[https://nvbugs/5648685][fix] Fix openAI server waiting time to avoid large model weight loading out time (#9254)

Signed-off-by: Wangshanshan <30051912+dominicshanshan@users.noreply.github.com>
Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com>
Signed-off-by: Mike Iovine <miovine@nvidia.com>
Signed-off-by: Wangshanshan <30051912+dominicshanshan@users.noreply.github.com>

Author: dominicshanshan

tests/unittest/llmapi/apps/openai_server.py

@@ -16,7 +16,7 @@
 
 class RemoteOpenAIServer:
     DUMMY_API_KEY = "tensorrt_llm"
-    MAX_SERVER_START_WAIT_S = 600  # wait for server to start for 600 seconds
+    MAX_SERVER_START_WAIT_S = 7200  # wait for server to start for 7200 seconds (~ 2 hours) for LLM models weight loading
 
     def __init__(self,
                  model: str,