curl http://127.0.0.1:xxx/v1/completions
-H 'accept: application/json' -H 'Content-Type: application/json'
-X 'POST'
-d '{
"model": "Qwen3-8B",
"prompt": "hello xllm",
"stream": false,
"max_tokens": 30,
"temperature": 0.0,
"beam_width": 5,
"logprobs": 5
}‘
I20260212 10:59:00.951273 189 xllm_server.cpp:59] Brpc Server started on port 28000, idle_timeout_s: -1, num_threads: 8
I20260212 11:00:01.958679 633 request.cpp:80] x-request-id: , x-request-time: , request_id: chatcmpl-3098677694751925817-AawnQ2NkgowCVLna2XXa8u, sequence 0, max_tokens: 3, temperature: 0.01, finish_reason: length, prompt_tokens: 9, generated_tokens: 3, ttft: 1344.0ms, total_latency: 1450.9ms
I20260212 11:01:37.401417 634 request.cpp:80] x-request-id: , x-request-time: , request_id: cmpl-3098677694751925817-6QPcFwuvtNVMFn4NzdGogM, sequence 0, max_tokens: 30, temperature: 0, finish_reason: length, prompt_tokens: 4, generated_tokens: 30, ttft: 41.0ms, total_latency: 461.8ms
[ERROR] TBE Subprocess[task_distribute] raise error[], main process disappeared!
[ERROR] TBE Subprocess[task_distribute] raise error[], main process disappeared!
[ERROR] TBE Subprocess[task_distribute] raise error[], main process disappeared!
[ERROR] TBE Subprocess[task_distribute] raise error[], main process disappeared!
[ERROR] TBE Subprocess[task_distribute] raise error[], main process disappeared!
[ERROR] TBE Subprocess[task_distribute] raise error[], main process disappeared!
[ERROR] TBE Subprocess[task_distribute] raise error[], main process disappeared!
[ERROR] TBE Subprocess[task_distribute] raise error[], main process disappeared!
/usr/lib64/python3.11/multiprocessing/resource_tracker.py:254: UserWarning: resource_tracker: There appear to be 30 leaked semaphore objects to clean up at shutdown
warnings.warn('resource_tracker: There appear to be %d '
Your environment
xllm-0.8.0-release-hb-rc2-arm
910B3
🐛 Describe the bug
xllm不开beam search可以正常推理,尝试使用beam search时,xllm进程会崩溃退出。
请问是配置问题还是当前配套不支持呢?
请求:
打印的日志: