CentOS Build Redis Cluster

Cluster principle-redis-cluster architecture Diagram

Architectural details :

(1) all the Redis nodes connected to each other (ping-pong mechanism ), internal use of binary protocols to optimize transmission speed and bandwidth .

(2) node's fail is not valid until more than half of the nodes in the cluster are detected as invalid .

(3) Client and Redis Node Direct Connect , no need for intermediate Proxy Layer . client does not need to connect all nodes of the cluster , Connect to any of the available nodes in the cluster

(4) Redis-cluster maps all the physical nodes to the [0-16383]slot , cluster is responsible for maintaining node< ->slot<->value

Redisthe cluster has built- in16384a hash slot, when needed in theRedisin the cluster, place aKey-valuewhen theRedisfirst toKeyUseCRC16The algorithm calculates a result and then turns the result to16384to find the remainder, so that eachKeywill correspond to a number in the0-16383between the hash slots,RedisThe hash slots are mapped to different nodes roughly equal to the number of nodes

Redis-cluster votes:Fault Tolerance

(1) The voting process is that all master in the cluster participates , if more than half of the Master node and master node communication exceeds (cluster-node-timeout) and the current Master node is considered to be dead .

(2): When is the entire cluster unavailable (cluster_state:fail)?

A:if the cluster arbitrarilyMasterHang off,and the currentMasterNoslave.Cluster EntryfailStatus,It can also be understood as a clusterSlotsMapping[0-16383]enter when not completedfailStatus. Ps:redis-3.0.0.rc1JoinCluster-require-full-coverageParameters,default off,failed to open cluster compatibility section.

B: if more than half of the cluster master hangs, regardless of whether there is a slave cluster into the fail State .

PS: when the cluster is unavailable , All operations on the cluster are not available, received ((Error) Clusterdown the cluster is down) error

1. Setting up the environment required for the cluster

Building a cluster requires the use of an official Ruby script.

You need to install Ruby 's environment.

Install Ruby

Yum Install Ruby

Yum Install RubyGems

Redis Cluster Management tool REDIS-TRIB.RB

[Email protected] ~]# CD redis-3.0.0

[[Email protected] redis-3.0.0]# CD src

[email protected] src]# ll *.RB

-rwxrwxr-x. 1 root root 48141 Apr 1 07:01 redis-trib.rb

[Email protected] src]#

redis-trib.rb script requires a ruby Package:

Needs to be uploaded to the Linux service.

To install Ruby Packages:

Gem Install Redis-3.0.0.gem

[Email protected] ~]# gem install Redis-3.0.0.gem

Successfully installed redis-3.0.0

1 Gem installed

Installing RI documentation for redis-3.0.0 ...

Installing RDOC documentation for redis-3.0.0 ...

gemhttp://download.csdn.net/detail/u012599988/8597787

2. Building a Cluster

Step One: Create 6 redis instances with port numbers from 7001~7006

Create a directory First

Dump.rdb is a snapshot of the state to save, and the other is aof save the command.

Step Two: Modify the redis configuration file

1. Modify the port number

2 . Open the comment in front of cluster-enable.

Next Replicate 6 nodes

Description

Redis cluster requires at least 3 primary nodes, each master node has a total of 6 nodes from the node

Replicas specified as 1 indicates that each master node has a slave node

Don't forget it. The port number of the 6 nodes has been changed.

Step three: Copy the Ruby script that created the cluster into the redis-cluster directory.

Then you need to copy the REDIS-TRIB.RB Cluster management tool under the Redis src directory to Redis-cluster

Fourth step: Launch 6 redis instances

Write a script to start a batch of 6 Redis nodes

Startall the script content

Files that are set to executable

Fifth step: Create the cluster.

./REDIS-TRIB.RB Create--replicas 1 192.168.188.131:7001 192.168.188.131:7002 192.168.188.131:7003 192.168.188.131:7004 192.168.188.131:7005 192.168.188.131:7006

[Email protected] redis-cluster]#/redis-trib.rb create--replicas 1 192.168.188.131:7001 192.168.188.131:7002 192.168.188.131:7003 192.168.188.131:7004 192.168.188.131:7005 192.168.188.131:7006
>>> Creating Cluster
Connecting to Node 192.168.188.131:7001:ok
Connecting to Node 192.168.188.131:7002:ok
Connecting to Node 192.168.188.131:7003:ok
Connecting to Node 192.168.188.131:7004:ok
Connecting to Node 192.168.188.131:7005:ok
Connecting to Node 192.168.188.131:7006:ok
>>> performing hash slots allocation on 6 nodes ...Connect 6 nodes successfully
Using 3 Masters:7001-7003 as the master node
192.168.188.131:7001
192.168.188.131:7002
192.168.188.131:7003
Adding replica 192.168.188.131:7004 to 192.168.188.131:70017004 as a slave node of 7001
Adding replica 192.168.188.131:7005 to 192.168.188.131:70027005 as a slave node of 7002
Adding replica 192.168.188.131:7006 to 192.168.188.131:70037006 as a slave node of 7003
m:5a8523db7e12ca600dc82901ced06741b3010076 192.168.188.131:7001
slots:0-5460 (5461 slots) Masterallocate 0-5460 slots to 7001 master nodes
M:bf6f0929044db485dea9b565bb51e0c917d20a53 192.168.188.131:7002
slots:5461-10922 (5462 slots) Masterallocate 5461-10922 slots to 7002 master nodes
M:c5e334dc4a53f655cb98fa3c3bdef8a808a693ca 192.168.188.131:7003
slots:10923-16383 (5461 slots) Masterallocate 10923-16383 slots to 7003 master nodes
s:2a61b87b49e5b1c84092918fa2467dd70fec115f 192.168.188.131:7004The 7004,7005,7006 is the same as the node slot and the master node, from which the node is backed up
Replicates 5a8523db7e12ca600dc82901ced06741b3010076
S:14848b8c813766387cfd77229bd2d1ffd6ac8d65 192.168.188.131:7005
Replicates bf6f0929044db485dea9b565bb51e0c917d20a53
s:3192cbe437fe67bbde9062f59d5a77dabcd0d632 192.168.188.131:7006
Replicates C5E334DC4A53F655CB98FA3C3BDEF8A808A693CA
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 192.168.188.131:7001)
m:5a8523db7e12ca600dc82901ced06741b3010076 192.168.188.131:7001
slots:0-5460 (5461 slots) Master
M:bf6f0929044db485dea9b565bb51e0c917d20a53 192.168.188.131:7002
slots:5461-10922 (5462 slots) Master
M:c5e334dc4a53f655cb98fa3c3bdef8a808a693ca 192.168.188.131:7003
slots:10923-16383 (5461 slots) Master
m:2a61b87b49e5b1c84092918fa2467dd70fec115f 192.168.188.131:7004
Slots: (0 slots) Master
Replicates 5a8523db7e12ca600dc82901ced06741b3010076
M:14848b8c813766387cfd77229bd2d1ffd6ac8d65 192.168.188.131:7005
Slots: (0 slots) Master
Replicates bf6f0929044db485dea9b565bb51e0c917d20a53
m:3192cbe437fe67bbde9062f59d5a77dabcd0d632 192.168.188.131:7006
Slots: (0 slots) Master
Replicates C5E334DC4A53F655CB98FA3C3BDEF8A808A693CA
[OK] All nodes agree about slots configuration.
>>> Check for open Slots ...
>>> Check Slots Coverage ...
[OK] All 16384 slots covered.
[Email protected] redis-cluster]#

3. Test the cluster

[[email protected] redis-cluster]# redis01/redis-cli-h 192.168.188.131-p 7002-c- h is the specified ip-p specified port-C represents the cluster to be connected

Note: The cluster must be added with-C, otherwise set will appear (error) moved error

192.168.188.131:7002>
[Email protected] redis-cluster]# redis01/redis-cli-h 192.168.188.131-p 7002-c
192.168.188.131:7002> set a 100
-Redirected to Slots [15495] located at 192.168.188.131:7003
Ok
192.168.188.131:7003>

CentOS Build Redis Cluster