配置rollback_hosts_linux.sh脚本

根据提供的脚本示例，结合实际情况，修改示例中的相关配置。

前提条件

已完成准备工作。

操作步骤

在执行脚本的主机上创建一个名为“rollback_hosts_linux.sh”的文件，然后将以下脚本示例的内容复制到文件中。如果您已经通过SSH连接到源端Linux主机，可以直接使用vim创建和编辑脚本文件，步骤如下：

在Vim编辑器中按 i 进入插入模式。
复制并粘贴脚本代码，完成后按 Esc。
输入 :wq 保存并退出。

#!/bin/bash
 
# Configuration
 
# Log directory path: Used to store run logs, error logs, and summary logs.
# If the directory doesn't exist, the script will create it automatically.
LOG_DIR="/var/log/update_hosts"
 
# Run log file path: Records detailed information about the script's execution.
RUN_LOG="$LOG_DIR/run.log"
 
# Error log file path: Records any errors that occur during the script's execution.
ERROR_LOG="$LOG_DIR/error.log"
 
# Summary log file path: Records a summary of the script's execution, including the number of successful and failed servers.
SUMMARY_LOG="$LOG_DIR/summary.log"
 
# CSV file path: Contains information about the target hosts (must be manually created and configured).
CSV_FILE="target_servers.csv"
DEFAULT_PORT=22
SSH_TIMEOUT=10
 
# Initialize log directory and files
initialize_logs() {
    mkdir -p "$LOG_DIR"
    echo "========================================" >> "$RUN_LOG"
    echo "[INFO] $(date '+%Y-%m-%d %H:%M:%S') - Starting new rollback execution" >> "$RUN_LOG"
    echo "========================================" >> "$RUN_LOG"
 
    echo "========================================" >> "$ERROR_LOG"
    echo "[INFO] $(date '+%Y-%m-%d %H:%M:%S') - Starting new rollback execution" >> "$ERROR_LOG"
    echo "========================================" >> "$ERROR_LOG"
 
    echo "========================================" > "$SUMMARY_LOG"
    echo "[INFO] $(date '+%Y-%m-%d %H:%M:%S') - Starting new rollback execution" >> "$SUMMARY_LOG"
    echo "========================================" >> "$SUMMARY_LOG"
}
 
# Log info function
log_info() {
    echo "[INFO] $(date '+%Y-%m-%d %H:%M:%S') - $1" | tee -a "$RUN_LOG"
}
 
# Log error function
log_error() {
    echo "[ERROR] $(date '+%Y-%m-%d %H:%M:%S') - $1" | tee -a "$RUN_LOG" "$ERROR_LOG"
}
 
# Read server information from CSV file
read_servers_from_csv() {
    local csv_file="$1"
    local servers=()
    local header_skipped=false
 
    if [ ! -f "$csv_file" ]; then
        log_error "CSV file '$csv_file' not found."
        exit 1
    fi
 
    # Ensure file ends with a newline character
    sed -i -e '$a\' "$csv_file"
 
    while IFS=, read -r username ip port password; do
        # Skip header row
        if [ "$header_skipped" = false ]; then
            header_skipped=true
            continue
        fi
        # Skip empty and invalid rows
        if [[ -z "$username" || -z "$ip" ]]; then
            continue
        fi
        port=${port:-$DEFAULT_PORT} # Use default port 22
        # Ensure port is numeric
        if ! [[ "$port" =~ ^[0-9]+$ ]]; then
            log_error "Invalid port '$port' for $username@$ip. Skipping this server."
            continue
        fi
        servers+=("$username@$ip:$port:$password")
    done < "$csv_file"
 
    echo "${servers[@]}"
}
 
# Initialize log files
initialize_logs
 
# Read server information from CSV file
servers=($(read_servers_from_csv "$CSV_FILE"))
 
# Counters for success and failure
success_count=0
failure_count=0
failed_servers=()
 
# Iterate over each server and execute rollback
for server in "${servers[@]}"; do
  # Extract user, IP, port, and password information
  IFS=':' read -r user_host port pass <<< "$server"
  IFS='@' read -r user ip <<< "$user_host"
  
  log_info "Starting rollback for $user@$ip:$port"
 
  # Create temporary script and SSH_ASKPASS script
  tmp_script=$(mktemp)
  askpass_script=$(mktemp)
 
  cat <<EOF > "$tmp_script"
#!/bin/bash
# Backup hosts file
if [ ! -f /etc/hosts.bak ]; then
  cp /etc/hosts /etc/hosts.bak
fi
 
# Remove old Migration-proxy section
sed -i '/#Migration-proxy-start/,/#Migration-proxy-end/d' /etc/hosts
EOF
 
  cat <<EOF > "$askpass_script"
#!/bin/bash
echo "$pass"
EOF
 
  chmod +x "$tmp_script" "$askpass_script"
 
  # Set SSH_ASKPASS environment variable and use ssh to connect to the target machine and execute the temporary script
  export SSH_ASKPASS="$askpass_script"
  export DISPLAY=:0
 
  ssh_output=$(mktemp)
  setsid ssh -o BatchMode=no -o ConnectTimeout=$SSH_TIMEOUT -o StrictHostKeyChecking=no -p "$port" "$user@$ip" 'bash -s' < "$tmp_script" 2> "$ssh_output"
  ssh_status=$?
 
  if [ $ssh_status -eq 0 ]; then
    log_info "Rolled back hosts on $ip:$port successfully"
    ((success_count++))
  else
    ssh_error=$(cat "$ssh_output")
    case $ssh_status in
      1)
        log_error "General error occurred while rolling back hosts on $ip:$port: $ssh_error"
        ;;
      2)
        log_error "Misuse of shell builtins while rolling back hosts on $ip:$port: $ssh_error"
        ;;
      255)
        if [[ "$ssh_error" == *"Permission denied"* ]]; then
          log_error "SSH login failed for $user@$ip:$port: Permission denied (password may be incorrect or username is wrong)"
        elif [[ "$ssh_error" == *"Connection refused"* ]]; then
          log_error "SSH login failed for $user@$ip:$port: Connection refused (port may be incorrect or SSH service not running on target)"
        elif [[ "$ssh_error" == *"No route to host"* ]]; then
          log_error "SSH login failed for $user@$ip:$port: No route to host (network unreachable)"
        elif [[ "$ssh_error" == *"Host key verification failed"* ]]; then
          log_error "SSH login failed for $user@$ip:$port: Host key verification failed"
        elif [[ "$ssh_error" == *"Connection timed out"* ]]; then
          log_error "SSH login failed for $user@$ip:$port: Connection timed out"
        else
          log_error "SSH login failed for $user@$ip:$port: $ssh_error"
        fi
        ;;
      *)
        log_error "An unknown error occurred while rolling back hosts on $ip:$port: $ssh_error"
        ;;
    esac
    failed_servers+=("$user@$ip:$port")
    ((failure_count++))
  fi
 
  # Remove temporary scripts and SSH output file
  rm -f "$tmp_script" "$askpass_script" "$ssh_output"
done
 
# Calculate failure and success percentages
total_count=${#servers[@]}
failure_percentage=$(echo "scale=2; ($failure_count / $total_count) * 100" | bc)
success_percentage=$(echo "scale=2; ($success_count / $total_count) * 100" | bc)
 
# Output summary result and log to file
summary_content=$(cat <<EOF
========================================
[SUMMARY] $(date '+%Y-%m-%d %H:%M:%S') - Execution Rollback Summary
========================================
Total number of servers: $total_count
Number of successful rollbacks: $success_count
Number of failed rollbacks: $failure_count
Success rate: $success_percentage%
Failure rate: $failure_percentage%
----------------------------------------
EOF
)
 
if [ $failure_count -gt 0 ]; then
    summary_content+="Failed servers:\n"
    for server in "${failed_servers[@]}"; do
        summary_content+="  - $server\n"
    done
fi
summary_content+="========================================"
 
# Output summary result to log file and terminal
echo -e "$summary_content" | tee -a "$SUMMARY_LOG"
 
log_info "Script execution completed. Check $SUMMARY_LOG for summary."

根据实际情况，修改脚本中的以下配置项参数：
- LOG_DIR="/var/log/rollback_hosts"
  - 描述：日志目录路径。
  - 参数默认值：/var/log/rollback_hosts
  - 修改建议：修改为当前用户具有写入权限的目录路径。
  - 修改示例：LOG_DIR="/home/username/rollback_hosts_logs"
- CSV_FILE="target_servers.csv"
  - 描述：CSV 文件路径，包含源端主机信息。
  - 参数默认值：target_servers.csv
  - 修改建议：使用绝对路径，或确保相对路径是正确的。
  - 修改示例：CSV_FILE="/home/username/configs/servers.csv"
配置项参数修改完成并保存后，在终端窗口（如果有图形化界面可以按Ctrl+Alt+T打开终端）执行脚本：
```
./rollback_hosts_linux.sh
```
脚本会在终端窗口中输出日志信息，并在执行完毕后生成一个执行结果报告，可以在LOG_DIR 指定目录中的 summary.log 文件中查看。

父主题： 配置脚本

上一篇：配置update_hosts_linux.sh脚本

下一篇：批量修改与还原源端Windows主机的Host配置

意见反馈

文档内容是否对您有帮助？

有帮助没帮助

提供反馈

提交成功！非常感谢您的反馈，我们会继续努力做到更好！您可在我的云声建议查看反馈及问题处理状态。

系统繁忙，请稍后重试

在使用文档中是否遇到以下问题

内容与产品页面不一致

内容不易理解

缺失示例代码

步骤不可操作

搜不到想要的内容

缺少最佳实践

意见反馈（选填）

0/500

请至少选择一项反馈信息并填写问题反馈

字符长度不能超过500

直接提交取消

如您有其它疑问，您也可以通过华为云社区问答频道来与我们联系探讨

智能客服提问云社区提问

配置rollback_hosts_linux.sh脚本

前提条件

操作步骤

相关文档

意见反馈

文档内容是否对您有帮助？

7*24

备案

专业服务

退订

建议反馈

售前咨询热线