如何修改GitHub Codespaces配置让postCreate脚本在新环境正常运行？

问题场景 #

当前.devcontainer/devcontainer.json配置及.devcontainer/postCreate.sh脚本在Rebuild Container操作下可正常完成端口转发与可见性修改，但创建新Codespace时触发容器启动失败，报错提示找不到codespace用户，最终系统自动创建恢复容器。

原devcontainer.json配置 #

{
    "image": "mcr.microsoft.com/devcontainers/universal:2",
    "build": {
        "context": ".."
    },
    "features": {
      "ghcr.io/devcontainers/features/python": {
        "version": "3.11"
      },
      "ghcr.io/devcontainers-contrib/features/poetry": {}
    },
    "forwardPorts": [
      8000
    ],
    "postCreateCommand": "bash .devcontainer/postCreate.sh"
}

原postCreate.sh脚本 #

#!/bin/bash
echo "POSTCREATECOMMAND"
echo "Elephant: Current directory: $(pwd)"
echo "Giraffe: Environment variables: $(env)"

# Check if .env file exists or needs to be created
if [ ! -f .env ]; then
    # Check if .env-template exists and copy it to .env
    if [ -f .env-template ]; then
        cp .env-template .env
    else
        echo "Tiger: Error: .env-template file not found. Please create or copy it to the current directory."
    fi
fi

# Get and replace CODESPACE_NAME in .env file if it exists
if [ -f .env ]; then
    CODESPACE_NAME=$(eval echo "$CODESPACE_NAME")
    sed -i "s|NEXT_PUBLIC_API_HOSTNAME=http://localhost:8000|NEXT_PUBLIC_API_HOSTNAME=https://$CODESPACE_NAME-8000.app.github.dev|g" .env
fi

port=8000
visibility=public
max_attempts=5
backoff_seconds=2 # Initial backoff time in seconds

# Check gh command and its output
gh_output=$(gh --version 2>&1)
gh_exit_code=$?
echo "Koala: GitHub CLI Output: $gh_output"

if [ $gh_exit_code -ne 0 ]; then
    echo "Penguin: Error: 'gh' command not found. Please install GitHub CLI."
    exit 1
fi

# Update port visibility with exponential backoff
attempt=1
echo "Lion: Updating port $port visibility to $visibility (attempt $attempt/$max_attempts)"
while true; do
    if gh codespace ports visibility $port:$visibility -c $CODESPACE_NAME; then
        break # Exit the loop if successful
    fi

    echo "Leopard: Error: Failed to update port visibility. Retrying in $backoff_seconds seconds..."
    sleep $backoff_seconds
    backoff_seconds=$((backoff_seconds * 2)) # Double the backoff time for each retry
    attempt=$((attempt + 1))

    if [ $attempt -gt $max_attempts ]; then
        echo "Kangaroo: Exceeded maximum attempts. Failed to update port visibility."
        exit 1
    fi
done

echo "Dolphin: Script execution completed at: $(date)"

错误日志 #

2024-07-22 18:16:17.858Z: unable to find user codespace: no matching entries in passwd file
2024-07-22 18:16:17.858Z: {"outcome":"error","message":"An error occurred setting up the container.","description":"An error occurred setting up the container.","containerId":"23c5269db6ae2d019296edd124bb68fcb4abafa9ac9b00046d9fb58196446b9c"}

2024-07-22 18:16:17.871Z: Failed to create container.
2024-07-22 18:16:17.874Z: Error code: 1302 (UnifiedContainersErrorFatalCreatingContainer)
2024-07-22 18:16:17.884Z: Container creation failed.
2024-07-22 18:16:18.128Z: Creating recovery container.

修复方案 #

1. 修正devcontainer.json配置 #

添加remoteUser字段明确指定容器运行用户，确保用户上下文存在：

{
    "image": "mcr.microsoft.com/devcontainers/universal:2",
    "build": {
        "context": ".."
    },
    "features": {
      "ghcr.io/devcontainers/features/python": {
        "version": "3.11"
      },
      "ghcr.io/devcontainers-contrib/features/poetry": {}
    },
    "forwardPorts": [
      8000
    ],
    "remoteUser": "codespace",
    "postCreateCommand": "bash .devcontainer/postCreate.sh"
}

2. 优化postCreate.sh脚本 #

简化CODESPACE_NAME的获取方式，避免不必要的eval操作，同时确保脚本执行稳定性：

#!/bin/bash
echo "POSTCREATECOMMAND"
echo "Elephant: Current directory: $(pwd)"
echo "Giraffe: Environment variables: $(env)"

# Check if .env file exists or needs to be created
if [ ! -f .env ]; then
    if [ -f .env-template ]; then
        cp .env-template .env
    else
        echo "Tiger: Error: .env-template file not found. Please create or copy it to the current directory."
    fi
fi

# Get and replace CODESPACE_NAME in .env file if it exists
if [ -f .env ]; then
    # 直接使用环境变量，无需eval
    CODESPACE_NAME="$CODESPACE_NAME"
    sed -i "s|NEXT_PUBLIC_API_HOSTNAME=http://localhost:8000|NEXT_PUBLIC_API_HOSTNAME=https://${CODESPACE_NAME}-8000.app.github.dev|g" .env
fi

port=8000
visibility=public
max_attempts=5
backoff_seconds=2

# Check gh command availability
gh_output=$(gh --version 2>&1)
gh_exit_code=$?
echo "Koala: GitHub CLI Output: $gh_output"

if [ $gh_exit_code -ne 0 ]; then
    echo "Penguin: Error: 'gh' command not found. Please install GitHub CLI."
    exit 1
fi

# Update port visibility with exponential backoff
attempt=1
echo "Lion: Updating port $port visibility to $visibility (attempt $attempt/$max_attempts)"
while true; do
    if gh codespace ports visibility $port:$visibility -c "$CODESPACE_NAME"; then
        break
    fi

    echo "Leopard: Error: Failed to update port visibility. Retrying in $backoff_seconds seconds..."
    sleep $backoff_seconds
    backoff_seconds=$((backoff_seconds * 2))
    attempt=$((attempt + 1))

    if [ $attempt -gt $max_attempts ]; then
        echo "Kangaroo: Exceeded maximum attempts. Failed to update port visibility."
        exit 1
    fi
done

echo "Dolphin: Script execution completed at: $(date)"

3. 备选方案（若基础镜像无codespace用户） #

如果添加remoteUser后仍报错用户不存在，可在postCreateCommand中先创建用户：

"postCreateCommand": "useradd -m codespace && bash .devcontainer/postCreate.sh"

（注：官方devcontainer通用镜像默认已包含该用户，此步骤通常无需执行）

修复逻辑 #

• 新Codespace创建时，容器默认未绑定有效用户上下文，添加remoteUser强制指定预存在的codespace用户，避免启动时用户不存在的错误；
• 移除脚本中风险较高的eval操作，直接使用环境变量保证稳定性；
• 确保postCreate脚本在正确的用户权限下执行，避免后续操作因权限或用户缺失失败。

内容的提问来源于stack exchange，提问作者Yann Stoneman