4 Commits

Include renames

Author	SHA1	Message	Date
poimenandrock	4169effd8a	feat: complete observability stack (O1-O13) (#52) ... ## Complete Observability Stack (O1-O13) Implements all 13 observability issues in a single PR. 119 metrics total. ### Commits (one per issue) | Issue | Title | Metrics | |-------|-------|---------| | **O10** | Prometheus metrics module + /metrics endpoint | Foundation | | **O1** | Instrument ingest handler | I1-I12 (12) | | **O2** | Instrument query handler | Q1-Q12 (12) | | **O3** | Instrument context endpoint | C1-C8 (8) | | **O4** | Relevance judge | R1-R9 (9) | | **O5** | Write volume + storage metrics | W1-W12 (12) | | **O6** | Pod resource observability | P1-P13 | | **O7** | Availability + dependency health | A1-A10 (10) | | **O8** | Ingest rate pattern tracking | IR1-IR10 (10) | | **O9** | Postgres internal observability | PG1-PG33 | | **O11** | Grafana dashboard | 12 panels | | **O12** | Prometheus alerting rules | 11 alerts | | **O13** | Relevance evaluation CronJob | K8s manifest | ### Key Changes - **metrics.rs**: Zero-dependency Prometheus metrics (Counter, Gauge, Histogram, Timer) - **GET /metrics**: Prometheus text exposition format endpoint - **Ingest/Query/Context handlers**: Instrumented with latency, errors, auth failures - **Health check**: DB dependency check with latency tracking - **Background task**: Periodic DB stats collection (entity/edge counts, pool stats) - **Relevance judge**: Threshold-based eval with precision/recall/F1 tracking - **Grafana dashboard**: 12 panels covering all metric groups - **Alert rules**: 11 PrometheusRule alerts (availability, latency, errors, quality) - **CronJob**: Periodic relevance evaluation with sample queries ### Testing - 506 tests passing (0 failures) - All metrics modules have unit tests - Relevance judge: 4 tests ### Deploy ```bash # Grafana dashboard kubectl apply -f k8s/infra/grafana-dashboard.json # Prometheus alerts kubectl apply -f k8s/infra/prometheus-alerts.yaml # Relevance eval CronJob kubectl apply -f k8s/infra/relevance-eval-cronjob.yaml ``` Closes #27 #28 #29 #30 #31 #32 #33 #34 #35 #36 #37 #38 #39 --------- Co-authored-by: rock <[email protected]> Reviewed-on: #52 Co-authored-by: poimen <[email protected]>	2026-09-13 13:53:50 +00:00
rock	b43baf8147	feat: Configurable embeddings models via EMBEDDINGS_MODEL env var ... Allow customers to choose embedding model without schema changes. All models standardized to 768-dim (matching pgvector schema): - nomic-ai/nomic-embed-text-v2-moe (default, fast, multilingual) - nomic-ai/nomic-embed-text-v1.5 (slower but better quality) - all-MiniLM-L6-v2 (very fast, English-only) - BAAI/bge-small-en-v1.5 (fast retrieval) - BAAI/bge-base-en-v1.5 (best English quality) Changes: - EmbeddingsClient::from_env() reads EMBEDDINGS_MODEL env var - New validate_model() checks model is supported and 768-compatible - New model_name() getter for logging - Startup validation prevents unsupported models Configuration: EMBEDDINGS_MODEL=nomic-ai/nomic-embed-text-v1.5 LLM_API_BASE=https://api.riotpiao.com LLM_API_KEY=<optional> Documentation: - docs/EMBEDDINGS_MODELS.md (performance comparison, troubleshooting) - Kubernetes example for switching models - Migration guide for re-embedding existing chunks - Custom model integration instructions Performance impact: - Default (v2-moe): ~200 texts/sec - Fast (all-MiniLM): ~330 texts/sec - Quality (bge-base): ~165 texts/sec	2026-08-28 13:16:52 -07:00
Story Crater Bot	e83b8ef3da	feat: Implement M2.1 Embeddings client (768-dim batching @32) ... M2.1 Complete: TEI embeddings via api.riotpiao.com gateway Implementation (crates/mem-llm/src/embeddings.rs): - EmbeddingsClient::embed(texts) batches at ≤32 per request - Preserves input order across batch boundaries - Asserts 768-dim vectors, errors loudly with model name on mismatch - Sends apikey header (future-proofing for auth plugin enablement) - 30s timeout, retry on 5xx via reqwest Client - Constants: EMBEDDINGS_DIM=768, BATCH_SIZE=32 (single source for schema migration) Tests (tests/it_embeddings.rs): 8 tests 1. a1_batches_at_32 — 100 inputs → 4 requests (32+32+32+4) 2. a2_order_preserved — identifiable vectors, cross-batch order assertion 3. a3_dimension_asserted — 512-dim response → error naming model & dimensions 4. a4_apikey_sent — header present even when route doesn't require auth 5. a5_live_dims — #[ignore] live gateway test (768-dim confirmation) 6. test_empty_input — empty batch → empty output 7. test_batch_boundary_32 — exact 32 inputs = 1 batch 8. test_batch_boundary_33 — 33 inputs = 2 batches (32+1) All tests pass locally. Builds cleanly: Updated INDEX.md: - Added M2.x row to progress table (6/8 ✅, 2 ⬜) - Updated total: 73 tasks, 48✅ + 2🟡 + 23⬜ (was 65 tasks) - Updated gate count: 6/11 green (was 5/10) - Test count: 247 passing, 2 ignored (was 239) Blocks: M1.1 ✅ (already complete, unblocked)	2026-08-27 20:36:57 -07:00
rock	af6f22217d	feat(core): implement full memory pipeline (#11)	2026-08-24 01:37:16 +00:00