46 lines
2.7 KiB
Markdown
46 lines
2.7 KiB
Markdown
# Implementation Plan — Findings Ledger (Sprint 120)
|
||
|
||
## Phase 1 – Observability baselines (LEDGER-29-007)
|
||
- Instrument writer/projector with metrics listed in `observability.md` (`ledger_write_latency_seconds`, `ledger_events_total`, `ledger_projection_lag_seconds`, etc.).
|
||
- Emit structured logs (Serilog JSON) including chain/sequence/hash metadata.
|
||
- Wire OTLP exporters, ensure `/metrics` endpoint exposes histogram buckets with exemplars.
|
||
- Publish Grafana dashboards + alert rules (Policy SLO pack).
|
||
- Deliver doc updates + sample Grafana JSON in repo (`docs/observability/dashboards/findings-ledger/`).
|
||
|
||
## Phase 2 – Determinism harness (LEDGER-29-008)
|
||
- Finalize NDJSON fixtures for ≥5 M findings/tenant (per tenant/test scenario).
|
||
- Implement `tools/LedgerReplayHarness` CLI as specified in `replay-harness.md`.
|
||
- Add GitHub/Gitea pipeline job(s) running nightly (1 M) + weekly (5 M) harness plus DSSE signing.
|
||
- Capture CPU/memory/latency metrics and commit signed reports for validation.
|
||
- Provide runbook for QA + Ops to rerun harness in their environments.
|
||
|
||
## Phase 3 – Deployment & backup collateral (LEDGER-29-009)
|
||
- Integrate ledger service into Compose (`docker-compose.prod.yaml`) and Helm values.
|
||
- Automate PostgreSQL migrations (DatabaseMigrator invocation pre-start).
|
||
- Document backup cadence (pg_dump + WAL archiving) and projection rebuild process (call harness).
|
||
- Ensure Offline Kit packaging pulls binaries, migrations, harness, and default dashboards.
|
||
|
||
## Phase 4 – Provenance & air-gap extensions
|
||
- LEDGER-34-101: ingest orchestrator run export metadata, index by artifact hash, expose audit endpoint.
|
||
- LEDGER-AIRGAP-56/57/58: extend ledger events to capture bundle provenance, staleness metrics, timeline events.
|
||
- LEDGER-ATTEST-73-001: store attestation pointers (DSSE IDs, Rekor metadata) for explainability.
|
||
- For each extension, update schema doc + workflow inference doc to describe newly recorded fields and tenant-safe defaults.
|
||
|
||
## Dependencies & sequencing
|
||
1. AdvisoryAI Sprint 110.A completion (raw findings parity).
|
||
2. Observability schema approval (Nov 15) to unblock Phase 1 instrumentation.
|
||
3. QA lab capacity for 5 M replay (Nov 18 checkpoint).
|
||
4. DevOps review of Compose/Helm overlays (Nov 20).
|
||
5. Orchestrator export schema freeze (Nov 25) for provenance linkage.
|
||
|
||
## Deliverables checklist
|
||
- [ ] Metrics/logging/tracing implementation merged, dashboards exported.
|
||
- [ ] Harness CLI + fixtures + signed reports committed.
|
||
- [ ] Compose/Helm overlays + backup/restore runbooks validated.
|
||
- [ ] Air-gap provenance fields documented + implemented.
|
||
- [ ] Sprint tracker and release notes updated after each phase.
|
||
|
||
---
|
||
|
||
*Draft: 2025-11-13. Update when sequencing or dependencies change.*
|