fix(replication): apply receiver-side LWW to inbound metadata categories

Metadata-only replication reuses the whole-object PUT/multipart
transports, so in active-active topologies an inbound authorized
replication write carried the source's tags, retention, and legal hold
verbatim and unconditionally overwrote a category the destination had
modified more recently — both sites ended permanently diverged while
reporting COMPLETED (rustfs/backlog#1953, audit A4/P1-6).

The receiver now judges each category independently under the object
write lock: when the destination version's stored internal timestamp is
strictly newer than the inbound source timestamp, the local category
values and timestamp are kept; the rest of the write proceeds per the
inbound metadata and the object-level result stays successful, so a
local win never feeds an MRF retry loop. When the inbound category wins,
its internal timestamp key is pinned to the source-authored time instead
of the receiver-now() value stamped by the object-lock eval_metadata
path. No stored timestamp (pre-P1-6 data) or no inbound timestamp keeps
today's overwrite behavior.

The put_object hook reuses the existing commit-lock WORM-gate read (no
extra fanout); complete_multipart_upload adds one gated read under its
held lock, and the sender's complete options now carry the three
category timestamps so the multipart transport gets the same receiver
behavior. Local wins restore the timestamp via insert_str so a
MinIO-written single-key version still yields both compatibility keys.
This commit is contained in:
唐小鸭
2026-08-21 19:29:26 +08:00
parent adb90fc6e1
commit 81ef2882df
7 changed files with 800 additions and 6 deletions
@@ -3868,6 +3868,7 @@ async fn replicate_object_with_multipart<S: ReplicationObjectIO>(ctx: MultipartR
actual_size,
object_info.etag.clone().unwrap_or_default(),
object_info.mod_time,
&put_opts.internal,
),
)
.await
@@ -472,6 +472,7 @@ pub(crate) fn replication_complete_multipart_options(
actual_size: String,
source_etag: String,
source_mtime: Option<OffsetDateTime>,
source_internal: &AdvancedPutOptions,
) -> PutObjectOptions {
let mut user_metadata = HashMap::new();
insert_header_map(&mut user_metadata, SUFFIX_REPLICATION_ACTUAL_OBJECT_SIZE, actual_size);
@@ -484,6 +485,14 @@ pub(crate) fn replication_complete_multipart_options(
// mtime must degrade to epoch so header() suppresses the header
// instead of asserting the replication time as the object's mtime.
source_mtime: source_mtime.unwrap_or(OffsetDateTime::UNIX_EPOCH),
// Carry the per-category LWW timestamps on the complete request as
// well: the receiver's CompleteMultipartUpload options builder
// parses the same headers, so the multipart transport gets the
// same receiver-side LWW as the single-PUT transport
// (rustfs/backlog#1953). Epoch values keep the headers suppressed.
tagging_timestamp: source_internal.tagging_timestamp,
retention_timestamp: source_internal.retention_timestamp,
legalhold_timestamp: source_internal.legalhold_timestamp,
replication_status: ReplicationStatusType::Replica,
replication_request: true,
..Default::default()
@@ -649,20 +658,39 @@ mod tests {
#[test]
fn replication_complete_multipart_options_sets_actual_size() {
let source_mtime = OffsetDateTime::from_unix_timestamp(1_716_170_000).expect("valid test timestamp");
let source_internal = AdvancedPutOptions {
tagging_timestamp: OffsetDateTime::from_unix_timestamp(1_716_170_100).expect("valid test timestamp"),
retention_timestamp: OffsetDateTime::from_unix_timestamp(1_716_170_200).expect("valid test timestamp"),
legalhold_timestamp: OffsetDateTime::from_unix_timestamp(1_716_170_300).expect("valid test timestamp"),
..Default::default()
};
let options = replication_complete_multipart_options(
"1024".to_string(),
"0123456789abcdef0123456789abcdef-3".to_string(),
Some(source_mtime),
&source_internal,
);
assert_eq!(options.internal.source_etag, "0123456789abcdef0123456789abcdef-3");
assert_eq!(options.internal.source_mtime, source_mtime);
// The complete request must carry the same per-category LWW timestamps
// as the initiate request; the receiver reads them from the complete
// headers (rustfs/backlog#1953).
assert_eq!(options.internal.tagging_timestamp, source_internal.tagging_timestamp);
assert_eq!(options.internal.retention_timestamp, source_internal.retention_timestamp);
assert_eq!(options.internal.legalhold_timestamp, source_internal.legalhold_timestamp);
// Absent source mtime must degrade to epoch (header suppressed), not
// the AdvancedPutOptions default of now_utc() — that default would
// stamp the replication time as the replica's mtime and break the
// multipart HEAD convergence.
let options_no_mtime = replication_complete_multipart_options("1024".to_string(), String::new(), None);
// multipart HEAD convergence. Unset category timestamps stay epoch so
// header() keeps suppressing them.
let options_no_mtime =
replication_complete_multipart_options("1024".to_string(), String::new(), None, &AdvancedPutOptions::default());
assert_eq!(options_no_mtime.internal.source_mtime.unix_timestamp(), 0);
assert_eq!(options_no_mtime.internal.tagging_timestamp.unix_timestamp(), 0);
assert_eq!(options_no_mtime.internal.retention_timestamp.unix_timestamp(), 0);
assert_eq!(options_no_mtime.internal.legalhold_timestamp.unix_timestamp(), 0);
assert_eq!(
get_header_map(&options.user_metadata, SUFFIX_REPLICATION_ACTUAL_OBJECT_SIZE).as_deref(),