A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 6, 23:00 UTC
fix: check HTTP status in batch read_file to prevent silent failures (#36397)
Signed-off-by: gambletan <ethanchang32@gmail.com> Co-authored-by: gambletan <ethanchang32@gmail.com> Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Alvin Tang committed
Mar 10, 2026 at 22:22 UTC
cf88b23749187b9a31406925d3f9e966fc4c566b
1 file changed
+1
vllm/entrypoints/openai/run_batch.py
+1
@@ -320,6 +320,7 @@ class BatchProgressTracker:
320
async def read_file(path_or_url: str) -> str:
321
if path_or_url.startswith("http://") or path_or_url.startswith("https://"):
322
async with aiohttp.ClientSession() as session, session.get(path_or_url) as resp:
323
+ resp.raise_for_status()
324
return await resp.text()
325
else:
326
with open(path_or_url, encoding="utf-8") as f: