105、Replicated Mode VS Global Mode (Swarm12)

 
Swarm 可以在 Service 创建和运行过程中灵活的通过 --replicas 调整容器副本数量,内部调度器则会根据当前集群的资源使用情况在不同的node 上启停容器,这就是 Service 默认的 replicated mode 。在此模式下,node 上运行的副本数有多有少,一般情况下,资源更丰富的node 运行的副本数更多,反之亦然。
 
除了 replicated mode ,Service 还提供了一个 global mode ,其作用是强制在每个node上都运行一个且最多一个副本。
 
global mode 特别适合需要运行daemon 的集群环境。比如要收集所有容器的日志,就可以用 global mode 创建 Service,在所有 node 上都运行 gliderlabs/logspout 容器,及时之后有新的node加入,swarm 也会自动在新的node上启动一个 gliderlabs/logspout 副本
 
root@host03:~# docker service create --mode global --name bbox busybox sleep 99999
fvbo6fpu7gb29qi0kl2ausunt
overall progress: 2 out of 2 tasks
asn5ufnogzky: running   
hvt2ez9e7zvq: running   
verify: Service converged
root@host03:~# docker service ps bbox
ID                  NAME                             IMAGE               NODE                DESIRED STATE       CURRENT STATE            ERROR               PORTS
9zo4pzhl7ysf        bbox.asn5ufnogzkyqigk4mizatoer   busybox:latest      host02              Running             Running 10 seconds ago                       
xuxima4u6am4        bbox.hvt2ez9e7zvqm2hz8nix1eke7   busybox:latest      host01              Running             Running 10 seconds ago                       
root@host03:~# docker node ls
ID                            HOSTNAME            STATUS              AVAILABILITY        MANAGER STATUS      ENGINE VERSION
hvt2ez9e7zvqm2hz8nix1eke7     host01              Ready               Active                                  18.09.6
asn5ufnogzkyqigk4mizatoer     host02              Ready               Active                                  18.09.6
h6rzavsz2vjxstwj3pytiebjb *   host03              Ready               Drain               Leader              18.09.6
root@host03:~# docker node update host03 --availability active
host03
root@host03:~# docker node ls
ID                            HOSTNAME            STATUS              AVAILABILITY        MANAGER STATUS      ENGINE VERSION
hvt2ez9e7zvqm2hz8nix1eke7     host01              Ready               Active                                  18.09.6
asn5ufnogzkyqigk4mizatoer     host02              Ready               Active                                  18.09.6
h6rzavsz2vjxstwj3pytiebjb *   host03              Ready               Active              Leader              18.09.6
root@host03:~# docker service ps bbox
ID                  NAME                             IMAGE               NODE                DESIRED STATE       CURRENT STATE                ERROR               PORTS
j4bmhr5yqokk        bbox.h6rzavsz2vjxstwj3pytiebjb   busybox:latest      host03              Running             Preparing 5 seconds ago                          
9zo4pzhl7ysf        bbox.asn5ufnogzkyqigk4mizatoer   busybox:latest      host02              Running             Running about a minute ago                       
xuxima4u6am4        bbox.hvt2ez9e7zvqm2hz8nix1eke7   busybox:latest      host01              Running             Running about a minute ago          
 
root@host03:~# docker node update host03 --availability drain
host03
root@host03:~# docker node ls
ID                            HOSTNAME            STATUS              AVAILABILITY        MANAGER STATUS      ENGINE VERSION
hvt2ez9e7zvqm2hz8nix1eke7     host01              Ready               Active                                  18.09.6
asn5ufnogzkyqigk4mizatoer     host02              Ready               Active                                  18.09.6
h6rzavsz2vjxstwj3pytiebjb *   host03              Ready               Drain               Leader              18.09.6
root@host03:~# docker service ps bbox
ID                  NAME                             IMAGE               NODE                DESIRED STATE       CURRENT STATE                     ERROR               PORTS
j4bmhr5yqokk        bbox.h6rzavsz2vjxstwj3pytiebjb   busybox:latest      host03              Shutdown            Shutdown less than a second ago                       
9zo4pzhl7ysf        bbox.asn5ufnogzkyqigk4mizatoer   busybox:latest      host02              Running             Running 3 minutes ago                                 
xuxima4u6am4        bbox.hvt2ez9e7zvqm2hz8nix1eke7   busybox:latest      host01              Running             Running 3 minutes ago                                 
 
上面实验验证了 global mode 下,当有新的node加入时(这里用host03 解除 Drain演示),会自动启动一个副本。而当host03再次标记为 Drain时,host03上的副本会被shutdown
 
无论我们采用哪种mode ,我们都无法精细的控制副本运行在哪个节点上。
posted @ 2019-05-15 22:59  三角形  阅读(350)  评论(0编辑  收藏  举报