[Benchmark] fix request loss if "ping" is returned (#19535)

Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

benchmarks/backend_request_func.py

@@ -404,8 +404,14 @@ async def async_request_openai_chat_completions(
                         chunk_bytes = chunk_bytes.strip()
                         if not chunk_bytes:
                             continue
+                        chunk_bytes = chunk_bytes.decode("utf-8")
+                        # NOTE: SSE comments (often used as pings) start with a colon.
+                        # These are not JSON data payload and should be skipped.
+                        if chunk_bytes.startswith(":"):
+                            continue
+
+                        chunk = chunk_bytes.removeprefix("data: ")
 
-                        chunk = chunk_bytes.decode("utf-8").removeprefix("data: ")
                         if chunk != "[DONE]":
                             timestamp = time.perf_counter()
                             data = json.loads(chunk)

vllm/benchmarks/endpoint_request_func.py

@@ -104,9 +104,15 @@ async def async_request_openai_completions(
                         chunk_bytes = chunk_bytes.strip()
                         if not chunk_bytes:
                             continue
+                        chunk_bytes = chunk_bytes.decode("utf-8")
+                        # NOTE: SSE comments (often used as pings) start with
+                        # a colon. These are not JSON data payload and should
+                        # be skipped.
+                        if chunk_bytes.startswith(":"):
+                            continue
+
+                        chunk = chunk_bytes.removeprefix("data: ")
 
-                        chunk = chunk_bytes.decode("utf-8").removeprefix(
-                            "data: ")
                         if chunk != "[DONE]":
                             data = json.loads(chunk)
 
@@ -213,9 +219,15 @@ async def async_request_openai_chat_completions(
                         chunk_bytes = chunk_bytes.strip()
                         if not chunk_bytes:
                             continue
+                        chunk_bytes = chunk_bytes.decode("utf-8")
+                        # NOTE: SSE comments (often used as pings) start with
+                        # a colon. These are not JSON data payload and should
+                        # be skipped.
+                        if chunk_bytes.startswith(":"):
+                            continue
+
+                        chunk = chunk_bytes.removeprefix("data: ")
 
-                        chunk = chunk_bytes.decode("utf-8").removeprefix(
-                            "data: ")
                         if chunk != "[DONE]":
                             timestamp = time.perf_counter()
                             data = json.loads(chunk)