Keysat
8d839e3714
- Add redaction gateway (redaction_gateway.py, redaction/ scrub + tests) - Add embeddings proxy and spark_embed service (Dockerfile + main.py) - Expand audio_proxy with speaker-aware handling; deep_health/health/server updates - Package: configureSparks action + sparkConfig model updates, manifest/main wiring - Docs: AUDIO_API, EMBEDDINGS, REDACTION_GATEWAY; HANDOFF and runbook/known-issues refresh
107 lines
4.1 KiB
Python
107 lines
4.1 KiB
Python
from __future__ import annotations
|
|
import httpx
|
|
from .config import Settings
|
|
|
|
|
|
_TIMEOUT = 3.0
|
|
|
|
|
|
async def check_vllm(settings: Settings) -> dict:
|
|
base_url = (
|
|
f"http://{settings.spark1_host}:{settings.vllm_port}/v1"
|
|
if settings.spark1_host
|
|
else None
|
|
)
|
|
if not settings.spark1_host:
|
|
return {"ok": False, "error": "spark1 not configured", "base_url": base_url}
|
|
try:
|
|
async with httpx.AsyncClient(timeout=_TIMEOUT) as c:
|
|
r = await c.get(f"http://{settings.spark1_host}:{settings.vllm_port}/v1/models")
|
|
r.raise_for_status()
|
|
ids = [m["id"] for m in r.json().get("data", [])]
|
|
return {
|
|
"ok": True,
|
|
"current_model": ids[0] if ids else None,
|
|
"all": ids,
|
|
"base_url": base_url,
|
|
}
|
|
except Exception as e:
|
|
return {"ok": False, "error": str(e), "base_url": base_url}
|
|
|
|
|
|
async def check_parakeet(settings: Settings) -> dict:
|
|
base_url = (
|
|
f"http://{settings.parakeet_host}:{settings.parakeet_port}"
|
|
if settings.parakeet_host
|
|
else None
|
|
)
|
|
if not settings.parakeet_host:
|
|
return {"ok": False, "error": "parakeet host not configured", "base_url": base_url}
|
|
try:
|
|
async with httpx.AsyncClient(timeout=_TIMEOUT) as c:
|
|
r = await c.get(f"http://{settings.parakeet_host}:{settings.parakeet_port}/health")
|
|
r.raise_for_status()
|
|
return {"ok": True, "detail": r.json(), "base_url": base_url}
|
|
except Exception as e:
|
|
return {"ok": False, "error": str(e), "base_url": base_url}
|
|
|
|
|
|
async def check_kokoro(settings: Settings) -> dict:
|
|
base_url = (
|
|
f"http://{settings.kokoro_host}:{settings.kokoro_port}"
|
|
if settings.kokoro_host
|
|
else None
|
|
)
|
|
if not settings.kokoro_host:
|
|
return {"ok": False, "error": "kokoro host not configured", "base_url": base_url}
|
|
try:
|
|
async with httpx.AsyncClient(timeout=_TIMEOUT) as c:
|
|
r = await c.get(f"http://{settings.kokoro_host}:{settings.kokoro_port}/health")
|
|
r.raise_for_status()
|
|
return {
|
|
"ok": True,
|
|
"detail": r.json() if r.headers.get("content-type", "").startswith("application/json") else r.text,
|
|
"base_url": base_url,
|
|
}
|
|
except Exception as e:
|
|
return {"ok": False, "error": str(e), "base_url": base_url}
|
|
|
|
|
|
async def check_embeddings(settings: Settings) -> dict:
|
|
base_url = (
|
|
f"http://{settings.embed_host}:{settings.embed_port}"
|
|
if settings.embed_host
|
|
else None
|
|
)
|
|
if not settings.embed_host:
|
|
return {"ok": False, "error": "embedding host not configured", "base_url": base_url}
|
|
try:
|
|
async with httpx.AsyncClient(timeout=_TIMEOUT) as c:
|
|
r = await c.get(f"{base_url}/health")
|
|
r.raise_for_status()
|
|
detail = r.json() if r.headers.get("content-type", "").startswith("application/json") else r.text
|
|
# spark-embed reports {"status":"ready"|"loading", ...} — only "ready" is healthy.
|
|
ready = isinstance(detail, dict) and detail.get("status") == "ready"
|
|
return {"ok": ready, "detail": detail, "base_url": base_url,
|
|
"model": detail.get("dense_model") if isinstance(detail, dict) else None}
|
|
except Exception as e:
|
|
return {"ok": False, "error": str(e), "base_url": base_url}
|
|
|
|
|
|
async def check_qdrant(settings: Settings) -> dict:
|
|
base_url = (
|
|
f"http://{settings.qdrant_host}:{settings.qdrant_port}"
|
|
if settings.qdrant_host
|
|
else None
|
|
)
|
|
if not settings.qdrant_host:
|
|
return {"ok": False, "error": "qdrant host not configured", "base_url": base_url}
|
|
try:
|
|
async with httpx.AsyncClient(timeout=_TIMEOUT) as c:
|
|
# /readyz returns 200 "all shards are ready" when serving.
|
|
r = await c.get(f"{base_url}/readyz")
|
|
r.raise_for_status()
|
|
return {"ok": True, "detail": r.text.strip()[:120], "base_url": base_url}
|
|
except Exception as e:
|
|
return {"ok": False, "error": str(e), "base_url": base_url}
|