size is the compressed
size while the stream carries the decompressed body. UploadPartCopy then
uses src_info.size as the copy length and truncates the part.
Fix: gate the hook probe with should_probe_body_cache_hook, refusing
raw_data_movement_read, data_movement, and compressed objects, mirroring
the conditions get_small_object_direct_memory_decision already applies.
ODC-33 (backlog#1138): build_get_object_body_cache_plan lacked the
is_remote() exclusion the ecstore hook enforces; add it so transitioned
(remote-tier) objects are excluded uniformly.
ODC-C1 (backlog#1142): zero-length bodies save no I/O (ecstore returns an
empty body before the hook probe) yet the planner admitted them; change
the guard to response_content_length <= 0 so they plan Skip, mirroring
should_buffer_get_object_in_memory_with_threshold.
Tests: body_cache_hook_gate_tests (4) cover plain-probe plus
raw/data-movement/compressed skips; planner gains
plan_skips_remote_transitioned_objects and plan_skips_zero_length_objects.
Co-Authored-By: heihutu <heihutu@gmail.com>
* fix(cache): allow compressed bodies via a fail-closed read allow-list
The body-cache hook probe was gated by a deny-list that refused compressed
objects outright, which cost the cache every compressed body — a growing
share of stored data. Replace it with an allow-list that returns the exact
plaintext length a hit may serve, or None.
full_object_plaintext_len() answers a single question: would the normal
ReadPlan produce this object's complete plaintext, and under which size?
Compressed objects now qualify, and the hit site publishes the returned
length as object_info.size, reproducing the contract ReadTransform::
Compressed establishes. A hit whose body length disagrees is refused and
falls through to the erasure read.
This also closes a gate the deny-list only covered by accident: a restore
read forces ReadPlan down the Plain branch, so a compressed object yields
STORED bytes under its compressed size. Refusing compressed objects hid
that; admitting them exposes it, so restore reads are refused explicitly.
Being fail-closed, a newly added ReadPlan branch bypasses the cache by
default rather than silently serving the wrong representation — the
structural defect behind both backlog#1108 and backlog#1109.
Refs: backlog#1108, backlog#1109
Co-Authored-By: heihutu <heihutu@gmail.com>
---------
Co-authored-by: heihutu <heihutu@gmail.com>
Two correctness defects on the opt-in codec-streaming GET path.
ECA-02 (#943): ErasureDecodeReader only decremented `remaining` for the
main fill buffer. Under the default DualInFlight policy each fill also
produces a queued stripe that is delivered to the client via
`prefetched_bufs.pop_front()` without touching `remaining`, so any object
larger than one erasure block finished with `remaining > 0` and the GET
terminated with LessData despite delivering all bytes. The inflated
`remaining` was also fed back into the fill worker, which used it to trim
the final stripe and to decide whether to read past EOF. Account for the
queued-stripe bytes when they enter the prefetch queue; queued buffers
come only from `Ok(true)` decodes so they are non-empty and bounded by
`remaining - main_buf.len()`, ruling out underflow.
ECA-04 (#945): the codec-streaming gate did not inspect `opts.part_number`.
A partNumber GET carries `range == None`, so it was not classified as a
Range request and reached the full-object codec-streaming reader, which
drops the storage offset/length returned by GetObjectReader::new. A
partNumber >= 2 request would then stream the whole object. Mirror the
direct-memory part_number fallback and route any partNumber request back
to the legacy duplex path, which applies the offset/length correctly.
Regression tests: DualInFlight read_to_end on a multi-block object and on
a non-block-aligned object; SingleInFlight vs DualInFlight byte-identical
output; gate fallback on partNumber requests.
Co-authored-by: heihutu <heihutu@gmail.com>
fix(lock): fence write commit on lock loss (backlog#899 Phase 2)
Phase 0+1 (#4388) made object write locks refreshable and marks the guard
lost when the heartbeat can no longer refresh a quorum, but does not act on
it. Under a partition a long write's lock can expire on an unreachable node
and be reclaimed, letting a third party re-acquire it; the original writer
keeps going and both commit -- a double write.
Expose the loss signal through NamespaceLockGuard::is_lock_lost() and
ObjectLockDiagGuard::is_lock_lost(), and fence the commit in put_object and
complete_multipart_upload: immediately before rename_data (the atomic commit
point), abort with a retryable NamespaceLockQuorumUnavailable (503) if the
lock was lost. In multipart the check precedes cleanup_multipart_path so a
lost lock leaves the upload intact and retryable. A write that already
reached rename_data Ok is durable and never aborted.
The loss criterion is unchanged (reacts to Phase 1's signal). Heal and the
long-GET read side are deferred follow-ups.
* fix(ecstore): make post-commit old data dir cleanup best-effort (backlog#898)
A write is authoritatively committed once rename_data returns Ok (the new
version is durable on >= write_quorum disks and immediately readable). The
subsequent reclamation of the now-dereferenced old object/<data_dir> is pure
space reclamation, yet commit_rename_data_dir propagated a below-quorum GC
failure via `?` into ErasureWriteQuorum -> 503, producing a false-negative
ACK for an already-persisted write. This is a deliberate divergence from
MinIO (erasure-object.go:1577), which couples the two; the divergence is
justified by durability semantics, not parity.
Changes:
- commit_rename_data_dir now returns a structured OldDataDirCleanup receipt
and never returns Err. Adds an old==committed-dir anti-misdelete guard and
a committed_data_dir parameter. Classification is extracted into pure
functions (classify_old_data_dir_cleanup / map_cleanup_join_result /
is_cleanup_not_found) so it is unit-testable. Task panic/cancel is mapped to
a non-ignored DiskError::other (never DiskNotFound), and not-found is
normalized to reclaimed.
- object.rs / multipart.rs consume the receipt instead of `?`. The result
reverts to Ok, so the invalidate_get_object_metadata_cache self-heal and the
capacity/compression accounting that a `?` early-return previously skipped
now run on the cleanup-failure path too.
- On residue, report_old_data_dir_cleanup emits leak metrics and enqueues an
object heal over the existing heal channel (disk-health signal replacing the
503). heal_object -> reclaim_orphan_data_dirs already reclaims unreferenced
local data dirs, closing the loop end to end.
- Adds rustfs_old_data_dir_* counters (attempted/reclaimed/leaked/below_quorum)
as the operator-visible backstop for leaked residue.
- Adds a test-only (#[cfg(test)]) delete fault-injection seam; in production it
inlines to a no-op None and has no behavioral effect.
Tests: pure-function A/C group + join-error mapping + actions decision; A5/A5b
real-disk guard/reclaim integration; end-to-end overwrite returning 200 while
old-data-dir cleanup fails. #864 rollback guard test remains green.
* fix(ecstore): resolve merge conflicts with origin/main in io_primitives.rs
---------
Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: houseme <housemecn@gmail.com>
cxymds