Shulou(Shulou.com)06/02 Report--

I. brief introduction

Zookeeper is an efficient distributed coordination service, which can provide configuration information management, naming, distributed synchronization, cluster management, database switching and other services. It is not suitable to store a large amount of information, but can be used to store a small amount of information such as configuration, publication, subscription and so on. Hadoop, Storm, message middleware, RPC service framework and distributed database synchronization system are all application scenarios of Zookeeper.

The number of nodes in a Zookeeper cluster is generally odd (> = 3). If the Master in the cluster is dropped and the number of remaining nodes is more than half, you can recommend a new master node to continue to provide services.

1. Typical application scenarios

1) distributed Notification / Coordination

Task distribution and task execution for distributed systems, line result feedback

2) Cluster management

-can monitor the survival status of nodes

3) Master election

In a system with Master/slave structure, to avoid a single point of failure, multiple master are required, in which

Only one active master,ZooKeeper helps to select active master

4) distributed lock

-monopolization: only one client can be obtained at a certain time

-Control timing: some processes of multiple clients execute rows sequentially

5) distributed queue

Https://blog.csdn.net/king866/article/details/53992653/

2. Architecture

2.1, Node

-2N+1 consists of 1 server. When one node is available, the whole system remains usable.

2.2, server Corner Color Typ

-Server: store data (in-memory)

-Leader: do not accept the client request, be responsible for initiating and deciding the voting of the advanced bank, and finally update the status-Follower: receive the customer request and return the customer result, participate in the vote initiated by Leader

3. ObServer node

problem

When the client becomes more, the ZK cluster becomes larger, and the nodes are all follower, the updated version is larger.

Solution scheme

ObServer node: interacts with client, keeps a copy of the data, and does not participate in voting

4. Role

5. ZooKeeper commitment

5.1, sequentiality

Client's update requests are processed sequentially according to the order in which they are sent.

5.2. Atomicity

A update operation either succeeds or fails, and there are no other possible results

5.3. Strong consistency

No matter which server client is connected to, the display shows it the same view.

5.4. Reliability

Once the update succeeds, it is persisted, except for a update request to update the current value

5.5, real-time

For each client, its system view is up to date

II. Preparatory work

1. System environment

Hostnam

System

Service version

IP

Master

CentOS7.4

3.4.12

192.168.56.129

Slave1

CentOS7.4

3.4.12

192.168.56.130

Slave2

CentOS7.4

3.4.12

192.168.56.131

2. Temporarily turn off the firewall and selinux

3. Software download

Download address: http://mirrors.hust.edu.cn/apache/zookeeper/

4. Java environment

Third, service building

1. Decompress the files of three nodes

[root@master ~] # cd / home/

[root@master home] # tar-zxvf zookeeper-3.4.12.tar.gz

[root@master home] # ln-s zookeeper-3.4.12 zookeeper

Note: three nodes have the same operation

2. Create a zoo.cfg file under the conf directory, and the configuration of the three nodes is the same, as follows

[root@master ~] # cd / home/zookeeper/conf/

[root@master conf] # cp zoo_sample.cfg zoo.cfg

[root@master conf] # vim zoo.cfg

TickTime=2000

InitLimit=10

SyncLimit=5

DataDir=/var/lib/zookeeper

ClientPort=2181

MaxClientCnxns=50

Server.1=192.168.56.129:2888:3888

Server.2=192.168.56.130:2888:3888

Server.3=192.168.56.131:2888:3888

3. The contents of creating files / var/lib/zookeeper/myid,3 nodes are 1, 2 and 3 respectively.

[root@master conf] # mkdir-p / var/lib/zookeeper

[root@master conf] # cd / var/lib/zookeeper/

[root@master zookeeper] # vim myid

4. Start three nodes

[root@master zookeeper] # bin/zkServer.sh start

[root@master zookeeper] # jps

[root@master zookeeper] # bin/zkServer.sh status

5. Test

5.1. in any node in zookeeper, execute the instruction zkCli.sh

[root@master zookeeper] #. / bin/zkCli.sh

5.2. enter: help # help

Ls / find root directory

Create / test abc creates nodes and assigns values

Get / test gets the value of the specified node

Set / test cb sets the value of an existing node

Rmr / test Recursive deletion Node

Delete / test/test01 deletes nodes that do not have child nodes

5.3. Related operation commands

New operation

Delete operation

5.4. Viewing tools

View Zookeeper cluster nodes in Eclipse

You can also use ZooInspector to view

Use the ZooInspector tool to view it here

Download address: https://issues.apache.org/jira/secure/attachment/12436620/ZooInspector.zip

Extract it, enter the directory ZooInspector\ build, and double-click to run zookeeper-dev-ZooInspector.jar

Note: only if Java is installed in the Windows environment

Click the connection button in the upper left corner and enter the zk service address: ip:2181

Click OK to view the ZK node information

6. Exception

CONNECTION_LOSS (connection lost)

-eg: the connected server down is dropped

6.2. SESSION_EXPIRED (session invalidation)

-managed by ZooKeeper cluster and not managed by client

-ZooKeeper clears the information related to the session.-this will not happen in the ZK of normal operation.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.