1：前言
在docker1.12版本之前，众所周知dokcer本身只能单机上运行，而集群则要依赖mesos、kubernetes、swarm等集群管理方案。其中swarm是docker公司自己的容器集群管理工具，在当时的热度还是低于前两者。docker1.12.0版本发布中，Docker公司出于战略眼光，将swarm集成到docker-engine中，使docker内置了集群解决方案。于是swarm这个“亲儿子”的江湖地位迅速提升，在docker集群方案中与mesos，k8s形成三足鼎立之势，在未来则大有赶超之势。
2：规划
IP 主机名 角色
10.10.32.245 swarm1 swarm manager
10.10.32.246 swarm2 worker node
10.10.32.247 swarm3 worker node
10.10.32.248 swarm4 worker node
3：集群通信原理
通过/usr/lib/system.d/system/docker.service配置开启2375管理端口
ExecStart=/usr/bin/dockerd -s overlay --insecure-registry registry.cntv.net -H tcp://0.0.0.0:2375 -H unix:///va
每个node节点上，都开启2375远程管理端口，swarm集群管理通过2375管理端口来执行
4：swarm调度原理
swarm中实现调度功能的主要是scheduler模块，其中包括Filter和strategy。
Filter：根据下发的任务的资源需求，用来过滤节点，从集群所有节点中找出满足条件的节点（比如资源足够，节点正常等等）
Strategy：用来在过滤出的节点中根据策略选择一个最优的节点（比如对找出的节点进行对比，找到资源最多的节点等等）
5：swarm集群创建
（1）初始化manager节点
-
10.10.32.245
$docker swarm init --advertise-addr 10.10.32.245
Swarm initialized: current node (b73yii7s7rn321ejz8n3ch7ay) is now a manager.
To add a worker to this swarm, run the following
command
:
docker swarm
join
\
--token SWMTKN-1-2z3obrzww2zcnwaiq5dzocpmvz9kty25usroy3gh3xux0l32uo-8dxurdbms8yli2crdxbmyk7y2 \
10.10.32.245:2377
To add a manager to this swarm, run
'docker swarm join-token manager'
and follow the instructions.
（2）manage查看节点swarm状态
-
10.10.32.245
$docker info
Swarm: active
NodeID: ax13h89zlb5dqw980ddo1ox4f
Is Manager:
true
ClusterID: d5nam0mink4wdk506o1ybbs9e
Managers: 1
Nodes: 1
（3）manage查看swarm 节点
-
10.10.32.245
$ docker node
ls
ID HOSTNAME STATUS AVAILABILITY MANAGER STATUS
ax13h89zlb5dqw980ddo1ox4f * swarm1 Ready Active Leader
（4）node节点加入swarm集群
-
10.10.32.246 10.10.32.247 10.10.32.248
$ docker swarm
join
-token manager
#manage上查询加入命令及token
To add a manager to this swarm, run the following
command
:
docker swarm
join
\
--token SWMTKN-1-2z3obrzww2zcnwaiq5dzocpmvz9kty25usroy3gh3xux0l32uo-8dxurdbms8yli2crdxbmyk7y2 \
10.10.32.245:2377
$ docker swarm
join
\
--token SWMTKN-1-2z3obrzww2zcnwaiq5dzocpmvz9kty25usroy3gh3xux0l32uo-8dxurdbms8yli2crdxbmyk7y2 \
10.10.32.245:2377
This node joined a swarm as a worker
-
（5）manage查看swarm节点
$ docker node
ls
ID HOSTNAME STATUS AVAILABILITY MANAGER STATUS
03b5w49n7uz663xwjuof3dsk0 swarm4 Ready Active
0g039jomhpad819a39yvkow6y swarm2 Ready Active
4kos85v5z6jh6lahwv8yjpku8 swarm3 Ready Active
ax13h89zlb5dqw980ddo1ox4f * swarm1 Ready Active Leader
-
（6）使节点离开集群
$ docker swarm leave
Node left the swarm.
6：在swarm集群上创建服务
-
（1）创建服务
$docker service create --network net3 --replicas 4 -p 8099:80 --name
time
-php registry.cntv.net
/heqin/tvtime-php
:v0.85dongsi
bdkwsgchcydz5ixatrbiyr1fx
-
（2）查看服务
$ docker service
ls
#查看服务
ID NAME REPLICAS IMAGE COMMAND
bdkwsgchcydz
time
-php 4
/4
registry.cntv.net
/heqin/tvtime-php
:v0.85dongsi
$docker service
ps
time
-php
#查看服务的实例
ID NAME IMAGE NODE DESIRED STATE CURRENT STATE ERROR
d15ambrdd7pyi22951vi497xa
time
-php.1 registry.cntv.net
/heqin/tvtime-php
:v0.85dongsi swarm4 Running Running 5 minutes ago
3wtmmggo65cmyutg7zwikb94s
time
-php.2 registry.cntv.net
/heqin/tvtime-php
:v0.85dongsi swarm3 Running Running 5 minutes ago
5ulbhndwgswpicx8g62af8k5z
time
-php.3 registry.cntv.net
/heqin/tvtime-php
:v0.85dongsi swarm2 Running Running 5 minutes ago
4b9rh8aiwfi88xrdgmyss9ili
time
-php.4 registry.cntv.net
/heqin/tvtime-php
:v0.85dongsi swarm2 Running Running 5 minutes ago
$docker service inspect
time
-php
#查看服务的详细信息
[
{
"ID"
:
"bdkwsgchcydz5ixatrbiyr1fx"
,
"Version"
: {
"Index"
: 3074
},
"CreatedAt"
:
"2016-08-31T01:41:18.892483521Z"
,
"UpdatedAt"
:
"2016-08-31T01:41:18.909164422Z"
,
"Spec"
: {
"Name"
:
"time-php"
,
"TaskTemplate"
: {
"ContainerSpec"
: {
"Image"
:
"registry.cntv.net/heqin/tvtime-php:v0.85dongsi"
},
"Resources"
: {
"Limits"
: {},
"Reservations"
: {}
},
"RestartPolicy"
: {
"Condition"
:
"any"
,
"MaxAttempts"
: 0
},
"Placement"
: {}
},
"Mode"
: {
"Replicated"
: {
"Replicas"
: 4
}
},
"UpdateConfig"
: {
"Parallelism"
: 1,
"FailureAction"
:
"pause"
},
"Networks"
: [
{
"Target"
:
"d83qc9rgkkj1ws8kvmod4x759"
}
],
"EndpointSpec"
: {
"Mode"
:
"vip"
,
"Ports"
: [
{
"Protocol"
:
"tcp"
,
"TargetPort"
: 80,
"PublishedPort"
: 8099
}
]
}
},
"Endpoint"
: {
"Spec"
: {
"Mode"
:
"vip"
,
"Ports"
: [
{
"Protocol"
:
"tcp"
,
"TargetPort"
: 80,
"PublishedPort"
: 8099
}
]
},
"Ports"
: [
{
"Protocol"
:
"tcp"
,
"TargetPort"
: 80,
"PublishedPort"
: 8099
}
],
"VirtualIPs"
: [
{
"NetworkID"
:
"biu7m9hi8fgbbihfzg48whxqi"
,
"Addr"
:
"10.255.0.19/16"
},
{
"NetworkID"
:
"d83qc9rgkkj1ws8kvmod4x759"
,
"Addr"
:
"10.88.0.2/24"
}
]
},
"UpdateStatus"
: {
"StartedAt"
:
"0001-01-01T00:00:00Z"
,
"CompletedAt"
:
"0001-01-01T00:00:00Z"
}
}
]
-
（3）扩展服务实例数
$ docker service scale
time
-php=10
time
-php scaled to 10
$ docker service
ls
ID NAME REPLICAS IMAGE
bdkwsgchcydz
time
-php 6
/6
registry.cntv.net
/heqin/tvtime-php
:v0.85dongsi
-
（4）更新服
$ docker service update --image registry.cntv.net
/heqin/tvtime-php
:v0.84xidan --log-driver=syslog
time
-php
time
-php
$ docker service
ls
ID NAME REPLICAS IMAGE
bdkwsgchcydz
time
-php 6
/6
registry.cntv.net
/heqin/tvtime-php
:v0.84xidan
-
（5）删除服务
$docker service
rm
time
-php
time
-php
7：网络
（1）查看容器网络
$docker network
ls
NETWORK ID NAME DRIVER SCOPE
a17f16650bf9 bridge bridge
local
#容器使用独立网络Namespace，连接docker0虚拟网卡（默认模式）
f1c102babcf8 host host
local
#容器与主机共享网络Namespace，拥有主机相同网卡
b6a2efce65ef none null
local
#容器没有任何网卡，适合不需要与外部通信的容器
asa2hv41mtci ingress overlay swarm
#swarm集群的的overlay网络，容器可以跨主机网络通信
注：上面三个网络是docker安装后默认带的三个网络，第四个overlay网络是启动swarm之后默认创建的overlay网络。
（2）查看overlay网络信息
$docker network inspect ingress
[
{
"Name"
:
"ingress"
,
"Id"
:
"asa2hv41mtci2qzjkaonusnc8"
,
"Scope"
:
"swarm"
,
"Driver"
:
"overlay"
,
"EnableIPv6"
:
false
,
"IPAM"
: {
"Driver"
:
"default"
,
"Options"
: null,
"Config"
: [
{
"Subnet"
:
"10.255.0.0/16"
,
"Gateway"
:
"10.255.0.1"
}
]
},
"Internal"
:
false
,
"Containers"
: null,
"Options"
: {
"com.docker.network.driver.overlay.vxlanid_list"
:
"258"
},
"Labels"
: null
}
]
（3）创建自定义的overlay网络
$docker network create --driver=overlay --subnet=10.88.0.0
/24
--gateway=10.88.0.1 net3
c654pb76q9jnni5bpdvd34rg4
$docker network
ls
NETWORK ID NAME DRIVER SCOPE
c654pb76q9jn net3 overlay swarm
$docker network inspect net3
[
{
"Name"
:
"net3"
,
"Id"
:
"c654pb76q9jnni5bpdvd34rg4"
,
"Scope"
:
"swarm"
,
"Driver"
:
"overlay"
,
"EnableIPv6"
:
false
,
"IPAM"
: {
"Driver"
:
"default"
,
"Options"
: null,
"Config"
: [
{
"Subnet"
:
"10.88.0.0/24"
,
"Gateway"
:
"10.88.0.1"
}
]
},
"Internal"
:
false
,
"Containers"
: null,
"Options"
: {
"com.docker.network.driver.overlay.vxlanid_list"
:
"258"
},
"Labels"
: null
}
注：在manage上创建overlay网络后，各node上并没有同步创建该网络，只有当使用该overlay网络的容器在node节点上运行时，才会在node节点上自动创建,容器删除后，网络也会在node节点上删除。
（4）将容器绑定到自定义的网络中
$docker service create --network net3 --replicas 4 --name
time
-php registry.cntv.net
/heqin/tvtime-php
:v0.85dongsi
5qiv4hhv7ra5p65fpq2v6ok0x
$docker service
ls
5qiv4hhv7ra5
time
-php 4
/4
registry.cntv.net
/heqin/tvtime-php
:v0.85dongsi
$docker service
ps
time
-php
ID NAME IMAGE NODE DESIRED STATE CURRENT STATE ERROR
529kuyazgxxl8l0eds4f740nu
time
-php.1 registry.cntv.net
/heqin/tvtime-php
:v0.85dongsi swarm3 Running Running 32 seconds ago
brdelpbuwm3p9cujy6k78cpzo
time
-php.2 registry.cntv.net
/heqin/tvtime-php
:v0.85dongsi swarm3 Running Running 33 seconds ago
3csp6fmrsg5cvski270732q98
time
-php.3 registry.cntv.net
/heqin/tvtime-php
:v0.85dongsi swarm2 Running Running 34 seconds ago
2ismj5zt9o3t2hfe1j1lx7dcw
time
-php.4 registry.cntv.net
/heqin/tvtime-php
:v0.85dongsi swarm4 Running Running 33 seconds ago
进入swarm3 ，可以看到net3网络下有两个容器。
$docker docker network inspect net3
[
{
"Name"
:
"net3"
,
"Id"
:
"d83qc9rgkkj1ws8kvmod4x759"
,
"Scope"
:
"swarm"
,
"Driver"
:
"overlay"
,
"EnableIPv6"
:
false
,
"IPAM"
: {
"Driver"
:
"default"
,
"Options"
: null,
"Config"
: [
{
"Subnet"
:
"10.88.0.0/24"
,
"Gateway"
:
"10.88.0.1"
}
]
},
"Internal"
:
false
,
"Containers"
: {
"d465611683b69f20c28ec9eb276a9e4088867e35508ecf2bf28c07c5d3e1b33a"
: {
"Name"
:
"time-php.1.529kuyazgxxl8l0eds4f740nu"
,
"EndpointID"
:
"ca667c89550f88eabf54eeaaaa3c55c678ec8184f121fc845bbaab1cd4be20ff"
,
"MacAddress"
:
"02:42:0a:58:00:05"
,
"IPv4Address"
:
"10.88.0.5/24"
,
"IPv6Address"
:
""
},
"d8fa64c4f8357418304ed1f19a049e4aac196c94ef50a27a0686ebe4ee734225"
: {
"Name"
:
"time-php.2.brdelpbuwm3p9cujy6k78cpzo"
,
"EndpointID"
:
"ab45bcf8e73d114e9bdf97cac128750a6c7a073149783746a9962c6cd4c599a8"
,
"MacAddress"
:
"02:42:0a:58:00:06"
,
"IPv4Address"
:
"10.88.0.6/24"
,
"IPv6Address"
:
""
}
},
"Options"
: {
"com.docker.network.driver.overlay.vxlanid_list"
:
"257"
},
"Labels"
: {}
}
用swarm3的容器去ping swarm2主机上的容器，是可以ping通的
$docker
exec
-it d8fa64c4f835
ping
10.88.0.3
PING 10.88.0.3 (10.88.0.3) 56(84) bytes of data.
64 bytes from 10.88.0.3: icmp_seq=1 ttl=64
time
=0.611 ms
64 bytes from 10.88.0.3: icmp_seq=2 ttl=64
time
=5.45 ms
64 bytes from 10.88.0.3: icmp_seq=3 ttl=64
time
=0.439 ms
64 bytes from 10.88.0.3: icmp_seq=4 ttl=64
time
=0.586 ms
64 bytes from 10.88.0.3: icmp_seq=5 ttl=64
time
=0.603 ms
8：swarm基本命令
-
swarm集群创建与管理
docker swarm Command
Commands:
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 swarm
-
swarm服务创建与管理
docker service Command
Commands:
create Create a new service
inspect Display detailed information on one or more services
ps List the tasks of a service
ls List services
rm Remove one or more services
scale Scale one or multiple services
update Update a service
-
swarm服务创建选项
docker service create [OPTIONS] IMAGE
Options:
--constraint value Placement constraints (default [])
--container-label value Container labels (default [])
--endpoint-mode string Endpoint mode (vip or dnsrr)
-e, --env value Set environment variables (default [])
--help Print usage
-l, --label value Service labels (default [])
--limit-cpu value Limit CPUs (default 0.000)
--limit-memory value Limit Memory (default 0 B)
--log-driver string Logging driver for service
--log-opt value Logging driver options (default [])
--mode string Service mode (replicated or global) (default "replicated")
--mount value Attach a mount to the service
--name string Service name
--network value Network attachments (default [])
-p, --publish value Publish a port as a node port (default [])
--replicas value Number of tasks (default none)
--reserve-cpu value Reserve CPUs (default 0.000)
--reserve-memory value Reserve Memory (default 0 B)
--restart-condition string Restart when condition is met (none, on-failure, or any)
--restart-delay value Delay between restart attempts (default none)
--restart-max-attempts value Maximum number of restarts before giving up (default none)
--restart-window value Window used to evaluate the restart policy (default none)
--stop-grace-period value Time to wait before force killing a container (default none)
--update-delay duration Delay between updates
--update-failure-action string Action on update failure (pause|continue) (default "pause")
--update-parallelism uint Maximum number of tasks updated simultaneously (0 to update all at once) (default 1)
-u, --user string Username or UID
9：利用2375端口远程管理
注：-H 指定远程地址和端口，即可在任意一台有docker客户端的机器远程操作swarm集群
$docker -H 10.00.32.245:2375 service
ls
ID NAME REPLICAS IMAGE COMMAND
3e0ihi2lnnaa test_api 1
/1
registry.cntv.net
/heqin/api-web
:1.1
5mdqw6r53tom test_apimem 1
/1
registry.cntv.net
/heqin/memcache-1
.4
5ucxo9wqywo9 test_tomcatmem 1
/1
registry.cntv.net
/heqin/memcache-1
.4
cjd1z14ug28g HQtest_tvtime 5
/5
registry.cntv.net
/heqin/tvtime-php
:v0.85dongsi
ex4mahzt5k8u test_tomcat 1
/1
registry.cntv.net
/heqin/jdk7-tomcat7
.0.52:201608011wq
10：swarm与mesos对比
（1）：部署配置简单，集群管理便利，但是没有web管理界面
（2）：每个节点无需手动安装agent，集群间只需要通过暴露tcp：//2375端口来保证节点间的通信。
（3）：无需bamboo来做负载均衡，swarm集群使用IPVS负载均衡
（4）：无需marathon、chrons等调度组件，swarm内置调度模块。
对于上述的VIP负载均衡算法做下补充说明：docker1.12使用的是linux自身的IPVS作为负载均衡方式。IPVS实则linux内核中一个叫做ip_vs的负载均衡模块。不同于DNS负载均衡将IP列表顺序轮询，IPVS会将负载均匀的分发到每个容器。IPVS是四层的转发者，能够转发TCP、UDP、DNS并且支持八种负载均衡算法。
2万+
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
