#!/bin/bash
#===========================================
# 自动恢复机制脚本
# 监控关键进程，失败时自动重启
#===========================================

# 配置文件
CONFIG_FILE="/root/.openclaw/config/auto-recovery.conf"
LOG_FILE="/root/.openclaw/logs/auto-recovery.log"
ALERT_LOG="/root/.openclaw/logs/auto-recovery-alerts.log"

mkdir -p /root/.openclaw/logs

#===========================================
# 加载配置
#===========================================
load_config() {
    if [ -f "$CONFIG_FILE" ]; then
        source "$CONFIG_FILE"
    else
        # 默认配置
        CHECK_INTERVAL=60
        RESTART_DELAY=5
        MAX_RESTART=3
    fi
}

#===========================================
# 记录日志
#===========================================
log() {
    local message=$1
    local timestamp=$(date "+%Y-%m-%d %H:%M:%S")
    echo "[$timestamp] $message" >> "$LOG_FILE"
}

#===========================================
# 检查进程状态
#===========================================
check_process() {
    local process_name=$1
    local restart_cmd=$2
    
    if pgrep -f "$process_name" > /dev/null; then
        echo "✅ $process_name: 运行中"
        return 0
    else
        echo "❌ $process_name: 未运行"
        log "进程 $process_name 未运行，准备重启..."
        
        # 尝试重启
        local restart_count=1
        while [ $restart_count -le $MAX_RESTART ]; do
            log "尝试重启 $process_name (第 $restart_count 次)..."
            
            eval "$restart_cmd" >> "$LOG_FILE" 2>&1
            
            sleep $RESTART_DELAY
            
            if pgrep -f "$process_name" > /dev/null; then
                log "✅ $process_name 重启成功"
                echo "✅ $process_name: 已重启"
                return 0
            fi
            
            restart_count=$((restart_count + 1))
        done
        
        log "❌ $process_name 重启失败"
        echo "❌ $process_name: 重启失败"
        
        # 发送告警
        echo "[$(date)] [CRITICAL] $process_name 自动恢复失败" >> "$ALERT_LOG"
        return 1
    fi
}

#===========================================
# 检查服务端口
#===========================================
check_port() {
    local name=$1
    local port=$2
    
    if curl -s --noproxy 127.0.0.1 --max-time 5 http://127.0.0.1:$port/ > /dev/null 2>&1; then
        echo "✅ $name 端口 $port: 正常"
        return 0
    else
        echo "❌ $name 端口 $port: 无响应"
        return 1
    fi
}

#===========================================
# 自动恢复主逻辑
#===========================================
auto_recovery() {
    echo "=== 自动恢复检查 $(date '+%Y-%m-%d %H:%M:%S') ==="
    log "开始自动恢复检查..."
    
    local failed=0
    
    # 1. 检查 Gateway
    echo ""
    echo "检查 Gateway..."
    if ! check_process "openclaw-gateway" "openclaw gateway restart"; then
        ((failed++))
    fi
    
    # 2. 检查端口
    echo ""
    echo "检查端口..."
    check_port "Gateway" "18789"
    
    # 3. 检查关键 OpenClaw 进程
    echo ""
    echo "检查 OpenClaw 进程..."
    check_process "openclaw-gateway" "openclaw gateway restart"
    
    # 总结
    echo ""
    if [ $failed -eq 0 ]; then
        echo "✅ 自动恢复检查完成 - 所有服务正常"
        log "自动恢复检查完成 - 所有服务正常"
    else
        echo "⚠️ 自动恢复检查完成 - $failed 项需要关注"
        log "自动恢复检查完成 - $failed 项需要关注"
    fi
}

#===========================================
# 持续监控模式
#===========================================
monitor_loop() {
    load_config
    
    echo "=== 自动恢复监控启动 ==="
    echo "检查间隔: ${CHECK_INTERVAL}秒"
    echo "最大重启次数: $MAX_RESTART"
    log "自动恢复监控启动..."
    
    while true; do
        auto_recovery
        sleep $CHECK_INTERVAL
    done
}

#===========================================
# 命令行接口
#===========================================
case "${1:-check}" in
    check)
        load_config
        auto_recovery
        ;;
    monitor)
        monitor_loop
        ;;
    list)
        load_config
        echo "=== 自动恢复配置 ==="
        echo "检查间隔: ${CHECK_INTERVAL}s"
        echo "重启延迟: ${RESTART_DELAY}s"
        echo "最大重启次数: $MAX_RESTART"
        ;;
    *)
        echo "用法: $0 {check|monitor|list}"
        ;;
esac