A cold rust-cache (every Cargo.lock change) means a full workspace +
failpoints-feature build on the 2-core runner, which now exceeds 45
minutes on slow runner days — and because a timed-out run never saves its
cache, an undersized budget self-perpetuates: every retry starts cold and
dies identically (observed four consecutive 45-minute cancellations on
main and PR #188 after #186's lock bump). Warm-cache runs stay ~15
minutes; 75 is headroom matching the rustfs job's budget, not a target.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
beta.4+ refuses the rustfsadmin/rustfsadmin test credentials unless
RUSTFS_ALLOW_INSECURE_DEFAULT_CREDENTIALS=true is set — acceptable for the
ephemeral CI container and the local bootstrap script (which already passed
it). The three S3 suites were validated against the beta.8 binary locally
before this bump. The pin stays explicit, never `latest`, so future
upgrades remain deliberate.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The apply-side coverage the implementation spec's hard gate requires before
Phase 4 graph-moving apply:
- crash after the payload phase: state.json byte-identical, blobs inert on
disk, lock released, no phantom statuses, nothing acknowledged; a plain
re-run repairs via skip-if-exists blob reuse.
- CAS race: a cfg_callback rewrites state.json at the exact read->write
window (the state.lock:false concurrent-writer scenario); apply surfaces
state_cas_mismatch, acknowledges nothing, reports the persisted status
snapshot, leaves the concurrent writer's state on disk; a re-run converges.
CI's failpoints step now runs both the engine and cluster suites.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The `test_windows_binaries` job ran a full Windows --release build +
smoke test on every code PR. It was a non-required (non-blocking) check,
so it never gated a merge — it only burned the slowest/most expensive
runner (windows-latest, --release, 75-min ceiling) on every code change.
Windows binary validation is already covered (better) on release tags:
release.yml's `smoke_windows_installer` (on v* tags) builds the release
binaries, installs via scripts/install.ps1, and smoke-runs
`omnigraph.exe version` + `omnigraph-server.exe --help` — the same smoke
test plus the real installer path. Nothing `needs:` the removed job.
Trade-off (accepted): a PR that breaks the Windows build or install.ps1
syntax is now caught at release-cut rather than at PR time. install.ps1
and platform-specific code change rarely; the cost savings on every PR
outweigh the earlier signal.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The container entrypoint's URI and config branches were mutually
exclusive, so a deployment driven by OMNIGRAPH_TARGET_URI could never
load a policy file. Forward --config alongside the positional URI when
OMNIGRAPH_CONFIG is also set (the URI still wins via resolve_target_uri),
enabling Cedar policy without changing how the URI is provided.
Add docker/entrypoint_test.sh (arg-composition cases) + a CI job, and
document the env-var contract in docs/user/deployment.md.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* exec/query: pushdown IR filters via DataFusion Expr (Scanner::filter_expr)
Switches `execute_node_scan` from string-flattened Lance SQL pushdown
(`build_lance_filter` + `scanner.filter(&str)`) to structured DataFusion
Expr pushdown (`build_lance_filter_expr` + `scanner.filter_expr(Expr)`).
## What this enables
1. **`CompOp::Contains` now pushes down.** `ir_filter_to_sql` returned
`None` for list-contains (the comment said *"Can't pushdown list
contains"*) because string SQL can't easily express it. With Expr,
it lowers to DataFusion's `array_has(col, value)` builtin via the
`nested_expressions` feature, and pushes down to Lance's scan layer
the same way Eq/Lt/etc. do. Pinned by the new regression test
`end_to_end::ir_filter_with_list_contains_pushes_down`.
2. **DataFusion 53's optimizer rules now reach our predicates.** Once
the Expr lands at the Lance scanner, DF's planner runs:
- `IN`-list vectorized eq kernel (DF #20528)
- `PhysicalExprSimplifier` (DF #20111)
- CASE WHEN x THEN y ELSE NULL shortcut (DF #20097)
- Push limit into hash join (DF #20228)
None of these were applicable before because the string SQL path
short-circuited the optimizer.
## Scope
This is one of three string-flattened pushdown sites; the other two
(`hydrate_nodes`/Expand pushdown at query.rs:771-796 and the mutation
delete path in `exec/mutation.rs::predicate_to_sql`) stay on the SQL
string path for now:
- The Expand pushdown still serializes through `hydrate_nodes`'s
`extra_filter_sql: Option<&str>` parameter. Migrating it changes the
`TableStorage` trait surface (`scan_stream(filter: Option<&str>)` →
`Option<Expr>`) and the cascading call sites — out of scope for this
MR.
- The mutation delete predicate still goes through `Dataset::delete(&str)`
in Lance 6.0.1. MR-A (delete two-phase via Lance #6658, gated on the
Lance v7 bump per issue #112) will migrate that path to
`DeleteBuilder::execute_uncommitted` taking an Expr.
The existing `ir_filter_to_sql` / `ir_expr_to_sql` / `literal_to_sql`
helpers stay in place to serve the remaining string-SQL consumers
(mutation predicates). They get retired when the other call sites
migrate.
## Cargo
Enables the `nested_expressions` feature on the `datafusion` workspace
dep. Lance already pulls in `datafusion-functions-nested` transitively
(it's listed in their feature set), so this just exposes the
`datafusion::functions_nested::expr_fn::array_has` re-export. No
transitive dep change (Cargo.lock unchanged).
## Tests
- New: `ir_filter_with_list_contains_pushes_down` — pins the case that
was previously impossible (`ir_filter_to_sql` returning `None`).
- 906/906 workspace tests still pass.
- 417/417 engine integration tests pass (was 416 + the new one).
- 19/19 failpoints (recovery canary).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* ci: pin rustfs/rustfs to 1.0.0-beta.3 (last known-good before creds-policy break)
The RustFS S3 Integration job started failing 2026-05-23 with all 3
tests panicking on the first PUT:
HTTP error: error sending request
The "Dump RustFS logs on failure" step revealed the container was
dying at startup:
[FATAL] Server encountered an error and is shutting down:
Default root credentials are not allowed on non-loopback listeners;
set RUSTFS_ACCESS_KEY and RUSTFS_SECRET_KEY to non-default values,
bind to loopback, or set RUSTFS_ALLOW_INSECURE_DEFAULT_CREDENTIALS=true
for local development only
`rustfs/rustfs:latest` was updated 2026-05-21 (1.0.0-beta.4) with a
credentials-policy check that rejects `rustfsadmin`/`rustfsadmin` as
"default" values. PR #111 passed yesterday because it ran against
beta.3; today's runs against beta.4 fail at container startup.
This is unrelated to PR #113's Expr-pushdown refactor — the bump
just happened to hit the same week.
Pin to 1.0.0-beta.3 (2026-05-14, last tag before the change). The
right long-term fix is one of:
- Rotate the CI creds to less-default values (less coupling to
RustFS's "default" set definition)
- Set `RUSTFS_ALLOW_INSECURE_DEFAULT_CREDENTIALS=true` per the
error message
- Use a workflow service container with controlled lifecycle
Deferred — pinning is the minimal restore. Also incidentally
documents *which* version we tested against, which `:latest` never
did.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The standalone test_failpoints_feature job took 21min on first run
(cold cache; the omnigraph-engine crate has lance + datafusion deps
that make any fresh build expensive). Folding into Test Workspace
shares the warm cache so the failpoints invocation is incremental —
~30s vs 21min on subsequent runs, and within the workspace job's
existing budget.
The failpoints feature is gated behind a Cargo flag and only adds
the small `fail` crate dep + a few feature-gated code paths; it
doesn't change the dep tree of any other crate.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Three new findings from Cubic on commit 3223b51:
* **Pending edge cardinality counted within-input duplicates** (P2):
count_src_per_edge's pending walk added every row to the count,
including duplicate rows that finalize will collapse via
dedupe_merge_batches_by_id. A LoadMode::Merge with the same edge id
twice would over-count → spurious @card violation. Fix: when
dedupe_key_column is Some, walk pending in reverse, track seen keys
via HashSet, count only the kept (last-occurrence) rows. Mirrors
finalize-time dedupe so cardinality counts what stage_merge_insert
actually publishes.
* **scan_with_pending silently disabled merge-shadow when projection
omitted key_column** (P2): if a caller passed Some("id") as
key_column but their projection didn't include "id", the
filter_out_rows_where_string_in helper passed batches through
unchanged — silently degrading to union semantics. Fix: validate
up front that projection contains key_column when both are Some;
return a typed Lance error otherwise. Tightened the helper too:
missing column is now an internal error (was a silent passthrough).
* **Cascade-vs-explicit delete test was too weak** (P2): asserted
only that edge count decreased after delete. The cascade alone
could satisfy that even if the explicit second-delete silently
no-op'd. Strengthened: assert post_knows == 0, which only holds
when both ops landed (Bob→Diana would survive if op-2 no-op'd).
CI gap: also added test_failpoints_feature job to .github/workflows/ci.yml.
The workspace test runs without --features failpoints (the feature is
behind a Cargo flag), so the failpoints test suite was never exercised
by CI before now. The new job builds + runs
`cargo test -p omnigraph-engine --features failpoints --test failpoints`
on every full CI run, mirroring the test_aws_feature pattern.
New tests on tests/runs.rs:
* load_merge_mode_dedupes_within_pending_for_cardinality_count
(Cubic P2 #2 — pending-vs-pending dedup, distinct from the
load_merge_mode_dedupes_edge_for_cardinality_count test which
covers committed-vs-pending dedup).
* scan_with_pending_rejects_key_column_missing_from_projection
(Cubic P2 #3 — verifies the up-front validation rejects bad
callers and that the happy path still works correctly).
Local test results:
* tests/runs.rs: 23/23 passed
* tests/failpoints.rs --features failpoints: 7/7 passed (includes the
two new finalize→publisher residual tests landed in 3223b51).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Splits the 990-line AGENTS.md into a 184-line map (architecture,
where-to-find index, always-on invariants, capability matrix,
maintenance contract) plus 18 new docs/*.md files holding the deep
content per topic (storage, schema and query languages, indexes,
embeddings, branches/commits, runs, merge, changes, execution, policy,
server, CLI reference, audit, errors, CI, constants, v0.3.1 notes).
Adds scripts/check-agents-md.sh and a check_agents_md CI job that
verifies every docs/ link in AGENTS.md resolves and every doc in the
canonical set is linked. CLAUDE.md remains a symlink to AGENTS.md.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Restore the default pull_request checkout (refs/pull/N/merge) so tests
see the merged state. The openapi.json auto-commit now uses a separate
shallow clone of the PR branch, so the pushed commit contains only the
spec change rather than the merge-commit tree.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The separate openapi-sync workflow was duplicating the workspace build
(~15 min cold-cache compile), paying the cost twice per PR. Fold the
regen + auto-commit into the existing test job: one compile, shared
rust-cache, same drift-check semantics.
- Same-repo PRs: OMNIGRAPH_UPDATE_OPENAPI=1 during the test run, then
commit the regenerated spec back to the PR branch
- Fork PRs / pushes: env var empty, test stays in strict drift-check mode
- openapi_spec_is_up_to_date treats empty env value as unset, so the
conditional workflow env expression works
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Introduces an opt-in AWS Secrets Manager backend for bearer tokens,
behind the `aws` Cargo feature. Default builds (on-prem, local dev)
don't pull in the AWS SDK and don't pay its compile cost.
- New Cargo feature `aws` gates the `aws-config` + `aws-sdk-secretsmanager`
optional deps. Default features remain empty.
- New `auth::aws::SecretsManagerTokenSource` implements `TokenSource` by
fetching a JSON `{"actor_id": "token", ...}` payload from a named
Secrets Manager secret. Credentials resolve via the AWS default chain
(env, shared config, IMDSv2 instance role, ECS task role) so no
explicit plumbing is needed under an IAM role.
- New `resolve_token_source()` dispatches based on the
`OMNIGRAPH_SERVER_BEARER_TOKENS_AWS_SECRET` env var. If the var is set
but the binary was built without `--features aws`, returns a clear
rebuild instruction rather than silently falling back.
- `serve()` now uses `resolve_token_source()` and logs which source was
selected at startup.
- `parse_json_secret_payload()` is factored out as a free function so
the payload validation (trim whitespace, reject blank actor/token,
reject non-object) is unit-testable without the AWS SDK.
- New CI job `test_aws_feature` builds + tests with `--features aws`.
Not in this PR (follow-ups):
- Background refresh loop for rotation. `SecretsManagerTokenSource`
advertises `supports_refresh: true` but the AppState-level refresh
task isn't wired yet.
- Config-YAML dispatch (today the AWS source is selected via env var
only; eventually `server.bearer_tokens.source` in `omnigraph.yaml`).
Tests:
- Default-feature build: 33 lib + 41 integration + 64 openapi.
- `--features aws` build: 32 lib (one test is cfg-gated) + 41 + 64.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
