?swarm是docker公司自己的容器集群管理工具，本文介紹了使用docker 1.12 搭建多主機docker swarm集群，分享給大家
準備
- 準備至少兩臺的centos 7 主機(全新最小安裝, 可以使用虛擬機安裝)
- 開放端口2377 tcp端口, 7946 4789 tcp udp 端口
- 本文使用192.168.99.101(hostname:centos-node4) 作為swarm manager
- 192.168.99.102(hostname:centos-node5) 作為swarm agent1
安裝docker engine 1.12
在每臺機器上執行以下命令
|
1
2
3
4
5
6
7
8
9
10
11
12
|
# sudo tee /etc/yum.repos.d/docker.repo <<-'EOF'[dockerrepo]name=Docker Repository baseurl=https://yum.dockerproject.org/repo/main/centos/7/enabled=1 gpgcheck=1 gpgkey=https://yum.dockerproject.org/gpgEOF# sudo yum install docker-engine# sudo systemctl enable docker# sudo systemctl start docker |
安裝完后查看Docker 版本
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
|
[root@centos-node4 ~]# docker version Client: Version: 1.12.0 API version: 1.24 Go version: go1.6.3 Git commit: 8eab29e Built: OS/Arch: linux/amd64Server: Version: 1.12.0 API version: 1.24 Go version: go1.6.3 Git commit: 8eab29e Built: OS/Arch: linux/amd64 |
開放端口相關命令
|
1
2
3
4
5
6
|
firewall-cmd --zone=public --add-port=2377/tcp --permanent firewall-cmd --zone=public --add-port=7946/tcp --permanent firewall-cmd --zone=public --add-port=7946/udp --permanent firewall-cmd --zone=public --add-port=4789/tcp --permanent firewall-cmd --zone=public --add-port=4789/udp --permanent firewall-cmd --reload |
新版docker swarm 命令詳情
有關集群的docker命令如下：
- docker swarm：集群管理，子命令有init, join,join-token, leave, update
- docker node：節點管理，子命令有demote, inspect,ls, promote, rm, ps, update
- docker service：服務管理，子命令有create, inspect, ps, ls ,rm , scale, update
- docker stack/deploy：試驗特性，用于多應用部署
創建swarm 集群
查看docker swarm 命令說明
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
|
[root@centos-node4 ~]# docker swarm -hFlag shorthand -h has been deprecated, please use --helpUsage: docker swarm COMMANDManage Docker SwarmOptions: --help Print usageCommands: init Initialize a swarm join Join a swarm as a node and/or manager join-token Manage join tokens update Update the swarm leave Leave a swarmRun 'docker swarm COMMAND --help' for more information on a command. |
在swarm manager(centos-node4:192.168.99.101)初始化swarm集群
用--listen-addr指定監聽的ip與端口
|
1
2
3
|
#命令格式: docker swarm init --listen-addr <MANAGER-IP>:<PORT>[root@centos-node4 ~]# docker swarm init --listen-addr 192.168.99.101:2377Swarm initialized: current node (a60d5c3ttymvtozr46uvk17q4) is now a manager. |
查看結果
|
1
2
3
|
[root@centos-node4 ~]# docker node lsID HOSTNAME MEMBERSHIP STATUS AVAILABILITY MANAGER STATUS a60d5c3ttymvtozr46uvk17q4 * centos-node4 Accepted Ready Active Leader |
把swarm-agent1(centos-node5: 192.168.99.102)添加到swarm集群
在swarm-agent1上執行:
|
1
2
3
|
#命令格式: docker swarm join <MANAGER-IP>:<PORT>[root@centos-node5 ~]# docker swarm join 192.168.99.101:2377This node joined a Swarm as a worker. |
在swarm manager查看結果
|
1
2
3
4
|
[root@centos-node4 ~]# docker node lsID HOSTNAME MEMBERSHIP STATUS AVAILABILITY MANAGER STATUS 0ypcw58hjlcvr0xqbtizmau62 centos-node5 Accepted Ready Active a60d5c3ttymvtozr46uvk17q4 * centos-node4 Accepted Ready Active Leader |
創建一個overlay 跨主機網絡
查看原有網絡
|
1
2
3
4
5
6
7
|
[root@centos-node4 ~]# docker network lsNETWORK ID NAME DRIVER SCOPE abec77415f48 bridge bridge locale2fff9d572a6 docker_gwbridge bridge local166bd71f7d0e host host local9gr6bfff1rv9 ingress overlay swarm 1d2bfc590294 none null local |
可以看到在swarm上默認已有一個名為ingress的overlay 網絡,默認在swarm里使用，本文會創建一個新的
創建一個新的overlay網絡
|
1
2
3
4
5
6
7
8
9
10
|
[root@centos-node4 ~]# docker network create --driver overlay docker-netaoqs3p835s5glx69hi46ou2dw [root@centos-node4 ~]# docker network lsNETWORK ID NAME DRIVER SCOPE abec77415f48 bridge bridge localaoqs3p835s5g docker-net overlay swarm e2fff9d572a6 docker_gwbridge bridge local166bd71f7d0e host host local9gr6bfff1rv9 ingress overlay swarm 1d2bfc590294 none null local |
新的網絡(docker-net)已創建
在新的跨主機overlay 網絡(docker-net)上創建應用
部署
用alpine鏡像在docker-net網絡上啟動兩個實例, 并編排為一組服務
|
1
2
|
[root@centos-node4 ~]# docker service create --replicas 2 --name helloworld --network=docker-net alpine ping docker.com5lgdq3ihiez0o7h2uegu4fgd3 |
查看部署結果
|
1
2
3
4
5
6
7
|
[root@centos-node4 ~]# docker service lsID NAME REPLICAS IMAGE COMMAND 5lgdq3ihiez0 helloworld 0/2 alpine ping docker.com [root@centos-node4 ~]# docker service tasks helloworldID NAME SERVICE IMAGE LAST STATE DESIRED STATE NODE eul3bus45qz3b555wekotdmo5 helloworld.1 helloworld alpine Running 14 seconds Running centos-node5 55uhq6xxcv53xlkqv2f0be9b9 helloworld.2 helloworld alpine Running 14 seconds Running centos-node4 |
可以看到兩個實例分別運行在兩個節點上
測試兩個主機的網絡是否能互通
分別查看兩個實例的名稱
|
1
2
3
4
5
6
7
8
9
|
# 在swarm-manager 上執行[root@centos-node4 ~]# docker ps -aCONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES f4a197abdb0b alpine:latest "ping docker.com" 42 minutes ago Up 42 minutes helloworld.2.55uhq6xxcv53xlkqv2f0be9b9# 在swarm-agnet1 上執行[root@centos-node5 ~]# docker ps -aCONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 39cc35cd54b5 alpine:latest "ping docker.com" 50 seconds ago Up 49 seconds helloworld.1.eul3bus45qz3b555wekotdmo5 |
在swarm-manager 上測試
|
1
2
3
4
5
6
7
8
9
10
11
|
[root@centos-node4 ~]# docker exec -ti helloworld.2.55uhq6xxcv53xlkqv2f0be9b9 sh/ # ping helloworld.1.eul3bus45qz3b555wekotdmo5PING helloworld.1.eul3bus45qz3b555wekotdmo5 (10.0.9.3): 56 data bytes 64 bytes from 10.0.9.3: seq=0 ttl=64 time=0.514 ms 64 bytes from 10.0.9.3: seq=1 ttl=64 time=0.508 ms 64 bytes from 10.0.9.3: seq=2 ttl=64 time=0.381 ms 64 bytes from 10.0.9.3: seq=3 ttl=64 time=0.408 ms ^C--- helloworld.1.eul3bus45qz3b555wekotdmo5 ping statistics ---4 packets transmitted, 4 packets received, 0% packet loss round-trip min/avg/max = 0.381/0.452/0.514 ms |
在swarm-agent1 上測試
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
|
[root@centos-node5 ~]# docker ps -aCONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 39cc35cd54b5 alpine:latest "ping docker.com" 50 seconds ago Up 49 seconds helloworld.1.eul3bus45qz3b555wekotdmo5 [root@centos-node5 ~]# docker exec -ti helloworld.1.eul3bus45qz3b555wekotdmo5 sh/ # ping helloworld.2.55uhq6xxcv53xlkqv2f0be9b9PING helloworld.2.55uhq6xxcv53xlkqv2f0be9b9 (10.0.9.4): 56 data bytes 64 bytes from 10.0.9.4: seq=0 ttl=64 time=0.892 ms 64 bytes from 10.0.9.4: seq=1 ttl=64 time=0.463 ms 64 bytes from 10.0.9.4: seq=2 ttl=64 time=0.462 ms 64 bytes from 10.0.9.4: seq=3 ttl=64 time=0.478 ms 64 bytes from 10.0.9.4: seq=4 ttl=64 time=0.468 ms 64 bytes from 10.0.9.4: seq=5 ttl=64 time=0.459 ms ^C--- helloworld.2.55uhq6xxcv53xlkqv2f0be9b9 ping statistics ---6 packets transmitted, 6 packets received, 0% packet loss round-trip min/avg/max = 0.459/0.537/0.892 ms |
現在新版的docker swarm 管理非常簡單, 可以快速的搭建起一個跨主機的集群并部署應用
dokcer swarm自帶的負載均衡
創建一組服務
復制代碼 代碼如下:
docker service create --replicas 2 --name whoami -p 8080:80 daocloud.io/nginx:alpine
訪問服務(可以多執行幾次)
|
1
2
|
# 訪問地址格式: swarm-manager.ip + (-p映射的端口)curl -v 192.168.99.101:8080 |
然后用docker logs 查看容器中nginx的訪問日志, 可以現兩個容器都有訪問記錄
以上就是本文的全部內容，希望對大家的學習有所幫助，也希望大家多多支持服務器之家。
原文鏈接：http://blog.csdn.net/houdaiye/article/details/52457190
