準(zhǔn)備工作：

centos7 鏡像地址：https://developer.aliyun.com/mirror/

redis 下載地址：http://download.redis.io/releases/redis-6.0.6.tar.gz

• 首先查看防火墻的狀態(tài)

export PATH=/usr/local/bin:$PATH
systemctl status firewalld.service

• 然后執(zhí)行命令進(jìn)行關(guān)閉

systemctl stop firewalld service

• systemctl disable firewalld.service，開機(jī)禁止防火墻服務(wù)器
• systemctl enable firewalld.service橄务，開機(jī)啟動(dòng)防火墻服務(wù)器

配置redis過程中可能遇到的問題藏杖，前置配置解決

Increased maximum number of open files to 10032 (it was originally set to 1024).

• 查看文件打開數(shù)配置

ulimit -a

• 配置方案

ulimit -n 65535
echo ulimit -n 65535 >>/etc/profile

WARNING: The TCP backlog setting of 511 cannot be enforced because /proc/sys/net/core/somaxconn is set to the lower value of 128.

• 臨時(shí)方案

echo 1024 > /proc/sys/net/core/somaxconn

• 配置方案

echo net.core.somaxconn= 1024 >> /etc/sysctl.conf && sysctl -p

WARNING overcommit_memory is set to 0! Background save may fail under low memory condition. To fix this issue add 'vm.overcommit_memory = 1' to /etc/sysctl.conf and then reboot or run the command 'sysctl vm.overcommit_memory=1' for this to take effect.

• 配置方案

echo vm.overcommit_memory = 1 >> /etc/sysctl.conf && sysctl -p

安裝

如果大家在centos7下編譯redis6.0版本凡泣，如果出現(xiàn)以下錯(cuò)誤

In file included from server.c:30:0:
server.h:1022:5: error: expected specifier-qualifier-list before ‘_Atomic’
_Atomic unsigned int lruclock; /* Clock for LRU eviction */

請(qǐng)先檢查gcc的版本是否低于5箫措，如果是請(qǐng)先升級(jí)桦山，可以使用以下命令：

sudo yum install centos-release-scl
sudo yum install devtoolset-7-gcc*
scl enable devtoolset-7 bash

開始安裝

sudo su root ## 先切換到 root 身份下
sudo mkdir /opt/src /opt/redis
cd /opt/src
sudo wget http://download.redis.io/releases/redis-6.0.6.tar.gz
sudo tar xzf redis-6.0.6.tar.gz
cd redis-6.0.6
sudo make && make install

修改安裝腳本

vim ./utils/install_server.sh
-- 注釋掉這段代碼
#bail if this system is managed by systemd
#_pid_1_exe="$(readlink -f /proc/1/exe)"
#if [ "${_pid_1_exe##*/}" = systemd ]
#then
#       echo "This systems seems to use systemd."
#       echo "Please take a look at the provided example service unit files in this directory, and adapt and install them. Sorry!"
#       exit 1
#fi
#unset _pid_1_exe

執(zhí)行安裝

sudo ./utils/install_server.sh

配置文件目錄輸入這個(gè)地址：/opt/redis/6379/6379.conf
其他的都可以默認(rèn)尘分，一直回車

修改啟動(dòng)腳本（不設(shè)置連接密碼的排拷，可以不修改）

sudo vim /etc/init.d/redis_6379
REDISPORT=6379 下加入
PASSWORD=$(cat $CONF|grep '^\s*requirepass'|awk '{print $2}'|sed 's/"http://g')
替換：

    stop)
        if [ ! -f $PIDFILE ]
        then
            echo "$PIDFILE does not exist, process is not running"
        else
            PID=$(cat $PIDFILE)
            echo "Stopping ..."
            if [ -z $PASSWORD ]
            then 
                $CLIEXEC -p $REDISPORT shutdown
            else
                $CLIEXEC -a $PASSWORD -p $REDISPORT shutdown
            fi
            while [ -x /proc/${PID} ]
            do
                echo "Waiting for Redis to shutdown ..."
                sleep 1
            done
            echo "Redis stopped"
        fi
        ;;

修改配置文件

• 修改成允許所有ip連接
• AOF開啟
• 重啟一下Redis，安裝就好了

sed -i 's/# bind 127.0.0.0/bind 0.0.0.0/g' /opt/redis/6379/6379.conf
sed -i 's/bind 127.0.0.0/bind 0.0.0.0/g' /opt/redis/6379/6379.conf
sed -i 's/appendonly no/appendonly yes/g' /opt/redis/6379/6379.conf
service redis_6379 restart

• 拷貝兩份Redis配置文件和啟動(dòng)文件缺前，一會(huì)用來做集群的搭建

mkdir /opt/redis/cluster
cp /opt/redis/6379/6379.conf /opt/redis/cluster/6380.conf
cp /opt/redis/6379/6379.conf /opt/redis/cluster/6381.conf

• 配置sentinel哨兵文件

cp /opt/src/redis-6.0.6/sentinel.conf /opt/redis/6379/
sed -i 's/daemonize no/daemonize yes/g' /opt/redis/6379/sentinel.conf
sed -i 's/logfile ""/logfile "\/var\/log\/sentinel.log"/g' /opt/redis/6379/sentinel.conf
sed -i 's/monitor mymaster 127.0.0.1 6379 2/monitor mymaster [你的master ip] 6379 2/g' /opt/redis/6379/sentinel.conf
echo "bind 0.0.0.0" >> /opt/redis/6379/sentinel.conf
echo "sentinel down-after-milliseconds mymaster 5000" >> /opt/redis/6379/sentinel.conf

## sentinel auth-pass mymaster <password> 配置密碼的話需要填寫密碼

• 配置sentinel哨兵啟動(dòng)文件

cp /etc/init.d/redis_6379 /etc/init.d/sentinel
sed -i 's/redis_6379.pid/redis-sentinel.pid/g' /etc/init.d/sentinel
sed -i 's/REDISPORT="6379"/REDISPORT="26379"/g' /etc/init.d/sentinel
sed -i 's/6379.conf/sentinel.conf/g' /etc/init.d/sentinel
sed -i 's/redis_6379/sentinel/g' /etc/init.d/sentinel
sed -i 's/Redis/Sentinel/g' /etc/init.d/sentinel
sed -i 's/redis-server/redis-sentinel/g' /etc/init.d/sentinel

• 修改 6380 配置

mkdir /var/lib/redis/6380
sed -i 's/port 6379/port 6380/g' /opt/redis/cluster/6380.conf
sed -i 's/redis_6379/redis_6380/g' /opt/redis/cluster/6380.conf
sed -i 's/dir \/var\/lib\/redis\/6379/dir \/var\/lib\/redis\/6380/g' /opt/redis/cluster/6380.conf
sed -i 's/# cluster-enabled yes/cluster-enabled yes/g' /opt/redis/cluster/6380.conf
sed -i 's/# cluster-node-timeout 15000/cluster-node-timeout 15000/g' /opt/redis/cluster/6380.conf

• 修改 6381 配置

/var/lib/redis/6381
sed -i 's/port 6379/port 6381/g' /opt/redis/cluster/6381.conf
sed -i 's/redis_6379/redis_6381/g' /opt/redis/cluster/6381.conf
sed -i 's/dir \/var\/lib\/redis\/6379/dir \/var\/lib\/redis\/6381/g' /opt/redis/cluster/6381.conf
sed -i 's/# cluster-enabled yes/cluster-enabled yes/g' /opt/redis/cluster/6381.conf
sed -i 's/# cluster-node-timeout 15000/cluster-node-timeout 15000/g' /opt/redis/cluster/6381.conf

• 配置cluster啟動(dòng)服務(wù)

cp /etc/init.d/redis_6379 /etc/init.d/redis_6380; cp /etc/init.d/redis_6379 /etc/init.d/redis_6381
sed -i 's/6379/6380/g' /etc/init.d/redis_6380
sed -i 's/\/opt\/redis\/6380/\/opt\/redis\/cluster/g' /etc/init.d/redis_6380
sed -i 's/6379/6381/g' /etc/init.d/redis_6381
sed -i 's/\/opt\/redis\/6381/\/opt\/redis\/cluster/g' /etc/init.d/redis_6381

• 軟鏈一下

ln -s /usr/local/bin/redis-cli /usr/bin/redis-cli

虛擬機(jī)的話蛀醉，克隆當(dāng)前機(jī)器出兩臺(tái)來，用來搭建集群

主從配置

• slave (兩臺(tái)機(jī)器)上執(zhí)行

sed -i 's/# replicaof <masterip> <masterport>/replicaof [你的master ip] 6379/g' /opt/redis/6379/6379.conf

• 重啟 slave (兩臺(tái)機(jī)器)上的redis

service redis_6379 restart

• 查看日志是否連接成功

tail -f /var/log/redis_6379.log

• 登錄redis 查看是否有主從信息

redis-cli
127.0.0.1:6379> info replication

主從-主-info-replication.png

• master 主上設(shè)置值

127.0.0.1:6379> set master master

slave 機(jī)器上

• 登錄redis 查看是否有主從信息衅码，并獲取master設(shè)置的值

redis-cli
127.0.0.1:6379> info replication
127.0.0.1:6379> get master

主從-從-info-replication.png

sentinel 哨兵配置

• master 機(jī)器上重啟sentinel

service sentinel restart

• [如果是克隆出來的鏡像的話] slave機(jī)器上拯刁，注釋掉myid 不能重復(fù)，并執(zhí)行重啟逝段，自動(dòng)生成

sed -i 's/sentinel myid/# sentinel myid/g' /opt/redis/6379/sentinel.conf
service sentinel restart

• [主備切換] master 機(jī)器上重啟停掉redis

service redis_6379 stop

• 查看日志是否有自動(dòng)切換master的日志

tail -f /var/log/sentinel.log

• 成功切換master之后垛玻，重新啟動(dòng)redis

service redis_6379 start

• 登錄redis

redis-cli
127.0.0.1:6379> info replication

• 查看是否已經(jīng)從 master -> slave

集群配置

• 主從機(jī)器，全部啟動(dòng) 6380 6381

service redis_6380 start
service redis_6381 start

• 創(chuàng)建集群 [ip替換自己的集群ip]

redis-cli --cluster create 10.211.55.7:6380 10.211.55.7:6381 10.211.55.8:6380 10.211.55.8:6381 10.211.55.9:6380 10.211.55.9:6381 --cluster-replicas 1

• 登錄集群 [-c 集群方式登錄]

redis-cli -c -p 6380

• 查看集群信息

127.0.0.1:6380> cluster info

cluster-cluster-info.png

• 列出節(jié)點(diǎn)信息

127.0.0.1:6380> cluster nodes

cluster-cluster-nodes.png

• 設(shè)置一個(gè)值

127.0.0.1:6380> set cluster cluster

• 在集群中任一節(jié)點(diǎn)進(jìn)去后奶躯，是否都能獲取到值

127.0.0.1:6380> get cluster

寫在最后：結(jié)合官方的安裝腳本帚桩，理論上能實(shí)現(xiàn)一個(gè)集群安裝腳本的