From 21f7d56a770e85b40a0b0d8632de88a6bd44e7c0 Mon Sep 17 00:00:00 2001
From: Junyi Xu <219237550+JunyiXu-nv@users.noreply.github.com>
Date: Wed, 3 Dec 2025 03:16:06 +0000
Subject: [PATCH 1/6] [None][fix] Preserving ip:port for trtllm-serve before
 initializing llm

Signed-off-by: Junyi Xu <219237550+JunyiXu-nv@users.noreply.github.com>
---
 tensorrt_llm/commands/serve.py      | 66 +++++++++++++++--------------
 tensorrt_llm/serve/openai_server.py |  5 ++-
 2 files changed, 38 insertions(+), 33 deletions(-)

diff --git a/tensorrt_llm/commands/serve.py b/tensorrt_llm/commands/serve.py
index affa88dc4e26..0eb0d1901a42 100644
--- a/tensorrt_llm/commands/serve.py
+++ b/tensorrt_llm/commands/serve.py
@@ -3,6 +3,7 @@
 import json
 import os
 import signal  # Added import
+import socket
 import subprocess  # nosec B404
 import sys
 from pathlib import Path
@@ -176,37 +177,40 @@ def launch_server(
 
     backend = llm_args["backend"]
     model = llm_args["model"]
-    if backend == 'pytorch':
-        llm_args.pop("build_config", None)
-        llm = PyTorchLLM(**llm_args)
-    elif backend == '_autodeploy':
-        from tensorrt_llm._torch.auto_deploy import LLM as AutoDeployLLM
-
-        # AutoDeploy does not support build_config
-        llm_args.pop("build_config", None)
-        llm = AutoDeployLLM(**llm_args)
-    elif backend == 'tensorrt' or backend == 'trt':
-        llm_args.pop("backend")
-        llm = LLM(**llm_args)
-    else:
-        raise click.BadParameter(
-            f"{backend} is not a known backend, check help for available options.",
-            param_hint="backend")
-
-    server = OpenAIServer(llm=llm,
-                          model=model,
-                          tool_parser=tool_parser,
-                          server_role=server_role,
-                          metadata_server_cfg=metadata_server_cfg,
-                          disagg_cluster_config=disagg_cluster_config,
-                          multimodal_server_config=multimodal_server_config,
-                          chat_template=chat_template)
-
-    # Optionally disable GC (default: not disabled)
-    if os.getenv("TRTLLM_SERVER_DISABLE_GC", "0") == "1":
-        gc.disable()
-
-    asyncio.run(server(host, port))
+    with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:
+        s.bind((host, port))
+
+        if backend == 'pytorch':
+            llm_args.pop("build_config", None)
+            llm = PyTorchLLM(**llm_args)
+        elif backend == '_autodeploy':
+            from tensorrt_llm._torch.auto_deploy import LLM as AutoDeployLLM
+
+            # AutoDeploy does not support build_config
+            llm_args.pop("build_config", None)
+            llm = AutoDeployLLM(**llm_args)
+        elif backend == 'tensorrt' or backend == 'trt':
+            llm_args.pop("backend")
+            llm = LLM(**llm_args)
+        else:
+            raise click.BadParameter(
+                f"{backend} is not a known backend, check help for available options.",
+                param_hint="backend")
+
+        server = OpenAIServer(llm=llm,
+                              model=model,
+                              tool_parser=tool_parser,
+                              server_role=server_role,
+                              metadata_server_cfg=metadata_server_cfg,
+                              disagg_cluster_config=disagg_cluster_config,
+                              multimodal_server_config=multimodal_server_config,
+                              chat_template=chat_template)
+
+        # Optionally disable GC (default: not disabled)
+        if os.getenv("TRTLLM_SERVER_DISABLE_GC", "0") == "1":
+            gc.disable()
+
+        asyncio.run(server(host, port, sockets=[s]))
 
 
 def launch_mm_encoder_server(
diff --git a/tensorrt_llm/serve/openai_server.py b/tensorrt_llm/serve/openai_server.py
index dcaa23494980..5886a7674c4f 100644
--- a/tensorrt_llm/serve/openai_server.py
+++ b/tensorrt_llm/serve/openai_server.py
@@ -3,6 +3,7 @@
 import os
 import re
 import signal
+import socket
 import traceback
 from collections import deque
 from contextlib import asynccontextmanager
@@ -990,7 +991,7 @@ async def create_stream_response(generator, request: ResponsesRequest, sampling_
         return JSONResponse(content={"detail": "None"})
 
 
-    async def __call__(self, host, port):
+    async def __call__(self, host, port, sockets: list[socket.socket] = None):
         # Store the binding address for server registration
         self.binding_addr = f"http://{host}:{port}"
         self.host = host
@@ -1000,4 +1001,4 @@ async def __call__(self, host, port):
                                 port=port,
                                 log_level="info",
                                 timeout_keep_alive=TIMEOUT_KEEP_ALIVE)
-        await uvicorn.Server(config).serve()
+        await uvicorn.Server(config).serve(sockets=sockets)

From d21b90ce916d1c59c6dae37058454d6e4f230f95 Mon Sep 17 00:00:00 2001
From: Junyi Xu <219237550+JunyiXu-nv@users.noreply.github.com>
Date: Fri, 5 Dec 2025 06:29:14 +0000
Subject: [PATCH 2/6] Update error message

Signed-off-by: Junyi Xu <219237550+JunyiXu-nv@users.noreply.github.com>
---
 tensorrt_llm/commands/serve.py | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/tensorrt_llm/commands/serve.py b/tensorrt_llm/commands/serve.py
index 0eb0d1901a42..716e27bda4fc 100644
--- a/tensorrt_llm/commands/serve.py
+++ b/tensorrt_llm/commands/serve.py
@@ -178,7 +178,10 @@ def launch_server(
     backend = llm_args["backend"]
     model = llm_args["model"]
     with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:
-        s.bind((host, port))
+        try:
+            s.bind((host, port))
+        except OSError as e:
+            raise RuntimeError(f"Failed to bind socket to {host}:{port}: {e}")
 
         if backend == 'pytorch':
             llm_args.pop("build_config", None)

From 319e1ef1f4c70dd135b891b4e2e2fab62a014579 Mon Sep 17 00:00:00 2001
From: Junyi Xu <219237550+JunyiXu-nv@users.noreply.github.com>
Date: Fri, 5 Dec 2025 07:25:45 +0000
Subject: [PATCH 3/6] Fix new tests

Signed-off-by: Junyi Xu <219237550+JunyiXu-nv@users.noreply.github.com>
---
 tensorrt_llm/serve/openai_server.py           |  2 +-
 .../accuracy/test_disaggregated_serving.py    | 19 +------------------
 tests/integration/defs/common.py              | 19 +++++++++++++++++++
 .../defs/disaggregated/test_disaggregated.py  |  6 ++++--
 4 files changed, 25 insertions(+), 21 deletions(-)

diff --git a/tensorrt_llm/serve/openai_server.py b/tensorrt_llm/serve/openai_server.py
index 5886a7674c4f..e64c5d20df69 100644
--- a/tensorrt_llm/serve/openai_server.py
+++ b/tensorrt_llm/serve/openai_server.py
@@ -991,7 +991,7 @@ async def create_stream_response(generator, request: ResponsesRequest, sampling_
         return JSONResponse(content={"detail": "None"})
 
 
-    async def __call__(self, host, port, sockets: list[socket.socket] = None):
+    async def __call__(self, host, port, sockets: list[socket.socket] | None = None):
         # Store the binding address for server registration
         self.binding_addr = f"http://{host}:{port}"
         self.host = host
diff --git a/tests/integration/defs/accuracy/test_disaggregated_serving.py b/tests/integration/defs/accuracy/test_disaggregated_serving.py
index e00e8850ee0f..3d7ed84dfdfb 100644
--- a/tests/integration/defs/accuracy/test_disaggregated_serving.py
+++ b/tests/integration/defs/accuracy/test_disaggregated_serving.py
@@ -13,8 +13,8 @@
 import pytest
 import requests
 import yaml
+from defs.common import revise_disaggregated_server_config_urls_with_free_ports
 
-from tensorrt_llm._utils import get_free_port
 from tensorrt_llm.executor.result import GenerationResultBase
 from tensorrt_llm.llmapi import CompletionOutput, RequestOutput, SamplingParams
 from tensorrt_llm.llmapi.llm_args import LlmArgs
@@ -68,23 +68,6 @@ def __exit__(self, exc_type, exc_val, exc_tb):
         return False
 
 
-def revise_disaggregated_server_config_urls_with_free_ports(
-        disaggregated_server_config: Dict[str, Any]) -> Dict[str, Any]:
-    num_ctx_ports = len(disaggregated_server_config["context_servers"]["urls"])
-    num_gen_ports = len(
-        disaggregated_server_config["generation_servers"]["urls"])
-
-    disaggregated_server_config['port'] = get_free_port()
-    disaggregated_server_config["context_servers"]["urls"] = [
-        f"localhost:{get_free_port()}" for _ in range(num_ctx_ports)
-    ]
-    disaggregated_server_config["generation_servers"]["urls"] = [
-        f"localhost:{get_free_port()}" for _ in range(num_gen_ports)
-    ]
-
-    return disaggregated_server_config
-
-
 @contextlib.contextmanager
 def launch_disaggregated_llm(
         disaggregated_server_config: Dict[str, Any],
diff --git a/tests/integration/defs/common.py b/tests/integration/defs/common.py
index 9b78e1bdee57..4986f7617cfb 100644
--- a/tests/integration/defs/common.py
+++ b/tests/integration/defs/common.py
@@ -20,10 +20,12 @@
 import time
 from difflib import SequenceMatcher
 from pathlib import Path
+from typing import Any
 
 from packaging import version
 
 from tensorrt_llm import LLM as LLM_torch
+from tensorrt_llm._utils import get_free_port
 from tensorrt_llm.executor.request import LoRARequest
 from tensorrt_llm.lora_manager import LoraConfig
 from tensorrt_llm.sampling_params import SamplingParams
@@ -1147,3 +1149,20 @@ def wait_for_server(host, port, timeout_seconds=180):
         except (socket.error, ConnectionRefusedError, OSError):
             time.sleep(2)
     return False
+
+
+def revise_disaggregated_server_config_urls_with_free_ports(
+        disaggregated_server_config: dict[str, Any]) -> dict[str, Any]:
+    num_ctx_ports = len(disaggregated_server_config["context_servers"]["urls"])
+    num_gen_ports = len(
+        disaggregated_server_config["generation_servers"]["urls"])
+
+    disaggregated_server_config['port'] = get_free_port()
+    disaggregated_server_config["context_servers"]["urls"] = [
+        f"localhost:{get_free_port()}" for _ in range(num_ctx_ports)
+    ]
+    disaggregated_server_config["generation_servers"]["urls"] = [
+        f"localhost:{get_free_port()}" for _ in range(num_gen_ports)
+    ]
+
+    return disaggregated_server_config
diff --git a/tests/integration/defs/disaggregated/test_disaggregated.py b/tests/integration/defs/disaggregated/test_disaggregated.py
index 3be534531ab3..d9d90c6383ae 100644
--- a/tests/integration/defs/disaggregated/test_disaggregated.py
+++ b/tests/integration/defs/disaggregated/test_disaggregated.py
@@ -22,7 +22,8 @@
 
 import pytest
 import yaml
-from defs.common import wait_for_server
+from defs.common import (
+    revise_disaggregated_server_config_urls_with_free_ports, wait_for_server)
 from defs.conftest import (get_sm_version, llm_models_root, skip_arm,
                            skip_no_hopper)
 from defs.trt_test_alternative import check_call, check_output, popen
@@ -277,7 +278,8 @@ def get_test_config(test_desc, example_dir, test_root):
         raise ValueError(f"Invalid test description: {test_desc}, "
                          f"valid descriptions are: {config_map.keys()}")
 
-    return config_map[test_desc]
+    return revise_disaggregated_server_config_urls_with_free_ports(
+        config_map[test_desc])
 
 
 def get_extra_llm_config(config, suffix, cwd):

From 9964511fae6f8498cd23497c7281e52e0649c61d Mon Sep 17 00:00:00 2001
From: Junyi Xu <219237550+JunyiXu-nv@users.noreply.github.com>
Date: Fri, 5 Dec 2025 10:20:57 +0000
Subject: [PATCH 4/6] update

Signed-off-by: Junyi Xu <219237550+JunyiXu-nv@users.noreply.github.com>
---
 .../defs/disaggregated/test_disaggregated.py        | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/tests/integration/defs/disaggregated/test_disaggregated.py b/tests/integration/defs/disaggregated/test_disaggregated.py
index d9d90c6383ae..2be9a6cf998f 100644
--- a/tests/integration/defs/disaggregated/test_disaggregated.py
+++ b/tests/integration/defs/disaggregated/test_disaggregated.py
@@ -278,8 +278,17 @@ def get_test_config(test_desc, example_dir, test_root):
         raise ValueError(f"Invalid test description: {test_desc}, "
                          f"valid descriptions are: {config_map.keys()}")
 
-    return revise_disaggregated_server_config_urls_with_free_ports(
-        config_map[test_desc])
+    # Revise the config file to use free ports
+    new_config = None
+    with open(config_map[test_desc][1], 'r') as f:
+        config = yaml.safe_load(f)
+        new_config = revise_disaggregated_server_config_urls_with_free_ports(
+            config)
+
+    with open(config_map[test_desc][1], 'w') as f:
+        yaml.dump(new_config, f)
+
+    return config_map[test_desc]
 
 
 def get_extra_llm_config(config, suffix, cwd):

From b5859a71b5d09937b98045690bfe36b10e5f9dff Mon Sep 17 00:00:00 2001
From: Junyi Xu <219237550+JunyiXu-nv@users.noreply.github.com>
Date: Fri, 5 Dec 2025 15:45:26 +0000
Subject: [PATCH 5/6] Fix more tests

Signed-off-by: Junyi Xu <219237550+JunyiXu-nv@users.noreply.github.com>
---
 tests/integration/defs/common.py              | 26 ++++++++-----
 .../defs/disaggregated/test_disaggregated.py  | 37 +++++++++++--------
 2 files changed, 37 insertions(+), 26 deletions(-)

diff --git a/tests/integration/defs/common.py b/tests/integration/defs/common.py
index 4986f7617cfb..243801a8c9e9 100644
--- a/tests/integration/defs/common.py
+++ b/tests/integration/defs/common.py
@@ -1153,16 +1153,22 @@ def wait_for_server(host, port, timeout_seconds=180):
 
 def revise_disaggregated_server_config_urls_with_free_ports(
         disaggregated_server_config: dict[str, Any]) -> dict[str, Any]:
-    num_ctx_ports = len(disaggregated_server_config["context_servers"]["urls"])
-    num_gen_ports = len(
-        disaggregated_server_config["generation_servers"]["urls"])
-
+    # Revise serve port
     disaggregated_server_config['port'] = get_free_port()
-    disaggregated_server_config["context_servers"]["urls"] = [
-        f"localhost:{get_free_port()}" for _ in range(num_ctx_ports)
-    ]
-    disaggregated_server_config["generation_servers"]["urls"] = [
-        f"localhost:{get_free_port()}" for _ in range(num_gen_ports)
-    ]
+
+    # Revise context and generation server urls
+    ctx_urls = disaggregated_server_config["context_servers"]["urls"]
+    gen_urls = disaggregated_server_config["generation_servers"]["urls"]
+    url_map = dict()
+    for url in set(ctx_urls + gen_urls):
+        url_map[url] = (url.split(':')[0], get_free_port())
+
+    for i, url in enumerate(ctx_urls):
+        disaggregated_server_config["context_servers"]["urls"][
+            i] = f"{url_map[url][0]}:{url_map[url][1]}"
+
+    for i, url in enumerate(gen_urls):
+        disaggregated_server_config["generation_servers"]["urls"][
+            i] = f"{url_map[url][0]}:{url_map[url][1]}"
 
     return disaggregated_server_config
diff --git a/tests/integration/defs/disaggregated/test_disaggregated.py b/tests/integration/defs/disaggregated/test_disaggregated.py
index 2be9a6cf998f..07a5adc9874b 100644
--- a/tests/integration/defs/disaggregated/test_disaggregated.py
+++ b/tests/integration/defs/disaggregated/test_disaggregated.py
@@ -28,7 +28,7 @@
                            skip_no_hopper)
 from defs.trt_test_alternative import check_call, check_output, popen
 
-from tensorrt_llm._utils import mpi_disabled
+from tensorrt_llm._utils import get_free_port, mpi_disabled
 from tensorrt_llm.logger import logger
 
 
@@ -144,12 +144,12 @@ def validate_timing_metrics(perf_metrics_item, request_context=""):
     return True
 
 
-def get_disagg_server_url_from_cfg(config_file: str) -> str:
+def get_disagg_server_url_from_cfg(config_file: str) -> tuple[str, int]:
     with open(config_file, 'r') as file:
         config = yaml.safe_load(file)
     server_host = config.get('hostname', 'localhost')
     server_port = config.get('port', 8000)
-    return f"http://{server_host}:{server_port}"
+    return server_host, server_port
 
 
 def get_test_config(test_desc, example_dir, test_root):
@@ -285,10 +285,11 @@ def get_test_config(test_desc, example_dir, test_root):
         new_config = revise_disaggregated_server_config_urls_with_free_ports(
             config)
 
-    with open(config_map[test_desc][1], 'w') as f:
+    temp_fd, new_config_file = tempfile.mkstemp(suffix=f'_{test_desc}.yaml')
+    with os.fdopen(temp_fd, 'w') as f:
         yaml.dump(new_config, f)
 
-    return config_map[test_desc]
+    return (config_map[test_desc][0], new_config_file)
 
 
 def get_extra_llm_config(config, suffix, cwd):
@@ -492,7 +493,8 @@ def run_disaggregated_test(example_dir,
         'trtllm-serve', 'disaggregated', '--server_start_timeout',
         str(server_start_timeout), '-c', config_file
     ]
-    server_url = get_disagg_server_url_from_cfg(config_file)
+    server_host, server_port = get_disagg_server_url_from_cfg(config_file)
+    server_url = f"http://{server_host}:{server_port}"
 
     try:
         if not use_ray:
@@ -549,8 +551,8 @@ def run_disaggregated_test(example_dir,
                           env=run_env,
                           cwd=cwd))
 
-                if not wait_for_server("localhost",
-                                       8000,
+                if not wait_for_server(server_host,
+                                       server_port,
                                        timeout_seconds=server_start_timeout):
                     raise RuntimeError(
                         f"Disaggregated server failed to start within {server_start_timeout} seconds"
@@ -1580,6 +1582,7 @@ def run_disaggregated_benchmark(example_dir,
         'trtllm-serve', 'disaggregated', '--server_start_timeout',
         str(server_start_timeout), '-c', config_file
     ]
+    server_host, server_port = get_disagg_server_url_from_cfg(config_file)
     try:
         with (  # Start workers
                 open('output_workers.log', 'w') as output_workers,
@@ -1633,9 +1636,9 @@ def run_disaggregated_benchmark(example_dir,
                 '--max-concurrency',
                 str(max_concurrency),
                 '--host',
-                'localhost',
+                server_host,
                 '--port',
-                '8000',
+                str(server_port),
                 '--ignore-eos',
                 '--no-test-input',
                 '--percentile-metrics',
@@ -1677,7 +1680,7 @@ def get_config_for_benchmark(model_root, backend):
     serve_config = {
         "model": model_root,
         "hostname": "localhost",
-        "port": 8000,
+        "port": get_free_port(),
         "backend": "pytorch",
         "context_servers": {
             "num_instances": 1,
@@ -1691,7 +1694,7 @@ def get_config_for_benchmark(model_root, backend):
                 "backend": backend,
                 "max_tokens_in_buffer": 512,
             },
-            "urls": ["localhost:8001"]
+            "urls": [f"localhost:{get_free_port()}"]
         },
         "generation_servers": {
             "num_instances": 1,
@@ -1704,7 +1707,7 @@ def get_config_for_benchmark(model_root, backend):
                 "backend": backend,
                 "max_tokens_in_buffer": 512,
             },
-            "urls": ["localhost:8002"]
+            "urls": [f"localhost:{get_free_port()}"]
         }
     }
     return serve_config
@@ -1735,6 +1738,7 @@ def run_disaggregated_genai_perf(config_file,
     ]
 
     artifact_dir = os.path.join(cwd or ".", "benchmark-results")
+    server_host, server_port = get_disagg_server_url_from_cfg(config_file)
 
     try:
         with (open('output_workers.log', 'w') as output_workers,
@@ -1751,8 +1755,9 @@ def run_disaggregated_genai_perf(config_file,
                     cwd=cwd) as server_proc):
 
             # Wait for server to be ready
-            if not wait_for_server(
-                    "localhost", 8000, timeout_seconds=server_start_timeout):
+            if not wait_for_server(server_host,
+                                   server_port,
+                                   timeout_seconds=server_start_timeout):
                 raise RuntimeError(
                     f"Disaggregated server did not become ready within {server_start_timeout} seconds"
                 )
@@ -1762,7 +1767,7 @@ def run_disaggregated_genai_perf(config_file,
                 'genai-perf', 'profile', '--model', model_path, '--tokenizer',
                 model_path, '--endpoint-type', 'chat', '--endpoint',
                 '/v1/chat/completions', '--streaming', '--url',
-                'localhost:8000', '--synthetic-input-tokens-mean',
+                f'{server_host}:{server_port}', '--synthetic-input-tokens-mean',
                 str(input_tokens), '--synthetic-input-tokens-stddev', '0',
                 '--output-tokens-mean',
                 str(output_tokens), '--output-tokens-stddev', '0',

From 40f33d752fe9a5178ed61fc0d703628d98ff555b Mon Sep 17 00:00:00 2001
From: Junyi Xu <219237550+JunyiXu-nv@users.noreply.github.com>
Date: Sat, 6 Dec 2025 04:20:08 +0000
Subject: [PATCH 6/6] Fix more tests

Signed-off-by: Junyi Xu <219237550+JunyiXu-nv@users.noreply.github.com>
---
 tests/integration/defs/common.py               | 17 +++++++++++++++++
 .../defs/disaggregated/test_disaggregated.py   | 18 ++++--------------
 .../defs/disaggregated/test_workers.py         |  2 ++
 3 files changed, 23 insertions(+), 14 deletions(-)

diff --git a/tests/integration/defs/common.py b/tests/integration/defs/common.py
index 243801a8c9e9..5ff8799fb3b4 100644
--- a/tests/integration/defs/common.py
+++ b/tests/integration/defs/common.py
@@ -17,11 +17,13 @@
 import platform
 import re
 import socket
+import tempfile
 import time
 from difflib import SequenceMatcher
 from pathlib import Path
 from typing import Any
 
+import yaml
 from packaging import version
 
 from tensorrt_llm import LLM as LLM_torch
@@ -1172,3 +1174,18 @@ def revise_disaggregated_server_config_urls_with_free_ports(
             i] = f"{url_map[url][0]}:{url_map[url][1]}"
 
     return disaggregated_server_config
+
+
+def revise_disagg_config_file_with_free_ports(disagg_config_file: str) -> str:
+    # Revise the config file to use free ports
+    new_config = None
+    with open(disagg_config_file, 'r') as f:
+        config = yaml.safe_load(f)
+        new_config = revise_disaggregated_server_config_urls_with_free_ports(
+            config)
+
+    temp_fd, new_config_file = tempfile.mkstemp(suffix='.yaml')
+    with os.fdopen(temp_fd, 'w') as f:
+        yaml.dump(new_config, f)
+
+    return new_config_file
diff --git a/tests/integration/defs/disaggregated/test_disaggregated.py b/tests/integration/defs/disaggregated/test_disaggregated.py
index 07a5adc9874b..b2db88f0d24e 100644
--- a/tests/integration/defs/disaggregated/test_disaggregated.py
+++ b/tests/integration/defs/disaggregated/test_disaggregated.py
@@ -22,8 +22,8 @@
 
 import pytest
 import yaml
-from defs.common import (
-    revise_disaggregated_server_config_urls_with_free_ports, wait_for_server)
+from defs.common import (revise_disagg_config_file_with_free_ports,
+                         wait_for_server)
 from defs.conftest import (get_sm_version, llm_models_root, skip_arm,
                            skip_no_hopper)
 from defs.trt_test_alternative import check_call, check_output, popen
@@ -278,18 +278,8 @@ def get_test_config(test_desc, example_dir, test_root):
         raise ValueError(f"Invalid test description: {test_desc}, "
                          f"valid descriptions are: {config_map.keys()}")
 
-    # Revise the config file to use free ports
-    new_config = None
-    with open(config_map[test_desc][1], 'r') as f:
-        config = yaml.safe_load(f)
-        new_config = revise_disaggregated_server_config_urls_with_free_ports(
-            config)
-
-    temp_fd, new_config_file = tempfile.mkstemp(suffix=f'_{test_desc}.yaml')
-    with os.fdopen(temp_fd, 'w') as f:
-        yaml.dump(new_config, f)
-
-    return (config_map[test_desc][0], new_config_file)
+    return (config_map[test_desc][0],
+            revise_disagg_config_file_with_free_ports(config_map[test_desc][1]))
 
 
 def get_extra_llm_config(config, suffix, cwd):
diff --git a/tests/integration/defs/disaggregated/test_workers.py b/tests/integration/defs/disaggregated/test_workers.py
index c4fb51f63db3..b1b537ec35d8 100644
--- a/tests/integration/defs/disaggregated/test_workers.py
+++ b/tests/integration/defs/disaggregated/test_workers.py
@@ -9,6 +9,7 @@
 import aiohttp
 import pytest
 import yaml
+from defs.common import revise_disagg_config_file_with_free_ports
 from defs.conftest import skip_no_hopper
 from defs.trt_test_alternative import popen
 from transformers import AutoTokenizer
@@ -42,6 +43,7 @@ def run_disaggregated_workers(
     num_ranks: Optional[int] = None
 ) -> Tuple[Generator[subprocess.Popen, None, None], List[str], List[str]]:
 
+    config_file = revise_disagg_config_file_with_free_ports(config_file)
     ctx_servers, gen_servers = get_ctx_gen_server_urls_from_cfg(config_file)
 
     # TODO: auto detect num_ranks