主从Redis及哨兵部署(docker)

一、主从Redis部署(docker)

首先，我准备了两台linux，一台准备当作master，ip是192.168.241.128，另一台是当作slave，ip是192.168.241.129。

1. 安装redis

docker pull redis

2. 下载对应版本的redis.conf

可以从github上下载。新建配置环境目录和持久化数据目录。

mkdir -p /home/redis/data
mkdir -p /home/redis/conf
mkdir -p /home/redis/log

修改配置redis.conf:

logfile "/log/redis.log"
dir /data

将redis.conf放到mkdir -p /home/redis/conf下。

3. 创建主redis容器

docker run --restart=always --log-opt max-size=100m --log-opt max-file=2 -p 6379:6379 --name redis -v /home/redis/data:/data -v /home/redis/conf/redis.conf:/etc/redis/redis.conf -v /home/redis/log/redis.log:/log/redis.log -d redis redis-server /etc/redis/redis.conf  --appendonly yes 

创建容器后，调用redis-server启动用挂载的路径的redis.conf启动。并且开启AOF持久化。

docker ps -a 查看，发现已成功启动。

如果报这个错：Can’t open the log file: Is a directory，可能得手动建一个空文件 /home/redis/log/redis.log

4. 创建第从redis容器

和上面的1，2，3步骤一样，在第二台linux服务器上创建第二个redis。稍微不同的是，在第2个步骤配置redis.conf时，要做修改。

redis.conf加上配置:

192.168.241.128 是我上面配置的第一台linux的地址，意思是第二台是第一台的slave库

slaveof 192.168.241.128 6379

依样部署好第二台redis就可以了。

5. 验证

在redis客户端中执行命令info replication

128(主):

> info replication
# Replication
role:master
connected_slaves:1
slave0:ip=192.168.241.129,port=6379,state=online,offset=0,lag=0
master_failover_state:no-failover
master_replid:8fade8b7a32d08413701d8816661604e773c6cd3
master_replid2:0000000000000000000000000000000000000000
master_repl_offset:0
second_repl_offset:-1
repl_backlog_active:1
repl_backlog_size:1048576
repl_backlog_first_byte_offset:1
repl_backlog_histlen:0

129(从):

> 192.168.241.129@6379 connected!
> info replication
# Replication
role:slave
master_host:192.168.241.128
master_port:6379
master_link_status:up
master_last_io_seconds_ago:5
master_sync_in_progress:0
slave_read_repl_offset:14
slave_repl_offset:14
slave_priority:100
slave_read_only:1
replica_announced:1
connected_slaves:0
master_failover_state:no-failover
master_replid:8fade8b7a32d08413701d8816661604e773c6cd3
master_replid2:0000000000000000000000000000000000000000
master_repl_offset:14
second_repl_offset:-1
repl_backlog_active:1
repl_backlog_size:1048576
repl_backlog_first_byte_offset:1
repl_backlog_histlen:14

很明显，一个role 是master，一个是slave。

验证一下数据同步:

一开始129尝试get “h” ，发现数据不存在。想要写入，因为是从节点，所以提示无法写入。

然后我们往128加入”h”->“hello”, 切换到129发现这条数据已经同步过来了，接着128删除这个键，129也是跟着删除了数据。

二、Redis 哨兵部署(docker)

准备好三台服务器，已经部署了redis，其中一台主，两台从。（可以按上面步骤再部署一台从节点）

1. 配置文件

准备好sentinel.conf配置文件。可以从上面的那个github分支下载对应版本的config文件。

拷到路径 /home/redis-sentinel/conf/sentinel.conf

mkdir -p /home/redis-sentinel/data
mkdir -p /home/redis-sentinel/conf

2. 创建容器

docker run --restart=always -p 26379:26379 --name redis-sentinel -v /home/redis-sentinel/data:/data -v /home/redis-sentinel/conf/sentinel.conf:/etc/redis/sentinel.conf -d redis redis-sentinel /etc/redis/sentinel.conf 

3. 验证

当三台都装好之后，依次在redis客户端打印主从信息： info replication

可以看出此时主节点是128，两个从节点是129和131

> 192.168.241.128@6379 connected!
> info replication
# Replication
role:master
connected_slaves:2
slave0:ip=192.168.241.129,port=6379,state=online,offset=136208,lag=0
slave1:ip=192.168.241.131,port=6379,state=online,offset=136208,lag=0
master_failover_state:no-failover
master_replid:8fade8b7a32d08413701d8816661604e773c6cd3
master_replid2:0000000000000000000000000000000000000000
master_repl_offset:136348
second_repl_offset:-1
repl_backlog_active:1
repl_backlog_size:1048576
repl_backlog_first_byte_offset:1
repl_backlog_histlen:136348



> 192.168.241.129@6379 connected!
> info replication
# Replication
role:slave
master_host:192.168.241.128
master_port:6379
master_link_status:up
master_last_io_seconds_ago:1
master_sync_in_progress:0
slave_read_repl_offset:138042
slave_repl_offset:138042
slave_priority:100
slave_read_only:1
replica_announced:1
connected_slaves:0
master_failover_state:no-failover
master_replid:8fade8b7a32d08413701d8816661604e773c6cd3
master_replid2:0000000000000000000000000000000000000000
master_repl_offset:138042
second_repl_offset:-1
repl_backlog_active:1
repl_backlog_size:1048576
repl_backlog_first_byte_offset:85875
repl_backlog_histlen:52168



> 192.168.241.131@6379 connected!
> info replication
# Replication
role:slave
master_host:192.168.241.128
master_port:6379
master_link_status:up
master_last_io_seconds_ago:1
master_sync_in_progress:0
slave_read_repl_offset:138742
slave_repl_offset:138742
slave_priority:100
slave_read_only:1
replica_announced:1
connected_slaves:0
master_failover_state:no-failover
master_replid:8fade8b7a32d08413701d8816661604e773c6cd3
master_replid2:0000000000000000000000000000000000000000
master_repl_offset:138742
second_repl_offset:-1
repl_backlog_active:1
repl_backlog_size:1048576
repl_backlog_first_byte_offset:86738
repl_backlog_histlen:52005

此时，将128主节点的redis关掉。等三十秒(这个时间是sentinel.conf默认配置的)，重新打印剩下两台从节点的信息：

> 192.168.241.129@6379 connected!
> info replication
# Replication
role:master
connected_slaves:1
slave0:ip=192.168.241.131,port=6379,state=online,offset=253237,lag=1
master_failover_state:no-failover
master_replid:94187cf053b9394bd53e13bd0334ab19f376a6d2
master_replid2:47fa536c9702cd72db072304c3904265e52b52de
master_repl_offset:253517
second_repl_offset:178390
repl_backlog_active:1
repl_backlog_size:1048576
repl_backlog_first_byte_offset:175950
repl_backlog_histlen:77568



> 192.168.241.131@6379 connected!
> info replication
# Replication
role:slave
master_host:192.168.241.129
master_port:6379
master_link_status:up
master_last_io_seconds_ago:0
master_sync_in_progress:0
slave_read_repl_offset:254371
slave_repl_offset:254371
slave_priority:100
slave_read_only:1
replica_announced:1
connected_slaves:0
master_failover_state:no-failover
master_replid:94187cf053b9394bd53e13bd0334ab19f376a6d2
master_replid2:0000000000000000000000000000000000000000
master_repl_offset:254371
second_repl_offset:-1
repl_backlog_active:1
repl_backlog_size:1048576
repl_backlog_first_byte_offset:180513
repl_backlog_histlen:73859

可以发现，129从原本的slave角色变为master角色了。而131从节点的master_host也改为128了。