phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-03 18:00:36 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

feat(tests): enable MCP tests in server mode (#4146)

Browse source 
We would like to run all OpenAI compatibility tests using only the
openai-client library. This is most friendly for contributors since they
can run tests without needing to update the client-sdks (which is
getting easier but still a long pole.)

This is the first step in enabling that -- no using "library client" for
any of the Responses tests. This seems like a reasonable trade-off since
the usage of an embeddeble library client for Responses (or any
OpenAI-compatible) behavior seems to be not very common. To do this, we
needed to enable MCP tests (which only worked in library client mode)
for server mode.
This commit is contained in:
Ashwin Bharambe 2025-11-13 07:23:23 -08:00 • committed by GitHub
parent 9eb81439d2
commit 1e81056a22
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
29 changed files with 13388 additions and 127 deletions
Download patch file Download diff file Expand all files Collapse all files

10
tests/common/mcp.py
View file

@ -244,8 +244,14 @@ def make_mcp_server(required_auth_token: str | None = None, tools: dict[str, Cal
timeout = 2
start_time = time.time()
server_url = f"http://localhost:{port}/sse"
logger.debug(f"Waiting for MCP server thread to start on port {port}")
# Determine the appropriate host for the server URL based on test environment
# - For library client and server mode: use localhost (both on same host)
# - For docker mode: use host.docker.internal (container needs to reach host)
import os
mcp_host = os.environ.get("LLAMA_STACK_TEST_MCP_HOST", "localhost")
server_url = f"http://{mcp_host}:{port}/sse"
logger.debug(f"Waiting for MCP server thread to start on port {port} (accessible via {mcp_host})")
while time.time() - start_time < timeout:
if server_thread.is_alive():