前言:
这里只是介绍一下Win10下的集群部署,实际上在 Linux下部署集群是更快的,也是合理有意义的,对 docker熟悉的大佬基本上可以一键部署完成!
目的:
1、为了缓解读取sql压力,可以采用便携式redis进行缓存,减少数据库压力。
2、在并发压力大的情况下,我们需要采用集群避免性能受到影响,而且集群更不容易出现崩溃现象。
1.Go-Redis库的安装以及Redis的安装
Go-Redis库的安装
可以自己go get 或者 git 亦或直接下载放到对应文件夹里即可。
Redis的安装
找最新的zip直接下载即可。
2.简单使用
在代码跑之前需要打开服务端,也就是刚才下载的redis文件夹里面找到redis-server.exe,打开即可。
redis客户端:
client := redis.NewClient(&redis.Options{
Addr: "127.0.0.1:6379",//Redis的端口地址
Password: "",//不用密码
DB: 0,//采用默认DB
})
var ctx = context.Background()
这里的Addr是默认的6379,如果要改的话 (百度吧),这里的ctx是各种操作需要的第一个参数。
只要Addr一样,就可以直接访问之前存放的数据,与client在哪个go文件使用无关,因此可以存放一些即时数据不需要用到MySQL。
Redis的存放是按照哈希存的,和Map数据结构类似。
Redis-Set
ping, err := client.Ping(ctx).Result()//检测连接
fmt.Println(ping, err)
//第二个参数为Key,第三个参数为Value,第四个参数为过期时间,0为不过期
err = client.Set(ctx,"key","myKey",0).Err()
if err != nil {
panic(err)
}
Redis-Get
//第二个参数为Key
val, err := client.Get(ctx,"key").Result()
if err != nil {
panic(err)
}
fmt.Println("key", val)
Redis-Del
//第二个参数为Key
client.Del(ctx,"key")
一些容器的使用
List:
//List容器
client.RPush(ctx,"list","val1")
client.RPush(ctx,"list","val2")
client.RPush(ctx,"list","val3")
//取List长度
rLen, err := client.LLen(ctx,"list").Result()
//遍历
lists,_ := client.LRange(ctx,"list",0, rLen - 1).Result()
fmt.Println("list",lists)
//修改下标为0位置的值
client.LSet(ctx,"list",2,"change")
lists,_ = client.LRange(ctx,"list",0, rLen - 1).Result()
fmt.Println("list",lists)
//清空List
for ; rLen > 0 ; rLen-- {
client.LPop(ctx, "list")
}
其他容器大致和List的操作是一样的,比如Map,Set等等。
它们的对应函数不一样罢了,对应参数顺序大致一样。
Demo
func main(){
client := redis.NewClient(&redis.Options{
Addr: "127.0.0.1:7000",
Password: "",
DB: 0,
})
var ctx = context.Background()
ping, err := client.Ping(ctx).Result()//检测连接
fmt.Println(ping, err)
//第二个参数为Key,第三个参数为Value,第四个参数为过期时间,0为不过期
err = client.Set(ctx,"key","myKey",0).Err()
if err != nil {
panic(err)
}
//第二个参数为Key
val, err := client.Get(ctx,"key").Result()
if err != nil {
panic(err)
}
fmt.Println("key", val)
val2, err := client.Get(ctx,"key1").Result()
if err == redis.Nil {
fmt.Println("key1 does not exists")
} else if err != nil {
panic(err)
} else {
fmt.Println("key1", val2)
}
//第二个参数为Key
client.Del(ctx,"key")
val3, err := client.Get(ctx,"key").Result()
if err == redis.Nil {
fmt.Println("key does not exists")
} else if err != nil {
panic(err)
} else {
fmt.Println("key", val3)
}
//List容器
//第二个参数为Key,第三个参数为Value值,可以输入多个value值一次性Push
//client.RPush(ctx,"list","val1","val2","val3")
client.RPush(ctx,"list","val1")
client.RPush(ctx,"list","val2")
client.RPush(ctx,"list","val3")
//取List长度
rLen, err := client.LLen(ctx,"list").Result()
//遍历
lists,_ := client.LRange(ctx,"list",0, rLen - 1).Result()
fmt.Println("list",lists)
//修改下标为0位置的值
client.LSet(ctx,"list",2,"change")
lists,_ = client.LRange(ctx,"list",0, rLen - 1).Result()
fmt.Println("list",lists)
//清空List
for ; rLen > 0 ; rLen-- {
client.LPop(ctx, "list")
}
rLen, err = client.LLen(ctx,"list").Result()
lists,_ = client.LRange(ctx,"list",0, rLen - 1).Result()
fmt.Println("list",lists)
client.RPush(ctx,"list","val1","val2","val3")
rLen, err = client.LLen(ctx,"list").Result()
lists,_ = client.LRange(ctx,"list",0, rLen - 1).Result()
fmt.Println("list",lists)
client.Close()
}
输出结果:
PONG <nil>
key myKey
key1 does not exists
key does not exists
list [val1 val2 val3]
list [val1 val2 change]
list []
list [val1 val2 val3]
3.Redis集群配置
因为Redis集群是半数淘汰,因此至少需要3个节点才可以使用。
这里采用三主三从,可以更好的避免节点失误导致集群不可用的情况发生。
6个节点配置文件需要修改的地方
port 你要设置的端口号 #(例如7000)
appendonly yes #数据的保存为aof格式
appendfilename "appendonly.7000.aof" #数据保存文件
cluster-enabled yes #是否开启集群
cluster-config-file nodes.7000.conf #集群配置文件
cluster-node-timeout 15000 #节点超时时间,如果一个master节点不可到达超过了指定时间,则认为它失败了。
cluster-slave-validity-factor 10 #如果设置为0,则一个slave将总是尝试故障转移一个master。如果设置为一个正数,那么最大失去连接的时间是node timeout乘以这个factor。
cluster-migration-barrier 1 #一个master和slave保持连接的最小数量(即:最少与多少个slave保持连接)
cluster-require-full-coverage yes #如果设置为yes,这也是默认值,如果key space没有达到百分之多少时停止接受写请求。如果设置为no,将仍然接受查询请求,即使它只是请求部分key。
Windows启动集群服务
将上述配置文件保存到Redis目录下,并使用这些配置文件安装服务并启动
安装:
Redis路径\redis-server --service-install Redis路径\conf文件名.conf --service-name redis文件端口号
例如:
D:\Redis\redis-server --service-install D:\Redis\redis.7000.conf --service-name redis7000
启动:
Redis路径\redis-server --service-start --service-name Redis端口号
例如:
D:\Redis\redis-server --service-start --service-name Redis7000
启动后图片:
下载ruby:
选择without devkit的最新版本
下载redis驱动:
选择兼容版本即可,我用的是3.2.0
安装驱动:
gem install --local gem文件路径\filename.gem
而现在的cluster安装方式不推荐使用redis-trib.rb,而是使用redis-cli
下载正确的redis-trib.rb:
链接:https://pan.baidu.com/s/1WMXEMpWVC6piBD88rqynvw
提取码:2b8s
开始cluster配置:
打开命令行,移动到redis-trib.rb目录之后,输入命令:
redis-trib.rb create --replicas 1 127.0.0.1:7000 127.0.0.1:7001 127.0.0.1:7002 127.0.0.1:7003 127.0.0.1:7004 127.0.0.1:7005
会弹出:
Can I set the above configuration? (type 'yes' to accept):
输入yes回车即可
Mine:
C:\Users\WIN10\Desktop\Redis>redis-trib.rb create --replicas 1 127.0.0.1:7000 127.0.0.1:7001 127.0.0.1:7002 127.0.0.1:7003 127.0.0.1:7004 127.0.0.1:7005
>>> Creating cluster
>>> Performing hash slots allocation on 6 nodes...
Using 3 masters:
127.0.0.1:7000
127.0.0.1:7001
127.0.0.1:7002
Adding replica 127.0.0.1:7003 to 127.0.0.1:7000
Adding replica 127.0.0.1:7004 to 127.0.0.1:7001
Adding replica 127.0.0.1:7005 to 127.0.0.1:7002
M: d0e194b5a53a79782c49ef44e3643ffce3842a67 127.0.0.1:7000
slots:0-5460 (5461 slots) master
M: 543e52694edf6db069b912fe0f1252f0275189f6 127.0.0.1:7001
slots:5461-10922 (5462 slots) master
M: 595906624067ffff9f204a5ccd2fb1c1911344a7 127.0.0.1:7002
slots:10923-16383 (5461 slots) master
S: 633c5b3ae785086c6c6041724c0b81be92e00ed0 127.0.0.1:7003
replicates d0e194b5a53a79782c49ef44e3643ffce3842a67
S: fe62faa1ecb26ca968433feab937c5833baf1db0 127.0.0.1:7004
replicates 543e52694edf6db069b912fe0f1252f0275189f6
S: 20d97ec5b8a129366f7a7496a85404efb8e71371 127.0.0.1:7005
replicates 595906624067ffff9f204a5ccd2fb1c1911344a7
Can I set the above configuration? (type 'yes' to accept): yes
>>> Nodes configuration updated
>>> Assign a different config epoch to each node
>>> Sending CLUSTER MEET messages to join the cluster
Waiting for the cluster to join..
>>> Performing Cluster Check (using node 127.0.0.1:7000)
M: d0e194b5a53a79782c49ef44e3643ffce3842a67 127.0.0.1:7000
slots:0-5460 (5461 slots) master
M: 543e52694edf6db069b912fe0f1252f0275189f6 127.0.0.1:7001
slots:5461-10922 (5462 slots) master
M: 595906624067ffff9f204a5ccd2fb1c1911344a7 127.0.0.1:7002
slots:10923-16383 (5461 slots) master
M: 633c5b3ae785086c6c6041724c0b81be92e00ed0 127.0.0.1:7003
slots: (0 slots) master
replicates d0e194b5a53a79782c49ef44e3643ffce3842a67
M: fe62faa1ecb26ca968433feab937c5833baf1db0 127.0.0.1:7004
slots: (0 slots) master
replicates 543e52694edf6db069b912fe0f1252f0275189f6
M: 20d97ec5b8a129366f7a7496a85404efb8e71371 127.0.0.1:7005
slots: (0 slots) master
replicates 595906624067ffff9f204a5ccd2fb1c1911344a7
[OK] All nodes agree about slots configuration.
>>> Check for open slots...
>>> Check slots coverage...
[OK] All 16384 slots covered.
验证连接:
输入:
redis-cli -p 7000 cluster nodes
响应:
20d97ec5b8a129366f7a7496a85404efb8e71371 127.0.0.1:7005@17005 slave 595906624067ffff9f204a5ccd2fb1c1911344a7 0 1637803028599 6 connected
633c5b3ae785086c6c6041724c0b81be92e00ed0 127.0.0.1:7003@17003 slave d0e194b5a53a79782c49ef44e3643ffce3842a67 0 1637803027508 4 connected
595906624067ffff9f204a5ccd2fb1c1911344a7 127.0.0.1:7002@17002 master - 0 1637803028000 3 connected 10923-16383
d0e194b5a53a79782c49ef44e3643ffce3842a67 127.0.0.1:7000@17000 myself,master - 0 1637803028000 1 connected 0-5460
543e52694edf6db069b912fe0f1252f0275189f6 127.0.0.1:7001@17001 master - 0 1637803026413 2 connected 5461-10922
fe62faa1ecb26ca968433feab937c5833baf1db0 127.0.0.1:7004@17004 slave 543e52694edf6db069b912fe0f1252f0275189f6 0 1637803029690 5 connected
或者输入:
redis-trib.rb check 127.0.0.1:7000
响应:
>>> Performing Cluster Check (using node 127.0.0.1:7000)
M: d0e194b5a53a79782c49ef44e3643ffce3842a67 127.0.0.1:7000
slots:0-5460 (5461 slots) master
1 additional replica(s)
S: 20d97ec5b8a129366f7a7496a85404efb8e71371 127.0.0.1:7005@17005
slots: (0 slots) slave
replicates 595906624067ffff9f204a5ccd2fb1c1911344a7
S: 633c5b3ae785086c6c6041724c0b81be92e00ed0 127.0.0.1:7003@17003
slots: (0 slots) slave
replicates d0e194b5a53a79782c49ef44e3643ffce3842a67
M: 595906624067ffff9f204a5ccd2fb1c1911344a7 127.0.0.1:7002@17002
slots:10923-16383 (5461 slots) master
1 additional replica(s)
M: 543e52694edf6db069b912fe0f1252f0275189f6 127.0.0.1:7001@17001
slots:5461-10922 (5462 slots) master
1 additional replica(s)
S: fe62faa1ecb26ca968433feab937c5833baf1db0 127.0.0.1:7004@17004
slots: (0 slots) slave
replicates 543e52694edf6db069b912fe0f1252f0275189f6
[OK] All nodes agree about slots configuration.
>>> Check for open slots...
>>> Check slots coverage...
[OK] All 16384 slots covered.
集群demo:
C:\Users\WIN10\Desktop\Redis>redis-cli -c -p 7000
127.0.0.1:7000> set foo bar
-> Redirected to slot [12182] located at 127.0.0.1:7002
OK
127.0.0.1:7002> set hello world
-> Redirected to slot [866] located at 127.0.0.1:7000
OK
127.0.0.1:7000> set good nice
OK
127.0.0.1:7000> set apple iphone
-> Redirected to slot [7092] located at 127.0.0.1:7001
OK
127.0.0.1:7001> set banana orange
OK
127.0.0.1:7001> get foo
-> Redirected to slot [12182] located at 127.0.0.1:7002
"bar"
127.0.0.1:7002> get apple
-> Redirected to slot [7092] located at 127.0.0.1:7001
"iphone"
127.0.0.1:7001> get good
-> Redirected to slot [2195] located at 127.0.0.1:7000
"nice"
127.0.0.1:7000> set gay shit
OK
127.0.0.1:7000> get gay
"shit"
127.0.0.1:7000> flush all
(error) ERR unknown command `flush`, with args beginning with: `all`,
127.0.0.1:7000> flushall
OK
127.0.0.1:7000> get gay
(nil)
127.0.0.1:7000> get doo
(nil)
127.0.0.1:7000> get foo
-> Redirected to slot [12182] located at 127.0.0.1:7002
"bar"
127.0.0.1:7002> flush all
(error) ERR unknown command `flush`, with args beginning with: `all`,
127.0.0.1:7002> flushall
OK
127.0.0.1:7002> get apple
-> Redirected to slot [7092] located at 127.0.0.1:7001
"iphone"
127.0.0.1:7001> flushall
OK
127.0.0.1:7001> get gay
-> Redirected to slot [2398] located at 127.0.0.1:7000
(nil)
127.0.0.1:7000> get apple
-> Redirected to slot [7092] located at 127.0.0.1:7001
(nil)
127.0.0.1:7001> get banana
(nil)
127.0.0.1:7001> get good
-> Redirected to slot [2195] located at 127.0.0.1:7000
(nil)
127.0.0.1:7000> get key
-> Redirected to slot [12539] located at 127.0.0.1:7002
(nil)
demo解释
Redis集群中有16384个hash slots,为了计算给定的key应该在哪个hash slot上,我们简单地用这个key的CRC16值来对16384取模。(即:key的CRC16 % 16384)
Redis集群中的每个节点负责一部分hash slots,假设你的集群有3个节点,那么:
Node A contains hash slots from 0 to 5500
Node B contains hash slots from 5501 to 11000
Node C contains hash slots from 11001 to 16383
因此对不同的Key,存储的节点位置是不一样的,但会自动切换到对应的节点位置。
4、注意事项
主要是需要注意文件配置,以及端口的开放。只要注意配置问题,基本上在Windos本机上按照步骤是不会出错的。(全是亲测过程)