10.40.100.143 docker-node0.venic.com

10.40.100.144 docker-node1.venic.com

 

 

原用swarm镜像

 

直接启动

 

 

10.40.100.143上

 docker run -d  --restart=always --name node0 docker-registry.venic.com:5000/swarm:latest join --advertise=10.40.100.143:2375 consul://docker-consul.venic.com:8500

 

10.40.100.144上

 docker run -d  --restart=always --name node1   docker-registry.venic.com:5000/swarm:latest join --advertise=10.40.100.144:2375 consul://docker-consul.venic.com:8500

 

 

在Manager节点上查看集群状态，多了2个node节点

 

 

 

# docker -H :4000 info

Containers: 2

 Running: 2

 Paused: 0

 Stopped: 0

Images: 6

Server Version: swarm/1.2.5

Role: primary

Strategy: spread --默认的均衡模式

Filters: health, port, containerslots, dependency, affinity, constraint

Nodes: 2

 docker-node0.venic.com: 10.40.100.143:2375

  └ ID: IP3U:TRP7:ZJTS:L3SX:UU4R:VVB5:B4LE:QNOL:MALC:A4BX:GFC7:WJPV

  └ Status: Healthy

  └ Containers: 1 (1 Running, 0 Paused, 0 Stopped)

  └ Reserved CPUs: 0 / 1

  └ Reserved Memory: 0 B / 2.052 GiB

  └ Labels: kernelversion=3.10.0-514.2.2.el7.x86_64, operatingsystem=CentOS Linux 7 (Core), storagedriver=devicemapper

  └ UpdatedAt: 2016-12-22T11:26:27Z

  └ ServerVersion: 1.12.5

 docker-node1.venic.com: 10.40.100.144:2375

  └ ID: TXYV:2VVM:H32T:DWDN:AD5M:Z22S:RRLJ:KYN2:SKNP:YFLI:4RUI:GRPM

  └ Status: Healthy

  └ Containers: 1 (1 Running, 0 Paused, 0 Stopped)

  └ Reserved CPUs: 0 / 1

  └ Reserved Memory: 0 B / 2.052 GiB

  └ Labels: kernelversion=3.10.0-514.2.2.el7.x86_64, operatingsystem=CentOS Linux 7 (Core), storagedriver=devicemapper

  └ UpdatedAt: 2016-12-22T11:26:35Z

  └ ServerVersion: 1.12.5

Plugins:

 Volume:

 Network:

Swarm:

 NodeID:

 Is Manager: false

 Node Address:

Security Options:

Kernel Version: 3.10.0-514.2.2.el7.x86_64

Operating System: linux

Architecture: amd64

CPUs: 2

Total Memory: 4.103 GiB

Name: 27fbbfe0fae4

Docker Root Dir:

Debug Mode (client): false

Debug Mode (server): false

WARNING: No kernel memory limit support

 

 

Swarm调度策略

---

Swarm在schedule节点运行容器的时候，会根据指定的策略来计算最适合运行容器的节点，目前支持的策略有：spread, binpack, random.

Random顾名思义，就是随机选择一个Node来运行容器，一般用作调试用，spread和binpack策略会根据各个节点的可用的CPU, RAM以及正在运行的容器的数量来计算应该运行容器的节点。

在同等条件下，Spread策略会选择运行容器最少的那台节点来运行新的容器，binpack策略会选择运行容器最集中的那台机器来运行新的节点(The binpack strategy causes Swarm to optimize for the Container which is most packed.)。

使用Spread策略会使得容器会均衡的分布在集群中的各个节点上运行，一旦一个节点挂掉了只会损失少部分的容器。