[PD][Bugfix]修复0151pd分离d卡住问题

vllm/envs.py

@@ -180,6 +180,7 @@ if TYPE_CHECKING:
     VLLM_XGRAMMAR_CACHE_MB: int = 0
     VLLM_MSGPACK_ZERO_COPY_THRESHOLD: int = 256
     VLLM_ALLOW_INSECURE_SERIALIZATION: bool = False
+    VLLM_DISABLE_REQUEST_ID_RANDOMIZATION: bool = False
     VLLM_NIXL_SIDE_CHANNEL_HOST: str = "localhost"
     VLLM_NIXL_SIDE_CHANNEL_PORT: int = 5600
     VLLM_MOONCAKE_BOOTSTRAP_PORT: int = 8998
@@ -1344,6 +1345,11 @@ environment_variables: dict[str, Callable[[], Any]] = {
     "VLLM_ALLOW_INSECURE_SERIALIZATION": lambda: bool(
         int(os.getenv("VLLM_ALLOW_INSECURE_SERIALIZATION", "0"))
     ),
+    # Temporary: skip adding random suffix to internal request IDs. May be
+    # needed for KV connectors that match request IDs across instances.
+    "VLLM_DISABLE_REQUEST_ID_RANDOMIZATION": lambda: bool(
+        int(os.getenv("VLLM_DISABLE_REQUEST_ID_RANDOMIZATION", "1"))
+    ),
     # IP address used for NIXL handshake between remote agents.
     "VLLM_NIXL_SIDE_CHANNEL_HOST": lambda: os.getenv(
         "VLLM_NIXL_SIDE_CHANNEL_HOST", "localhost"

vllm/v1/engine/input_processor.py

@@ -6,6 +6,7 @@ import time
 from collections.abc import Mapping
 from typing import Any, Literal, cast
 
+import vllm.envs as envs
 from vllm.config import VllmConfig
 from vllm.exceptions import VLLMValidationError
 from vllm.inputs import (
@@ -474,7 +475,14 @@ class InputProcessor:
                 " passed to vLLM; use the request_id field."
             )
         request.external_req_id = request.request_id
-        request.request_id = f"{request.external_req_id}-{random_uuid():.8}"
+        if envs.VLLM_DISABLE_REQUEST_ID_RANDOMIZATION:
+            logger.warning_once(
+                "VLLM_DISABLE_REQUEST_ID_RANDOMIZATION is set and will be "
+                "removed in a future release. Duplicate externally-provided "
+                "request IDs may cause failures and/or subtle correctness errors."
+            )
+        else:
+            request.request_id = f"{request.external_req_id}-{random_uuid():.8}"
 
     def process_inputs(
         self,