P2.2+P2.3: NATS offload node-worker + router offload integration

Node Worker (services/node-worker/):
- NATS subscriber for node.{NODE_ID}.llm.request / vision.request
- Canonical JobRequest/JobResponse envelope (Pydantic)
- Idempotency cache (TTL 10min) with inflight dedup
- Deadline enforcement (DEADLINE_EXCEEDED on expired jobs)
- Concurrency limiter (semaphore, returns busy)
- Ollama + Swapper vision providers

Router offload (services/router/offload_client.py):
- NATS req/reply with configurable retries
- Circuit breaker per node+type (3 fails/60s → open 120s)
- Concurrency semaphore for remote requests

Model selection (services/router/model_select.py):
- exclude_nodes parameter for circuit-broken nodes
- force_local flag for fallback re-selection
- Integrated circuit breaker state awareness

Router /infer pipeline:
- Remote offload path when NCS selects remote node
- Automatic fallback: exclude failed node → force_local re-select
- Deadline propagation from router to node-worker

Tests: 17 unit tests (idempotency, deadline, circuit breaker)
Docs: ops/offload_routing.md (subjects, envelope, verification)
Made-with: Cursor
This commit is contained in:
Apple
2026-02-27 02:44:05 -08:00
parent a92c424845
commit c4b94a327d
19 changed files with 1075 additions and 6 deletions

View File

@@ -0,0 +1,53 @@
"""Node Worker — NATS offload executor for cross-node inference."""
import logging
import os
from fastapi import FastAPI
import config
import worker
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger("node-worker")
app = FastAPI(title="Node Worker", version="1.0.0")
_nats_client = None
@app.get("/healthz")
async def healthz():
connected = _nats_client is not None and _nats_client.is_connected if _nats_client else False
return {
"status": "ok" if connected else "degraded",
"node_id": config.NODE_ID,
"nats_connected": connected,
"max_concurrency": config.MAX_CONCURRENCY,
}
@app.on_event("startup")
async def startup():
global _nats_client
try:
import nats as nats_lib
_nats_client = await nats_lib.connect(config.NATS_URL)
logger.info(f"✅ NATS connected: {config.NATS_URL}")
await worker.start(_nats_client)
logger.info(f"✅ Node Worker ready: node={config.NODE_ID} concurrency={config.MAX_CONCURRENCY}")
except Exception as e:
logger.error(f"❌ Startup failed: {e}")
@app.on_event("shutdown")
async def shutdown():
if _nats_client:
try:
await _nats_client.close()
except Exception:
pass
if __name__ == "__main__":
import uvicorn
uvicorn.run(app, host="0.0.0.0", port=config.PORT)