mirror of
https://github.com/meta-llama/llama-stack.git
synced 2025-06-27 18:50:41 +00:00
feat: add health to all providers through providers endpoint (#1418)
The `/v1/providers` now reports the health status of each provider when implemented. ``` curl -L http://127.0.0.1:8321/v1/providers|jq % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 4072 100 4072 0 0 246k 0 --:--:-- --:--:-- --:--:-- 248k { "data": [ { "api": "inference", "provider_id": "ollama", "provider_type": "remote::ollama", "config": { "url": "http://localhost:11434" }, "health": { "status": "OK" } }, { "api": "vector_io", "provider_id": "faiss", "provider_type": "inline::faiss", "config": { "kvstore": { "type": "sqlite", "namespace": null, "db_path": "/Users/leseb/.llama/distributions/ollama/faiss_store.db" } }, "health": { "status": "Not Implemented", "message": "Provider does not implement health check" } }, { "api": "safety", "provider_id": "llama-guard", "provider_type": "inline::llama-guard", "config": { "excluded_categories": [] }, "health": { "status": "Not Implemented", "message": "Provider does not implement health check" } }, { "api": "agents", "provider_id": "meta-reference", "provider_type": "inline::meta-reference", "config": { "persistence_store": { "type": "sqlite", "namespace": null, "db_path": "/Users/leseb/.llama/distributions/ollama/agents_store.db" } }, "health": { "status": "Not Implemented", "message": "Provider does not implement health check" } }, { "api": "telemetry", "provider_id": "meta-reference", "provider_type": "inline::meta-reference", "config": { "service_name": "llama-stack", "sinks": "console,sqlite", "sqlite_db_path": "/Users/leseb/.llama/distributions/ollama/trace_store.db" }, "health": { "status": "Not Implemented", "message": "Provider does not implement health check" } }, { "api": "eval", "provider_id": "meta-reference", "provider_type": "inline::meta-reference", "config": { "kvstore": { "type": "sqlite", "namespace": null, "db_path": "/Users/leseb/.llama/distributions/ollama/meta_reference_eval.db" } }, "health": { "status": "Not Implemented", "message": "Provider does not implement health check" } }, { "api": "datasetio", "provider_id": "huggingface", "provider_type": "remote::huggingface", "config": { "kvstore": { "type": "sqlite", "namespace": null, "db_path": "/Users/leseb/.llama/distributions/ollama/huggingface_datasetio.db" } }, "health": { "status": "Not Implemented", "message": "Provider does not implement health check" } }, { "api": "datasetio", "provider_id": "localfs", "provider_type": "inline::localfs", "config": { "kvstore": { "type": "sqlite", "namespace": null, "db_path": "/Users/leseb/.llama/distributions/ollama/localfs_datasetio.db" } }, "health": { "status": "Not Implemented", "message": "Provider does not implement health check" } }, { "api": "scoring", "provider_id": "basic", "provider_type": "inline::basic", "config": {}, "health": { "status": "Not Implemented", "message": "Provider does not implement health check" } }, { "api": "scoring", "provider_id": "llm-as-judge", "provider_type": "inline::llm-as-judge", "config": {}, "health": { "status": "Not Implemented", "message": "Provider does not implement health check" } }, { "api": "scoring", "provider_id": "braintrust", "provider_type": "inline::braintrust", "config": { "openai_api_key": "********" }, "health": { "status": "Not Implemented", "message": "Provider does not implement health check" } }, { "api": "tool_runtime", "provider_id": "brave-search", "provider_type": "remote::brave-search", "config": { "api_key": "********", "max_results": 3 }, "health": { "status": "Not Implemented", "message": "Provider does not implement health check" } }, { "api": "tool_runtime", "provider_id": "tavily-search", "provider_type": "remote::tavily-search", "config": { "api_key": "********", "max_results": 3 }, "health": { "status": "Not Implemented", "message": "Provider does not implement health check" } }, { "api": "tool_runtime", "provider_id": "code-interpreter", "provider_type": "inline::code-interpreter", "config": {}, "health": { "status": "Not Implemented", "message": "Provider does not implement health check" } }, { "api": "tool_runtime", "provider_id": "rag-runtime", "provider_type": "inline::rag-runtime", "config": {}, "health": { "status": "Not Implemented", "message": "Provider does not implement health check" } }, { "api": "tool_runtime", "provider_id": "model-context-protocol", "provider_type": "remote::model-context-protocol", "config": {}, "health": { "status": "Not Implemented", "message": "Provider does not implement health check" } }, { "api": "tool_runtime", "provider_id": "wolfram-alpha", "provider_type": "remote::wolfram-alpha", "config": { "api_key": "********" }, "health": { "status": "Not Implemented", "message": "Provider does not implement health check" } } ] } ``` Per providers too: ``` curl -L http://127.0.0.1:8321/v1/providers/ollama {"api":"inference","provider_id":"ollama","provider_type":"remote::ollama","config":{"url":"http://localhost:11434"},"health":{"status":"OK"}} ``` Signed-off-by: Sébastien Han <seb@redhat.com>
This commit is contained in:
parent
ff14773fa7
commit
69554158fa
15 changed files with 244 additions and 76 deletions
|
@ -4,6 +4,7 @@
|
|||
# This source code is licensed under the terms described in the LICENSE file in
|
||||
# the root directory of this source tree.
|
||||
|
||||
import asyncio
|
||||
import time
|
||||
from typing import Any, AsyncGenerator, AsyncIterator, Dict, List, Optional, Union
|
||||
|
||||
|
@ -60,7 +61,7 @@ from llama_stack.apis.vector_io import Chunk, QueryChunksResponse, VectorIO
|
|||
from llama_stack.log import get_logger
|
||||
from llama_stack.models.llama.llama3.chat_format import ChatFormat
|
||||
from llama_stack.models.llama.llama3.tokenizer import Tokenizer
|
||||
from llama_stack.providers.datatypes import RoutingTable
|
||||
from llama_stack.providers.datatypes import HealthResponse, HealthStatus, RoutingTable
|
||||
from llama_stack.providers.utils.telemetry.tracing import get_current_span
|
||||
|
||||
logger = get_logger(name=__name__, category="core")
|
||||
|
@ -580,6 +581,29 @@ class InferenceRouter(Inference):
|
|||
provider = self.routing_table.get_provider_impl(model_obj.identifier)
|
||||
return await provider.openai_chat_completion(**params)
|
||||
|
||||
async def health(self) -> Dict[str, HealthResponse]:
|
||||
health_statuses = {}
|
||||
timeout = 0.5
|
||||
for provider_id, impl in self.routing_table.impls_by_provider_id.items():
|
||||
try:
|
||||
# check if the provider has a health method
|
||||
if not hasattr(impl, "health"):
|
||||
continue
|
||||
health = await asyncio.wait_for(impl.health(), timeout=timeout)
|
||||
health_statuses[provider_id] = health
|
||||
except asyncio.TimeoutError:
|
||||
health_statuses[provider_id] = HealthResponse(
|
||||
status=HealthStatus.ERROR,
|
||||
message=f"Health check timed out after {timeout} seconds",
|
||||
)
|
||||
except NotImplementedError:
|
||||
health_statuses[provider_id] = HealthResponse(status=HealthStatus.NOT_IMPLEMENTED)
|
||||
except Exception as e:
|
||||
health_statuses[provider_id] = HealthResponse(
|
||||
status=HealthStatus.ERROR, message=f"Health check failed: {str(e)}"
|
||||
)
|
||||
return health_statuses
|
||||
|
||||
|
||||
class SafetyRouter(Safety):
|
||||
def __init__(
|
||||
|
|
Loading…
Add table
Add a link
Reference in a new issue