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 14, 05:30 UTC
Update troubleshooting.md and remind VLLM_TRACE_FUNCTION usage (#27069)
Signed-off-by: cong-meta <prowindy@hotmail.com>
cong-meta committed
Oct 17, 2025 at 01:53 UTC
bbc1b29665ad8b1574ccfcfab7af7da492aa4d3b
1 file changed
+1
-1
docs/usage/troubleshooting.md
+1
-1
@@ -38,7 +38,7 @@ If other strategies don't solve the problem, it's likely that the vLLM instance
38
- `export VLLM_LOG_STATS_INTERVAL=1.` to get log statistics more frequently for tracking running queue, waiting queue and cache hit states.
39
- `export CUDA_LAUNCH_BLOCKING=1` to identify which CUDA kernel is causing the problem.
40
- `export NCCL_DEBUG=TRACE` to turn on more logging for NCCL.
41
-- `export VLLM_TRACE_FUNCTION=1` to record all function calls for inspection in the log files to tell which function crashes or hangs. Do not use this flag unless absolutely needed for debugging, it will cause significant delays in startup time.
41
+- `export VLLM_TRACE_FUNCTION=1` to record all function calls for inspection in the log files to tell which function crashes or hangs. (WARNING: This flag will slow down the token generation by **over 100x**. Do not use unless absolutely needed.)
42
43
## Breakpoints
44