1. 更新allinone脚本提示，增加中文提示;2. 新增服务一键停止脚本shutdown_all.sh;3. 更新readme说明;3. 修复命令行bug

README.md

@@ -351,7 +351,7 @@ $ streamlit run webui.py --server.port 666
 
 ### 6 一键启动
 
-**!!!注意，一键启动脚本仅适用于linux和mac设备!!!**
+**!!!注意，一键启动脚本仅适用于linux和mac设备,且加载需要命令行指定模型，不会读取model_config.py配置!!!**
 
 #### 6.1 api服务一键启动脚本
 
@@ -411,6 +411,12 @@ $ python webui_alline.py --model-path-address model@host@port --num-gpus 2 --gpu
 
 其他参数详见各脚本及fastchat服务说明。
 
+上述两个一键启动脚本会后台运行多个服务，如要停止所有服务，可使用shutdown_all.sh脚本：
+
+```shell
+bash shutdown_all.sh
+```
+
 ## 常见问题
 
 参见 [常见问题](docs/FAQ.md)。

server/api_allinone.py

@@ -22,9 +22,7 @@ parser.add_argument("--api-host", type=str, default="0.0.0.0")
 parser.add_argument("--api-port", type=int, default=7861)
 parser.add_argument("--ssl_keyfile", type=str)
 parser.add_argument("--ssl_certfile", type=str)
-# 初始化消息
-args = parser.parse_args()
-args_dict = vars(args)
+
 
 api_args = ["api-host","api-port","ssl_keyfile","ssl_certfile"]
 
@@ -43,6 +41,9 @@ def run_api(host, port, **kwargs):
 if __name__ == "__main__":
     print("Luanching api_allinone，it would take a while, please be patient...")
     print("正在启动api_allinone，LLM服务启动约3-10分钟，请耐心等待...")
+    # 初始化消息
+    args = parser.parse_args()
+    args_dict = vars(args)
     launch_all(args=args,controller_args=controller_args,worker_args=worker_args,server_args=server_args)
     run_api(
             host=args.api_host,

server/llm_api_launch.py

@@ -159,17 +159,7 @@ server_args = ["server-host", "server-port", "allow-credentials", "api-keys",
                "controller-address"
                ]
 
-args = parser.parse_args()
-# 必须要加http//:，否则InvalidSchema: No connection adapters were found
-args = argparse.Namespace(**vars(args),
-                          **{"controller-address": f"http://{args.controller_host}:{str(args.controller_port)}"})
 
-if args.gpus:
-    if len(args.gpus.split(",")) < args.num_gpus:
-        raise ValueError(
-            f"Larger --num-gpus ({args.num_gpus}) than --gpus {args.gpus}!"
-        )
-    os.environ["CUDA_VISIBLE_DEVICES"] = args.gpus
 
 # 0,controller, model_worker, openai_api_server
 # 1, 命令行选项
@@ -211,13 +201,13 @@ def string_args(args, args_list):
     return args_str
 
 
-def launch_worker(item,args=args,worker_args=worker_args):
+def launch_worker(item,args,worker_args=worker_args):
     log_name = item.split("/")[-1].split("\\")[-1].replace("-", "_").replace("@", "_").replace(".", "_")
     # 先分割model-path-address,在传到string_args中分析参数
     args.model_path, args.worker_host, args.worker_port = item.split("@")
     args.worker_address = f"http://{args.worker_host}:{args.worker_port}"
     print("*" * 80)
-    print(f"worker启动视设备不同而不同，约需3-10分钟，如长时间未启动，请到{LOG_PATH}/{log_name}下查看日志")
+    print(f"如长时间未启动，请到{LOG_PATH}{log_name}.log下查看日志")
     worker_str_args = string_args(args, worker_args)
     print(worker_str_args)
     worker_sh = base_launch_sh.format("model_worker", worker_str_args, LOG_PATH, f"worker_{log_name}")
@@ -226,7 +216,7 @@ def launch_worker(item,args=args,worker_args=worker_args):
     subprocess.run(worker_check_sh, shell=True, check=True)
 
 
-def launch_all(args=args,
+def launch_all(args,
                controller_args=controller_args,
                worker_args=worker_args,
                server_args=server_args
@@ -238,7 +228,7 @@ def launch_all(args=args,
     controller_check_sh = base_check_sh.format(LOG_PATH, "controller", "controller")
     subprocess.run(controller_sh, shell=True, check=True)
     subprocess.run(controller_check_sh, shell=True, check=True)
-
+    print(f"worker启动时间视设备不同而不同，约需3-10分钟，请耐心等待...")
     if isinstance(args.model_path_address, str):
         launch_worker(args.model_path_address,args=args,worker_args=worker_args)
     else:
@@ -255,4 +245,15 @@ def launch_all(args=args,
     print("LLM服务启动完毕。")
 
 if __name__ == "__main__":
-    launch_all()
+    args = parser.parse_args()
+    # 必须要加http//:，否则InvalidSchema: No connection adapters were found
+    args = argparse.Namespace(**vars(args),
+                            **{"controller-address": f"http://{args.controller_host}:{str(args.controller_port)}"})
+
+    if args.gpus:
+        if len(args.gpus.split(",")) < args.num_gpus:
+            raise ValueError(
+                f"Larger --num-gpus ({args.num_gpus}) than --gpus {args.gpus}!"
+            )
+        os.environ["CUDA_VISIBLE_DEVICES"] = args.gpus
+    launch_all(args=args)

shutdown_all.sh

@@ -0,0 +1 @@
+ps -eo pid,user,cmd|grep -P 'server/api.py|webui.py|fastchat.serve'|grep -v grep|awk '{print $1}'|xargs kill -9

webui_allinone.py

@@ -34,14 +34,14 @@ parser.add_argument("--theme.secondaryBackgroundColor",type=str,default='"#f5f5f
 parser.add_argument("--theme.textColor",type=str,default='"#000000"')
 web_args = ["server.port","theme.base","theme.primaryColor","theme.secondaryBackgroundColor","theme.textColor"]
 
-args = parser.parse_args()
 
-def launch_api(args=args,args_list=api_args,log_name=None):
+def launch_api(args,args_list=api_args,log_name=None):
     print("Launching api ...")
     print("启动API服务...")
     if not log_name:
         log_name = f"{LOG_PATH}api_{args.api_host}_{args.api_port}"
     print(f"logs on api are written in {log_name}")
+    print(f"API日志位于{log_name}下，如启动异常请查看日志")
     args_str = string_args(args,args_list)
     api_sh = "python  server/{script} {args_str} >{log_name}.log 2>&1 &".format(
         script="api.py",args_str=args_str,log_name=log_name)
@@ -49,14 +49,16 @@ def launch_api(args=args,args_list=api_args,log_name=None):
     print("launch api done!")
     print("启动API服务完毕.")
 
-def launch_webui(args=args,args_list=web_args,log_name=None):
+def launch_webui(args,args_list=web_args,log_name=None):
     print("Launching webui...")
     print("启动webui服务...")
     if not log_name:
         log_name = f"{LOG_PATH}webui"
-    print(f"logs on api are written in {log_name}")
+
     args_str = string_args(args,args_list)
     if args.nohup:
+        print(f"logs on api are written in {log_name}")
+        print(f"webui服务日志位于{log_name}下，如启动异常请查看日志")
         webui_sh = "streamlit run webui.py {args_str} >{log_name}.log 2>&1 &".format(
         args_str=args_str,log_name=log_name)
     else:
@@ -70,6 +72,9 @@ def launch_webui(args=args,args_list=web_args,log_name=None):
 if __name__ == "__main__":
     print("Starting webui_allineone.py, it would take a while, please be patient....")
     print(f"开始启动webui_allinone,启动LLM服务需要约3-10分钟，请耐心等待，如长时间未启动，请到{LOG_PATH}下查看日志...")
+    args = parser.parse_args()
+
+    print("*"*80)
     if not args.use_remote_api:
         launch_all(args=args,controller_args=controller_args,worker_args=worker_args,server_args=server_args)
     launch_api(args=args,args_list=api_args)