$ cat > disaster_recovery_tech_eval.txt << EOF 评估因素 | 权重 | 备份技术 | 复制技术 | 虚拟化容灾 | 云容灾 ---------|------|----------|----------|------------|-------- RTO要求 | 30% | 低 | 高 | 高 | 中 RPO要求 | 30% | 低 | 高 | 高 | 中 成本 | 20% | 低 | 高 | 中 | 中 实施复杂度 | 10% | 低 | 高 | 中 | 低 维护难度 | 10% | 低 | 中 | 中 | 低 总分 | 100% | 60 | 85 | 80 | 75 EOF # 技术选择建议 $ cat > tech_selection_recommendation.txt << EOF 业务类型 | 推荐技术 | 理由 ---------|----------|------ 关键业务 | 复制技术 | RTO和RPO要求高，需要实时数据保护 重要业务 | 虚拟化容灾 | 平衡成本和容灾能力 一般业务 | 备份技术 | 成本低，满足基本容灾需求 弹性业务 | 云容灾 | 按需扩展，灵活性高 EOF

# 步骤1：配置备份策略
$ cat /etc/backup.conf
# 备份配置文件
BACKUP_DIR=”/backup”
RETENTION_DAYS=”30″
FULL_BACKUP_SCHEDULE=”0 2 * * 0″ # 每周日凌晨2点
INCREMENTAL_BACKUP_SCHEDULE=”0 2 * * 1-6″ # 周一至周六凌晨2点

# 步骤2：创建备份脚本
$ cat /usr/bin/backup.sh
#!/bin/bash

source /etc/backup.conf
TIMESTAMP=$(date +”%Y%m%d_%H%M%S”)

# 创建备份目录
mkdir -p “$BACKUP_DIR/$(date +”%Y%m%d”)”

# 执行备份
if [ “$1” = “full” ]; then
echo “执行全量备份…”
tar -czf “$BACKUP_DIR/$(date +”%Y%m%d”)/full_$TIMESTAMP.tar.gz” /data
echo “全量备份完成: $BACKUP_DIR/$(date +”%Y%m%d”)/full_$TIMESTAMP.tar.gz”
# 更新最新全量备份链接
ln -sf “$BACKUP_DIR/$(date +”%Y%m%d”)/full_$TIMESTAMP.tar.gz” “$BACKUP_DIR/latest_full.tar.gz”
elif [ “$1” = “incremental” ]; then
echo “执行增量备份…”
tar -czf “$BACKUP_DIR/$(date +”%Y%m%d”)/incremental_$TIMESTAMP.tar.gz” /data –newer=”$BACKUP_DIR/latest_full.tar.gz”
echo “增量备份完成: $BACKUP_DIR/$(date +”%Y%m%d”)/incremental_$TIMESTAMP.tar.gz”
fi

# 清理过期备份
find “$BACKUP_DIR” -name “*.tar.gz” -mtime +$RETENTION_DAYS -delete

# 步骤1：配置MySQL主从复制
$ cat /etc/my.cnf
[mysqld]
server-id = 1
log-bin = mysql-bin
binlog-format = ROW
sync-binlog = 1
innodb_flush_log_at_trx_commit = 1

# 步骤2：配置从服务器
$ cat /etc/my.cnf
[mysqld]
server-id = 2
relay-log = relay-bin
read-only = 1

# 步骤3：初始化主从复制
$ mysql -u root -p
# 在主库创建复制用户
CREATE USER ‘repl’@’%’ IDENTIFIED BY ‘password’;
GRANT REPLICATION SLAVE ON *.* TO ‘repl’@’%’;
FLUSH PRIVILEGES;

# 锁定主库并获取备份
FLUSH TABLES WITH READ LOCK;
SHOW MASTER STATUS;
# 记录File和Position值

# 备份主库
$ mysqldump -u root -p –all-databases > full_backup.sql

# 解锁主库
UNLOCK TABLES;

# 在从库恢复备份
$ mysql -u root -p < full_backup.sql # 配置从库连接主库 $ mysql -u root -p CHANGE MASTER TO MASTER_HOST='master_ip', MASTER_USER='repl', MASTER_PASSWORD='password', MASTER_LOG_FILE='mysql-bin.000001', MASTER_LOG_POS=107; START SLAVE; # 检查复制状态 $ mysql -u root -p -e "SHOW SLAVE STATUS\G"

# 步骤1：配置VMware Site Recovery Manager (SRM)

# 安装SRM服务器
$ ./VMware-srm-8.3.0.exe

# 配置SRM站点对
$ srm-admin.exe configure

# 步骤2：创建保护组
$ srm-admin.exe create-protection-group –name “Production VMs” –vms “VM1,VM2,VM3”

# 步骤3：创建恢复计划
$ srm-admin.exe create-recovery-plan –name “DR Plan” –protection-group “Production VMs”

# 步骤4：测试恢复计划
$ srm-admin.exe test-recovery-plan –name “DR Plan”

# 步骤5：执行故障转移
$ srm-admin.exe execute-recovery-plan –name “DR Plan” –failover

# 步骤1：配置AWS Backup

# 创建备份计划
$ aws backup create-backup-plan –backup-plan ‘{“BackupPlanName”: “Production-Backup-Plan”, “Rules”: [{“RuleName”: “Daily-Backup”, “TargetBackupVaultName”: “Production-Backup-Vault”, “ScheduleExpression”: “cron(0 2 * * ? *)”, “StartWindowMinutes”: 60, “CompletionWindowMinutes”: 10080, “Lifecycle”: {“DeleteAfterDays”: 30}}]}’

# 步骤2：配置AWS Disaster Recovery

# 创建CloudFormation堆栈
$ aws cloudformation create-stack –stack-name dr-stack –template-url https://s3.amazonaws.com/aws-quickstart/quickstart-aws-disaster-recovery/templates/aws-disaster-recovery-master.template.yaml –parameters ParameterKey=PrimaryRegion,ParameterValue=us-east-1 ParameterKey=SecondaryRegion,ParameterValue=us-west-2

# 步骤3：配置自动扩展组
$ aws autoscaling create-auto-scaling-group –auto-scaling-group-name dr-asg –launch-configuration-name dr-lc –min-size 1 –max-size 3 –desired-capacity 1 –availability-zones us-west-2a

# 配置Prometheus监控容灾状态
$ cat /etc/prometheus/prometheus.yml
scrape_configs:
– job_name: ‘disaster_recovery’
static_configs:
– targets: [‘dr-monitor:9100’]
metrics_path: ‘/metrics’

# 配置Grafana仪表板
$ cat > dr-dashboard.json << EOF { "dashboard": { "id": null, "title": "Disaster Recovery Status", "panels": [ { "title": "Replication Status", "type": "gauge", "targets": [ { "expr": "mysql_slave_running" } ] }, { "title": "Backup Status", "type": "gauge", "targets": [ { "expr": "backup_success" } ] } ] } } EOF # 导入Grafana仪表板 $ curl -X POST -H "Content-Type: application/json" -d @dr-dashboard.json http://grafana:3000/api/dashboards/db

# 配置Ansible playbook
$ cat dr-playbook.yml
—
– hosts: dr_servers
tasks:
– name: 检查复制状态
shell: mysql -u root -p{{ mysql_password }} -e “SHOW SLAVE STATUS\G” | grep “Slave_IO_Running\|Slave_SQL_Running”
register: replication_status

– name: 启动故障转移
shell: /usr/bin/failover.sh
when: “‘Yes’ not in replication_status.stdout”

– name: 发送告警
mail:
to: admin@fgedu.net.cn
subject: “容灾系统故障转移”
body: “检测到主系统故障，已启动故障转移流程”
when: “‘Yes’ not in replication_status.stdout”

# 配置Cron定时执行
$ crontab -e
*/5 * * * * ansible-playbook /etc/ansible/dr-playbook.yml

# 步骤1：制定测试计划
$ cat > dr_test_plan.txt << EOF 测试名称: 容灾系统完整性测试 测试目的: 验证容灾系统的恢复能力 测试步骤: 1. 模拟主系统故障 2. 执行故障转移 3. 验证备用系统运行状态 4. 验证数据一致性 5. 执行回切操作 测试时间: 每月第一个周末 测试人员: 系统管理员、数据库管理员 EOF # 步骤2：执行测试 $ cat /usr/bin/dr_test.sh #!/bin/bash # 记录测试开始时间 start_time=$(date +%s) echo "[$(date)] 开始容灾测试" # 步骤1：模拟主系统故障 echo "[$(date)] 模拟主系统故障" systemctl stop mysql # 步骤2：执行故障转移 echo "[$(date)] 执行故障转移" /usr/bin/failover.sh # 步骤3：验证备用系统运行状态 echo "[$(date)] 验证备用系统运行状态" sleep 60 if systemctl is-active mysql; then echo "[$(date)] 备用系统启动成功" else echo "[$(date)] 备用系统启动失败" exit 1 fi # 步骤4：验证数据一致性 echo "[$(date)] 验证数据一致性" mysql -u root -p -e "SELECT COUNT(*) FROM test_db.test_table;" # 步骤5：执行回切操作 echo "[$(date)] 执行回切操作" /usr/bin/failback.sh # 记录测试结束时间 end_time=$(date +%s) test_duration=$((end_time - start_time)) echo "[$(date)] 容灾测试完成，耗时: $test_duration 秒" # 生成测试报告 echo "测试完成，耗时: $test_duration 秒" > dr_test_report.txt

$ cat > dr_maintenance_plan.txt << EOF 日常维护（每日）: - 检查复制状态 - 验证备份完成情况 - 检查容灾系统日志 周维护（每周）: - 测试备份恢复 - 检查容灾系统资源使用情况 - 更新容灾系统配置 月维护（每月）: - 执行完整容灾测试 - 清理过期备份 - 检查容灾系统性能 季度维护（每季度）: - 评估容灾系统有效性 - 更新容灾计划 - 培训相关人员 年度维护（每年）: - 全面容灾演练 - 容灾系统升级 - 容灾策略评估与更新 EOF