frozenleaves/vllm
1
0
Fork 0
mirror of https://github.com/vllm-project/vllm.git synced 2025-10-20 14:53:52 +08:00
Code Issues Packages Projects Releases Wiki Activity

Update troubleshooting.md and remind VLLM_TRACE_FUNCTION usage (#27069)

Browse Source 
Signed-off-by: cong-meta <prowindy@hotmail.com>
This commit is contained in:
cong-meta
2025-10-17 01:53:06 -07:00
committed by GitHub
parent acb1bfa601
commit bbc1b29665
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
docs/usage/troubleshooting.md
View File

@ -38,7 +38,7 @@ If other strategies don't solve the problem, it's likely that the vLLM instance
- `export VLLM_LOG_STATS_INTERVAL=1.` to get log statistics more frequently for tracking running queue, waiting queue and cache hit states.
- `export CUDA_LAUNCH_BLOCKING=1` to identify which CUDA kernel is causing the problem.
- `export NCCL_DEBUG=TRACE` to turn on more logging for NCCL.
- `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.
- `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.)
## Breakpoints