`NewerNoncurrentVersions` had no lower bound at PUT, and evaluation read a
negative count through `usize::try_from(...).unwrap_or(usize::MAX)`. An
HTTP-accepted rule therefore retained (almost) everything and silently
stopped expiring versions — the one outcome a retention rule must never
produce by accident.
Reject a negative count during validation, and stop reading one as
"retain everything" anywhere it can still arrive from older persistence
or an import: evaluation takes no action for such a rule and says so in a
diagnostic, the batch limit path yields no event, and `Evaluator::eval`
reports a typed corruption error to callers that can surface one.
A count-only noncurrent expiration is a MinIO extension, not an AWS form.
It used to be rejected as an actionless rule and was never executed. It
is now accepted and honoured with the semantics MinIO gives it: the
newest N noncurrent versions are kept and every older one is due as soon
as it became noncurrent. Zero keeps the meaning the batch limit path has
always given it — no count constraint — so a zero-count rule with no age
condition still has no action.
`LifecycleRuleFilter` is an all-`Option` DTO, so the schema constraints
were not checked anywhere: validate at most one top-level predicate, an
`And` that combines at least two, no repeated tag key, tag key/value
limits, non-negative sizes, and `ObjectSizeGreaterThan <
ObjectSizeLessThan`. An empty filter stays valid — AWS documents it as
"every object in the bucket".
Schema-shape violations are reported with a distinct `ErrorKind` so the
S3 boundary answers them with `MalformedXML`; rejected values keep the
`InvalidArgument` this path has always returned.
backlog#2201
`GET /v3/tier-stats` answered from whichever process received the
request, returning that node's rolling 24-hour transition counters as
if they were cluster totals, and the `TierRequestsSuccess` and
`TierRequestsFailure` metric names had no producer at all.
The body now separates the two quantities a tier carries. Stored
inventory comes from the persisted scanner usage snapshot, which is
already cluster-wide; rolling activity is summed over every member
through a new read-only `TierDailyStats` peer RPC. Rings are merged
rather than added, so an idle node's expired hours age out, and each
node counts only its own committed transitions, so a retry is counted
once. Coverage travels with the numbers: `activity.status` names the
reporting members and the ones that could not be asked, timed out, or
answered with a ring this build refuses to merge, and per-tier
inventory is absent rather than zero when the snapshot has no
accounting. The version 1 body stays reachable at `?format=legacy`.
Tier request counters are recorded at the two seams every remote
request passes through, so a new provider is counted by construction,
with a closed operation/outcome label set that can never grow a tier
name, endpoint or object key.
Closesrustfs/backlog#2207
Co-authored-by: cxymds <cxymds@gmail.com>
* feat(odm): merge the source listing into ListObjectsV2
Adds policy.list_through: ListObjectsV2 merges the local and source
listings into one ordered page so clients see the whole namespace during
an on-demand migration. Local entries win a key both sides hold,
CommonPrefixes are unioned under a delimiter, and the continuation token
is an opaque versioned envelope carrying both cursors.
A source listing failure or an open breaker follows policy.source_error:
propagate answers 424, not_found answers from local state and marks the
response x-rustfs-on-demand-migration-list: local_only. Source listings
are capped at 10 per second per bucket.
* test(odm): refresh the e2e-full darwin selection digest
The list-through e2e module adds seven cases to the merge lane.
* fix(odm): declare the remote client retry policy per consumer
The SDK retry policy was an inherited default: one logical call could cost
three wire requests, so the migration breaker counted logical calls on top
of a threefold amplification against a source that was already failing.
Make it an explicit RemoteS3EndpointSpec field. Replication targets declare
today's standard three attempts and keep their behaviour; the on-demand
migration source and its admin probe declare a disabled policy, so one
counted failure is exactly one source request and pull.rs owns the only
retry budget.
* fix(odm): count a stalled inline source as a source timeout
The inline tee wraps its source body in the idle guard, but the tee turns a
stalled source into an ordinary body read error, so the write-back reported
it as a local write failure. Hand commit_inline the guard so the pull is
counted under source_timeout instead.
The background pump now enforces the idle budget through the same guard
rather than a second copy of the timeout loop.
* test(odm): cover a stalled source body end to end
The fake target can now deliver a GetObject body in slices with a pause
between them, so the inline abort can be driven by a stalled source instead
of a truncated one. Two fault cases drop the workarounds they carried for
the SDK's retries: the scripted fault count and the observed source request
count now have to agree.
The operations guide records the retry and idle-timeout guarantees.
* fix(replication): send an integrity header on Object Lock replication PUTs
AWS S3, MinIO and most compatible targets reject a PutObject that carries
x-amz-object-lock-* headers unless it also carries Content-MD5 or an
x-amz-checksum-* header. Since rustfs#6895 the replication client sends
plain signed payloads with no SDK checksum, so every replicated object
with a retention period or legal hold failed against such targets.
TargetClient::put_object now decides per request through the pure
rustfs_replication::object_lock_put_integrity: a plaintext single-part
object whose source ETag is its MD5 gets Content-MD5 derived from the
ETag (no body pass, framing unchanged); a multipart-layout ETag, managed
SSE or SSE-C passthrough falls back to an SDK CRC32; a forwarded source
checksum or an unlocked PUT is left alone.
The outbound target matrix flips its two KnownFailing(rustfs#7082) cells
to Completed and every Completed cell now asserts that a locked
PutObject carried an integrity header.
Fixes rustfs#7082.
* test(e2e): keep the matrix expectation table clippy-clean under -D warnings
The CI lint runs cargo clippy --all-targets -- -D warnings. With every cell
green the single-arm match tripped match_single_binding and the unused
KnownFailing variant tripped dead_code, and the target-client tests tripped
field_reassign_with_default. Drive the expectation table from a
KNOWN_FAILING_CELLS constant (so the variant stays live and adding a red
cell is a one-line entry), build the test options as struct literals, and
refresh the e2e-repl-nightly selection digest for the renamed table test.
* feat(odm): enable on-demand migration by default
The module switch RUSTFS_ON_DEMAND_MIGRATION_ENABLED now defaults to true,
so the feature is reachable without an opt-in; setting it to false still
keeps the module out of the read path entirely. A bucket without an
on-demand-migration.json is never resolved by the runtime and makes no
source call, so the flip changes nothing for unconfigured buckets.
The admin plane now reads the switch through the predicate published by
module_switches.rs instead of its own duplicated env constant; the
behaviour (an environment read per call) is unchanged.
* test(e2e): wire three on-demand migration cases into e2e-smoke
The PR smoke lane gains one case per user-visible contract: a GET miss
that pulls and persists, a HEAD miss that answers from the source and
stores nothing, and the admin config/status pair that must redact the
source secret. The HEAD case did not exist outside the nightly
real-source lane, so it is added to get_basic_test.
Measured on darwin: the lane goes from 168 tests in 101.98 s to 171
tests in 101.92 s, since the three cases overlap the lane's existing
work. The darwin selection digests for e2e-smoke and e2e-full are
regenerated; the e2e-full linux digest still needs a Linux runner.
* docs(changelog): record the on-demand migration feature
* fix(ecstore): stop scan_dir emitting entries past a limit hit inside a subdirectory
scan_dir's flush loop recurses into a pending subdirectory when the
current sibling entry's page limit is reached mid-recursion, but kept
writing the current (later-sorting) entry regardless. gather_results
then builds the next page's continuation marker from that later entry,
which permanently skips the still-unscanned tail of the subdirectory
on resume instead of just deferring it to the next page.
Add a limit re-check right after the flush loop, before the current
entry is written, so scan_dir stops cleanly at the true last-written
key. Reproduces and fixes the rc.5 recursive ListObjectsV2 data-loss
report (7826/7881 keys, contiguous 55-key block silently dropped).
Adds scan_dir_does_not_emit_entries_past_a_limit_hit_inside_a_subdirectory.
* fix(ecstore): re-check the page limit on every dir_stack flush iteration
The flush loop that drains dir_stack can pop and recurse into more than
one pending subdirectory per outer iteration (whenever more than one
stack entry sorts below the current sibling entry). The limit re-check
added in the previous commit only ran once, after the whole flush loop
exited - so if the first recursive scan_dir call already exhausted the
page limit, the loop's next pop+recurse still went ahead and scanned
(and emitted entries for) another subdirectory beyond where the page
was supposed to stop.
Confirmed against production data: a bucket with ~1.17M objects under
one prefix still cut a recursive ListObjectsV2 listing short (825 of an
expected much larger next page, IsTruncated=false) even with the first
fix deployed, at a two-level-nested subdirectory. Move the check inside
the while loop so it runs before every pop, not just once after.
---------
Co-authored-by: Claude Agent <agent@local>
* feat(ecstore): add on-demand migration backfill job core
Add the background backfill job for on-demand migration
(rustfs/backlog#2159): a durable checkpoint under
buckets/<bucket>/on-demand-migration-backfill.json saved by If-Match
compare-and-set every 1000 keys or 10 s, a 60 s owner lease renewed by
every save, a recovery pass that takes over expired leases (or jobs this
node owned before a restart) and cancels jobs whose config changed, and a
main loop over the source ListObjectsV2 pages with the skip_existing
policy, dry runs, bounded outstanding pulls and wait-on-full enqueueing.
The pull queue gains per-job completion reports so the job can count
pulled/failed keys (hashes only), and pull permits become two-tier so an
online miss is never queued behind a backfill pull.
* feat(admin): expose on-demand migration backfill job
Wire the ODM-12 backfill job (rustfs/backlog#2159) to its operators:
POST /v3/on-demand-migration/{bucket}/backfill?op=start|cancel and
GET .../backfill return the checkpoint document, GET .../status gains a
backfill summary, and the recovery loop plus the process-wide runner are
installed at startup. Backfill control reuses
Set/GetBucketOnDemandMigrationAction and is recorded in the route policy,
the registration matrix and the admin route snapshot.
Add the rustfs-madmin wire types and client methods with golden fixtures
shared by the server tests, the backfill_* metric descriptors and their
collector, and three e2e scenarios: a full backfill across list pages,
cancellation, and resuming from the persisted continuation token after a
server restart.
The guard added in #7021 fails a >5 GiB single-PutObject replication up
front instead of streaming the body to a target that must reject it. Its
message asserted a conclusion: "was not written as multipart on the
source ... re-upload it with multipart". That text is only as right as
the transport decision feeding it, and until #7047 that decision was
wrong for multipart objects carrying a full-object checksum. On 1.0.0-rc.5
a 768-part object was misrouted to the single-PUT path, and the new
default-level error line told the operator to re-upload as multipart an
object whose own ETag ended in -768.
State the evidence instead of the conclusion. The message now quotes the
ETag the decision was read from and says what was read from it (no
part-count suffix), so an operator can check the line against the
object's listing. A misroute then reads as a visible contradiction --
a suffixed ETag on a single-PUT line -- and the message says that case is
a transport-selection defect to report, not something to fix by
re-uploading. A missing or empty ETag is printed as <none> rather than
hidden.
The routing itself is already fixed by #7047; this changes only what the
guard says when it fires.
* fix(ecstore): persist merged checksum type for full-object multipart
complete_multipart_upload built the object-level checksum record from a
ChecksumType copied before the MULTIPART / INCLUDES_MULTIPART flags were
merged in. ChecksumType::merge takes &mut self, so the merge updated the
local variable while the copy already inside the Checksum struct stayed
behind. The composite branch rebuilt the Checksum from the merged type
and was unaffected; the full-object branch never rebuilt it, so those
flags never reached disk.
rustfs_rio::read_checksums only sets its multipart flag and only emits
the "x-amz-checksum-type" = "FULL_OBJECT" entry inside its MULTIPART
branch, so a full-object multipart object read back as non-multipart with
no type entry, and GetObject and HeadObject answered with no
x-amz-checksum-type header at all where AWS returns FULL_OBJECT.
Hand the full-object branch the merged type instead of rebuilding the
Checksum: the value must stay the running merge produced by add_part,
because hashing the concatenated part digests would yield the COMPOSITE
value, a different number than the one the client sent. The serialization
now lives in multipart_object_checksum_record so both shapes are covered
by unit tests.
Records written by earlier builds carry the bare algorithm type with no
MULTIPART flags and no trailing part block; they keep reading back to the
same checksum value, and the FULL_OBJECT reader arm predates this change
so older peers parse the new record shape correctly too.
Found while root-causing rustfs#6825.
* fix(s3): reject contradicting multipart checksum type as client error
A CompleteMultipartUpload declaring an x-amz-checksum-type that
contradicts the type recorded at CreateMultipartUpload answered 500
InternalError, telling the caller to retry a request that can only ever
fail. The storage layer does refuse the combination, but through a
generic error that maps to InternalError.
Validate the header against the recorded type in the usecase, where the
upload metadata returned by get_multipart_info is already in hand, and
answer InvalidRequest naming both types, matching AWS. The storage-layer
check stays as a backstop for non-HTTP callers.
Uploads created without a checksum algorithm record no type, so there is
nothing to contradict and the header is left alone rather than newly
rejected. Replication is unaffected: replication_put_object_options
already excludes x-amz-checksum-type from the metadata it forwards.
* test(e2e): cover full-object multipart checksum type round-trip
Adds an end-to-end test that a CRC32 FULL_OBJECT multipart upload reports
x-amz-checksum-type: FULL_OBJECT and the unsuffixed full-object value on
both GetObject and HeadObject, and one that a CompleteMultipartUpload
contradicting the recorded type is rejected as InvalidRequest while
leaving the upload intact. Extends the existing CRC64NVME multipart test
with the same checksum-type assertion.
* fix(s3): keep checksum-type validation off the s3s error macro
The s3s footprint ratchet (scripts/check_s3s_footprint.sh) counts
s3_error! invocation lines and is lower-only: new code must route
through the gateway abstractions rather than widen the direct s3s
surface the s3gate migration is shrinking.
Raise the contradiction through ApiError::invalid_request instead. The
response is byte-for-byte identical -- From<ApiError> for S3Error carries
the InvalidRequest code and the message through unchanged -- and the
usecase already returns ApiError elsewhere, so this is the idiomatic
path rather than a way around the counter.
The explanatory comment deliberately says "the s3s error macro" instead
of naming the macro: the ratchet counts raw matches, so spelling it out
in a comment tripped the same check.
* feat(ecstore): add on-demand migration pull queue and write-back pipeline
Background pull queue per bucket (bounded by pull_queue_capacity, concurrency via the state's pull slot), OdmWriteBack/PullSource traits, single-part and multipart write-back with a pumped body that enforces idle timeout, cancel and content length, retry policy for retryable source errors, inline commit helper, and stats accounting (rustfs/backlog#2153).
* feat(object): implement on-demand migration write-back over internal put
OdmWriteBack impl mapping source heads onto InternalPutContext (content-header allowlist, x-amz-meta copy, tags, dual-prefix odm-* provenance, ETag policy), injected into OnDemandMigrationSys at startup; removes the dead-code gates left by ODM-06a (rustfs/backlog#2153).
* feat(ecstore): add on-demand migration bucket config model
Introduce OnDemandMigrationConfig (deny_unknown_fields, version 1) with typed validation, credential redaction, a secret-free Debug impl, and the OnceLock publish hook the runtime registers into. Exported through the api facade.
* feat(ecstore): persist on-demand migration config in bucket metadata
Store the config as a RustFS extension entry (on-demand-migration.json) with its update time in .metadata.bin, add the typed BucketMetadataSys accessor, and publish the config through the hook on every cache-install path alongside the durability sync.
* refactor(ecstore): extract shared remote S3 client builder
Move the aws_sdk_s3 client construction out of bucket_target_sys into
bucket/remote_s3_client.rs: endpoint assembly, credential provider,
path-style selection, custom CA / skip-TLS transports and the outbound
SSRF gate now build from a neutral RemoteS3EndpointSpec so replication
targets and the upcoming on-demand migration source client share one
policy. Replication builds its client through From<&BucketTarget>; the
gate keeps its relaxed semantics (private allowed, loopback only behind
RUSTFS_REPLICATION_ALLOW_LOOPBACK_TARGET) verbatim. The builder also
gains optional connect/read timeouts and a User-Agent suffix
interceptor, both unset for replication.
Refs rustfs/backlog#2149
* feat(ecstore): add on-demand migration SourceClient
Add bucket/on_demand_migration/source_client.rs on top of the shared
remote S3 builder: HEAD, ranged streaming GET, ListObjectsV2 with
source-prefix mapping, GetObjectTagging and an admin probe. Every request
carries the x-rustfs-/x-minio-source-proxy-request anti-loop markers and
a RustFS-OnDemandMigration/<version> User-Agent suffix; SSE-C source
objects are rejected as unsupported. SourceError classifies SDK failures
(not found, access denied, throttled, timeout, connect, server error)
with retryability and a stable metrics label. Debug output redacts
credentials.
Refs rustfs/backlog#2149
* docs(operations): point outbound policy at shared remote S3 client builder
* chore: integrate ODM-01 and ODM-02 as B1 base (fix facade merge)
* feat(ecstore): add on-demand migration runtime OnDemandMigrationSys
Per-node runtime for On-Demand Migration (rustfs/backlog#2152): turns each
bucket's persisted config into a live SourceClient guarded by a three-state
circuit breaker, a TTL negative cache, per-key singleflight, a pull
concurrency semaphore and lock-free counters with a serializable snapshot.
- sys.rs: OnceLock singleton; `apply` installs/rebuilds/removes bucket state
(config compared by value, counters preserved across rebuilds, old
cancellation token fired); `publish` is the metadata publish-hook entry
(sync removal, spawned install, generation-ordered so a slow older install
cannot overwrite a newer one); `resolve(bucket, key)` judges module switch,
bucket state, prefix filter, client availability, negative cache, breaker.
- breaker.rs: Closed/Open/HalfOpen with fixed constants (5 failures / 30 s
window / 30 s open / 1 probe); NotFound resets, AccessDenied is neutral.
- negative_cache.rs: moka sync cache keyed by local key, ttl=0 disables.
- stats.rs: requests_total{op,outcome}, pulled_bytes/objects, pull_failures,
inflight/queue gauges, log-bucket latency histogram, last_source_error;
snake_case snapshot pinned by a golden JSON test.
- Anonymous sources surface as a typed `OdmStateError::AnonymousUnsupported`
until the shared client builder gains an anonymous mode.
- rustfs: `RUSTFS_ON_DEMAND_MIGRATION_ENABLED` module switch (default false)
published to module_switches and injected into ecstore before bucket
metadata loads; hook registered at the same point.
* feat(ecstore): add on-demand migration bucket config model
Introduce OnDemandMigrationConfig (deny_unknown_fields, version 1) with typed validation, credential redaction, a secret-free Debug impl, and the OnceLock publish hook the runtime registers into. Exported through the api facade.
* feat(ecstore): persist on-demand migration config in bucket metadata
Store the config as a RustFS extension entry (on-demand-migration.json) with its update time in .metadata.bin, add the typed BucketMetadataSys accessor, and publish the config through the hook on every cache-install path alongside the durability sync.
* test(e2e): rename stall timing variable flagged by typos
* test(storage): heap-pin the RestoreObject usecase future in the generation guard test
* refactor(ecstore): extract shared remote S3 client builder
Move the aws_sdk_s3 client construction out of bucket_target_sys into
bucket/remote_s3_client.rs: endpoint assembly, credential provider,
path-style selection, custom CA / skip-TLS transports and the outbound
SSRF gate now build from a neutral RemoteS3EndpointSpec so replication
targets and the upcoming on-demand migration source client share one
policy. Replication builds its client through From<&BucketTarget>; the
gate keeps its relaxed semantics (private allowed, loopback only behind
RUSTFS_REPLICATION_ALLOW_LOOPBACK_TARGET) verbatim. The builder also
gains optional connect/read timeouts and a User-Agent suffix
interceptor, both unset for replication.
Refs rustfs/backlog#2149
* feat(ecstore): add on-demand migration SourceClient
Add bucket/on_demand_migration/source_client.rs on top of the shared
remote S3 builder: HEAD, ranged streaming GET, ListObjectsV2 with
source-prefix mapping, GetObjectTagging and an admin probe. Every request
carries the x-rustfs-/x-minio-source-proxy-request anti-loop markers and
a RustFS-OnDemandMigration/<version> User-Agent suffix; SSE-C source
objects are rejected as unsupported. SourceError classifies SDK failures
(not found, access denied, throttled, timeout, connect, server error)
with retryability and a stable metrics label. Debug output redacts
credentials.
Refs rustfs/backlog#2149
* docs(operations): point outbound policy at shared remote S3 client builder
* fix(ilm): enqueue committed tier free versions
* fix(ilm): stabilize causal cleanup CI coverage
* test(ilm): make expire GET race deterministic
* test(ilm): synchronize expiry with active GET
A 6 GiB object uploaded to the source as a 768-part multipart upload was
replicated to a generic S3 target with a single PutObject, and the target
rejected the body with EntityTooLarge. No CreateMultipartUpload was ever
issued, so the multipart replication transport never ran for the object
it exists for.
`replication_put_object_options` seeded the transport from
`object_info.is_multipart()` and then overwrote it with the second
return value of `decrypt_checksums`. Those two booleans do not mean the
same thing: the first is the object's storage shape, read from the ETag,
while the second reports whether the stored *checksum record* carries
per-part data. A full-object checksum -- what `aws s3 cp` writes by
default for a CRC algorithm -- is serialized with no MULTIPART flag even
on a multipart upload, so the record reports false and the object was
routed as a single PUT. `decrypt_checksums` documents this in
object_api/types.rs: callers that need routing must consult
`is_multipart()`. Replication did the opposite.
Route on the object's own shape, and let the checksum record only add
multipart-ness, never take it away. Objects already stored with such a
record are fixed too: the ETag was always right.
This also repairs the diagnosis of rustfs#6825, where the single-PUT
5 GiB guard fired against an object that was multipart all along and
told the operator to re-upload it as multipart.
Tests cover the three shapes the router has to separate: a multipart
object with a full-object checksum record (the regression, which fails
without this change), a multipart object with a composite record, and a
single-part object that must not be promoted onto multipart.
DeleteBucket answers from a raw per-disk residue scan rather than from a
listing, so it can refuse for a reason no S3 request can observe: the
client drains every version the API will show, DeleteBucket still returns
BucketNotEmpty, and the client-visible message is the generic "The bucket
you tried to delete is not empty" for every blocker kind.
The server does know which residue blocked it, and where — that is what
`bucket_delete_blocked` carries. But it was emitted at `debug`, below
both the `error` DEFAULT_LOG_LEVEL and the `info` the CI s3-tests lane
runs at, so it was never actually written down. An intermittent
BucketNotEmpty in that lane leaves a server log with no trace of the
refusal at all, which is not a diagnosable state: confirmed against the
artifact log of a failing run, where the rejected bucket appears only in
span-close lines and the blocker event is absent entirely.
Split the blocker kinds by whether the client can still reach the
residue. A visible version or a tier free-version is an ordinary 409 —
the bucket really is not empty and the caller can list and delete what is
left — so that stays at `warn`. UnknownXlMeta, OrphanDirectory, and
DiagnosticBudgetExceeded are on-disk state no S3 request can remove; that
is a server-side integrity problem and is now reported at `error`, with
the blocker kind, the residue counts, and the sample path.
This does not change what DeleteBucket accepts or rejects, and does not
retry or suppress anything — it makes the existing diagnosis reachable.
Refs #7005, #7010
Replication could fail an object with nothing in the server log an
operator could act on. Every failure branch in the resyncer is quieter
than `error` on purpose — most sit on the hot path and fire once per
object per ARN — but `DEFAULT_LOG_LEVEL` is `error`, so on a stock
deployment a failed object produced no line at all. Raising those
branches to `warn` (#6840) did not close this: the default filter still
dropped them.
Report the terminal outcome instead of the branches. `replicate_object_
with_outcome` and `replicate_delete_with_outcome` now emit one `error`
per failed (object, target) once the per-target results are merged,
carrying the object key, version id, target ARN and endpoint, and the
target's own error, redacted through `sanitize_resync_error_detail` so
an echoed credential cannot reach the log. Volume is bounded by objects
that actually fail rather than by attempts inside a transfer.
Also state the single-PutObject size limit instead of discovering it at
the target. Replication picks its transport from the source object's
storage shape, not its size, so an object written with one PutObject
replicates with one PutObject however large it is — and S3 caps that at
5 GiB. Such an object could never reach a generic S3 target, and only
found out after streaming the whole body. `replication_single_put_size_
error` fails it up front with a message naming the size, the limit, and
the remedy.
Version-identity drift moves to `error` on a 10-minute per-ARN throttle.
It was `warn` deduped once per ARN per process, so the one line
explaining why a purged version is still on the target was both filtered
out by default and gone for good after it first fired.
Fixes#6825
Refs #6822
`build_metrics_summary` emitted a single metric entry for the local
deployment with `online` hardcoded to `true` and `last_online` stamped
with the current time, so `mc admin replicate status` reported "I am
online" rather than whether the remote site was reachable. A peer could
be down for minutes with replication failing while the status page
stayed green, leaving operators with no signal that the link had
dropped.
Emit an entry for every peer instead, deriving `online` from the
`reachable_peers` set the handler already computes by probing each peer,
and take `total_downtime`/`last_online` from the replication heartbeat's
existing `EpHealth` tracking. Node-local replication counters stay on
the local entry so a two-site cluster does not double-count its own
traffic.
The new `BucketTargetSys::endpoint_health` accessor deliberately does not
call `init_hc`: unlike `is_offline` it must not create health entries as
a side effect, or merely rendering the status page would mark an unknown
peer online.
Failure counters (`Errors`) are unchanged and still read zero; that is a
separate defect in the bucket-level statistics path and is not addressed
here.
Six set_disk::ops tests failed non-deterministically only under
concurrent full-suite load, rotating between runs while each passed in
isolation. All six share one root cause: a lock-owning put_object
quorum-acks once the rename fanout reaches write quorum and lets a
detached tail task finish the lagging disks, so a fixture that inspects
per-disk state immediately after PUT can observe a disk the tail has not
reached yet.
The two heal report fixtures, the inline-commit fixture, and the
transaction-fencing fixture read or delete physical shards right after
PUT, and hit FileNotFound on a lagging disk. The two metadata-cache
fixtures prime the cache after PUT, and the read fanout refuses to publish
a cache entry while any disk still reports an error, so the priming read
observably published nothing.
Keep every affected setup PUT on the full-fanout commit path with
no_lock: true, following the existing precedent in this module, so PUT
returns only after every disk has committed. The option only governs lock
acquisition, so it does not weaken what any of these fixtures assert; the
transaction-fencing gate in particular is driven by the fleet proof and
env vars, never by the lock option. Where a fixture also depends on cache
publication, re-prime until the current generation is observably cached
instead of asserting on a single read that a loaded host can stall past
the cache TTL. The heal race fixture's shard damage injection is
best-effort by construction, so it now skips injection when the previous
round's tail still lags rather than unwrapping a read that may
legitimately race.
No production code changes, and no retries or sleeps added.
* test(ecstore): retain final decommission capacity snapshot override
take_decommission_capacity_info_override_for_test used to pop the queue
to exhaustion, after which get_decommission_all_pool_capacity_infos
silently fell back to the host's real statfs numbers. Any new sampling
point added to the decommission start paths re-introduced that host
dependency and broke tests on some dev machines (#6989 patched one
instance by topping up snapshot counts, but the coupling remained).
Keep the final queued snapshot and replay it for every subsequent
sample so tests always observe injected capacity once an override is
installed. All existing injection patterns (single snapshot, repeated
identical snapshots, decreasing sequences ending at the post-operation
state) keep their semantics.
* test(ci): serialize load-sensitive heal and cache-generation tests
Under a heavily parallel nextest run (~792 ecstore tests), two tests of
set_disk::ops::heal::heal_result_report_tests failed nondeterministically
per round (different members each time; all 29 pass standalone). Every
test in the module builds a TempDir-backed 4-disk hermetic erasure set
and drives MiB-scale writes plus deep-scan heal: under load a single
disk's IO can fail while write quorum still holds, flipping per-disk
readback and aggregate-outcome assertions. The module's #[serial]
markers do not serialize across nextest's process-per-test boundary.
Verification also caught complete_multipart_generation_retires_cached_snapshot
failing once under the same load; it and its object.rs sibling carry
#[serial(metadata_cache_invalidation_probe)] and assert
get_object_metadata_cache generation semantics - the same shape that
forced the transition matrix tests into the serial group.
Add both families to the ecstore-serial-flaky test-group in the default
and ci profiles. Preventive serialization only, no retries. Three full
parallel rounds after the change: 792/792 passed each round.
A completed multipart upload's staging cleanup prunes empty parent
directories up to the volume root, which removes shared prefixes such as
`data-movement/` and the per-object `<sha>/` while a concurrent
new_multipart_upload builds its destination chain below them. The writer
holds a descriptor to the pruned component, so its next handle-relative
mkdirat fails NotFound. Because rename never retried NotFound, the cleanup
fan-out failed several disks in the same window and broke write quorum.
Give rename preparation its own retry rule: a NotFound is retried once per
component below the base directory, so a rebuilt walk outlasts a pruning
walk, which removes ancestors monotonically upward and stops at the base.
A destination whose parent is the base keeps NotFound terminal, so
speculative cleanup renames still fail fast, and the base is only ever
opened, never created, so a genuinely missing base still fails. The rename
itself keeps its own budget and its unchanged NotFound-is-terminal rule.
test(ecstore): assert inline fanout gate on deterministic scheduled metric
non_inline_data_read_early_stop_does_not_add_inline_fanout_on_unequal_layout
compared disk_call_counters::KIND_READ_VERSION totals between the two-phase
read-plan gate being off and on. That counter records inside each spawned
fanout task, so the single-pending inline hedge read races the early-stop
abort_all(): whether the hedge task gets its first poll before cancellation
decides a 4-vs-5 count per read. Under concurrent nextest load the two reads
can disagree (reproduced locally at ~5% when run beside one other test,
matching the CI failure on PR #6961).
Assert on the rustfs_io_get_object_metadata_fanout_scheduled histogram
instead, which records the scheduling decision synchronously in the fanout
loop and is deterministic, using the CapturingRecorder + current-thread
runtime pattern already used by the neighboring tests in this module.
overtrue