2. 首页 > Rancher教程 > 正文

Rancher教程FG028-Rancher灾备方案与故障切换演练实战

教程整理：风哥教程  |  更新时间：2026-03-13  |  教程分类：Rancher教程  |  文档学习：9

本篇文章详细介绍Rancher灾备方案设计与故障切换演练，包括备份策略、恢复流程、灾难演练、故障切换等实战内容。风哥教程参考Rancher官方文档备份恢复与高可用相关章节。

目录大纲

Part01-基础概念与理论知识

1.1 灾备架构设计

Rancher灾备架构包括主备架构、双活架构、多活架构。主备架构：主节点提供服务，备节点待命；双活架构：两个节点同时提供服务；多活架构：多个节点同时提供服务。灾备级别：RPO（恢复点目标）、RTO（恢复时间目标）。更多视频教程www.fgedu.net.cn

1.2 故障切换机制

故障切换包括自动切换和手动切换。自动切换：通过Keepalived、HAProxy等实现VIP漂移；手动切换：通过DNS切换、负载均衡配置实现。故障检测：心跳检测、健康检查、监控告警。切换流程：故障检测→决策→切换→验证。学习交流加群风哥微信: itpux-com

Part02-生产环境规划与建议

2.1 备份策略规划

备份策略包括：全量备份、增量备份、差异备份。备份频率：ETCD每小时一次，Rancher每天一次，应用数据根据业务需求。备份保留：保留7天、30天、90天备份。备份存储：本地存储、异地存储、云存储。学习交流加群风哥QQ113257174

2.2 灾备方案设计

灾备方案设计原则：高可用、可恢复、可验证。灾备级别：冷备（数据备份）、温备（系统备份）、热备（实时同步）。灾备演练：每季度进行一次灾备演练，验证备份恢复流程。灾备文档：建立详细的灾备文档和操作手册。更多学习教程公众号风哥教程itpux_com

Part03-生产环境项目实施方案

3.1 ETCD备份配置

配置ETCD自动备份。


















> /Rancher/fgdata/etcd-backup/etcd_backup.log 2>&1










Snapshot saved at /Rancher/fgdata/etcd-backup/etcd-backup-20260410-210000.db
Hash: 1234567890abcdef1234567890abcdef1234567890abcdef1234567890abcdef
Revision: 123456
Total key: 1234
Total size: 123456789
ETCD备份完成: /Rancher/fgdata/etcd-backup/etcd-backup-20260410-210000.db







total 500M
-rw-r--r-- 1 root root 120M Apr 10 21:00 etcd-backup-20260410-210000.db
-rw-r--r-- 1 root root 120M Apr 10 20:00 etcd-backup-20260410-200000.db
-rw-r--r-- 1 root root 120M Apr 10 19:00 etcd-backup-20260410-190000.db
-rw-r--r-- 1 root root 120M Apr 10 18:00 etcd-backup-20260410-180000.db


from Rancher视频:www.itpux.com


3.2 Rancher备份配置


配置Rancher数据备份。



 $BACKUP_FILE

# 验证备份
if [ -f $BACKUP_FILE ]; then
    echo “Rancher备份成功: $BACKUP_FILE”
    ls -lh $BACKUP_FILE
else
    echo “Rancher备份失败”
    exit 1
fi

# 清理30天前的备份
find $BACKUP_DIR -name “rancher-backup-*.tar.gz” -mtime +30 -delete




















> /Rancher/fgdata/rancher-backup/rancher_backup.log 2>&1










Rancher备份成功: /Rancher/fgdata/rancher-backup/rancher-backup-20260410-210000.tar.gz
-rw-r--r-- 1 root root 500M Apr 10 21:00 /Rancher/fgdata/rancher-backup/rancher-backup-20260410-210000.tar.gz







total 2.0G
-rw-r--r-- 1 root root 500M Apr 10 21:00 rancher-backup-20260410-210000.tar.gz
-rw-r--r-- 1 root root 500M Apr  9 02:00 rancher-backup-20260409-020000.tar.gz
-rw-r--r-- 1 root root 500M Apr  8 02:00 rancher-backup-20260408-020000.tar.gz
-rw-r--r-- 1 root root 500M Apr  7 02:00 rancher-backup-20260407-020000.tar.gz


3.3 故障切换配置


配置Keepalived实现VIP故障切换。







Loaded plugins: fastestmirror
Loading mirror speeds from cached hostfile
 * base: mirrors.aliyun.com
 * extras: mirrors.aliyun.com
 * updates: mirrors.aliyun.com
Resolving Dependencies
--> Running transaction check
---> Package keepalived.x86_64 0:2.1.5-8.el9 will be installed
--> Finished Dependency Resolution
Dependencies Resolved

=======================================================================================================================================================
 Package                        Arch                      Version                         Repository                 Size
=======================================================================================================================================================
Installing:
 keepalived                     x86_64                    2.1.5-8.el9                     base                       432 k

Transaction Summary
=======================================================================================================================================================
Install  1 Package

Total download size: 432 k
Installed size: 1.1 M
Downloading packages:
keepalived-2.1.5-8.el9.x86_64.rpm                                                                                         | 432 kB  00:00:00
Running transaction check
Running transaction test
Transaction test succeeded
Running transaction
  Installing : keepalived-2.1.5-8.el9.x86_64                                                                                             1/1
  Verifying  : keepalived-2.1.5-8.el9.x86_64                                                                                             1/1

Installed:
  keepalived.x86_64 0:2.1.5-8.el9

Complete!



 /etc/keepalived/keepalived.conf <






 /etc/keepalived/keepalived.conf <










Created symlink /etc/systemd/system/multi-user.target.wants/keepalived.service → /usr/lib/systemd/system/keepalived.service.







● keepalived.service - LVS and VRRP High Availability Monitor
   Loaded: loaded (/usr/lib/systemd/system/keepalived.service; enabled; vendor preset: disabled)
   Active: active (running) since Fri 2026-04-10 21:00:00 CST; 5s ago
  Process: 12345 ExecStart=/usr/sbin/keepalived $KEEPALIVED_OPTIONS (code=exited, status=0/SUCCESS)
 Main PID: 12346 (keepalived)
    Tasks: 3 (limit: 4915)
   Memory: 2.1M
   CGroup: /system.slice/keepalived.service
           ├─12346 /usr/sbin/keepalived -D
           ├─12347 /usr/sbin/keepalived -D
           └─12348 /usr/sbin/keepalived -D

Apr 10 21:00:00 fgedu.net.cn Keepalived[12346]: Starting VRRP child process, pid=12348
Apr 10 21:00:00 fgedu.net.cn Keepalived_vrrp[12348]: (VI_1) Entering MASTER STATE
Apr 10 21:00:00 fgedu.net.cn Keepalived_vrrp[12348]: (VI_1) setting VIPs.
Apr 10 21:00:00 fgedu.net.cn Keepalived_vrrp[12348]: (VI_1) Sending gratuitous ARPs on eth0 for 192.168.1.200
Apr 10 21:00:00 fgedu.net.cn Keepalived_vrrp[12348]: (VI_1) Sending gratuitous ARPs on eth0 for 192.168.1.200
Apr 10 21:00:00 fgedu.net.cn systemd[1]: Started Keepalived service.







2: eth0:  mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 00:11:22:33:44:55 brd ff:ff:ff:ff:ff:ff
    inet 192.168.1.100/24 brd 192.168.1.255 scope global eth0
       valid_lft forever preferred_lft forever
    inet 192.168.1.200/32 scope global eth0
       valid_lft forever preferred_lft forever
    inet6 fe80::211:22ff:fe33:4455/64 scope link
       valid_lft forever preferred_lft forever






Part04-生产案例与实战讲解


4.1 灾备演练实战


执行灾备演练验证备份恢复流程。














Apr 10 21:30:00 fgedu.net.cn Keepalived_vrrp[12348]: (VI_1) Master received advert with higher priority 100, ours 90
Apr 10 21:30:00 fgedu.net.cn Keepalived_vrrp[12348]: (VI_1) Entering BACKUP STATE
Apr 10 21:30:00 fgedu.net.cn Keepalived_vrrp[12348]: (VI_1) removing VIPs.
Apr 10 21:30:01 fgedu.net.cn Keepalived_vrrp[12348]: (VI_1) Master received advert with higher priority 100, ours 90
Apr 10 21:30:02 fgedu.net.cn Keepalived_vrrp[12348]: (VI_1) Entering BACKUP STATE
Apr 10 21:30:03 fgedu.net.cn Keepalived_vrrp[12348]: (VI_1) removing VIPs.







2: eth0:  mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 00:11:22:33:44:66 brd ff:ff:ff:ff:ff:ff
    inet 192.168.1.101/24 brd 192.168.1.255 scope global eth0
       valid_lft forever preferred_lft forever
    inet 192.168.1.200/32 scope global eth0
       valid_lft forever preferred_lft forever
    inet6 fe80::211:22ff:fe33:4466/64 scope link
       valid_lft forever preferred_lft forever







pong


4.2 故障切换实战


执行故障切换演练。














● rancher-server.service - Rancher Server
   Loaded: loaded (/usr/lib/systemd/system/rancher-server.service; enabled; vendor preset: disabled)
   Active: active (running) since Fri 2026-04-10 21:00:00 CST; 30min ago
  Process: 12345 ExecStart=/usr/local/bin/docker run --name rancher --restart=unless-stopped -d --privileged -v /var/lib/rancher:/var/lib/rancher -p 80:80 -p 443:443 rancher/rancher:latest (code=exited, status=0/SUCCESS)
 Main PID: 12346 (docker run)
    Tasks: 0 (limit: 4915)
   Memory: 500M
   CGroup: /system.slice/rancher-server.service
           └─12346 /usr/bin/docker run --name rancher --restart=unless-stopped -d --privileged -v /var/lib/rancher:/var/lib/rancher -p 80:80 -p 443:443 rancher/rancher:latest

Apr 10 21:00:00 fgedu.net.cn systemd[1]: Started Rancher Server.







NAME                                    READY   STATUS    RESTARTS   AGE
cattle-cluster-agent-abc123def456-ghi78   1/1     Running   0          30m
cattle-node-agent-jkl012mno345           1/1     Running   0          30m














● rancher-server.service - Rancher Server
   Loaded: loaded (/usr/lib/systemd/system/rancher-server.service; enabled; vendor preset: disabled)
   Active: active (running) since Fri 2026-04-10 21:35:00 CST; 5s ago
  Process: 12345 ExecStart=/usr/local/bin/docker run --name rancher --restart=unless-stopped -d --privileged -v /var/lib/rancher:/var/lib/rancher -p 80:80 -p 443:443 rancher/rancher:latest (code=exited, status=0/SUCCESS)
 Main PID: 12346 (docker run)
    Tasks: 0 (limit: 4915)
   Memory: 500M
   CGroup: /system.slice/rancher-server.service
           └─12346 /usr/bin/docker run --name rancher --restart=unless-stopped -d --privileged -v /var/lib/rancher:/var/lib/rancher -p 80:80 -p 443:443 rancher/rancher:latest

Apr 10 21:35:00 fgedu.net.cn systemd[1]: Started Rancher Server.


4.3 恢复验证实战


验证备份恢复流程。





















2026-04-10 21:40:00.123456 I | snapshot restored successfully














https://192.168.1.100:2379 is healthy: successfully committed proposal: took = 12.345678ms







NAME            STATUS   ROLES    AGE   VERSION
fgedu-node-1    Ready       30d   v1.28.5
fgedu-node-2    Ready       30d   v1.28.5
fgedu-node-3    Ready       30d   v1.28.5






Part05-风哥经验总结与分享


5.1 生产环境最佳实践


1. 定期执行备份和恢复演练

            2. 使用多个备份存储位置

            3. 配置监控告警及时发现问题

            4. 建立详细的灾备文档和操作手册

            5. 使用自动化工具简化灾备流程

            6. 定期测试故障切换流程

            7. 配置合理的RPO和RTO目标

            8. 建立灾备团队和应急响应流程


5.2 常见问题与解决方案


1. 备份失败：检查磁盘空间、验证备份脚本

            2. 恢复失败：验证备份完整性、检查恢复环境

            3. 故障切换失败：检查网络连通性、验证VIP配置

            4. 数据不一致：使用增量备份、验证数据完整性

            5. 演练影响生产：使用测试环境、选择低峰时段

            6. 恢复时间过长：优化恢复流程、增加资源

            7. 备份存储不足：清理旧备份、扩展存储容量

            8. 文档不完整：定期更新文档、建立知识库



													
						

							
本文由风哥教程整理发布,仅用于学习测试使用,转载注明出处:http://www.fgedu.net.cn/10327.html

						

					

					

						
      

						
相关推荐

						
					

				

			

		

		

	







  

    

      
联系我们

      
在线咨询：点击这里给我发消息

      
微信号：itpux-com

      
工作日：9:30-18:30，节假日休息