Docker下部署Consul集群和ACL权限配置

文章目录

  • Docker下部署Consul集群和ACL权限配置
    • 规划与准备
    • 搭建Consul集群
      • 创建server1配置文件,启动server1节点
      • 创建server2配置文件,启动server2节点
      • 创建server3配置文件,启动server3节点
      • 加入集群
      • 验证Consul集群选举机制
      • 配置join参数,节点自动加入集群
        • server节点退出集群
        • 节点自动加入集群
      • 节点之间加入通讯密钥
    • 增加ACL token权限配置
      • 启用ACL,配置master token
      • 配置agent token

Docker下部署Consul集群和ACL权限配置

规划与准备

本次计划部署的consul集群有3个节点,都是server类型

容器IP 节点 类型
172.17.0.2 server1 server
172.17.0.3 server2 server
172.17.0.4 server3 server

把consul的数据文件都映射到宿主机上,有利于备份数据以及方便以后重构容器。
宿主机建立目录server1、server2、server3,下面分别存放3个consul节点的信息:

1
2
3
4
5
6
7
8
9
10
11
[root@wuli-centOS7 ~]# mkdir -p /data/consul/server1/config
[root@wuli-centOS7 ~]# mkdir -p /data/consul/server1/data
[root@wuli-centOS7 ~]# mkdir -p /data/consul/server1/log

[root@wuli-centOS7 ~]# mkdir -p /data/consul/server2/config
[root@wuli-centOS7 ~]# mkdir -p /data/consul/server2/data
[root@wuli-centOS7 ~]# mkdir -p /data/consul/server2/log

[root@wuli-centOS7 ~]# mkdir -p /data/consul/server3/config
[root@wuli-centOS7 ~]# mkdir -p /data/consul/server3/data
[root@wuli-centOS7 ~]# mkdir -p /data/consul/server3/log

搭建Consul集群

创建server1配置文件,启动server1节点

  1. 创建server1的配置文件:
1
[root@wuli-centOS7 ~]# vim /data/consul/server1/config/config.json
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
{
    "datacenter": "dc1",
    "bootstrap_expect": 3,
    "data_dir": "/consul/data",
    "log_file": "/consul/log/",
    "log_level": "INFO",
    "node_name": "consul_server_1",
    "client_addr": "0.0.0.0",
    "server": true,
    "ui": true,
    "enable_script_checks": true,
    "addresses": {
        "https": "0.0.0.0",
        "dns": "0.0.0.0"
    }
}
  1. 启动节点consul_server_1
1
[root@wuli-centOS7 ~]# docker run -d -p 8510:8500 -v /data/consul/server1/data:/consul/data -v /data/consul/server1/config:/consul/config -e CONSUL_BIND_INTERFACE='eth0' --privileged=true --name=consul_server_1 consul agent -data-dir=/consul/data;

docker run命令说明:

  • Environment Variable(环境变量):
    CONSUL_BIND_INTERFACE=eth0:在容器启动时,自动绑定eth0端口的IP地址

  • docker参数:
    -e:将时区信息传入到容器内部
    -d:Daemon模式
    -p:绑定端口
    –privileged:表示以root权限运行
    –name:指定实例名称
    consul:consul启动命令

启动后,因为配置了bootstrap_expect=3,但只启动了一个server,所以会报错:没有集群领导者

1
2
3
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul monitor
2020-05-03T18:01:40.453Z [ERROR] agent.anti_entropy: failed to sync remote state: error="No cluster leader"
2020-05-03T18:01:57.802Z [ERROR] agent: Coordinate update error: error="No cluster leader"

创建server2配置文件,启动server2节点

  1. 创建server2的配置文件:
1
[root@wuli-centOS7 ~]# vim /data/consul/server2/config/config.json

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
{
    "datacenter": "dc1",
    "bootstrap_expect": 3,
    "data_dir": "/consul/data",
    "log_file": "/consul/log/",
    "log_level": "INFO",
    "node_name": "consul_server_2",
    "client_addr": "0.0.0.0",
    "server": true,
    "ui": true,
    "enable_script_checks": true,
    "addresses": {
        "https": "0.0.0.0",
        "dns": "0.0.0.0"
    }
}
  1. 启动节点consul_server_2
1
[root@wuli-centOS7 ~]# docker run -d -p 8520:8500 -v /data/consul/server2/data:/consul/data -v /data/consul/server2/config:/consul/config -e CONSUL_BIND_INTERFACE='eth0' --privileged=true --name=consul_server_2 consul agent -data-dir=/consul/data;

创建server3配置文件,启动server3节点

  1. 创建server3的配置文件:
1
[root@wuli-centOS7 ~]# vim /data/consul/server3/config/config.json
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
{
    "datacenter": "dc1",
    "bootstrap_expect": 3,
    "data_dir": "/consul/data",
    "log_file": "/consul/log/",
    "log_level": "INFO",
    "node_name": "consul_server_3",
    "client_addr": "0.0.0.0",
    "server": true,
    "ui": true,
    "enable_script_checks": true,
    "addresses": {
        "https": "0.0.0.0",
        "dns": "0.0.0.0"
    }
}
  1. 启动节点consul_server_3
1
[root@wuli-centOS7 ~]# docker run -d -p 8530:8500 -v /data/consul/server3/data:/consul/data -v /data/consul/server3/config:/consul/config -e CONSUL_BIND_INTERFACE='eth0' --privileged=true --name=consul_server_3 consul agent -data-dir=/consul/data;

加入集群

  1. 查看节点server2、server3的IP,然后通过join命令把全部节点加入集群
1
2
3
4
[root@wuli-centOS7 ~]# docker inspect --format '{{ .NetworkSettings.IPAddress }}' consul_server_2
172.17.0.3
[root@wuli-centOS7 ~]# docker inspect --format '{{ .NetworkSettings.IPAddress }}' consul_server_3
172.17.0.4
1
2
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul join 172.17.0.3
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul join 172.17.0.4
  1. 查看后台日志:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul monitor
2020-05-03T19:04:39.254Z [INFO]  agent: (LAN) joining: lan_addresses=[172.17.0.3]
2020-05-03T19:04:39.260Z [INFO]  agent.server.serf.lan: serf: EventMemberJoin: consul_server_2 172.17.0.3
2020-05-03T19:04:39.261Z [INFO]  agent: (LAN) joined: number_of_nodes=1
2020-05-03T19:04:39.261Z [INFO]  agent.server: Adding LAN server: server="consul_server_2 (Addr: tcp/172.17.0.3:8300) (DC: dc1)"
2020-05-03T19:04:39.265Z [INFO]  agent.server.serf.wan: serf: EventMemberJoin: consul_server_2.dc1 172.17.0.3
2020-05-03T19:04:39.266Z [INFO]  agent.server: Handled event for server in area: event=member-join server=consul_server_2.dc1 area=wan
2020-05-03T19:04:40.362Z [ERROR] agent: Coordinate update error: error="No cluster leader"
2020-05-03T19:04:44.019Z [INFO]  agent: (LAN) joining: lan_addresses=[172.17.0.4]
2020-05-03T19:04:44.021Z [INFO]  agent.server.serf.lan: serf: EventMemberJoin: consul_server_3 172.17.0.4
2020-05-03T19:04:44.021Z [INFO]  agent: (LAN) joined: number_of_nodes=1
2020-05-03T19:04:44.022Z [INFO]  agent.server: Adding LAN server: server="consul_server_3 (Addr: tcp/172.17.0.4:8300) (DC: dc1)"
2020-05-03T19:04:44.027Z [INFO]  agent.server: Found expected number of peers, attempting bootstrap: peers=172.17.0.2:8300,172.17.0.3:8300,172.17.0.4:8300
2020-05-03T19:04:44.049Z [INFO]  agent.server.serf.wan: serf: EventMemberJoin: consul_server_3.dc1 172.17.0.4
2020-05-03T19:04:44.049Z [INFO]  agent.server: Handled event for server in area: event=member-join server=consul_server_3.dc1 area=wan
2020-05-03T19:04:48.088Z [WARN]  agent.server.raft: heartbeat timeout reached, starting election: last-leader=
2020-05-03T19:04:48.088Z [INFO]  agent.server.raft: entering candidate state: node="Node at 172.17.0.2:8300 [Candidate]" term=2
2020-05-03T19:04:48.100Z [INFO]  agent.server.raft: election won: tally=2
2020-05-03T19:04:48.101Z [INFO]  agent.server.raft: entering leader state: leader="Node at 172.17.0.2:8300 [Leader]"
2020-05-03T19:04:48.101Z [INFO]  agent.server.raft: added peer, starting replication: peer=78293668-16a6-1de0-673f-455d594e7447
2020-05-03T19:04:48.101Z [INFO]  agent.server.raft: added peer, starting replication: peer=0b6169d8-7acc-ed24-682f-56ffd12b486c
2020-05-03T19:04:48.102Z [INFO]  agent.server: cluster leadership acquired
2020-05-03T19:04:48.103Z [INFO]  agent.server: New leader elected: payload=consul_server_1
2020-05-03T19:04:48.104Z [WARN]  agent.server.raft: appendEntries rejected, sending older logs: peer="{Voter 78293668-16a6-1de0-673f-455d594e7447 172.17.0.3:8300}" next=1
2020-05-03T19:04:48.107Z [INFO]  agent.server.raft: pipelining replication: peer="{Voter 0b6169d8-7acc-ed24-682f-56ffd12b486c 172.17.0.4:8300}"
2020-05-03T19:04:48.112Z [INFO]  agent.server.raft: pipelining replication: peer="{Voter 78293668-16a6-1de0-673f-455d594e7447 172.17.0.3:8300}"
2020-05-03T19:04:48.120Z [INFO]  agent.server: Cannot upgrade to new ACLs: leaderMode=0 mode=0 found=true leader=172.17.0.2:8300
2020-05-03T19:04:48.129Z [INFO]  agent.leader: started routine: routine="CA root pruning"
2020-05-03T19:04:48.129Z [INFO]  agent.server: member joined, marking health alive: member=consul_server_1
2020-05-03T19:04:48.146Z [INFO]  agent.server: member joined, marking health alive: member=consul_server_2
2020-05-03T19:04:48.156Z [INFO]  agent.server: member joined, marking health alive: member=consul_server_3
2020-05-03T19:04:48.720Z [INFO]  agent: Synced node info

  1. 查看成员
1
2
3
4
5
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul members
Node             Address          Status  Type    Build  Protocol  DC   Segment
consul_server_1  172.17.0.2:8301  alive   server  1.7.2  2         dc1  <all>
consul_server_2  172.17.0.3:8301  alive   server  1.7.2  2         dc1  <all>
consul_server_3  172.17.0.4:8301  alive   server  1.7.2  2         dc1  <all>
  1. 查看集群的选举情况,领导者为哪个节点等信息
1
2
3
4
5
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul operator raft list-peers    
Node             ID                                    Address          State     Voter  RaftProtocol
consul_server_1  0214e0b9-04fe-e8ad-c855-b5091dfc8e2e  172.17.0.2:8300  leader    true   3
consul_server_2  78293668-16a6-1de0-673f-455d594e7447  172.17.0.3:8300  follower  true   3
consul_server_3  0b6169d8-7acc-ed24-682f-56ffd12b486c  172.17.0.4:8300  follower  true   3
  1. 访问ui,节点正常,标星的节点表示是leader
    在这里插入图片描述

在这里插入图片描述

验证Consul集群选举机制

Consul中只有server节点会参与Raft算法并且作为peer set中的一部分。Raft中的节点总是处于以下三种状态之一: follower、candidate或leader。目前server1是leader,我们下面重启consul_server_1容器,观察consul集群变化情况。

  1. 重启consul_server_1
1
[root@wuli-centOS7 ~]# docker restart consul_server_1
  1. 观察server2和server3的日志,consul_server_2后台日志:
1
2
3
4
5
6
7
8
[root@wuli-centOS7 ~]# docker exec consul_server_2 consul monitor
2020-05-03T19:26:29.564Z [INFO]  agent.server.memberlist.lan: memberlist: Suspect consul_server_1 has failed, no acks received
2020-05-03T19:26:30.533Z [INFO]  agent.server.serf.wan: serf: EventMemberUpdate: consul_server_1.dc1
2020-05-03T19:26:30.533Z [INFO]  agent.server: Handled event for server in area: event=member-update server=consul_server_1.dc1 area=wan
2020-05-03T19:26:30.564Z [INFO]  agent.server.serf.lan: serf: EventMemberUpdate: consul_server_1
2020-05-03T19:26:30.565Z [INFO]  agent.server: Updating LAN server: server="consul_server_1 (Addr: tcp/172.17.0.2:8300) (DC: dc1)"
2020-05-03T19:26:33.542Z [WARN]  agent.server.raft: rejecting vote request since we have a leader: from=172.17.0.4:8300 leader=172.17.0.2:8300
2020-05-03T19:26:33.565Z [INFO]  agent.server: New leader elected: payload=consul_server_3
  1. consul_server_3后台日志:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
[root@wuli-centOS7 ~]# docker exec consul_server_3 consul monitor
2020-05-03T19:26:28.151Z [ERROR] agent.server.memberlist.lan: memberlist: Push/Pull with consul_server_1 failed: dial tcp 172.17.0.2:8301: connect: connection refused
2020-05-03T19:26:30.235Z [INFO]  agent.server.memberlist.lan: memberlist: Suspect consul_server_1 has failed, no acks received
2020-05-03T19:26:30.420Z [ERROR] agent: Coordinate update error: error="rpc error making call: stream closed"
2020-05-03T19:26:30.536Z [INFO]  agent.server.serf.wan: serf: EventMemberUpdate: consul_server_1.dc1
2020-05-03T19:26:30.536Z [INFO]  agent.server: Handled event for server in area: event=member-update server=consul_server_1.dc1 area=wan
2020-05-03T19:26:30.729Z [INFO]  agent.server.serf.lan: serf: EventMemberUpdate: consul_server_1
2020-05-03T19:26:30.729Z [INFO]  agent.server: Updating LAN server: server="consul_server_1 (Addr: tcp/172.17.0.2:8300) (DC: dc1)"
2020-05-03T19:26:32.234Z [WARN]  agent.server.memberlist.wan: memberlist: Was able to connect to consul_server_1.dc1 but other probes failed, network may be misconfigured
2020-05-03T19:26:33.536Z [WARN]  agent.server.raft: heartbeat timeout reached, starting election: last-leader=172.17.0.2:8300
2020-05-03T19:26:33.536Z [INFO]  agent.server.raft: entering candidate state: node="Node at 172.17.0.4:8300 [Candidate]" term=3
2020-05-03T19:26:33.546Z [INFO]  agent.server.raft: election won: tally=2
2020-05-03T19:26:33.546Z [INFO]  agent.server.raft: entering leader state: leader="Node at 172.17.0.4:8300 [Leader]"
2020-05-03T19:26:33.546Z [INFO]  agent.server.raft: added peer, starting replication: peer=0214e0b9-04fe-e8ad-c855-b5091dfc8e2e
2020-05-03T19:26:33.546Z [INFO]  agent.server.raft: added peer, starting replication: peer=78293668-16a6-1de0-673f-455d594e7447
2020-05-03T19:26:33.547Z [INFO]  agent.server: cluster leadership acquired
2020-05-03T19:26:33.548Z [INFO]  agent.server: New leader elected: payload=consul_server_3
2020-05-03T19:26:33.550Z [INFO]  agent.server.raft: pipelining replication: peer="{Voter 78293668-16a6-1de0-673f-455d594e7447 172.17.0.3:8300}"
2020-05-03T19:26:33.551Z [INFO]  agent.server.raft: pipelining replication: peer="{Voter 0214e0b9-04fe-e8ad-c855-b5091dfc8e2e 172.17.0.2:8300}"
2020-05-03T19:26:33.554Z [INFO]  agent.server: Cannot upgrade to new ACLs: leaderMode=0 mode=0 found=true leader=172.17.0.4:8300
2020-05-03T19:26:33.555Z [INFO]  agent.leader: started routine: routine="CA root pruning"
  1. consul_server_1重启后,查看领导者,此时consul_server_3已被推选为领导者:
1
2
3
4
5
[root@wuli-centOS7 ~]# docker exec -it consul_server_1 consul operator raft list-peers
Node             ID                                    Address          State     Voter  RaftProtocol
consul_server_3  0b6169d8-7acc-ed24-682f-56ffd12b486c  172.17.0.4:8300  leader    true   3
consul_server_1  0214e0b9-04fe-e8ad-c855-b5091dfc8e2e  172.17.0.2:8300  follower  true   3
consul_server_2  78293668-16a6-1de0-673f-455d594e7447  172.17.0.3:8300  follower  true   3

配置join参数,节点自动加入集群

在Consul集群中,一个节点优雅退出,不会影响到集群中其他节点的正常运行,而集群的数据中心会把离开的节点标志为left状态,等到该节点重新加入集群,状态会变为alive,默认情况下,数据中心会保留离开节点信息72小时,72小时候后如果仍然没有加入集群,则会把该节点的信息移除掉。

server节点退出集群

下面演示节点退出的情况:

  1. consul_server_2优雅退出
1
2
[root@wuli-centOS7 ~]# docker exec consul_server_2 consul leave
Graceful leave complete
  1. 查看成员状态,
1
2
3
4
5
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul members
Node             Address          Status  Type    Build  Protocol  DC   Segment
consul_server_1  172.17.0.2:8301  alive   server  1.7.2  2         dc1  <all>
consul_server_2  172.17.0.3:8301  left    server  1.7.2  2         dc1  <all>
consul_server_3  172.17.0.4:8301  alive   server  1.7.2  2         dc1  <all>
  1. 查看leader
1
2
3
4
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul operator raft list-peers
Node             ID                                    Address          State     Voter  RaftProtocol
consul_server_3  0b6169d8-7acc-ed24-682f-56ffd12b486c  172.17.0.4:8300  leader    true   3
consul_server_1  0214e0b9-04fe-e8ad-c855-b5091dfc8e2e  172.17.0.2:8300  follower  true   3

查看后台日志,没有报错,访问ui正常,说明配置文件的bootstrap_expect=3,只是在创建集群的时候期待的节点数量,如果达不到就不会初次创建集群,但节点数据量达到3后,集群初次创建成功,后面如果server通过优雅退出,不影响集群的健康情况,集群仍然会正常运行,而优雅退出的集群的状态会标志为“left”。

  1. 重新启动consul_server_2后,不会自动加入集群,因为配置文件没有start_join和retry_join参数,需要通过命令consul join加入集群,通过consul_server_2或者集群中的其他server发起命令都可以:
1
2
[root@wuli-centOS7 ~]# docker exec consul_server_2 consul join 172.17.0.2
Successfully joined cluster by contacting 1 nodes.

或者:

1
2
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul join 172.17.0.3
Successfully joined cluster by contacting 1 nodes.

节点自动加入集群

全部server的配置文件添加start_join和retry_join参数,可以在重启后自动集群。

  1. 参看各个server的ip:
1
2
3
4
5
6
[root@wuli-centOS7 server4]# docker inspect --format '{{ .NetworkSettings.IPAddress }}' consul_server_1
172.17.0.2
[root@wuli-centOS7 server4]# docker inspect --format '{{ .NetworkSettings.IPAddress }}' consul_server_2
172.17.0.3
[root@wuli-centOS7 server4]# docker inspect --format '{{ .NetworkSettings.IPAddress }}' consul_server_3
172.17.0.4
  1. 分别编辑各个server的配置文件
1
[root@wuli-centOS7 ~]# vim /data/consul/server1/config/config.json
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
{
    "datacenter": "dc1",
    "bootstrap_expect": 3,
    "data_dir": "/consul/data",
    "log_file": "/consul/log/",
    "log_level": "INFO",
    "node_name": "consul_server_3",
    "client_addr": "0.0.0.0",
    "server": true,
    "ui": true,
    "enable_script_checks": true,
    "addresses": {
        "https": "0.0.0.0",
        "dns": "0.0.0.0"
    },
    "start_join":[
        "172.17.0.2",
        "172.17.0.3",
        "172.17.0.4"
    ],
    "retry_join":[
        "172.17.0.2",
        "172.17.0.3",
        "172.17.0.4"
    ]
}
  1. 重载配置文件,验证配置是否正确
1
2
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul reload  
Configuration reload triggered
  1. 配置正确,以后重启consul server,或者优雅退出后再启动,会自动加入集群,无需使用命令join。

节点之间加入通讯密钥

增加通讯密钥,可以防止其他节点加入集群。步骤如下:

  1. 使用consul keygen命令生成通讯密钥
1
2
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul keygen
zVCGFgICqf5MAU61Wd/1wDP1hoQ37rQQFVvMkhzpM1c=
  1. 把密钥信息分别写入3个server的配置文件中
1
[root@wuli-centOS7 ~]# vim /data/consul/server1/config/config.json
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
{  
    "datacenter": "dc1",
    "bootstrap_expect": 3,
    "data_dir": "/consul/data",
    "log_file": "/consul/log/",
    "log_level": "INFO",
    "node_name": "consul_server_1",
    "client_addr": "0.0.0.0",
    "server": true,
    "ui": true,
    "enable_script_checks": true,
    "addresses": {
        "https": "0.0.0.0",
        "dns": "0.0.0.0"
    },
    "encrypt": "zVCGFgICqf5MAU61Wd/1wDP1hoQ37rQQFVvMkhzpM1c="
}
  1. consul重新加载配置文件
1
2
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul reload
Configuration reload triggered
  1. 实验中,发现重载配置文件后,其他节点仍然可以加入集群,需要重新启动consul
1
2
3
4
5
6
7
8
[root@wuli-centOS7 ~]# docker restart consul_server_1
consul_server_1

[root@wuli-centOS7 ~]# docker restart consul_server_2
consul_server_2

[root@wuli-centOS7 ~]# docker restart consul_server_3
consul_server_3

后面如果有新的节点,要加入集群中,必须提供encrypt才行。

  1. 再创建一个节点server4,尝试加入有密钥的集群:
    5.1 创建配置配置文件:
1
[root@wuli-centOS7 ~]# vim /data/consul/server4/config/config.json

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
{
    "datacenter": "dc1",
    "bootstrap_expect": 1,
    "data_dir": "/consul/data",
    "log_file": "/consul/log/",
    "log_level": "INFO",
    "node_name": "consul_server_4",
    "client_addr": "0.0.0.0",
    "server": true,
    "ui": true,
    "enable_script_checks": true,
    "addresses": {
        "https": "0.0.0.0",
        "dns": "0.0.0.0"
    }
}

5.2 启动server4

1
2
[root@wuli-centOS7 ~]# docker run -d -p 8540:8500 -v /data/consul/server4/data:/consul/data -v /data/consul/server4/config:/consul/config -e CONSUL_BIND_INTERFACE='eth0' --privileged=true --name=consul_server_4 consul agent -data-dir=/consul/data;
cdf6edee727f92baf2f7d324cb9522644c851f81fe62356fb2fb9aad126eaf13

5.3 尝试加入集群,失败:

1
2
3
4
5
6
[root@wuli-centOS7 ~]# docker exec consul_server_4 consul join 172.17.0.2        
Error joining address '172.17.0.2': Unexpected response code: 500 (1 error occurred:
        * Failed to join 172.17.0.2: Remote state is encrypted and encryption is not configured

)
Failed to join any nodes.

5.4 在配置文件中加入通讯密钥

1
[root@wuli-centOS7 ~]# vim /data/consul/server4/config/config.json
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
{
    "datacenter": "dc1",
    "bootstrap_expect": 1,
    "data_dir": "/consul/data",
    "log_file": "/consul/log/",
    "log_level": "INFO",
    "node_name": "consul_server_4",
    "client_addr": "0.0.0.0",
    "server": true,
    "ui": true,
    "enable_script_checks": true,
    "addresses": {
        "https": "0.0.0.0",
        "dns": "0.0.0.0"
    },
    "encrypt": "zVCGFgICqf5MAU61Wd/1wDP1hoQ37rQQFVvMkhzpM1c="
}

5.5 重载配置文件,然后尝试加入集群,仍然失败,说明添加密钥都需要重启consul才生效

1
2
3
4
5
6
7
8
9
[root@wuli-centOS7 ~]# docker exec consul_server_4 consul reload
Configuration reload triggered

[root@wuli-centOS7 ~]# docker exec consul_server_4 consul join 172.17.0.2
Error joining address '172.17.0.2': Unexpected response code: 500 (1 error occurred:
        * Failed to join 172.17.0.2: Remote state is encrypted and encryption is not configured

)
Failed to join any nodes.

5.6 重启server4,加入集群成功!

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
[root@wuli-centOS7 server4]# docker restart consul_server_4
consul_server_4
[root@wuli-centOS7 server4]# docker exec consul_server_4 consul join 172.17.0.2
Successfully joined cluster by contacting 1 nodes.
[root@wuli-centOS7 server4]# docker exec consul_server_1 consul members
Node             Address          Status  Type    Build  Protocol  DC   Segment
consul_server_1  172.17.0.2:8301  alive   server  1.7.2  2         dc1  <all>
consul_server_2  172.17.0.3:8301  alive   server  1.7.2  2         dc1  <all>
consul_server_3  172.17.0.4:8301  alive   server  1.7.2  2         dc1  <all>
consul_server_4  172.17.0.5:8301  alive   server  1.7.2  2         dc1  <all>
[root@wuli-centOS7 server4]# docker exec consul_server_1 consul operator raft list-peers
Node             ID                                    Address          State     Voter  RaftProtocol
consul_server_3  0b6169d8-7acc-ed24-682f-56ffd12b486c  172.17.0.4:8300  leader    true   3
consul_server_2  78293668-16a6-1de0-673f-455d594e7447  172.17.0.3:8300  follower  true   3
consul_server_1  0214e0b9-04fe-e8ad-c855-b5091dfc8e2e  172.17.0.2:8300  follower  true   3
consul_server_4  587e81c7-f1b5-5f19-311c-741c06ca446d  172.17.0.5:8300  follower  false  3

增加ACL token权限配置

配置master的token,master的token可以自由定义,但为了与其他token格式一致,官方建议使用64位的UUID。consul的配置文件可以有多个,文件后缀名可以是json或者hcl,我们这里使用hcl来演示。

启用ACL,配置master token

Consul的ACL功能需要显示启用,在配置文件中通过设置参数acl.enabled=true即可。
下面演示了两种方法来配置ACL,个人推荐方法二。
方法一:配置acl.enabled=true,然后通过命令consul acl bootstrap生成token,之后把改token作为master的token。

  1. 添加配置文件acl.hcl:
1
[root@wuli-centOS7 ~]# vim /data/consul/server4/config/acl.hcl
1
2
3
4
5
6
7
8
primary_datacenter = "dc1"
acl {
  enabled = true
  default_policy = "deny"
  enable_token_persistence = true
  tokens {
  }
}
  1. 重载配置文件,创建初始token,生成的SecretID就是token
1
2
3
4
5
6
7
8
9
10
11
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul reload
Configuration reload triggered

[root@wuli-centOS7 ~]# consul acl bootstrap
AccessorID:       b6676320-dbef-4020-ae69-8a47ae13dcef
SecretID:         474dcea7-ee4d-3f11-1af1-a38eb37d3f5d
Description:      Bootstrap Token (Global Management)
Local:            false
Create Time:      2020-04-28 11:30:42.2992871 +0800 CST
Policies:
   00000000-0000-0000-0000-000000000001 - global-management
  1. 修改配置文件acl.hcl,加入mater token
1
[root@wuli-centOS7 ~]# vim /data/consul/server4/config/acl.hcl
1
2
3
4
5
6
7
8
9
primary_datacenter = "dc1"
acl {
  enabled = true
  default_policy = "deny"
  enable_token_persistence = true
  tokens {
    master = "474dcea7-ee4d-3f11-1af1-a38eb37d3f5d  
  }
}
  1. 重启服务,验证

方法二:

  1. 使用linux的uuidgen命令生成一个64位UUID作为master token
1
2
[root@wuli-centOS7 ~]# uuidgen
dcb93655-0661-4ea1-bfc4-e5744317f99e
  1. 编写acl.hcl文件文件
1
[root@wuli-centOS7 ~]# vim /data/consul/server4/config/acl.hcl
1
2
3
4
5
6
7
8
9
primary_datacenter = "dc1"
acl {
  enabled = true
  default_policy = "deny"
  enable_token_persistence = true
  tokens {
    master = "dcb93655-0661-4ea1-bfc4-e5744317f99e"
  }
}

修改config.json配置,把bootstrap_expect修改成1

3.重载配置文件,验证是否正确。

1
2
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul reload
Configuration reload triggered
  1. 优雅关闭其他server,重启consul_server_1容器
1
2
3
4
5
[root@wuli-centOS7 ~]# docker exec consul_server_2 consul leave
[root@wuli-centOS7 ~]# docker exec consul_server_3 consul leave
[root@wuli-centOS7 ~]# docker exec consul_server_4 consul leave

[root@wuli-centOS7 ~]# docker restart consul_server_1
  1. 访问ui,提示需要输入token,输入我们上面的mater token即可
    在这里插入图片描述

  2. 此时,使用consul大部分命令,都需要带上token,否则报错:

1
2
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul info
Error querying agent: Unexpected response code: 403 (Permission denied)

带上token参数:

1
2
3
4
5
6
7
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul info -token=dcb93655-0661-4ea1-bfc4-e5744317f99e  
agent:
        check_monitors = 0
        check_ttls = 0
        checks = 0
        services = 0
……
  1. 不是在docker启动的consul,可以通过增加环境变量CONSUL_HTTP_TOKEN代替每次命令后带token参数
1
2
3
4
[root@wuli-centOS7 ~]# vi /etc/profile
export CONSUL_HTTP_TOKEN=dcb93655-0661-4ea1-bfc4-e5744317f99e

[root@wuli-centOS7 ~]# source /etc/profile

在docker运行中的consul容器,目前不清楚怎么修改环境变量永久生效,但可以rm移除旧容器,然后在run的时候添加上环境变量,这里就不演示了。

查看consul_server_1容器的全部环境变量

1
2
3
4
5
6
7
[root@wuli-centOS7 ~]# docker exec consul_server_1 env  
PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
HOSTNAME=2c2c68c4058f
CONSUL_BIND_INTERFACE=eth0
CONSUL_VERSION=1.7.2
HASHICORP_RELEASES=https://releases.hashicorp.com
HOME=/root

查资料CONSUL_BIND_INTERFACE是什么参数

配置agent token

agent token是每个集群都需要的token
没有配置agent token,查看日志报以下警告:

1
2
3
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul monitor -token=dcb93655-0661-4ea1-bfc4-e5744317f99e  

2020-05-06T02:40:09.386Z [WARN]  agent: Coordinate update blocked by ACLs: accessorID=
  1. 使用API生成全权限的token作为agent token,可以根据实际情况分配agent token的权限
1
2
3
4
5
6
[root@wuli-centOS7 ~]# curl -X PUT \
  http://localhost:8510/v1/acl/create \
  -H 'X-Consul-Token: dcb93655-0661-4ea1-bfc4-e5744317f99e' \
  -d '{"Name": "dc1","Type": "management"}'
 
{"ID":"7f587432-3650-9073-e3f4-445a2463b11f"}

  1. 把生成的token写入到配置文件acl.hcl中
1
[root@wuli-centOS7 ~]# vim /data/consul/server1/config/acl.hcl
1
2
3
4
5
6
7
8
9
10
primary_datacenter = "dc1"
acl {
  enabled = true
  default_policy = "deny"
  enable_token_persistence = true
  tokens {
    master = "dcb93655-0661-4ea1-bfc4-e5744317f99e"
    agent = "7f587432-3650-9073-e3f4-445a2463b11f"
  }
}
  1. 重载配置文件,这里无需重启consul或者容器
1
2
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul reload -token=dcb93655-0661-4ea1-bfc4-e5744317f99e
Configuration reload triggered
  1. 查看日志,重载配置文件后,已经不报任何警告了
1
2
3
4
[root@wuli-centOS7 ~]# docker exec consul_server_1 consul monitor -token=dcb93655-0661-4ea1-bfc4-e5744317f99e

2020-05-06T03:00:20.034Z [INFO]  agent: Caught: signal=hangup
2020-05-06T03:00:20.034Z [INFO]  agent: Reloading configuration...
  1. 在集群中acl的配置信息是一致的,所以直接把server1的acl.hcl配置文件复制到其他server节点的配置文件夹下即可
1
2
[root@wuli-centOS7 config]# cp /data/consul/server1/config/acl.hcl /data/consul/server2/config/
[root@wuli-centOS7 config]# cp /data/consul/server1/config/acl.hcl /data/consul/server3/config/
  1. 重启其他容器,查看集群情况
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
[root@wuli-centOS7 config]# docker restart consul_server_2
consul_server_2

[root@wuli-centOS7 config]# docker restart consul_server_3
consul_server_3

[root@wuli-centOS7 config]# docker exec consul_server_1 consul members -token=dcb93655-0661-4ea1-bfc4-e5744317f99e
Node             Address          Status  Type    Build  Protocol  DC   Segment
consul_server_1  172.17.0.2:8301  alive   server  1.7.2  2         dc1  <all>
consul_server_2  172.17.0.3:8301  alive   server  1.7.2  2         dc1  <all>
consul_server_3  172.17.0.4:8301  alive   server  1.7.2  2         dc1  <all>

[root@wuli-centOS7 config]# docker exec consul_server_1 consul operator raft list-peers -token=dcb93655-0661-4ea1-bfc4-e5744317f99e                
Node             ID                                    Address          State     Voter  RaftProtocol
consul_server_1  0214e0b9-04fe-e8ad-c855-b5091dfc8e2e  172.17.0.2:8300  leader    true   3
consul_server_2  78293668-16a6-1de0-673f-455d594e7447  172.17.0.3:8300  follower  true   3
consul_server_3  0b6169d8-7acc-ed24-682f-56ffd12b486c  172.17.0.4:8300  follower  true   3

可以看到,由于前面配置了join参数,所以节点会自动加入集群。
7. 分别访问各个server的ui,查看到的token信息是一致的
在这里插入图片描述

至此,consul集群搭建和ACL token权限配置完成!

其他联想:
如果第一个节点配置了bootstrap:true或者bootstrap_expect=1,那么自己就会是领导,后面的节点在启动前,就可以在配置文件中加入start_join和retry_join的配置信息,然后启动,自动加入集群;
如果第一个节点配置的是bootstrap_expect=n(n大于1),那么由于没有达到节点数,所以一直不会有leader节点,所以后面的节点在启动前,如果配置了start_join和retry_join就会报错,所以只能在启动后,通过consul join命令来加入集群。