kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-01-14 06:27:45 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

[nvbug/5347489][nvbug/5388036] increase timeout in disagg worker test (#6041)

Browse Source 
Signed-off-by: zhengd-nv <200704041+zhengd-nv@users.noreply.github.com>
This commit is contained in:
Zheng Duan 2025-07-16 13:52:13 +08:00 committed by GitHub
parent 509dc7c831
commit 385af53a4d
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 15 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
tests/integration/defs/disaggregated/test_workers.py
View File

@ -64,21 +64,26 @@ def run_disaggregated_workers(
return workers_proc, ctx_servers, gen_servers
DEFAULT_TIMEOUT_SERVER_START = 900
DEFAULT_TIMEOUT_REQUEST = 180
class BasicWorkerTester:
def __init__(self,
ctx_servers: List[str],
gen_servers: List[str],
req_timeout_secs: int = 180,
server_start_timeout_secs: int = 180):
req_timeout_secs: int = DEFAULT_TIMEOUT_REQUEST,
server_start_timeout_secs: int = DEFAULT_TIMEOUT_SERVER_START):
self.ctx_servers = ctx_servers
self.gen_servers = gen_servers
self.req_timeout_secs = req_timeout_secs
self.server_start_timeout_secs = server_start_timeout_secs
async def new_session(self):
session = aiohttp.ClientSession(timeout=aiohttp.ClientTimeout(
total=self.req_timeout_secs))
session = aiohttp.ClientSession(
connector=aiohttp.TCPConnector(force_close=True),
timeout=aiohttp.ClientTimeout(total=self.req_timeout_secs))
await OpenAIDisaggServer.wait_for_all_servers_ready(
session, self.ctx_servers, self.gen_servers,
self.server_start_timeout_secs)
@ -146,8 +151,8 @@ class ConditionalWorkerTester(BasicWorkerTester):
def __init__(self,
ctx_servers: List[str],
gen_servers: List[str],
req_timeout_secs: int = 180,
server_start_timeout_secs: int = 180,
req_timeout_secs: int = DEFAULT_TIMEOUT_REQUEST,
server_start_timeout_secs: int = DEFAULT_TIMEOUT_SERVER_START,
model_name: str = "TinyLlama/TinyLlama-1.1B-Chat-v1.0"):
super().__init__(ctx_servers, gen_servers, req_timeout_secs,
server_start_timeout_secs)
@ -199,8 +204,8 @@ class KvCacheEventWorkerTester(BasicWorkerTester):
def __init__(self,
ctx_servers: List[str],
gen_servers: List[str],
req_timeout_secs: int = 180,
server_start_timeout_secs: int = 240,
req_timeout_secs: int = DEFAULT_TIMEOUT_REQUEST,
server_start_timeout_secs: int = DEFAULT_TIMEOUT_SERVER_START,
model_name: str = "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
model_path: Optional[str] = None):
super().__init__(ctx_servers, gen_servers, req_timeout_secs,
@ -316,8 +321,8 @@ class KvCacheAwareRouterTester(BasicWorkerTester):
def __init__(self,
ctx_servers: List[str],
gen_servers: List[str],
req_timeout_secs: int = 180,
server_start_timeout_secs: int = 180,
req_timeout_secs: int = DEFAULT_TIMEOUT_REQUEST,
server_start_timeout_secs: int = DEFAULT_TIMEOUT_SERVER_START,
model_name: str = "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
tokens_per_block: int = 32):
super().__init__(ctx_servers, gen_servers, req_timeout_secs,