介绍
redis-shake是阿里云Redis团队开源的用于Redis数据迁移和数据过滤的工具
基本功能
redis-shake它支持解析、恢复、备份、同步四个功能
- 恢复restore：将RDB文件恢复到目的redis数据库。
- 备份dump：将源redis的全量数据通过RDB文件备份起来。
- 解析decode：对RDB文件进行读取，并以json格式解析存储。
- 同步sync：支持源redis和目的redis的数据同步，支持全量和增量数据的迁移，支持单节点、主从版、集群版之间的互相同步。
- 同步rump：支持源redis和目的redis的数据同步，仅支持全量的迁移,采用scan和restore命令进行迁移，支持不同云厂商不同redis版本的迁移。
基本原理
RedisShake同步原理
- 源Redis服务实例相当于主库，Redis-shake相当于从库，它会发送psync指令给源Redis服务实例。
- 源Redis实例先把RDB文件传输给 Redis-shake ，Redis-shake 会把RDB文件发送给目的实例。
- 源实例会再把增量命令发送给 Redis-shake ，Redis-shake负责把这些增量命令再同步给目的实例。
RedisShake执行过程
- 启动Redis-shake进程，这个进程模拟了一个 Redis 实例，Redis-shake的基本原理就是模拟一个Slave从节点加入源Redis集群，然后进行增量的拉取（通过psync命令）。
- Redis-shake进程和数据迁出的源实例进行数据的全量拉取同步，并回放，这个过程和 Redis 主从实例的全量同步是类似的
支持的Redis架构
- Standalone：单源拉取，主从版/单节点
- Sentinel：从Sentinel获取地址并拉取
- Cluster：开源Cluster模式
- Proxy：从Proxy拉取
版本迭代特性
现在 redis-shake 有两个主版本： redis-shake 2.x：持续更新 3 年，目前停止更新与答疑，遇到问题推荐尝试 3.x 版本 redis-shake 3.x：基于 redis-shake 2.x 重写，代码可读性高，性能较佳
应用场景
1、数据同步——SYNC/PSYNC
在SYNC模式下支持全量同步与增量同步，当然它有一个限制，需要源端支持PSYNC/SYNC，在向源端发送SYNC/PSYNC命令的时候，可以收到源端的回复。源端、目的端的形态可以是主从集群/Proxy/Cluster，迁移也可以用于云下到云上，云上到云下这种混合云迁移
2、多活
多活通常是用于解决因地域网络传输层面带来的问题，也就是异地多活 比如说业务层面想要在北京和上海两地的机房实时同步最新数据，同样反过来也可以，这样会需要做到一个多活
3、数据备份与恢复
作用
- 数据恢复restore
- 自建Redis迁移上云
- 云上数据迁移到云下
- 将数据库恢复至之前的状态，数据回滚
- 可以配合Filter使用，只恢复部分Key
4、多云厂商之间数据迁移
对于某些云上redis，比如部分云厂商不支持SYNC/PSYNC 命令，那么在进行云上和下云的时候如何进行行迁移呢
RedisShake对于这种场景也做了支持，比如绕过SYNC和PSYNC的同步方式。它是以Scan的方式从源端Redis获取到全量数据，再写入到目的端，实现数据迁移 如上图所示：
部署和使用
前提条件
安装golang 请浏览 go 官方地址：(https://go.dev/dl/?spm=a2c4e.10696291.0.0.24e019a4HrBgn5) 并且检查一下是否有新的版本可用
版本要求
go version >=go1.17
ubuntu：
1、下载 Go 压缩包
root@ubuntu20-171:~# wget -c https://go.dev/dl/go1.20.linux-amd64.tar.gz -O - | sudo tar -xz -C /usr/local
2、调整环境变量
通过将 Go 目录添加到$PATH环境变量，系统将会知道在哪里可以找到 Go 可执行文件
root@ubuntu20-171:~# export PATH=$PATH:/usr/local/go/bin
root@ubuntu20-171:~# export PATH=$PATH:/usr/local/go/bin
root@ubuntu20-171:~# source ~/.profile
3、验证 Go 安装过程
root@ubuntu20-171:~# go version
go version go1.20 linux/amd64
部署
root@ubuntu20-171:~# git clone https://github.com/alibaba/RedisShake
root@ubuntu20-171:~# cd RedisShake
root@ubuntu20-171:~# sh build.sh
root@ubuntu20-171:~#
root@ubuntu20-171:RedisShake# ll
total 80
drwxr-xr-x 10 root root 4096 Feb 6 16:36 ./
drwxr-xr-x 3 root root 4096 Feb 6 16:33 ../
drwxr-xr-x 4 root root 4096 Feb 6 16:40 bin/ ### build.sh成功生成bin目录
-rwxr-xr-x 1 root root 966 Feb 6 16:32 build.sh*
drwxr-xr-x 3 root root 4096 Feb 6 16:32 cmd/
drwxr-xr-x 2 root root 4096 Feb 6 16:32 filters/
drwxr-xr-x 8 root root 4096 Feb 6 16:34 .git/
drwxr-xr-x 4 root root 4096 Feb 6 16:32 .github/
-rw-r--r-- 1 root root 55 Feb 6 16:32 .gitignore
-rw-r--r-- 1 root root 424 Feb 6 16:32 go.mod
-rw-r--r-- 1 root root 3232 Feb 6 16:32 go.sum
drwxr-xr-x 13 root root 4096 Feb 6 16:32 internal/
-rw-r--r-- 1 root root 1078 Feb 6 16:32 license.txt
-rw-r--r-- 1 root root 3499 Feb 6 16:32 README.md
-rw-r--r-- 1 root root 1924 Feb 6 16:32 restore.toml
-rw-r--r-- 1 root root 1918 Feb 6 16:32 scan.toml
drwxr-xr-x 4 root root 4096 Feb 6 16:32 scripts/
-rw-r--r-- 1 root root 2031 Feb 6 16:32 sync.toml
drwxr-xr-x 5 root root 4096 Feb 6 16:32 test/
root@ubuntu20-171:RedisShake# cd bin/
root@ubuntu20-171:bin# ll
total 9760
drwxr-xr-x 4 root root 4096 Feb 6 16:40 ./
drwxr-xr-x 10 root root 4096 Feb 6 16:36 ../
drwxr-xr-x 2 root root 4096 Feb 6 16:34 cluster_helper/
drwxr-xr-x 2 root root 4096 Feb 6 16:34 filters/
-rwxr-xr-x 1 root root 9962342 Feb 6 16:34 redis-shake* ##可执行文件
-rw-r--r-- 1 root root 1968 Feb 6 16:40 restore.toml ## 参数模板
-rw-r--r-- 1 root root 1918 Feb 6 16:34 scan.toml
-rw-r--r-- 1 root root 2031 Feb 6 16:34 sync.toml
使用
基础用法
1、编辑 对应 参数sync.toml or restore.toml 或手动编辑自行创建后缀名为toml的迁移配置文件
2、构建并执行数据同步进程 redis-shake.
root@ubuntu20-171:RedisShake# ./bin/redis-shake redis-shake.toml
# or
root@ubuntu20-171:RedisShake# ./bin/redis-shake restore.toml
步骤流程
创建xx.toml配置文件 ---> 编写toml配置文件同步规则 ---> 执行redis-shake进程 ---> 核对数据同步情况
参数配置说明
type = "sync" # 同步机制实现
[source] # 源Redis服务实例
version = 5.0 # 填写Redis源服务版本, 例如：2.8, 4.0, 5.0, 6.0, 6.2, 7.0, ...。
address = "127.0.0.1:6379" # 源Redis服务实例 地址+端口
username = "" # 如果Redis没有配置ACL，则可以不填写，否则需要填写用户名
password = "" # 如果Redis没有配置ACL，则可以不填写，否则需要填写密码
tls = false # 是否开启tls安全机制
elasticache_psync = "" # 是否支持AWS的elasticache
[target]
type = "standalone" # 选择Redis的类型："standalone：单机模式" or "cluster：集群模式"
version = 5.0 # 填写Redis源服务版本, 例如：2.8, 4.0, 5.0, 6.0, 6.2, 7.0, ...。
# 如果目标Redis服务实例属于cluster集群模式, 那么可以写入其中一个节点的地址和端口.
# redis-shake 会通过`cluster nodes` 命令获取其他的节点地址和端口
address = "127.0.0.1:6380" # 填写的对应的ip加端口
username = "" # 如果Redis没有配置ACL，则可以不填写，否则需要填写用户名
password = "" # 如果Redis没有配置ACL，则可以不填写，否则需要填写密码
tls = false # 是否开启tls安全机制
[advanced]
dir = "data" # 数据同步的存储目录
# 设置使用的最大CPU核心数, 如果设置了0 代表着 使用 runtime.NumCPU() 实际的cpu cores数量
ncpu = 4
# 开启pprof性能检测的port, 0代表着禁用
pprof_port = 0
# 开启metric port端口, 0代表着禁用
metrics_port = 0
# log的相关设置
log_file = "redis-shake.log" # 设置对应的日志文件名称
log_level = "info" # debug, info or warn # 设置对应的日志级别
log_interval = 5 # in seconds # 日志打印频次
# redis-shake gets key and value from rdb file, and uses RESTORE command to
# create the key in target redis. Redis RESTORE will return a "Target key name
# is busy" error when key already exists. You can use this configuration item
# to change the default behavior of restore:
# panic: redis-shake will stop when meet "Target key name is busy" error.
# rewrite: redis-shake will replace the key with new value.
# ignore: redis-shake will skip restore the key when meet "Target key name is busy" error.
rdb_restore_command_behavior = "rewrite" # restore的操作类型：panic, rewrite or skip
# pipeline的大小数量阈值
pipeline_count_limit = 1024
# Client query buffers accumulate new commands. They are limited to a fixed
# amount by default. This amount is normally 1gb.
target_redis_client_max_querybuf_len = 1024_000_000
# In the Redis protocol, bulk requests, that are, elements representing single
# strings, are normally limited to 512 mb.
target_redis_proto_max_bulk_len = 512_000_000
同步日志注释
当打印的日志出现send RDB finished，表示完成全量数据迁移，接下来进入增量数据迁移阶段。
日志信息中各参数说明如下：
allowOps：表示每秒向目标库发送多少条命令。
说明 通常当allowOps为0时，表示数据迁移完成，可以停止redis-shake。但源库会定时发送PING命令，所以allowOps偶尔不为0。
disallowOps：表示每秒过滤的命令数。
entryId：从1开始计数，表示redis-shake共处理多少条命令。
InQueueEntriesCount：表示还剩余多少条命令待发送。
暂停向源库写入数据，等待返回日志中allowOps对应值连续多次为0时，使用Ctrl+C组合键停止运行redis-shake。
此时目标库的数据与源库完全一致，您可以将业务的数据库服务由自建Redis数据库切换至Tair或Redis实例。
toml文件的编写方式
单机 -> 单机 配置格式：
type = "sync"
[source]
address = "127.0.0.1:6379"
password = "123456" ## 根据自身密码填写
[target]
type = "standalone" #这里type属性设置一定为standalone
address = "127.0.0.1:6379"
password = "123456" ## 根据自身密码填写
启动 redis-shake：
root@ubuntu20-171:RedisShake# ./redis-shake sync.toml
单机->集群 配置格式：
type = "sync"
[source] #数据源配置
address = "127.0.0.1:6379"
password = "1234566"
[target] #目的源配置
type = "cluster" #这里type属性设置一定为cluster
address = "127.0.0.1:6379" # 这里写集群中的任意一个节点的地址即可
password = "1234566"
启动 redis-shake：
root@ubuntu20-171:RedisShake# ./redis-shake sync.toml
集群 -> 集群 配置格式：
在 RedisShake V3系列版本，也就是当前使用最多的版本中，集群与集群之间的redis数据同步，统一在单机到集群的基础上，为每一个数据源端的redis master节点建立一个shake进程
方法1：手动起多个 redis-shake
例：二个集群3主3从之间的数据同步只需为数据源端的每一个master建立一个shake进程，既需要建立3个单机至集群的shake进程用于二个Redis集群间的无感数据迁移与同步
集群 C 有3个节点：
192.168.0.1:6379
192.168.0.2:6379
192.168.0.3:6379
把 3个节点当成 3个单机实例，参照 单机到集群 部署 3个 redis-shake 进行数据同步。
⚠️注意：不要在同一个目录启动多个 redis-shake，因为 redis-shake 会在本地存储临时文件，多个 redis-shake 之间的临时文件会干扰，正确做法是建立多个目录。
方法2：借助 cluster_helper.py 启动
脚本 cluster_helper.py 可以方便启动多个 redis-shake 从集群迁移数据，效果等同于方法1 ⚠️注意：
- 源端有多少个分片，cluster_helper.py 就会起多少个 redis-shake 进程，所以如果源端分片数较多的时候，需要评估当前机器是否可以承担这么多进程。
- cluster_helper.py 异常退出的时候，可能没有正常退出 redis-shake 进程，需要 ps aux | grep redis-shake 检查
- 每个 redis-shake 进程的执行日志记录在 RedisShake/cluster_helper/data/xxxxx 中，反馈问题请提供相关日志。
依赖
Python 需要 python3.6 及以上版本，安装 Python 依赖：
root@ubuntu20-171:RedisShake# cd bin/cluster_helper
root@ubuntu20-171:cluster_helper# pip3 install -r requirements.txt
配置
修改 sync.toml:
type = "sync"
[source]
address = "192.168.0.1:6379" # 集群 C 中任意一个节点地址
password = "r-ccccc:xxxxx"
[target]
type = "cluster"
address = "192.168.1.1:6380" # 集群 D 中任意一个节点地址
password = "r-ddddd:xxxxx"
启动 redis-shake：
root@ubuntu20-171:RedisShake# cd bin/cluster_helper
root@ubuntu20-171:cluster_helper# python3 cluster_helper.py ../redis-shake ../sync.toml
参数 1 是 redis-shake 可执行程序的路径
参数 2 是配置文件路径
实操 (以sync模式为例)
root@ubuntu20-171:~# cd /data/tool/
root@ubuntu20-171:/data/tool/RedisShake/bin# cd RedisShake/bin
root@ubuntu20-171:bin# ll
total 9760
drwxr-xr-x 4 root root 4096 Feb 6 16:40 ./
drwxr-xr-x 10 root root 4096 Feb 6 16:36 ../
drwxr-xr-x 2 root root 4096 Feb 6 16:34 cluster_helper/
drwxr-xr-x 2 root root 4096 Feb 6 16:34 filters/
-rwxr-xr-x 1 root root 9962342 Feb 6 16:34 redis-shake*
-rw-r--r-- 1 root root 1968 Feb 6 16:40 restore.toml
-rw-r--r-- 1 root root 1918 Feb 6 16:34 scan.toml
-rw-r--r-- 1 root root 2031 Feb 6 16:34 sync.toml
root@ubuntu20-171:bin# vim sync.toml
type = "sync"
[source]
version = 4.0 # redis version, such as 2.8, 4.0, 5.0, 6.0, 6.2, 7.0, ...
address = "10.150.38.200:6379"
##username = "@a123" # keep empty if not using ACL
password = "********" # keep empty if no authentication is required
tls = false
[target]
type = "standalone" # "standalone" or "cluster"
version = 6.2 # redis version, such as 2.8, 4.0, 5.0, 6.0, 6.2, 7.0, ...
address = "10.150.3.38:6379"
password = "@a123" # keep empty if no authentication is required
tls = false
[advanced]
dir = "/data/redis"
ncpu = 2
pprof_port = 0
metrics_port = 0
# log
log_file = "/data/backup/redis-shake.log"
log_level = "info" # debug, info or warn
log_interval = 1 # in seconds
# pip
rdb_restore_command_behavior = "rewrite" # panic, rewrite or skip
pipeline_count_limit = 1024
target_redis_client_max_querybuf_len = 1024_000_000
target_redis_proto_max_bulk_len = 512_000_000
启动
root@xgsdk-dev-mysql8-transfer-01:/data/tool/RedisShake/bin# ./redis-shake sync.toml
日志信息
同步分为三个阶段：
等待源端save rdb完毕，日志如下：
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"GOOS: linux, GOARCH: amd64"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"Ncpu: 2, GOMAXPROCS: 2"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"pid: 245326"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"pprof_port: 0"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"No lua file specified, will not filter any cmd."}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"auth successful. address=[10.150.3.38:6379]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"redisWriter connected to redis successful. address=[10.150.3.38:6379]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"auth successful. address=[10.150.38.200:6379]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"psyncReader connected to redis successful. address=[10.150.38.200:6379]"}
{"level":"warn","time":"2023-02-02T17:17:42+08:00","message":"remove file. filename=[225789358.aof]"}
{"level":"warn","time":"2023-02-02T17:17:42+08:00","message":"remove file. filename=[6379.rdb]"}
{"level":"warn","time":"2023-02-02T17:17:42+08:00","message":"remove file. filename=[dump.rdb]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"start save RDB. address=[10.150.38.200:6379]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"send [replconf listening-port 10007]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"send [PSYNC ? -1]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"receive [FULLRESYNC fdb52e2bb825223140e9eca34e67d1cd234f0aa8 225842514]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"source db is doing bgsave. address=[10.150.38.200:6379]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"source db bgsave finished. timeUsed=[0.01]s, address=[10.150.38.200:6379]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"received rdb length. length=[62050]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"create dump.rdb file. filename_path=[dump.rdb]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"save RDB finished. address=[10.150.38.200:6379], total_bytes=[62050]"}
全量同步阶段
数据量大的话会显示进度百分比：(我dev环境数据量小)日志如下：
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"start send RDB. address=[10.150.38.200:6379]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"RDB version: 8"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"RDB AUX fields. key=[redis-ver], value=[4.0.9]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"start save AOF. address=[10.150.38.200:6379]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"RDB AUX fields. key=[redis-bits], value=[64]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"AOFWriter open file. filename=[225842514.aof]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"RDB AUX fields. key=[ctime], value=[1675329462]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"RDB AUX fields. key=[used-mem], value=[118817296]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"RDB repl-stream-db: 6"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"RDB AUX fields. key=[repl-id], value=[fdb52e2bb825223140e9eca34e67d1cd234f0aa8]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"RDB AUX fields. key=[repl-offset], value=[225842514]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"RDB AUX fields. key=[aof-preamble], value=[0]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"RDB resize db. db_size=[427], expire_size=[421]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"RDB resize db. db_size=[2], expire_size=[0]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"RDB resize db. db_size=[58], expire_size=[56]"}
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"send RDB finished. address=[10.150.38.200:6379], repl-stream-db=[6]"} ## 全量完成
## 显示 send RDB finished 说明全量完成
增量同步
日志如下：
{"level":"info","time":"2023-02-02T17:17:42+08:00","message":"send RDB finished. address=[10.150.38.200:6379], repl-stream-db=[6]"}
{"level":"info","time":"2023-02-02T17:17:43+08:00","message":"syncing aof. allowOps=[487.00], disallowOps=[0.00], entryId=[486], InQueueEntriesCount=[0], unansweredBytesCount=[0]bytes, diff=[225842655], aofReceivedOffset=[225842655], aofAppliedOffset=[0]"} ### 说明开始增量同步
{"level":"info","time":"2023-02-02T17:17:43+08:00","message":"AOFReader open file. aof_filename=[225842514.aof]"}
{"level":"info","time":"2023-02-02T17:17:43+08:00","message":"syncing aof. allowOps=[0.00], disallowOps=[0.00], entryId=[12], InQueueEntriesCount=[0], unansweredBytesCount=[0]bytes, diff=[0], aofReceivedOffset=[242], aofAppliedOffset=[242]
{"level":"info","time":"2023-02-02T17:17:44+08:00","message":"syncing aof. allowOps=[0.20], disallowOps=[0.00], entryId=[13], InQueueEntriesCount=[0], unansweredBytesCount=[0]bytes, diff=[0], aofReceivedOffset=[256], aofAppliedOffset=[256]
{"level":"info","time":"2023-02-02T17:17:45+08:00","message":"syncing aof. allowOps=[0.00], disallowOps=[0.00], entryId=[13], InQueueEntriesCount=[0], unansweredBytesCount=[0]bytes, diff=[0], aofReceivedOffset=[256], aofAppliedOffset=[256]
{"level":"info","time":"2023-02-02T17:17:46+08:00","message":"syncing aof. allowOps=[0.20], disallowOps=[0.00], entryId=[14], InQueueEntriesCount=[0], unansweredBytesCount=[0]bytes, diff=[0], aofReceivedOffset=[270], aofAppliedOffset=[270]
{"level":"info","time":"2023-02-02T17:17:47+08:00","message":"syncing aof. allowOps=[0.00], disallowOps=[0.00], entryId=[14], InQueueEntriesCount=[0], unansweredBytesCount=[0]bytes, diff=[0], aofReceivedOffset=[270], aofAppliedOffset=[270]
{"level":"info","time":"2023-02-02T17:17:48+08:00","message":"syncing aof. allowOps=[0.20], disallowOps=[0.00], entryId=[15], InQueueEntriesCount=[0], unansweredBytesCount=[0]bytes, diff=[0], aofReceivedOffset=[284], aofAppliedOffset=[284]
{"level":"info","time":"2023-02-02T17:17:49+08:00","message":"syncing aof. allowOps=[0.00], disallowOps=[0.00], entryId=[15], InQueueEntriesCount=[0], unansweredBytesCount=[0]bytes, diff=[0], aofReceivedOffset=[284], aofAppliedOffset=[284]
{"level":"info","time":"2023-02-02T17:17:50+08:00","message":"syncing aof. allowOps=[0.20], disallowOps=[0.00], entryId=[16], InQueueEntriesCount=[0], unansweredBytesCount=[0]bytes, diff=[0], aofReceivedOffset=[298], aofAppliedOffset=[298]
{"level":"info","time":"2023-02-02T17:17:51+08:00","message":"syncing aof. allowOps=[0.00], disallowOps=[0.00], entryId=[16], InQueueEntriesCount=[0], unansweredBytesCount=[0]bytes, diff=[0], aofReceivedOffset=[298], aofAppliedOffset=[298]
出现字样syncing aof. allowOps=[0.00] 或者出现syncing aof. allowOps=[0.00]的频率很高 说明 当前增量同步完成 可以ctrl+c停止redis-shake 进程，但是 一般不会为0 或者 因为有client在一直连接
查看源端和目标端的情况
源端
127.0.0.1:6379> info Replication
# Replication
role:master
connected_slaves:1 ### 源端会启动一个redis-shake的从节点
slave0:ip=127.0.0.1,port=10007,state=online,offset=1040,lag=0 ###shake进程
master_failover_state:no-failover
master_replid:b20dfd6f617c867cfaf0910c08be627fe7b922d7
master_replid2:0000000000000000000000000000000000000000
master_repl_offset:1040
second_repl_offset:-1
repl_backlog_active:1
repl_backlog_size:104857600
repl_backlog_first_byte_offset:229
repl_backlog_histlen:812
127.0.0.1:6379> info Keyspace ## 查看比对key数量
# Keyspace
db0:keys=6,expires=0,avg_ttl=0
db2:keys=3,expires=0,avg_ttl=0
db10:keys=3,expires=0,avg_ttl=0
目标端
127.0.0.1:7000> info Replication
# Replication
role:master ### 目标端 还是master角色
connected_slaves:0
master_failover_state:no-failover
master_replid:1017a92f7fd8447d304172796ea6044a1f801e26
master_replid2:0000000000000000000000000000000000000000
master_repl_offset:0
second_repl_offset:-1
repl_backlog_active:0
repl_backlog_size:104857600
repl_backlog_first_byte_offset:0
repl_backlog_histlen:0
127.0.0.1:7000> info Keyspace ## 对比key数量
# Keyspace
db0:keys=6,expires=0,avg_ttl=0
db2:keys=3,expires=0,avg_ttl=0
db10:keys=3,expires=0,avg_ttl=0
注意事项
- 如果目标库的数据淘汰策略（maxmemory-policy）配置为noeviction以外的值，可能导致目标库的数据与源库不一致
- 如果源库中的某些Key使用了过期（expire）机制，由于可能存在Key已过期但未被及时删除的情形，所以在目标库中查看（如通过info命令）到的Key数量会比源库的Key数量少
参考
官方文档:https://github.com/alibaba/RedisShake