alkstore/tasks/pg-fix-scheduler-tick-seam.md at 5abd351c86d6d0b0a761f3bf199ccaeb5dc3f1de

0 Fork 0

Files

T

glm-5.3-flash 5abd351c86 Wave-6 decomposition (mem engine): eleven tasks per the pg-wave rhythm — foundations (crate scaffold, MemStore::new isolation, guarded state core, wake routing, wasm32 gate from birth), seam-tx (the load-bearing tx overlay: eleven staged _tx methods, commit-merge + wake-at-commit — pg-fix-tx-wake's discipline structural from the start, drop = discard trivial, with_tx panic posture), mechanisms mutually parallel (notify-listen close-at-engine-drop/no-replay/never-coalesced; queues = ADR-010 machine driver-free with the equal-jitter curve and opts resolution as ADR-012 §2's third owner; streams log+offsets+trim; locks TTL registry, no busy-path analog), scheduler-outbox (tick/catch-up-64/leadership through the __alkstore_scheduler lock, engine-uniform per ADR-009 §4; outbox over queues with the 60/5/5 seam stamp), integration (suite third StoreFactory column green on host under a current-thread flavor, wasm32 acceptance items, engine-mem.md to implementation-complete), and the review-wave-6 gate (born-pinned audit: conformance read, discharge-map wiring, wasm items, three-way equivalence green, class-1 window discipline). Window riders: suite-harness-current-thread — the wave's one suite-side pre-task, past_stamp_sleep's blocking sleep goes async per review 003's properties.rs:1683 posture note, enabling the current-thread flavor; pg-fix-scheduler-tick-seam — review 003 Finding 1, the cfg(test) transient-fault seam driving the real tick_with_retries through retry-then-success and exhaustion arms (the last uncovered wave-4 hardening arm, sqlite-commit-error-arm pattern). Task hygiene: pg-fix-open-path frontmatter fixed (unquoted 'max_size: 0' colon-space inside the name scalar broke YAML parsing — taskgraph validate had been red one-error since the wave-4 fix-batch decomposition; also restores review-wave-4-fixes' dependency resolution). implementation.md: wave-6 table row → decomposed, task-set listing with the parallelism note (rider + pre-task independent of the mem chain; rider closes before the gate), wave-6 decomposition bullet in the review-rounds record. Docs-only change; verified taskgraph validate (60 tasks, no cycles), cargo fmt --check, cross-reference read

2026-10-10 09:57:01 +00:00

2.6 KiB

Raw Blame History

id, name, status, depends_on, scope, risk, impact, level, tags

name

status

depends_on

scope

risk

impact

level

Description

Review 003 Finding 1 (MEDIUM — docs/reviews/003-wave-5-general-review.md): the pg scheduler's transient-error retry loop (alkstore-postgres/src/scheduler.rs:680-704, tick_with_retries) has never executed under any test — pg-fix-scheduler-resilience pinned its constants (the_tick_retry_policy_is_pinned) but the loop that uses them (retry, backoff sleeps, exhaustion error, the eprintln diagnostic — lcov 171-184, 237-260) is dead code under test, with no fault seam anywhere in the pg scheduler for tick_once. This is the last uncovered hardening arm from the wave-4 fix batch, and the exact class that produced review 002's live bugs.

Fix shape (the review's recommendation): a cfg(test) transient- fault seam on the pg tick path, following the sqlite-commit-error-arm precedent — a one-shot armed flag feeding a fabricated transient error into the real retry loop (never a replica):

(a) one retried tick then success: the seam faults once, the next tick_once succeeds, the job processes — the retry + backoff-sleep arms execute for real;
(b) exhaustion: the seam faults the max retry count, the loop exhausts to the typed error shape and the eprintln diagnostic fires, the runner survives (the resilience property the fix promised).

Cross-test safety is per-store scoping (the commit-fault seam's pattern); no production-path changes — the seam is cfg-gated, born disarmed, consumed-on-use.

Acceptance Criteria

The seam lands cfg(test)-gated; the real tick_with_retries executes under test (no loop replication)
Arms (a) and (b) pinned as above; the runner survives exhaustion
cargo test -p alkstore-postgres green against the harness server; skips clean server-less
Clippy -D warnings, fmt clean; workspace gates green

References

docs/reviews/003-wave-5-general-review.md §Finding 1 (sites, lcov inventory, recommendation)
docs/architecture/decisions/024-mem-engine.md (the wave-6 window this rider rides — the plan's wave-6 section, "Riding the window")
tasks/sqlite-commit-error-arm.md (the seam pattern of record)
tasks/pg-fix-scheduler-resilience.md (the constants the loop uses)

Notes

To be filled by implementation agent

Summary

To be filled on completion

2.6 KiB Raw Blame History