hailin
/
sglang.0.4.8.post1
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

.

This commit is contained in:
hailin 2025-09-14 19:12:27 +08:00
parent c516e234c0
commit 29de4e1411
3 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
Dockerfile
View File

@ -172,7 +172,7 @@ RUN chmod +x /tini
ENTRYPOINT ["/tini", "--"]
# ---- 拷贝模型（路径可换） ----
COPY ./Alibaba/Qwen3-30B-A3B /root/.cradle/Alibaba/Qwen3-30B-A3B
# COPY ./Alibaba/Qwen3-30B-A3B /root/.cradle/Alibaba/Qwen3-30B-A3B
HEALTHCHECK --interval=30s --timeout=2s --start-period=300s --retries=5 CMD curl -fs http://localhost:30000/health || exit 1

2
Dockerfile.llm_lora_external → Dockerfile.tmp
View File

@ -172,7 +172,7 @@ RUN chmod +x /tini
ENTRYPOINT ["/tini", "--"]
# ---- 拷贝模型（路径可换） ----
# COPY ./Alibaba/Qwen3-30B-A3B /root/.cradle/Alibaba/Qwen3-30B-A3B
COPY ./Alibaba/Qwen3-30B-A3B /root/.cradle/Alibaba/Qwen3-30B-A3B
HEALTHCHECK --interval=30s --timeout=2s --start-period=300s --retries=5 CMD curl -fs http://localhost:30000/health || exit 1

2
supervisord.conf
View File

@ -5,7 +5,7 @@ logfile_maxbytes=0
loglevel=info
[program:sglang]
command=python3 -m sglang.launch_server --host 0.0.0.0 --port 30000 --model-path /root/.cradle/external/llm/ --lora-paths q3=/root/.cradle/external/lora/q3 --tp 4 --api-key token-abc123 --enable-metrics
command=python3 -m sglang.launch_server --host 0.0.0.0 --port 30000 --model-path /root/.cradle/external/llm/ --lora-paths q3=/root/.cradle/external/lora/q3 --disable-radix-cache --tp 4 --api-key token-abc123 --enable-metrics
autostart=true
autorestart=true
stdout_logfile=/dev/stdout