部署文档

目录

1 架构说明

2 安装mysql

2.1 下载软件包

2.2 安装

3 安装rabbitmq

3.1 安装erlang

3.2 部署rabbitmq

4 安装redis

4.1 下载软件包

4.2 编译部署

4.3 启动redis

5 安装mongodb

5.1 下载软件包

5.2 安装

5.3 主库操作

5.4 从库操作

5.5 延迟从库操作

5.6 测试

1 架构说明

操作系统Ubuntu-Server-18.04.x-LTS

注: 此部署文档为测试环境，为了节省机器，复用了些集群，生产环境各节点机器独立

服务器优化配置

# sudo su - root

(1) 所有节点时间保持一致
# ln -sf /usr/share/zoneinfo/Asia/Shanghai /etc/localtime
# apt-get install ntpdate -y
# if ! crontab -l | grep "ntpdate" &>/dev/null; then
    (echo "*/5 * * * * /usr/sbin/ntpdate ntp.aliyun.com >/dev/null 2>&1"; crontab -l) | crontab
fi

(2) 使用sudo命令时加载/etc/profile中的环境变量
# echo 'alias sudo="sudo env PATH=$PATH"' >> /etc/profile
# source /etc/profile

(3) 解决ssh连接速度慢的问题
# cat >> /etc/ssh/sshd_config << EOF
GSSAPIAuthentication no
UseDNS no
EOF
# systemctl restart sshd.service

(4) 系统内核优化
cat >> /etc/sysctl.conf << EOF
net.ipv4.tcp_tw_reuse = 1
net.ipv4.ip_local_port_range = 1024 65000
net.ipv4.tcp_syncookies = 1
net.ipv4.tcp_max_tw_buckets = 20480
net.ipv4.tcp_max_syn_backlog = 20480
net.core.netdev_max_backlog = 262144
net.ipv4.tcp_fin_timeout = 20
vm.swappiness = 0
EOF
# sysctl -p

(5) 设置最打开文件描述符数量
参考文档: https://askubuntu.com/questions/1049058/how-to-increase-max-open-files-limit-on-ubuntu-18-04
# cat >> /etc/security/limits.conf << EOF
* soft     nproc          65535
* hard     nproc          65535
* soft     nofile         65535
* hard     nofile         65535

root soft     nproc          65535
root hard     nproc          65535
root soft     nofile         65535
root hard     nofile         65535
EOF

# su -
# ulimit -n

(6) 创建lc用户

# sudo useradd lc -m -s /bin/bash

# echo "lc:123456"|sudo chpasswd

# sudo usermod -aG sudo lc

# su - lc

2 安装mysql

在172.16.1.81节点上操作

2.1 下载软件包

# wget https://mirrors.tuna.tsinghua.edu.cn/mysql/downloads/MySQL-8.0/mysql-8.0.26-linux-glibc2.12-x86_64.tar.xz

2.2 安装

1 前期准备

(1) 安装依赖包
# apt-get install libaio1 -y

(2) 解压安装包
# tar -xf mysql-8.0.26-linux-glibc2.12-x86_64.tar.xz
# mv mysql-8.0.26-linux-glibc2.12-x86_64/ /opt/mysql/

(3) 创建mysql 用户
# groupadd -g 1600 mysql
# useradd -g mysql -u 1600 -M -s /sbin/nologin mysql
# id mysql
uid=1600(mysql) gid=1600(mysql) 组=1600(mysql)

(4) 创建存放数据库文件的目录并赋予mysql安装目录的属主和属组都是mysql
# mkdir -p /opt/mysql/data/
# chown -R mysql.mysql /opt/mysql/

2 配置my.cnf 文件

# cat > /etc/my.cnf << EOF
[client]
port = 3306
default-character-set = utf8mb4
socket = /opt/mysql/data/mysql.sock
[mysql]
no-auto-rehash
[mysqld]
user = mysql
port = 3306
basedir = /opt/mysql
datadir = /opt/mysql/data
socket = /opt/mysql/data/mysql.sock
bind-address = 0.0.0.0
pid-file = /opt/mysql/data/mysqld.pid
character-set-server = utf8mb4
collation-server = utf8mb4_0900_ai_ci
log-error = /opt/mysql/data/mysqld.log
slow_query_log = ON
long_query_time = 2
slow_query_log_file = /opt/mysql/data/mysql-slow.log
max_connections = 1024
open_files_limit = 65535
innodb_buffer_pool_size = 2G
innodb_flush_log_at_trx_commit = 2
innodb_log_file_size = 256M
transaction_isolation = READ-COMMITTE
default-storage-engine = innodb
innodb_file_per_table = on
innodb_flush_method = O_DIRECT
interactive_timeout = 1800
wait_timeout = 1800
innodb_open_files=1500
innodb_io_capacity=5000
innodb_buffer_pool_instances=4
innodb_autoextend_increment=64
innodb_purge_threads=1
innodb_purge_batch_size=150
innodb_write_io_threads=4
innodb_read_io_threads=4
innodb_concurrency_tickets=2500
explicit_defaults_for_timestamp = 1
skip-name-resolve
lower_case_table_names = 1
server-id = 1
skip-log-bin
[mysqldump]
quick
max_allowed_packet = 32M
EOF

3 初始化mysql 数据库

# /opt/mysql/bin/mysqld --initialize --user=mysql --basedir=/opt/mysql --datadir=/opt/mysql/data

4 将mysql 服务加入到systemctl

# cat /lib/systemd/system/mysqld.service
[Unit]
Description=MySQL Server 8.0.26
Documentation=man:mysqld(8)
Documentation=http://dev.mysql.com/doc/refman/en/using-systemd.html
After=network.target
After=syslog.target
[Install]
WantedBy=multi-user.target
[Service]
User=mysql
Group=mysql
Type=forking
PIDFile=/opt/mysql/data/mysqld.pid
TimeoutSec=0
PermissionsStartOnly=true
# ExecStartPre=/usr/local/mysql/bin/mysqld_pre_systemd
ExecStart=/opt/mysql/bin/mysqld --defaults-file=/etc/my.cnf --daemonize --pid-file=/opt/mysql/data/mysqld.pid $MYSQLD_OPTS
EnvironmentFile=-/etc/sysconfig/mysql
LimitNOFILE = 65535
Restart=on-failure
RestartPreventExitStatus=1
PrivateTmp=false

5 启动mysql 数据库

# systemctl daemon-reload
# systemctl start mysqld.service
# systemctl enable mysqld.service
# cat >> /etc/profile << EOF
export PATH=/opt/mysql/bin:\$PATH
EOF
# source /etc/profile

6 修改数据库密码

# grep 'temporary password' /opt/mysql/data/mysqld.log
……[Note] [MY-010454] [Server] A temporary password is generated for root@localhost: dHZkG>feg6+A
# mysql -uroot -p'dHZkG>feg6+A'
mysql> ALTER USER 'root'@'localhost' IDENTIFIED BY '123456';
mysql> update mysql.user set host='127.0.0.1' where user='root';
mysql> flush privileges;
mysql> exit;
# mysql -h127.0.0.1 -uroot -p'123456'

3 安装rabbitmq

172.16.1.84-86节点

3.1 安装erlang

在172.16.1.84节点上操作

1 下载erlang软件包

# wget http://erlang.org/download/otp_src_24.0.tar.gz

2 安装依赖包

# apt-get update
# apt-get install build-essential libncurses5-dev libssl-dev -y

3 编译

# tar -xzf otp_src_24.0.tar.gz
# cd otp_src_24.0/
# ./configure --prefix=/opt/erlang --without-javac
# make && make install

验证
# /opt/erlang/bin/erl
Erlang/OTP 24 [erts-12.0] [source] [64-bit] [smp:1:1] [ds:1:1:10] [async-threads:1] [jit]

Eshell V12.0  (abort with ^G)
1> 10+20 .
30
2>
按两下Ctrl + c退出。

4 将编译好的erlang拷贝到172.16.1.85-86两台机器上

# scp -rp /opt/erlang root@172.16.1.85:/opt/
# scp -rp /opt/erlang root@172.16.1.86:/opt/

5 配置环境变量

在172.16.1.84-86节点上操作

# echo 'export PATH=/opt/erlang/bin:$PATH' >> /etc/profile
# source /etc/profile

3.2 部署rabbitmq

在172.16.1.84节点上操作

1 下载rabbitmq软件包

# wget https://github.com/rabbitmq/rabbitmq-server/releases/download/v3.8.22/rabbitmq-server-generic-unix-3.8.22.tar.xz

2 配置rabbitmq配置文件

(1) 解压包
# tar -xf rabbitmq-server-generic-unix-3.8.22.tar.xz
# mv rabbitmq_server-3.8.22/ /opt/rabbitmq/

(2) rabbitmq-server启动脚本中添加erlang环境变量，解决脚本找不到erl命令
# vim /opt/rabbitmq/sbin/rabbitmq-server +8
export PATH=$PATH:/opt/erlang/bin

(3) 添加rabbitmq配置文件
# echo '[{rabbit,[{tcp_listeners,[5672]}]}].' > /opt/rabbitmq/etc/rabbitmq/rabbitmq.config

3 将解压的rabbitmq软件包拷贝到172.16.1.85-86两台机器上

# scp -rp /opt/rabbitmq root@172.16.1.85:/opt/
# scp -rp /opt/rabbitmq root@172.16.1.86:/opt/

4 启动rabbitmq

分别在172.16.1.84-86节点上操作

(1) 添加环境变量
# cat >> /etc/profile << EOF
export RABBITMQ_HOME=/opt/rabbitmq
export PATH=/opt/rabbitmq/sbin:\$PATH
EOF
# source /etc/profile

(2) 创建用户并赋权
# useradd rabbitmq
# mkdir -p /home/rabbitmq/
# chown -R rabbitmq.rabbitmq /home/rabbitmq/
# chown -R rabbitmq.rabbitmq /opt/rabbitmq/

(3) 加入systemd管理
# cat rabbitmq-server.service
[Unit]
Description=RabbitMQ broker
After=syslog.target network.target

[Service]
Type=simple
User=rabbitmq
Group=rabbitmq
# UMask=0027
NotifyAccess=all
TimeoutStartSec=600
SuccessExitStatus=127
# To override LimitNOFILE, create the following file:
#
# /etc/systemd/system/rabbitmq-server.service.d/limits.conf
#
# with the following content:
#
# [Service]
# LimitNOFILE=65536

LimitNOFILE=32768

# Note: systemd on CentOS 7 complains about in-line comments,
# so only append them here
#
# Restart:
# The following setting will automatically restart RabbitMQ
# in the event of a failure. systemd service restarts are not a
# replacement for service monitoring. Please see
# https://www.rabbitmq.com/monitoring.html
Restart=on-failure
RestartSec=10
WorkingDirectory=/opt/rabbitmq
ExecStart=/opt/rabbitmq/sbin/rabbitmq-server
ExecStop=/opt/rabbitmq/sbin/rabbitmqctl shutdown
# See rabbitmq/rabbitmq-server-release#51
SuccessExitStatus=69
Restart=on-failure

[Install]
WantedBy=multi-user.target

(4) 启动
# systemctl daemon-reload
# systemctl start rabbitmq-server.service
# systemctl enable rabbitmq-server.service

(5) 开启rabbitmq web管理界面
# /opt/rabbitmq/sbin/rabbitmq-plugins enable rabbitmq_management
# systemctl restart rabbitmq-server.service

(6) 验证
可以通过端口进行访问，出现了rabbitmq UI界面，说明rabbitmq搭建成功。
http://172.16.1.84-86:15672

5 rabbitmq集群配置

分别在172.16.1.84-86节点上操作

(1) 修改hosts解析
# cat >> /etc/hosts << EOF
172.16.1.84 1-84
172.16.1.85 1-85
172.16.1.86 1-86
EOF

(2) 停止RabbitMQ服务
# systemctl stop rabbitmq-server.service

(3) 设置Erlang Cookie
设置不同节点间同一认证的Erlang Cookie，这里将1-84的Cookie传给1-85、1-86两台服务器
在172.16.1.84节点上操作
# scp -p /home/rabbitmq/.erlang.cookie root@1-85:/home/rabbitmq/
# scp -p /home/rabbitmq/.erlang.cookie root@1-86:/home/rabbitmq/
注:
cookie在所有节点上必须完全一样，且权限必须为400，属主和属组为rabbitmq，同步时一定要注意。erlang是
通过主机名来连接服务，必须保证各个主机名之间可以ping通。可以通过编辑/etc/hosts来手工添加主机名和IP
对应关系。如果主机名ping不通，rabbitmq服务启动会失败。

(4) 设置好cookie后重启rabbitmq
# systemctl restart rabbitmq-server.service

(5) 查看单节点的集群状态(任意一台都可，这里以172.16.1.84节点为例)
# cp -a /home/rabbitmq/.erlang.cookie ./
# chown root.root /root/.erlang.cookie
# rabbitmqctl cluster_status


注: 因为.erlang.cookie在/home/rabbitmq/目录下，要查看rabbitmq集群状态时要使用下面的命令拷贝
.erlang.cookie到当前用户的家目录下，并更改属主属组为当前用户。
# cp -a /home/rabbitmq/.erlang.cookie <当前用户的家目录>
# chown -R <当前用户名>:<当前用户名>  <当前用户的家目录>/.erlang.cookie

(6) 加入集群
这里将1-85和1-86加入到1-84当中，按照顺序执行先1-85再1-86
1) 在1-85上操作
# rabbitmqctl stop_app
# rabbitmqctl join_cluster rabbit@1-84
# rabbitmqctl start_app

2) 在1-86上操作
# rabbitmqctl stop_app
# rabbitmqctl join_cluster rabbit@1-84
# rabbitmqctl start_app

(7) 查看集群状态，任意一个节点有可以，这里以1-84为例子
# rabbitmqctl cluster_status

注: 集群状态正常

(8) 创建用户并授权(在任意节点都可)
# 创建用户
rabbitmqctl add_user admin '123456'
# 添加用户[管理员]标签
rabbitmqctl set_user_tags admin administrator
# 为用户授权可以访问的[virtual /]和操作类型[配置、读、写]
rabbitmqctl set_permissions -p / admin '.*' '.*' '.*'
# 查看所有用户
# rabbitmqctl list_users

(9) 在UI界面中查看集群状态
http://<RabbitmqIP>:15672/

4 安装redis

172.16.1.81-86节点

4.1 下载软件包

# wget https://download.redis.io/releases/redis-5.0.13.tar.gz

4.2 编译部署

在172.16.1.81节点上操作

1 安装依赖包

# apt-get update
# apt-get install build-essential libncurses5-dev libssl-dev -y

2 编译

# tar -xzf redis-5.0.13.tar.gz
# cd redis-5.0.13/
# make
# make install PREFIX=/opt/redis
# mkdir -p /opt/redis/{cfg,logs,data}
# cp -a /root/redis-5.0.13/redis.conf /opt/redis/cfg/

3 修改配置文件

# cat > /opt/redis/cfg/redis.conf << EOF
bind 172.16.1.81
protected-mode yes
port 6379
tcp-backlog 511
timeout 0
tcp-keepalive 300
daemonize yes
supervised no
pidfile /opt/redis/cfg/redis_6379.pid
loglevel notice
logfile /opt/redis/logs/redis_6379.log
databases 16
always-show-logo yes
stop-writes-on-bgsave-error yes
rdbcompression yes
rdbchecksum yes
dbfilename dump.rdb
dir /opt/redis/data
masterauth 123456
replica-serve-stale-data yes
replica-read-only no
repl-diskless-sync no
repl-diskless-sync-delay 5
repl-disable-tcp-nodelay no
replica-priority 100
requirepass 123456
maxclients 10000
lazyfree-lazy-eviction no
lazyfree-lazy-expire no
lazyfree-lazy-server-del no
replica-lazy-flush no
appendonly yes
appendfilename "appendonly.aof"
appendfsync everysec
no-appendfsync-on-rewrite no
auto-aof-rewrite-percentage 100
auto-aof-rewrite-min-size 64mb
aof-load-truncated yes
aof-use-rdb-preamble yes
lua-time-limit 5000
cluster-enabled yes
cluster-config-file /opt/redis/cfg/nodes.conf
cluster-node-timeout 15000
slowlog-log-slower-than 10000
slowlog-max-len 128
latency-monitor-threshold 0
notify-keyspace-events ""
hash-max-ziplist-entries 512
hash-max-ziplist-value 64
list-max-ziplist-size -2
list-compress-depth 0
set-max-intset-entries 512
zset-max-ziplist-entries 128
zset-max-ziplist-value 64
hll-sparse-max-bytes 3000
stream-node-max-bytes 4096
stream-node-max-entries 100
activerehashing yes
client-output-buffer-limit normal 0 0 0
client-output-buffer-limit replica 256mb 64mb 60
client-output-buffer-limit pubsub 32mb 8mb 60
hz 10
dynamic-hz yes
aof-rewrite-incremental-fsync yes
rdb-save-incremental-fsync yes
EOF

4 将redis加入systemd管理

# cat > /lib/systemd/system/redis.service << EOF
[Unit]
Description=Redis persistent key-value database
After=network.target
After=network-online.target
Wants=network-online.target

[Service]
Type=forking
ExecStart=/opt/redis/bin/redis-server /opt/redis/cfg/redis.conf
# ExecStop=/usr/libexec/redis-shutdown
User=redis
Group=redis
Restart=on-failure

[Install]
WantedBy=multi-user.target
EOF

5 将redis软件包拷贝到172.16.1.82-86节点上

# scp -rp /opt/redis root@172.16.1.82:/opt/
# scp -rp /opt/redis root@172.16.1.83:/opt/
# scp -rp /opt/redis root@172.16.1.84:/opt/
# scp -rp /opt/redis root@172.16.1.85:/opt/
# scp -rp /opt/redis root@172.16.1.86:/opt/

# scp -p /lib/systemd/system/redis.service root@172.16.1.82:/lib/systemd/system/
# scp -p /lib/systemd/system/redis.service root@172.16.1.83:/lib/systemd/system/
# scp -p /lib/systemd/system/redis.service root@172.16.1.84:/lib/systemd/system/
# scp -p /lib/systemd/system/redis.service root@172.16.1.85:/lib/systemd/system/
# scp -p /lib/systemd/system/redis.service root@172.16.1.86:/lib/systemd/system/

4.3 启动redis

分别在172.16.1.81-86节点上操作

1  安装集群所需要的依赖包

# apt-get update
# apt-get install ruby rubygems -y
# gem install redis

2 将redis命令加入环境变量

# echo 'export PATH=/opt/redis/bin:$PATH' >> /etc/profile
# source /etc/profile

3 修改redis绑定的ip地址为当前的ip地址

# vim /opt/redis/cfg/redis.conf
bind 172.16.1.81 # 此配置修改为当前的主机的ip地址

4 启动redis服务

# useradd redis
# chown -R redis:redis /opt/redis/
# systemctl daemon-reload
# systemctl restart redis
# systemctl enable redis
# systemctl status redis

5 查看redis启动日志

(1) 警告项解决办法
# cat /opt/redis/logs/redis_6379.log
……

看到依次从上到下有三个错误。

解决办法如下
# cat >> /etc/sysctl.conf << EOF
net.core.somaxconn = 20480
vm.overcommit_memory = 1
EOF
# sysctl -p

# echo never > /sys/kernel/mm/transparent_hugepage/enabled
# 加入开机自启，但是Ubuntu已经弃用了/etc/rc.local配置文件，按照如下方法自己添加
# cat >> /lib/systemd/system/rc-local.service << EOF
[Install]
WantedBy=multi-user.target
Alias=rc-local.service
EOF

# systemctl daemon-reload
# systemctl enable rc-local

# cat > /etc/rc.local << EOF
#!/bin/bash
echo never > /sys/kernel/mm/transparent_hugepage/enabled
EOF

# chmod +x /etc/rc.local

(2) 重启redis
# systemctl restart redis

(3) 再次查看日志
# cat /opt/redis/logs/redis_6379.log

redis警告解决

6 创建集群

在172.16.1.81节点上操作

# redis-cli -a 123456 --cluster create \
172.16.1.81:6379 172.16.1.82:6379 \
172.16.1.83:6379 172.16.1.84:6379 \
172.16.1.85:6379 172.16.1.86:6379 \
--cluster-replicas 1

输入yes，然后回车，集群创建完毕。创建完成后默认前三个节点81-83为主节点，84-86为
从节点。

7 查看集群信息(访问任意一个节点都可以)

# redis-cli -h 172.16.1.81 -a 123456 -c -p 6379
172.16.1.81:6379> cluster info


172.16.1.81:6379> cluster nodes

5 安装mongodb

172.16.1.81-83节点

5.1 下载软件包

# wget https://fastdl.mongodb.org/linux/mongodb-linux-x86_64-3.6.23.tgz

5.2 安装

在172.16.1.81节点上操作

1 解压包

# tar -xzf mongodb-linux-x86_64-3.6.23.tgz
# mv mongodb-linux-x86_64-3.6.23/ /opt/mongodb/
# mkdir -p /opt/mongodb/{data,conf,log}
# touch /opt/mongodb/log/mongodb.log

2 修改配置文件

# cat > /opt/mongodb/conf/mongodb.conf << EOF
dbpath=/opt/mongodb/data
#数据库文件位置
bind_ip=172.16.1.81
#绑定地址，默认127.0.0.1，只能通过本地连接。
port=27017
#端口，默认27017，MongoDB的默认服务TCP端口，监听客户端连接。
journal=true
#启用日志文件，默认启用。
logpath=/opt/mongodb/log/mongodb.log
#日志文件位置，该日志文件必须存在，否则会报错
logappend=true
#以追加方式写入日志。
quiet=true
#这个选项可以过滤掉一些无用的日志信息，若需要调试使用请设置为false。
fork=true
#以守护进程方式运行
EOF

3 将mongodb加入systemd管理

# cat > /lib/systemd/system/mongodb.service << EOF
[Unit]
Description=mongodb 3.6.23
After=network.target

[Service]
Type=forking
#PIDFile=/opt/mongodb/mongodb.pid
ExecStart=/opt/mongodb/bin/mongod -f /opt/mongodb/conf/mongodb.conf
ExecReload=/bin/kill -s HUP \$MAINPID
#ExecStop=/bin/kill -s QUIT \$MAINPID
PrivateTmp=true
User=mongodb
Group=mongodb
Restart=on-failure
 
[Install]
WantedBy=multi-user.target
EOF

4 分别拷贝安装包到另外两个节点

# scp -rp /opt/mongodb root@172.16.1.82:/opt/
# scp -rp /opt/mongodb root@172.16.1.83:/opt/

# scp -p /lib/systemd/system/mongodb.service root@172.16.1.82:/lib/systemd/system/
# scp -p /lib/systemd/system/mongodb.service root@172.16.1.83:/lib/systemd/system/

5 启动mongodb

在172.16.1.81-83节点上操作

(1) 添加环境变量
# echo 'export PATH=/opt/mongodb/bin:$PATH' >> /etc/profile
# source /etc/profile

(2) 授权
# useradd mongodb
# chown -R mongodb:mongodb /opt/mongodb/

(3) 修改mongodb绑定的ip地址为本地ip地址
# vim /opt/mongodb/conf/mongodb.conf
bind_ip=172.16.1.81  # 指定各自服务的的ip地址

(4) 启动
# systemctl daemon-reload
# systemctl restart mongodb.service
# systemctl enable mongodb.service
# systemctl status mongodb.service

6 开启mongodb认证

在172.16.1.81-83节点上操作

(1) 创建管理用户root为root权限
# mongo --host 172.16.1.81 --port 27017
> use admin
> db.createUser({user: "root",pwd: "123456",roles: [ { role: "root", db: "admin" } ]})
> exit

(2) 开启认证
# cat >> /opt/mongodb/conf/mongodb.conf << EOF
#开启认证
auth = true
EOF

(3) 认证登录
# systemctl restart mongodb.service
# mongo --host 172.16.1.81 --port 27017 -uroot -p'123456' admin

5.3 主库操作

在172.16.1.81节操作

1 修改配置文件

# cat >> /opt/mongodb/conf/mongodb.conf << EOF
############主从复制配置####################################
# master节点
master = true
#集群的私钥的完整路径，只对于Replica Set架构有效，
#auth = true 配置此项。
keyFile = /opt/mongodb/conf/mongodb-keyfile
#主节点设置oplog的大小，主节点操作记录存储到local的oplog中，用于从库拉取。
#单位mb，新的操作会覆盖旧的操作。
oplogSize = 40960
EOF

2 创建集群私钥并发送到另外两个节点

# openssl rand -base64 741 > /opt/mongodb/conf/mongodb-keyfile
# chmod 600 /opt/mongodb/conf/mongodb-keyfile
# chown mongodb:mongodb /opt/mongodb/conf/mongodb-keyfile

# scp -p /opt/mongodb/conf/mongodb-keyfile root@172.16.1.82:/opt/mongodb/conf/
# scp -p /opt/mongodb/conf/mongodb-keyfile root@172.16.1.83:/opt/mongodb/conf/
# ssh -p 22 root@172.16.1.82 "chown mongodb:mongodb /opt/mongodb/conf/mongodb-keyfile"
# ssh -p 22 root@172.16.1.83 "chown mongodb:mongodb /opt/mongodb/conf/mongodb-keyfile"

3 重启mongodb

# systemctl restart mongodb.service

5.4 从库操作

在172.16.1.82节点上操作

1 修改配置文件

# cat >> /opt/mongodb/conf/mongodb.conf << EOF
##################从库配置###########################
#开启从服务器
slave = true
# 主数据库端口
source = 172.16.1.81:27017
#集群的私钥的完整路径，只对于Replica Set架构有效，
#auth = true 配置此项。
keyFile = /opt/mongodb/conf/mongodb-keyfile
#延迟复制，单位为秒
#slavedelay = 1800
#当发现从服务器的数据不是最新时，向主服务器请求同步数据
autoresync = true
EOF

2 重启mongodb

# systemctl restart mongodb.service
# systemctl status mongodb.service
# mongo --host 172.16.1.82 --port 27017 -uroot -p'123456' admin
> rs.secondaryOk()
> show dbs

5.5 延迟从库操作

在172.16.1.83节点上操作

1 修改配置文件

# cat >> /opt/mongodb/conf/mongodb.conf << EOF
##################从库配置###########################
#开启从服务器
slave = true
# 主数据库端口
source = 172.16.1.81:27017
#集群的私钥的完整路径，只对于Replica Set架构有效，
#auth = true 配置此项。
keyFile = /opt/mongodb/conf/mongodb-keyfile
#延迟复制，单位为秒
slavedelay = 1800
#当发现从服务器的数据不是最新时，向主服务器请求同步数据
autoresync = true
EOF

2 重启mongodb

# systemctl restart mongodb.service
# systemctl status mongodb.service
# mongo --host 172.16.1.83 --port 27017 -uroot -p'123456' admin
> rs.secondaryOk()
> show dbs

5.6 测试

1 在172.16.1.81主库上创建测试库和数据

# mongo --host 172.16.1.81 --port 27017 -uroot -p'123456' admin
> use master_slave_test
> function add(){var i = 0;for(;i<7;i++){db.persons.insert({"name":"master_slave_test"+i})}}
> add()
> db.persons.find()

2 在172.16.1.82从库上查看

# mongo --host 172.16.1.82 --port 27017 -uroot -p'123456' admin
> use master_slave_test
> rs.secondaryOk()
> db.persons.find()

3 同172.16.1.82节点操作步骤相同，30分钟后在172.16.1.83延迟从库上查看test数据也同步过来了。