docs(kms): correct the static backend's MinIO compatibility claim (#5596)

* docs(kms): correct the static backend's MinIO compatibility claim

`StaticConfig` claimed the backend derives DEKs via "HMAC-SHA256 +
AES-256-GCM, matching the MinIO builtin/static KMS wire format". Neither
half holds: the configured key is used directly as the AES-256-GCM key
with no derivation step, and wrapped DEKs are serialized as RustFS's own
`DataKeyEnvelope` JSON. MinIO's KMS ciphertext uses a different shape,
which `is_data_key_envelope` explicitly classifies as foreign (see the
`minio_legacy` case in encryption/dek.rs).

The claim as written tells a migrating operator that MinIO-written
ciphertext will open here, which it will not. Restate what the backend
actually does and point at rustfs/backlog#1638 for the real interop work.

Also refresh the neighbouring ciphertext-format note in static_kms.rs,
which still described a raw `ciphertext || nonce` layout that the JSON
envelope replaced.

* ci(minio-interop): fix the dead test selector and guard against empty runs

The job selected its tests with `-p rustfs-ecstore -E
'binary(minio_generated_read_test)'`. #5435 moved those reader tests from
crates/ecstore/tests/minio_generated_read_test.rs into the `rustfs` crate
as a `#[cfg(test)] mod`, which removed that test binary; the selector has
selected zero interop tests since. Point it at the tests where they now
live, verified locally:

  cargo nextest list --run-ignored all -p rustfs --features rio-v2 \
    -E 'test(minio_generated_read_test::)'   # 4 tests, was 0

Add a guard step in front of the run. The old `binary(...)` form happened
to fail loudly once its binary disappeared, but the name-based form that
replaces it is a valid filterset even when it matches nothing, so a later
rename would silently reduce this job to a pass that asserts nothing. The
guard counts the selection and fails with an explicit reason; the count
comes from `filter-match.status`, since the JSON's top-level `test-count`
is the package total and ignores `-E`. `--no-tests=fail` on the run step
covers the same case if the guard is ever dropped.

Also record in the header what this job does and does not prove: MinIO
wrapped-DEK envelopes are still rejected by both envelope parsers, and
that work is tracked in rustfs/backlog#1638.
This commit is contained in:
Zhengchao An
2026-08-02 05:31:30 +08:00
committed by GitHub
parent a12043f49f
commit da6fc5314d
3 changed files with 58 additions and 7 deletions
+45 -4
View File
@@ -18,7 +18,14 @@
# This is NOT a PR gate. The fixtures are real MinIO backend trees generated on
# the fly (they are gitignored, never committed), so the job regenerates them
# each run with Docker and then runs the `#[ignore]` reader tests in
# crates/ecstore/tests/minio_generated_read_test.rs.
# rustfs/src/storage/minio_generated_read_test.rs.
#
# Scope: end-to-end MinIO-to-RustFS SSE interop is NOT implemented yet. Both
# envelope parsers reject MinIO's own wrapped-DEK shape — see
# `is_data_key_envelope` in crates/kms/src/encryption/dek.rs and the
# `deny_unknown_fields` `LocalSseDekEnvelope` in rustfs/src/storage/sse.rs — and
# closing that gap is tracked in rustfs/backlog#1638. Treat this job as the
# harness for #1638, not as standing evidence that a MinIO migration reads back.
#
# Runner: GitHub-hosted `ubuntu-latest`. It reliably ships Docker + Python,
# unlike the self-hosted fleet, whose pods drift in Docker/pip availability
@@ -55,6 +62,19 @@ jobs:
env:
# Fixed 32-byte test KMS key baked into the fixture lab; not a secret.
RUSTFS_MINIO_STATIC_KMS_KEY_B64: IyqsU3kMFloCNup4BsZtf/rmfHVcTgznO2F25CkEH1g=
# Single definition of "the interop tests", shared by the guard step and
# the run step so the two cannot drift apart.
#
# These used to live in crates/ecstore/tests/minio_generated_read_test.rs
# and were selected with `-p rustfs-ecstore -E
# 'binary(minio_generated_read_test)'`. #5435 moved them into the `rustfs`
# crate as a `#[cfg(test)] mod`, which deleted that test binary; the
# selector was never updated and has selected zero interop tests ever
# since (cargo-nextest 0.9.140 now rejects it outright: "operator didn't
# match any binary names", exit 94).
INTEROP_PACKAGE: rustfs
INTEROP_FEATURES: rio-v2
INTEROP_FILTER: "test(minio_generated_read_test::)"
steps:
- name: Checkout repository
uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7
@@ -71,8 +91,29 @@ jobs:
- name: Generate real MinIO fixtures via Docker
run: bash crates/rio-v2/tests/minio_fixture_lab/capture_via_docker.sh
# `binary(...)` at least dies loudly when nothing matches, but `test(...)`
# is a perfectly valid filterset that matches zero tests, so the next
# rename or module move would leave this job selecting nothing and
# reporting success without executing a single interop assertion. Count
# the selection and fail with a reason instead.
#
# Count only `filter-match.status == "matches"`: the top-level
# `test-count` in the JSON is the package total and ignores `-E` entirely.
- name: Assert the interop selector still matches tests
run: |
set -euo pipefail
count="$(cargo nextest list --run-ignored all \
-p "$INTEROP_PACKAGE" --features "$INTEROP_FEATURES" \
-E "$INTEROP_FILTER" --message-format json \
| python3 -c 'import json,sys; d=json.load(sys.stdin); print(sum(1 for s in d.get("rust-suites", {}).values() for t in s.get("testcases", {}).values() if t.get("filter-match", {}).get("status") == "matches"))')"
echo "interop tests selected: ${count}"
if [ "${count}" -eq 0 ]; then
echo "::error::Selector '${INTEROP_FILTER}' in package '${INTEROP_PACKAGE}' matched 0 tests. The MinIO interop reader tests have moved or been renamed again; fix the selector instead of letting this job pass without running them. Context: rustfs/backlog#1638."
exit 1
fi
- name: Run MinIO interop reader tests
run: |
cargo nextest run --run-ignored ignored-only \
-p rustfs-ecstore --features rio-v2 \
-E 'binary(minio_generated_read_test)'
cargo nextest run --run-ignored ignored-only --no-tests=fail \
-p "$INTEROP_PACKAGE" --features "$INTEROP_FEATURES" \
-E "$INTEROP_FILTER"