098、Swarm 如何实现 Failover （Swarm05）

参考 https://www.cnblogs.com/CloudMan6/p/7898245.html

 

故障是在所难免的，容器可能崩溃，Docker Host 可能宕机，不过幸运的是，Swarm 已经内置了 failover策略。

 

创建Service 的时候，我们没有告诉 swarm 发生故障时该如何处理，只是说明了我们期望的状态（比如 3 份副本），swarm会尽最大努力达成这个期望的状态，无论发生什么状况。

 

root@host03:~# docker service ps web_server

ID                  NAME                IMAGE               NODE                DESIRED STATE       CURRENT STATE               ERROR               PORTS

tznw0lrn8wh8        web_server.1        httpd:latest        host01              Running             Running 28 minutes ago                          

jykct1jmfrte         \_ web_server.1    httpd:latest        host03              Shutdown            Shutdown 28 minutes ago                         

4x32c9x1hizg        web_server.2        httpd:latest        host01              Running             Running 28 minutes ago                          

n4afxlx16tny         \_ web_server.2    httpd:latest        host03              Shutdown            Shutdown 28 minutes ago                         

mlsb1ey4n65r        web_server.4        httpd:latest        host02              Running             Running about an hour ago    

 

当前 3 个副本运行在 host01（2副本） 和 host02 （1副本）上 ，现在我们测试swarm 的failover特性，host01 关机 

 

root@host03:~# docker service ps web_server    #    host01 关机前的状态

ID                  NAME                IMAGE               NODE                DESIRED STATE       CURRENT STATE             ERROR               PORTS

tznw0lrn8wh8        web_server.1        httpd:latest        host01              Running             Running 12 seconds ago                        

jykct1jmfrte         \_ web_server.1    httpd:latest        host03              Shutdown            Shutdown 44 minutes ago                       

4x32c9x1hizg        web_server.2        httpd:latest        host01              Running             Running 12 seconds ago                        

n4afxlx16tny         \_ web_server.2    httpd:latest        host03              Shutdown            Shutdown 44 minutes ago                       

mlsb1ey4n65r        web_server.4        httpd:latest        host02              Running             Running 2 hours ago                           

root@host03:~# docker service ps web_server    #    swarm检测到host01 关机，开始在host02上启动新的容器

ID                  NAME                IMAGE               NODE                DESIRED STATE       CURRENT STATE                  ERROR               PORTS

wtipc3fngioo        web_server.1        httpd:latest        host02              Ready               Ready less than a second ago                       

tznw0lrn8wh8         \_ web_server.1    httpd:latest        host01              Shutdown            Running 13 seconds ago                             

jykct1jmfrte         \_ web_server.1    httpd:latest        host03              Shutdown            Shutdown 44 minutes ago                            

vhp70jgq49j8        web_server.2        httpd:latest        host02              Ready               Ready less than a second ago                       

4x32c9x1hizg         \_ web_server.2    httpd:latest        host01              Shutdown            Running 13 seconds ago                             

n4afxlx16tny         \_ web_server.2    httpd:latest        host03              Shutdown            Shutdown 44 minutes ago                            

mlsb1ey4n65r        web_server.4        httpd:latest        host02              Running             Running 2 hours ago                                

root@host03:~# docker service ps web_server    #    故障转移完毕，3个副本都运行在了host02 上

ID                  NAME                IMAGE               NODE                DESIRED STATE       CURRENT STATE             ERROR               PORTS

wtipc3fngioo        web_server.1        httpd:latest        host02              Running             Running 5 seconds ago                         

tznw0lrn8wh8         \_ web_server.1    httpd:latest        host01              Shutdown            Running 24 seconds ago                        

jykct1jmfrte         \_ web_server.1    httpd:latest        host03              Shutdown            Shutdown 44 minutes ago                       

vhp70jgq49j8        web_server.2        httpd:latest        host02              Running             Running 5 seconds ago                         

4x32c9x1hizg         \_ web_server.2    httpd:latest        host01              Shutdown            Running 24 seconds ago                        

n4afxlx16tny         \_ web_server.2    httpd:latest        host03              Shutdown            Shutdown 44 minutes ago                       

mlsb1ey4n65r        web_server.4        httpd:latest        host02              Running             Running 2 hours ago                           

root@host03:~# docker node ls    #    host01 已被标记为 Down

ID                            HOSTNAME            STATUS              AVAILABILITY        MANAGER STATUS      ENGINE VERSION

hvt2ez9e7zvqm2hz8nix1eke7     host01              Down                Active                                  18.09.6

asn5ufnogzkyqigk4mizatoer     host02              Ready               Active                                  18.09.6

h6rzavsz2vjxstwj3pytiebjb *   host03              Ready               Drain               Leader              18.09.6

 

 

转载于:https://www.cnblogs.com/www1707/p/10872735.html