Ansible部署kafka集群

节点	IP	主机名
ansible	172.25.253.137	ansible
node1	172.25.253.138	zookeeper1
node2	172.25.253.139	zookeeper2
node3	172.25.253.140	zookeeper3

查看文件树

[root@ansible ansible]# tree role/ 
role/
├── kafka
│   ├── file
│   │   └── run.sh
│   ├── tasks
│   │   ├── kafka.yaml
│   │   └── main.yaml
│   ├── templates
│   │   ├── server.properties1.j2
│   │   ├── server.properties2.j2
│   │   └── server.properties3.j2
│   └── vars
│       └── main.yaml
├── playbook.yaml
└── zookeeper
    ├── file
    │   ├── hosts
    │   ├── local.repo
    │   └── run.sh
    ├── tasks
    │   ├── main.yaml
    │   ├── start.yaml
    │   └── zookeeper.yaml
    ├── templates
    │   ├── myid1.j2
    │   ├── myid2.j2
    │   ├── myid3.j2
    │   └── zoo.cfg.j2
    ├── vars
    │   └── main.yaml
    └── zookeeper-3.4.14.tar.gz

Ansible部署Kafka集群

• 部署kafka集群需要提前部署zookeeper集群，所以要部署zookeeper集群
• 使用role的两个角色扮演的部署集群方式
• 角色分别是zookeeper角色和kafka角色
• zookeeper角色主要是负责zookeeper环境的部署
• kafka角色主要是负责kafka集群的部署

配置ansible的主机清单

[root@ansible ansible]# cat /etc/ansible/hosts  

[server]
172.25.253.138 
172.25.253.140
172.25.253.139

创建角色和角色需要的文件

• 分别创建file tasks vars templates
• 并且上传两个角色相对应的压缩包

[root@ansible ansible]# mkdir  role/{kafka,zookeeper}
[root@ansible ansible]# touch role/kafka/{file,vars,tasks,templates}
[root@ansible ansible]# touch role/zookeeper/{file,vars,tasks,templates}

[root@ansible ansible]# ll role/zookeeper/ 
drwxr-xr-x 2 root root       51 Oct 13 22:12 file
drwxr-xr-x 2 root root       63 Oct 14 13:33 tasks
drwxr-xr-x 2 root root       72 Oct 14 13:40 templates
drwxr-xr-x 2 root root       23 Oct 14 13:38 vars
-rw-r--r-- 1 root root 37676320 Nov 12  2020 zookeeper-3.4.14.tar.gz

[root@ansible ansible]# ll role/kafka/ 
drwxr-xr-x 2 root root 20 Oct 14 11:39 file
drwxr-xr-x 2 root root 41 Oct 14 13:15 tasks
drwxr-xr-x 2 root root 93 Oct 14 11:28 templates
drwxr-xr-x 2 root root 23 Oct 14 11:26 vars
-rw-r--r-- 1 root root 57471165 Nov 12  2020 kafka_2.11-1.1.1.tgz

Zookeeper角色

Zookeeper

1.基础环境部署

• 复制hosts文件到file目录下并且修改
• 复制local.repo到file目录下并且修改
• 编写run.sh脚本作为zookeeper集群的启动脚本

[root@ansible file]# cat hosts  
127.0.0.1   localhost localhost.localdomain localhost4 localhost4.localdomain4
::1         localhost localhost.localdomain localhost6 localhost6.localdomain6


172.25.253.138 zookeeper1 
172.25.253.139 zookeeper2 
172.25.253.140 zookeeper3

[root@ansible file]# cat  local.repo  
[centos]
name=centos
baseurl=http://172.25.253.144/file/centos/
gpgcheck=0
[gpmall-repo]
name=gpmall-repo
baseurl=http://172.25.253.144/file/gpmall-repo/
gpgcheck=0

[root@ansible file]# cat  run.sh  
#!/bin/bash
cd /opt/zookeeper-3.4.14/bin/
./zkServer.sh start
./zkServer.sh restart

2.编写playbook

• 在tasks目录里编写zookeeper的清单
• 编写zookeeper.yaml作为安装zookeeper的环境清单
• 编写start.yaml作为开启集群的清单
• 编写main.yaml指定顺序

[root@ansible tasks]# cat zookeeper.yaml  
- name: copy hosts 
  copy: src=./file/hosts dest=/etc/hosts 
- name: delete repo 
  shell: rm -rf /etc/yum.repos.d/*
- name: copy repo 
  copy: src=./file/local.repo  dest=/etc/yum.repos.d/local.repo 
- name: Install Packages 
  yum: 
    name: 
      - java-1.8.0-openjdk 
      - java-1.8.0-openjdk-devel 
    state: latest
- name: tar tar 
  unarchive: src=zookeeper-3.4.14.tar.gz dest=/opt/zookeeper-3.4.14/
- name: delete file 
  file: path=/opt/zookeeper-3.4.14/conf/zoo_sample.cfg state=absent 
- name: add config 
  template: src=zoo.cfg.j2 dest=/opt/zookeeper-3.4.14/conf/zoo.cfg  owner=2002 group=2002 mode=664
- name: create file 
  file: path=/tmp/zookeeper state=directory
- name: myid1 
  template: src=myid1.j2 dest=/tmp/zookeeper/myid
  when: ansible_fqdn == "zookeeper1"
- name: myid2
  template: src=myid2.j2 dest=/tmp/zookeeper/myid
  when: ansible_fqdn == "zookeeper2"
- name: myid3
  template: src=myid3.j2 dest=/tmp/zookeeper/myid
  when: ansible_fqdn == "zookeeper3"

[root@ansible tasks]# cat  start.yaml  
- name: start zookeeper 
  script: /root/ansible/role/zookeeper/file/run.sh 
  when: ansible_fqdn == {{ "item" }}
  with_items: 
    - zookeeper1 
    - zookeeper2 
    - zookeeper3

[root@ansible tasks]# cat main.yaml  
- include: zookeeper.yaml 
- include: start.yaml 
[root@ansible tasks]# 

3.添加变量vars

• 在vars的目录main.yaml添加如下变量
• 指定网段加端口号

[root@ansible vars]# cat main.yaml 
---

ip1: 172.25.253.138:2888:3888
ip2: 172.25.253.139:2888:3888
ip3: 172.25.253.140:2888:3888

4.修改templates的j2文件

• 添加myid的三个文件内容分别是1、2、3
• 配置zoo.cfg文件需要添加三个server的id以及ip

[root@ansible templates]# echo 1 > myid1.j2
[root@ansible templates]# echo 2 > myid2.j2
[root@ansible templates]# echo 3 > myid3.j2

[root@ansible templates]# cat zoo.cfg.j2  
# The number of milliseconds of each tick
tickTime=2000
# The number of ticks that the initial 
# synchronization phase can take
initLimit=10
# The number of ticks that can pass between 
# sending a request and getting an acknowledgement
syncLimit=5
# the directory where the snapshot is stored.
# do not use /tmp for storage, /tmp here is just 
# example sakes.
dataDir=/tmp/zookeeper
# the port at which the clients will connect
clientPort=2181
# the maximum number of client connections.
# increase this if you need to handle more clients
#maxClientCnxns=60
#
# Be sure to read the maintenance section of the 
# administrator guide before turning on autopurge.
#
# http://zookeeper.apache.org/doc/current/zookeeperAdmin.html#sc_maintenance
#
# The number of snapshots to retain in dataDir
#autopurge.snapRetainCount=3
# Purge task interval in hours
# Set to "0" to disable auto purge feature
#autopurge.purgeInterval=1

server.1={{ ip1 }}
server.2={{ ip2 }}
server.3={{ ip3 }}


[root@ansible templates]# ll
total 16
-rw-r--r-- 1 root root   2 Oct 14 13:35 myid1.j2
-rw-r--r-- 1 root root   2 Oct 14 13:35 myid2.j2
-rw-r--r-- 1 root root   2 Oct 13 21:38 myid3.j2
-rw-r--r-- 1 root root 982 Oct 14 13:40 zoo.cfg.j2

Kafka角色

kafka

1.编写playbook

• 在kafka角色中只需要传送压缩包解压缩包括替换文件
• 编排akfka的playboot

[root@ansible tasks]# cat kafka.yaml  
- name: tar  tar
  unarchive: src=kafka/kafka_2.11-1.1.1.tgz  dest=/media/
- name: change config1 
  template: src=server.properties1.j2 dest=/media/kafka_2.11-1.1.1/config/server.properties
  when: ansible_fqdn == "zookeeper1"
- name: change config2
  template: src=server.properties2.j2 dest=/media/kafka_2.11-1.1.1/config/server.properties
  when: ansible_fqdn == "zookeeper2"
- name: change config3
  template: src=server.properties3.j2 dest=/media/kafka_2.11-1.1.1/config/server.properties
  when: ansible_fqdn == "zookeeper3"
- name: run kafka 
  shell: cd /media/kafka_2.11-1.1.1/bin/  && ./kafka-server-start.sh -daemon ../config/server.properties
[root@ansible tasks]# cat main.yaml  
- include: kafka.yaml 

[root@ansible tasks]# cat main.yaml  
- include: kafka.yaml 

2.添加变量vars

• 在vars的目录中添加相对应的变量

[root@ansible vars]# cat main.yaml  
---

id1: 1 
id2: 2 
id3: 3 
cluster_ip: 172.25.253.138:2181,172.25.253.139:2181,172.25.253.140:2181
localip1: 172.25.253.138:9092
localip2: 172.25.253.139:9092
localip3: 172.25.253.140:9092

3.修改j2文件

• 修改templates的目录下的三个j2文件

在配置文件中找到以下两行并注释掉（在文本前加#）如下所示：
#broker.id=0
#zookeeper.connect=localhost:2181



然后在配置文件的底部添加如下 3 个配置。三个节点分别修改不同的变量 zookeeper1使用id1、zookeeper2使用id2
zookeeper1 节点：
broker.id= {{ id1  }}
zookeeper.connect= {{ cluster_ip }}
listeners=PLAINTEXT://{{ localip1 }}

zookeeper2 节点：
broker.id= {{ id2  }}
zookeeper.connect= {{ cluster_ip }}
listeners=PLAINTEXT://{{ localip2 }}

zookeeper3 节点：
broker.id= {{ id3  }}
zookeeper.connect= {{ cluster_ip }}
listeners=PLAINTEXT://{{ localip3 }}

[root@ansible templates]# ll
total 24
-rw-r--r-- 1 root root 6953 Oct 14 11:27 server.properties1.j2
-rw-r--r-- 1 root root 6952 Oct 14 11:28 server.properties2.j2
-rw-r--r-- 1 root root 6951 Oct 14 11:28 server.properties3.j2

Role的playbook

5.编写role的playbook分步执行

[root@ansible role]# cat playbook.yaml 
- hosts: server 
  remote_user: root 
  
  roles:
    - role: zookeeper
    - role: kafka

集群验证

1.执行脚本

[root@ansible role]# ansible-playbook playbook.yaml  

PLAY [server] **************************************************************************************************************

TASK [Gathering Facts] *****************************************************************************************************
ok: [172.25.253.139]
ok: [172.25.253.140]
ok: [172.25.253.138]

TASK [zookeeper : copy hosts] **********************************************************************************************
ok: [172.25.253.140]
ok: [172.25.253.139]
ok: [172.25.253.138]
·
·
·
PLAY RECAP *****************************************************************************************************************
172.25.253.138             : ok=14   changed=8    unreachable=0    failed=0    skipped=4    rescued=0    ignored=0   
172.25.253.139             : ok=14   changed=8    unreachable=0    failed=0    skipped=4    rescued=0    ignored=0   
172.25.253.140             : ok=14   changed=8    unreachable=0    failed=0    skipped=4    rescued=0    ignored=0   

2.验证集群是否正常

• 查看端口开放

[root@ansible role]# ansible server -m shell -a "ss -ntpl | grep 9092"
172.25.253.140 | CHANGED | rc=0 >>
LISTEN     0      50       ::ffff:172.25.253.140:9092                    :::*                   users:(("java",pid=6105,fd=99))
172.25.253.139 | CHANGED | rc=0 >>
LISTEN     0      50       ::ffff:172.25.253.139:9092                    :::*                   users:(("java",pid=2783,fd=99))
172.25.253.138 | CHANGED | rc=0 >>
LISTEN     0      50       ::ffff:172.25.253.138:9092                    :::*                   users:(("java",pid=5814,fd=99))

• 验证

zookeeper1：
[root@localhost bin]# ./kafka-topics.sh --create --zookeeper 172.25.253.138:2181 --replication-factor 1 --partitions 1 --topic test
Created topic "test".

zookeeper2：
[root@localhost bin]# ./kafka-topics.sh --list --zookeeper 172.25.253.139:2181   
test