[Bug] Docker container cannot be accessed after starting the program

[monkeyGeek369]

通过docker成功启动程序，也开启了webui，但是奇怪的是容器内可以访问127.0.0.1:7860但是宿主机无法访问。 1、已经做了端口映射，无效 2、容器采用host模式，无效 3、容器内可以ping通宿主机，但宿主机无法ping通容器ip 4、容器内可以访问页面

[monkeyGeek369]

求大恨帮忙，到底什么原因？

[monkeyGeek369]

起了两个容器，无论是端口映射还是host网络都是一样的

docker run -p 7860:7860 --name fish-speech-myself -v E:\gitProject\fish-speech:/exp -d --gpus all fish-speech-fish-speech tail -f /dev/null

docker run --network host --name fish-speech-myself -v E:\gitProject\fish-speech:/exp -d --gpus all fish-speech-fish-speech tail -f /dev/null

[AnyaCoder]

在命令行里输入

netstat -ano | findstr :7860

看看是否有占用。

[fallbernana123456]

https://speech.fish.audio/inference/#webui 这个问题跟gradio相关，要使用demo.launch(..., server_name="0.0.0.0") 或者设置环境变量GRADIO_SERVER_NAME="0.0.0.0"

[chaoqunxie]

https://speech.fish.audio/inference/#webui 这个问题跟gradio相关，要使用demo.launch(..., server_name="0.0.0.0") 或者设置环境变量GRADIO_SERVER_NAME="0.0.0.0"

这是正确的 但是我启动推理页面总是说端口占用 但是实际上端口并没有被使用 2024-07-17 06:05:50.748 | INFO | tools.llama.generate:generate_long:509 - Generating sentence 1/1 of sample 1/1 0%| | 0/2015 [00:00<?, ?it/s]/usr/local/lib/python3.10/site-packages/torch/backends/cuda/init.py:342: FutureWarning: torch.backends.cuda.sdp_kernel() is deprecated. In the future, this context manager will be removed. Please see, torch.nn.attention.sdpa_kernel() for the new context manager, with updated signature. warnings.warn( 0%| | 0/2015 [00:29<?, ?it/s] 2024-07-17 06:06:20.701 | INFO | main::507 - Warming up done, launching the web UI... Traceback (most recent call last): File "/exp/tools/webui.py", line 510, in app.launch(show_api=True) File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 2326, in launch ) = http_server.start_server( File "/usr/local/lib/python3.10/site-packages/gradio/http_server.py", line 154, in start_server raise OSError( OSError: Cannot find empty port in range: 10001-10100. You can specify a different port by setting the GRADIO_SERVER_PORT environment variable or passing the server_port parameter to launch().

[chaoqunxie]

[AnyaCoder]

0.0.0.0

[chaoqunxie]

0.0.0.0

感谢！ 手滑输错了~

[chaoqunxie]

起了两个容器，无论是端口映射还是host网络都是一样的

docker run -p 7860:7860 --name fish-speech-myself -v E:\gitProject\fish-speech:/exp -d --gpus all fish-speech-fish-speech tail -f /dev/null

docker run --network host --name fish-speech-myself -v E:\gitProject\fish-speech:/exp -d --gpus all fish-speech-fish-speech tail -f /dev/null

#!/bin/bash
# 设置字符编码
export LC_ALL=en_US.UTF-8

# 设置初始变量
USE_MIRROR=true

PYTHON_CMD=python
API_FLAG_PATH=$(dirname "$0")/API_FLAGS.txt

# 设置默认的端点
HF_ENDPOINT="https://huggingface.co"
no_proxy=""

# 如果指定了使用镜像，则设置镜像地址
if [ "$USE_MIRROR" == "true" ]; then
    HF_ENDPOINT="https://hf-mirror.com"
    no_proxy="localhost, 127.0.0.1, 0.0.0.0"
fi

echo "HF_ENDPOINT: $HF_ENDPOINT"
echo "NO_PROXY: $no_proxy"

# 运行模型下载脚本
$PYTHON_CMD ./tools/download_models.py

# 初始化API标志
API_FLAGS=""
flags=""

# 如果存在API标志文件，则处理文件中的标志
if [ -f "$API_FLAG_PATH" ]; then
    while IFS= read -r line; do
        if [ "${line:0:1}" != "#" ]; then
            line=$(echo "$line" | sed 's/ /<SPACE>/g' | sed 's/\\//g' | sed 's/<SPACE>/ /g')
            if [ ! -z "$line" ]; then
                API_FLAGS="$API_FLAGS$line "
            fi
        fi
    done < "$API_FLAG_PATH"
fi

# 去除API_FLAGS末尾的空格
API_FLAGS=$(echo "$API_FLAGS" | sed 's/ *$//')

# 检查模式标志并相应地设置模式
echo "$API_FLAGS" | grep -q -- "--api"
if [ $? -eq 0 ]; then
    echo
    echo "启动HTTP API..."
    mode="api"
    process_flags
fi

echo "$API_FLAGS" | grep -q -- "--infer"
if [ $? -eq 0 ]; then
    echo
    echo "启动WebUI推理..."
    mode="infer"
    process_flags
fi

# 处理剩余标志
process_flags() {
    for p in $API_FLAGS; do
        if [ "$p" != "--$mode" ]; then
            flags="$flags $p"
        fi
    done
    flags=$(echo "$flags" | sed 's/^ *//')
}

# 输出调试信息
echo "调试: flags = $flags"

# 设置环境变量 GRADIO_SERVER_NAME
export GRADIO_SERVER_NAME="0.0.0.0"

# 运行相应模式
if [ "$mode" == "api" ]; then
    $PYTHON_CMD -m tools.api $flags
elif [ "$mode" == "infer" ]; then
    $PYTHON_CMD -m tools.webui $flags
fi

echo
echo "接下来启动页面..."
$PYTHON_CMD fish_speech/webui/manage.py

# 脚本结束                       用这个启动脚本

[chaoqunxie]

0.0.0.0

Open Labeler WebUI 这个好像打不开2024-07-17 06:51:58.924 | INFO | main::507 - Warming up done, launching the web UI... 2024-07-17 07:16:25.574 | WARNING | main:change_label:121 - asr-label execution not found!

[AnyaCoder]

用在线版吧。这个要单独下载,解压后里面的文件放项目根目录下。 asr-label-linux-x64.zip.zip

[chaoqunxie]

用在线版吧。这个要单独下载,解压后里面的文件放项目根目录下。 asr-label-linux-x64.zip.zip

感谢大佬

[chaoqunxie]

用在线版吧。这个要单独下载,解压后里面的文件放项目根目录下。 asr-label-linux-x64.zip.zip

还是报错 2024-07-17 07:47:59.967 | WARNING | main:change_label:121 - asr-label execution not found!

[AnyaCoder]

可能是没有权限导致的, sudo chmod 777 asr-label-linux-x64. 不行的话用在线版，点击打标的时候会出现提示。

[Stardust-minus]

515 Change title to English.