fix(replication): fail SSE-C passthrough closed on targets that drop transport headers

SSE-C ciphertext passthrough replicates via X-Rustfs-Replication-* transport
headers. A MinIO/generic-S3 target silently discards them, storing bare
ciphertext with no decryption material — yet the PUT succeeded, so the object
reported COMPLETED with a silently unreadable replica (backlog#1675 N2).

Fail-closed design:
- SsecPassthroughCapability {Unknown, Supported, Unsupported} cached in
  BucketTargetSys per target ARN with a recording timestamp. Entries reset
  whenever the target is rebuilt, edited, or removed (arn_remotes_map
  lifecycle) and expire after SSEC_PASSTHROUGH_CAPABILITY_TTL (10 minutes):
  an expired verdict in either direction is re-earned through the audit, so
  an Unsupported target recovers automatically after an upgrade (at most one
  wasted PUT+HEAD audit per bad target per TTL window) and a Supported
  verdict cannot outlive a backend swapped behind the same endpoint.
- Replication worker (replicate_object and replicate_all): fresh Unsupported
  targets never receive the PUT — the attempt fails immediately into the
  normal MRF retry channel with a "run ?replication-check to re-probe" hint.
  Unknown or expired verdicts are audited: after the PUT the worker HEADs
  the replica back through the replication-check channel (source version id
  mapped through resolve_read_api_version_id, so null-version objects audit
  correctly) and requires SSE-C evidence (the echoed customer-algorithm
  header); missing evidence records Unsupported and fails the attempt.
  Convergence HEADs are audited the same way, so a broken ciphertext replica
  from an earlier attempt can never launder itself into COMPLETED via an
  ETag match. The gate/evidence policy is pure (replication_target_boundary,
  staleness folded in as an input) for the M2 worker migration.
- replication-check grows an SsecPassthrough probe phase: a probe PUT
  carrying the live transport-header shape, HEAD-back for evidence, and a
  machine-readable Code BucketRemoteSsecPassthroughUnsupported on failure.
  The probe verdict is synced into the runtime capability cache. Unlike
  VersionFidelity, a failed SsecPassthrough phase does NOT fail the target
  overall — it is a capability limit, not a broken replication contract,
  and a plaintext-only deployment against such a target must not turn red.
- fake_s3_target: default mode now models a RustFS target (stores the
  transport headers, echoes SSE-C evidence); the new
  drop_unlisted_replication_headers mode models MinIO. The journal records
  whether a request carried transport headers.

Receiver-echo verification: the replication-check HEAD exemption only skips
SSE-C key validation; the response has always built sse-customer-algorithm
from stored metadata (rustfs/src/app/object_usecase.rs), so no receiver
change was needed — pinned end to end by the replication-check e2e against
a real RustFS target.

Rolling-upgrade constraint: RustFS targets older than the replication-check
HEAD exemption (#5898) answer the audit HEAD without SSE-C evidence (or fail
it outright), so SSE-C replication to such targets reports FAILED. This is
deliberate — FAILED-and-retryable beats a silently undecryptable replica —
and self-heals: once the target is upgraded, the next TTL expiry (or a
manual ?replication-check re-probe) re-audits and records Supported.
Plaintext and managed-SSE replication are unaffected. The capability cache
is per-node; each node audits independently.

Known limitations:
- The audit judges evidence from the echoed customer-algorithm header only.
  A hypothetical target that preserves that one header while dropping other
  transport headers (partial-drop) would pass the audit; no known target
  behaves this way — observed targets drop the whole unknown-header family.
- A mixed-version target cluster can flap the verdict between audits routed
  to different target nodes until the rollout completes; the TTL bounds how
  long each stale verdict persists.

New e2e (backlog#1675 C1 + N2, red-first): fail-closed against a
header-dropping fake (FAILED + no second PUT via the capability cache,
journal-asserted; red run showed the old COMPLETED), replication-check
reports the SsecPassthrough phase Code while the target stays OK overall,
SSE-C heal convergence after a real target outage, and SSE-C
existing-object resync landing a REPLICA readable with the customer key.
TTL expiry in both directions is pinned at the cache and gate seams.
This commit is contained in:
唐小鸭
2026-08-18 00:54:26 +08:00
parent 9baa92563a
commit d20476a66c
8 changed files with 1300 additions and 31 deletions
+129 -2
View File
@@ -90,6 +90,13 @@ const SOURCE_LEGALHOLD_TIMESTAMP_HEADERS: [&str; 2] = [
"x-rustfs-source-replication-legalhold-timestamp",
"x-minio-source-replication-legalhold-timestamp",
];
/// Wire prefix of the SSE-C passthrough replication transport headers
/// (`X-Rustfs-Replication-*`). In the default mode the fake stores them like a
/// RustFS target and echoes SSE-C evidence back on HEAD/GET; with
/// [`FakeS3Target::drop_unlisted_replication_headers`] it models MinIO /
/// generic S3, which silently discard unknown x-* headers.
const REPLICATION_SSE_TRANSPORT_PREFIX: &str = "x-rustfs-replication-";
const REPLICATION_SSEC_ALGORITHM_TRANSPORT_HEADER: &str = "x-rustfs-replication-ssec-algorithm";
const RESERVED_BUCKET_PREFIXES: [&str; 3] = ["xn--", "sthree-", "amzn-s3-demo-"];
const RESERVED_BUCKET_SUFFIXES: [&str; 6] = ["-s3alias", "--ol-s3", ".mrap", "--x-s3", "--table-s3", "-an"];
@@ -166,6 +173,10 @@ pub struct ProxyHeaderSnapshot {
pub ssec_algorithm: Option<String>,
pub ssec_key_present: bool,
pub ssec_key_md5: Option<String>,
/// Whether the request carried any `X-Rustfs-Replication-*` SSE-C
/// passthrough transport header, so fail-closed tests can assert the
/// sender really shipped the material a dropping target discarded.
pub ssec_transport_present: bool,
}
impl ProxyHeaderSnapshot {
@@ -179,6 +190,9 @@ impl ProxyHeaderSnapshot {
.map(bounded_journal_value),
ssec_key_present: headers.contains_key("x-amz-server-side-encryption-customer-key"),
ssec_key_md5: header_value(headers, &["x-amz-server-side-encryption-customer-key-md5"]).map(bounded_journal_value),
ssec_transport_present: headers
.keys()
.any(|name| name.as_str().starts_with(REPLICATION_SSE_TRANSPORT_PREFIX)),
}
}
}
@@ -213,6 +227,10 @@ struct ControlState {
struct StoreState {
assign_own_version_ids: bool,
assign_own_multipart_version_ids: bool,
/// MinIO-like mode: silently discard non-whitelisted replication
/// transport headers instead of storing them (see
/// [`REPLICATION_SSE_TRANSPORT_PREFIX`]).
drop_unlisted_replication_headers: bool,
buckets: HashMap<String, BucketState>,
uploads: HashMap<String, MultipartState>,
total_bytes: usize,
@@ -237,6 +255,9 @@ struct ObjectVersion {
/// Object tags as ordered key/value pairs (PutObjectTagging replaces the
/// whole set, DeleteObjectTagging clears it).
tags: Vec<(String, String)>,
/// SSE-C passthrough transport headers stored with the version (RustFS
/// target behavior); empty when the drop mode discarded them.
replication_sse_headers: Vec<(String, String)>,
}
#[derive(Clone)]
@@ -246,6 +267,7 @@ struct MultipartState {
version_id: String,
content_type: Option<String>,
metadata: Option<HashMap<String, String>>,
replication_sse_headers: Vec<(String, String)>,
parts: BTreeMap<i32, MultipartPart>,
}
@@ -466,6 +488,15 @@ impl FakeS3Target {
/// Mint own version ids for the multipart path only — models a target
/// that adopts PutObject version ids but not CreateMultipartUpload ones.
/// MinIO-like mode: silently drop every `X-Rustfs-Replication-*` SSE-C
/// passthrough transport header instead of storing it. The default (off)
/// models a RustFS target, which preserves the headers and echoes SSE-C
/// evidence (`x-amz-server-side-encryption-customer-algorithm`) on
/// HEAD/GET of the replica.
pub fn drop_unlisted_replication_headers(&self, enabled: bool) {
lock(&self.backend.store).drop_unlisted_replication_headers = enabled;
}
pub fn assign_own_multipart_version_ids(&self, enabled: bool) {
lock(&self.backend.store).assign_own_multipart_version_ids = enabled;
}
@@ -842,6 +873,29 @@ fn new_version_id(headers: &HeaderMap, assign_own: bool) -> S3Result<String> {
Ok(version_id.to_string())
}
/// Capture the SSE-C passthrough transport headers a replication PUT carried.
/// Returns an empty set in the MinIO-like drop mode.
fn captured_replication_sse_headers(headers: &HeaderMap, drop_unlisted: bool) -> Vec<(String, String)> {
if drop_unlisted {
return Vec::new();
}
headers
.iter()
.filter(|(name, _)| name.as_str().starts_with(REPLICATION_SSE_TRANSPORT_PREFIX))
.filter_map(|(name, value)| Some((name.as_str().to_string(), value.to_str().ok()?.to_string())))
.collect()
}
/// SSE-C evidence a RustFS-like target echoes for a stored passthrough
/// replica: the customer algorithm restored from the transport headers.
fn stored_sse_customer_algorithm(version: &ObjectVersion) -> Option<String> {
version
.replication_sse_headers
.iter()
.find(|(name, _)| name == REPLICATION_SSEC_ALGORITHM_TRANSPORT_HEADER)
.map(|(_, value)| value.clone())
}
fn source_etag(headers: &HeaderMap) -> S3Result<Option<String>> {
header_value(headers, &SOURCE_ETAG_HEADERS)
.map(|value| validate_retained_identifier(value, "source ETag").map(|value| normalize_etag(&value)))
@@ -1312,7 +1366,10 @@ impl S3 for FakeBackend {
let input = req.input;
let body = collect_stream(input.body, input.content_length, fault.as_ref(), &self.control).await?;
validate_stored_metadata(&input.content_type, &input.metadata)?;
let assign_own = lock(&self.store).assign_own_version_ids;
let (assign_own, drop_unlisted) = {
let state = lock(&self.store);
(state.assign_own_version_ids, state.drop_unlisted_replication_headers)
};
let version_id = new_version_id(&headers, assign_own)?;
let e_tag = match source_etag(&headers)? {
Some(value) => value,
@@ -1330,6 +1387,7 @@ impl S3 for FakeBackend {
content_type: input.content_type,
metadata: input.metadata,
tags: Vec::new(),
replication_sse_headers: captured_replication_sse_headers(&headers, drop_unlisted),
};
upsert_version(&mut lock(&self.store), &input.bucket, input.key, version)?;
Ok(apply_response_fault(
@@ -1350,6 +1408,7 @@ impl S3 for FakeBackend {
let state = lock(&self.store);
find_version(&state, &input.bucket, &input.key, input.version_id.as_deref())?
};
let sse_customer_algorithm = stored_sse_customer_algorithm(&version);
Ok(apply_response_fault(
S3Response::new(GetObjectOutput {
body: Some(StreamingBlob::new(Body::from(version.body.clone()))),
@@ -1359,6 +1418,7 @@ impl S3 for FakeBackend {
e_tag: Some(ETag::Strong(version.e_tag)),
last_modified: Some(version.last_modified.clone()),
version_id: Some(version.version_id),
sse_customer_algorithm,
..Default::default()
}),
fault.as_ref(),
@@ -1373,6 +1433,7 @@ impl S3 for FakeBackend {
let state = lock(&self.store);
find_version(&state, &input.bucket, &input.key, input.version_id.as_deref())?
};
let sse_customer_algorithm = stored_sse_customer_algorithm(&version);
Ok(apply_response_fault(
S3Response::new(HeadObjectOutput {
content_length: Some(version.body.len() as i64),
@@ -1381,6 +1442,7 @@ impl S3 for FakeBackend {
e_tag: Some(ETag::Strong(version.e_tag)),
last_modified: Some(version.last_modified.clone()),
version_id: Some(version.version_id),
sse_customer_algorithm,
..Default::default()
}),
fault.as_ref(),
@@ -1530,6 +1592,7 @@ impl S3 for FakeBackend {
content_type: None,
metadata: None,
tags: Vec::new(),
replication_sse_headers: Vec::new(),
},
)?;
Ok(apply_response_fault(
@@ -1557,9 +1620,10 @@ impl S3 for FakeBackend {
ensure_upload_budget(&state)?;
validate_stored_metadata(&input.content_type, &input.metadata)?;
let upload_id = Uuid::new_v4().to_string();
// Read the flag before the mutable borrow of `state.uploads` below
// Read the flags before the mutable borrow of `state.uploads` below
// (and never re-lock the store: the mutex is not reentrant).
let mint_own = state.assign_own_version_ids || state.assign_own_multipart_version_ids;
let drop_unlisted = state.drop_unlisted_replication_headers;
let version_id = new_version_id(&headers, mint_own)?;
state.uploads.insert(
upload_id.clone(),
@@ -1569,6 +1633,7 @@ impl S3 for FakeBackend {
version_id,
content_type: input.content_type,
metadata: input.metadata,
replication_sse_headers: captured_replication_sse_headers(&headers, drop_unlisted),
parts: BTreeMap::new(),
},
);
@@ -1706,6 +1771,7 @@ impl S3 for FakeBackend {
version_id: upload.version_id.clone(),
content_type: upload.content_type.clone(),
metadata: upload.metadata.clone(),
replication_sse_headers: upload.replication_sse_headers.clone(),
parts: BTreeMap::new(),
},
selected,
@@ -1733,6 +1799,7 @@ impl S3 for FakeBackend {
content_type: upload.content_type,
metadata: upload.metadata,
tags: Vec::new(),
replication_sse_headers: upload.replication_sse_headers,
};
let mut state = lock(&self.store);
let current = state
@@ -1937,6 +2004,65 @@ mod tests {
Ok(())
}
/// Default mode is RustFS-like: SSE-C passthrough transport headers are
/// stored and the customer algorithm is echoed on HEAD/GET. Drop mode is
/// MinIO-like: the headers are silently discarded, so no evidence comes
/// back — the exact difference the N2 fail-closed audit keys on. Both
/// modes journal that the sender shipped the transport headers.
#[tokio::test]
async fn ssec_passthrough_headers_echo_and_drop_modes() -> Result<(), BoxError> {
let target = FakeS3Target::start().await?;
target.create_bucket("target-bucket");
let client = client(&target);
let put_with_transport_headers = |key: &'static str| {
client
.put_object()
.bucket("target-bucket")
.key(key)
.body(ByteStream::from_static(b"ciphertext"))
.customize()
.map_request(move |mut request| {
let headers = request.headers_mut();
headers.insert("x-rustfs-replication-ssec-algorithm", "AES256");
headers.insert("x-rustfs-replication-ssec-key-md5", "AAAAAAAAAAAAAAAAAAAAAA==");
Ok::<_, std::convert::Infallible>(request)
})
.send()
};
put_with_transport_headers("kept").await?;
let head = client.head_object().bucket("target-bucket").key("kept").send().await?;
assert_eq!(head.sse_customer_algorithm(), Some("AES256"));
let get = client.get_object().bucket("target-bucket").key("kept").send().await?;
assert_eq!(get.sse_customer_algorithm(), Some("AES256"));
target.drop_unlisted_replication_headers(true);
put_with_transport_headers("dropped").await?;
let head = client.head_object().bucket("target-bucket").key("dropped").send().await?;
assert_eq!(head.sse_customer_algorithm(), None, "drop mode must discard SSE-C evidence");
let requests = target.requests();
for key in ["kept", "dropped"] {
let record = requests
.iter()
.find(|record| record.operation == Operation::PutObject && record.key.as_deref() == Some(key))
.expect("PUT must be journaled");
assert!(
record.proxy_headers.ssec_transport_present,
"the journal must prove the sender shipped the transport headers for {key}"
);
}
let plain_head = requests
.iter()
.find(|record| record.operation == Operation::HeadObject)
.expect("HEAD must be journaled");
assert!(!plain_head.proxy_headers.ssec_transport_present);
target.shutdown().await;
Ok(())
}
macro_rules! assert_sdk_error {
($error:expr, $status:expr, $code:expr) => {{
let error = &$error;
@@ -3202,6 +3328,7 @@ mod tests {
version_id: index.to_string(),
content_type: None,
metadata: None,
replication_sse_headers: Vec::new(),
parts: BTreeMap::new(),
},
);