kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-02-04 18:21:52 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

[https://nvbugs/5754977][fix] Use free port for serve test (#10878)

Browse Source 
Signed-off-by: Junyi Xu <219237550+JunyiXu-nv@users.noreply.github.com>
Signed-off-by: Wangshanshan <30051912+dominicshanshan@users.noreply.github.com>
This commit is contained in:
JunyiXu-nv 2026-01-22 05:09:43 +08:00 committed by Yanchao Lu
parent 0306c0f12c
commit 2a5b8800e1
2 changed files with 9 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
tests/integration/defs/examples/serve/test_serve.py
View File

@ -7,13 +7,14 @@ import time
import pytest
import requests
import yaml
from defs.common import get_free_port_in_ci
from defs.conftest import llm_models_root, skip_no_hopper
from defs.trt_test_alternative import popen, print_error, print_info
from openai import OpenAI
from requests.exceptions import RequestException
def check_server_ready(http_port="8000", timeout_timer=600, sleep_interval=5):
def check_server_ready(http_port, timeout_timer=600, sleep_interval=5):
timer = 0
while timer <= timeout_timer:
try:
@ -49,7 +50,7 @@ def wait_for_log(log_queue, expected_log, timeout=10):
return False
def check_openai_chat_completion(http_port="8000",
def check_openai_chat_completion(http_port,
model_name="TinyLlama-1.1B-Chat-v1.0"):
"""
Test the launched trtllm-serve server using OpenAI client.
@ -127,6 +128,7 @@ def test_config_file_loading(serve_test_root, config_flag):
assert os.path.exists(
model_path), f"model_path does not exist: {model_path}"
port = get_free_port_in_ci()
cmd = [
"trtllm-serve",
"serve",
@ -134,7 +136,7 @@ def test_config_file_loading(serve_test_root, config_flag):
"--host",
"0.0.0.0",
"--port",
"8000",
str(port),
"--backend",
"pytorch",
config_flag,
@ -143,11 +145,11 @@ def test_config_file_loading(serve_test_root, config_flag):
print_info(f"Launching trtllm-serve with {config_flag}...")
with popen(cmd):
check_server_ready()
check_server_ready(http_port=port)
# Extract model name from the model path for consistency
model_name = model_path.split('/')[-1] # "Qwen3-30B-A3B-FP8"
# Test the server with OpenAI chat completion
check_openai_chat_completion(model_name=model_name)
check_openai_chat_completion(http_port=port, model_name=model_name)
@skip_no_hopper
@ -170,7 +172,7 @@ def test_env_overrides_pdl(tmp_path):
}
}))
port = 8001
port = get_free_port_in_ci()
env = os.environ.copy()
# Pre-load with 0 to verify override works
env.update({
@ -198,7 +200,7 @@ def test_env_overrides_pdl(tmp_path):
[output_queue.put(line) for line in iter(proc.stdout.readline, '')],
daemon=True).start()
check_server_ready(http_port=str(port), timeout_timer=300)
check_server_ready(http_port=port, timeout_timer=300)
response = OpenAI(base_url=f"http://localhost:{port}/v1",
api_key="tensorrt_llm").chat.completions.create(
model="TinyLlama-1.1B-Chat-v1.0",

1
tests/integration/test_lists/waives.txt
View File

@ -227,7 +227,6 @@ accuracy/test_llm_api_pytorch.py::TestLlama3_1_8B_Instruct_RocketKV::test_auto_d
unittest/_torch/sampler/test_return_logits.py SKIP (https://nvbugs/5764627)
unittest/_torch/sampler/test_logits_logprobs.py::test_generate_with_return_logits SKIP (https://nvbugs/5764627)
unittest/_torch/sampler/test_logits_logprobs.py::test_generate_async_with_return_logits SKIP (https://nvbugs/5764627)
examples/serve/test_serve.py::test_config_file_loading[--config] SKIP (https://nvbugs/5754977)
examples/test_ray.py::test_ray_disaggregated_serving[tp2] SKIP (https://nvbugs/5612502)
unittest/executor/test_rpc_proxy.py SKIP (https://nvbugs/5605741)
unittest/executor/test_rpc_worker.py SKIP (https://nvbugs/5605741)