ELK+Kafka+Zookeeper日志收集系统

环境准备

节点IP	节点规划	主机名
192.168.112.3	Elasticsearch + Kibana + Logstash + Zookeeper + Kafka + Nginx	elk-node1
192.168.112.3	Elasticsearch + Logstash + Zookeeper + Kafka	elk-node2
192.168.112.3	Elasticsearch + Logstash + Zookeeper + Kafka + Nginx	elk-node3

基础环境

systemctl disable firewalld --now && setenforce 0
sed -i 's/SELINUX=enforcing/SELINUX=disabled/g' /etc/selinux/config
mv /etc/yum.repos.d/CentOS-* /tmp/
curl -o /etc/yum.repos.d/centos.repo http://mirrors.aliyun.com/repo/Centos-7.repo
curl -o /etc/yum.repos.d/epel.repo http://mirrors.aliyun.com/repo/epel-7.repo
yum install -y vim net-tools wget unzip

修改主机名

[root@localhost ~]# hostnamectl set-hostname elk-node1
[root@localhost ~]# bash

[root@localhost ~]# hostnamectl set-hostname elk-node2
[root@localhost ~]# bash

[root@localhost ~]# hostnamectl set-hostname elk-node3
[root@localhost ~]# bash

配置映射

[root@elk-node1 ~]# cat /etc/hosts
127.0.0.1   localhost localhost.localdomain localhost4 localhost4.localdomain4
::1         localhost localhost.localdomain localhost6 localhost6.localdomain6
192.168.112.3 elk-node1
192.168.112.4 elk-node2
192.168.112.5 elk-node3

Elasticserach部署

安装Elasticserach

三台主机都需安装java及elasticserach

[root@elk-node1 ~]# yum install -y java-1.8.0-*

[root@elk-node1 ~]# wget https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-6.0.0.rpm

[root@elk-node1 ~]# rpm -ivh elasticsearch-6.0.0.rpm
### 参数含义：i表示安装，v表示显示安装过程，h表示显示进度

启动报错

### 二进制安装
[root@elk-node1 ~]# ln -s /opt/jdk1.8.0_391/bin/java /usr/bin/java

Elasticserach配置

elk1节点配置

[root@elk-node1 ~]# cat /etc/elasticsearch/elasticsearch.yml | grep -v ^# | grep -v ^$
cluster.name: ELK
node.name: elk-node-1
node.master: true
node.data: true
path.data: /var/lib/elasticsearch
path.logs: /var/log/elasticsearch
network.host: 192.168.112.3
http.port: 9200
discovery.zen.ping.unicast.hosts: ["elk-node1", "elk-node2","elk-node3"]

elk2节点配置

[root@elk-node2 ~]# cat /etc/elasticsearch/elasticsearch.yml | grep -v ^# | grep -v ^$
cluster.name: ELK
node.name: elk-node2
node.master: true
node.data: true
path.data: /var/lib/elasticsearch
path.logs: /var/log/elasticsearch
network.host: 192.168.112.4
http.port: 9200
discovery.zen.ping.unicast.hosts: ["elk-node1", "elk-node2","elk-node3"]

elk3节点配置

[root@elk-node3 ~]# cat /etc/elasticsearch/elasticsearch.yml | grep -v ^# | grep -v ^$
cluster.name: ELK
node.name: elk-node3
node.master: true
node.data: true
path.data: /var/lib/elasticsearch
path.logs: /var/log/elasticsearch
network.host: 192.168.112.5
http.port: 9200
discovery.zen.ping.unicast.hosts: ["elk-node1", "elk-node2","elk-node3"]

启动服务

[root@elk-node1 ~]# systemctl daemon-reload
[root@elk-node1 ~]# systemctl enable elasticsearch --now
Created symlink from /etc/systemd/system/multi-user.target.wants/elasticsearch.service to /usr/lib/systemd/system/elasticsearch.service.

检测进程和端口

[root@elk-node1 ~]# ps -ef | grep elasticsearch
elastic+  12663      1 99 22:28 ?        00:00:11 /bin/java -Xms1g -Xmx1g -XX:+UseConcMarkSweepGC -XX:CMSInitiatingOccupancyFraction=75 -XX:+UseCMSInitiatingOccupancyOnly -XX:+AlwaysPreTouch -server -Xss1m -Djava.awt.headless=true -Dfile.encoding=UTF-8 -Djna.nosys=true -XX:-OmitStackTraceInFastThrow -Dio.netty.noUnsafe=true -Dio.netty.noKeySetOptimization=true -Dio.netty.recycler.maxCapacityPerThread=0 -Dlog4j.shutdownHookEnabled=false -Dlog4j2.disable.jmx=true -XX:+HeapDumpOnOutOfMemoryError -XX:HeapDumpPath=/var/lib/elasticsearch -Des.path.home=/usr/share/elasticsearch -Des.path.conf=/etc/elasticsearch -cp /usr/share/elasticsearch/lib/* org.elasticsearch.bootstrap.Elasticsearch -p /var/run/elasticsearch/elasticsearch.pid --quiet
root      12720   1822  0 22:28 pts/0    00:00:00 grep --color=auto elasticsearch
[root@elk-node1 ~]# netstat -ntpl
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address           Foreign Address         State       PID/Program name    
tcp        0      0 0.0.0.0:22              0.0.0.0:*               LISTEN      1021/sshd           
tcp        0      0 127.0.0.1:25            0.0.0.0:*               LISTEN      1175/master         
tcp6       0      0 192.168.112.3:9200      :::*                    LISTEN      12663/java          
tcp6       0      0 192.168.112.3:9300      :::*                    LISTEN      12663/java          
tcp6       0      0 :::22                   :::*                    LISTEN      1021/sshd           
tcp6       0      0 ::1:25                  :::*                    LISTEN      1175/master

检测集群状态

[root@elk-node1 ~]# curl 'elk-node1:9200/_cluster/health?pretty'
{
  "cluster_name" : "ELK",   		//集群名称
  "status" : "green",   				//集群健康状态，green为健康，yellow或者red则是集群有问题
  "timed_out" : false   				//是否超时,
  "number_of_nodes" : 3,   			//集群中节点数
  "number_of_data_nodes" : 3,   //集群中data节点数量
  "active_primary_shards" : 0,
  "active_shards" : 0,
  "relocating_shards" : 0,
  "initializing_shards" : 0,
  "unassigned_shards" : 0,
  "delayed_unassigned_shards" : 0,
  "number_of_pending_tasks" : 0,
  "number_of_in_flight_fetch" : 0,
  "task_max_waiting_in_queue_millis" : 0,
  "active_shards_percent_as_number" : 100.0
}

Kibana部署

安装Kibana

[root@elk-node1 ~]# wget https://artifacts.elastic.co/downloads/kibana/kibana-6.0.0-x86_64.rpm

[root@elk-node1 ~]# rpm -ivh kibana-6.0.0-x86_64.rpm

Kibana配置

添加nginx源

[root@elk-node1 ~]# vim /etc/yum.repos.d/nginx.repo
[nginx]
name = nginx repo
baseurl = https://nginx.org/packages/mainline/centos/7/$basearch/
gpgcheck = 0
enabled = 1

安装nginx

[root@elk-node1 ~]# yum install -y nginx

启动服务

[root@elk-node1 ~]# systemctl enable nginx --now
Created symlink from /etc/systemd/system/multi-user.target.wants/nginx.service to /usr/lib/systemd/system/nginx.service.

配置nginx负载均衡

[root@elk-node1 ~]# cat /etc/nginx/nginx.conf

user  nginx;
worker_processes  auto;

error_log  /var/log/nginx/error.log notice;
pid        /var/run/nginx.pid;


events {
    worker_connections  1024;
}


http {
    include       /etc/nginx/mime.types;
    default_type  application/octet-stream;

    log_format  main  '$remote_addr - $remote_user [$time_local] "$request" '
                      '$status $body_bytes_sent "$http_referer" '
                      '"$http_user_agent" "$http_x_forwarded_for"';

    access_log  /var/log/nginx/access.log  main;

    sendfile        on;
    #tcp_nopush     on;

    keepalive_timeout  65;

    #gzip  on;


    upstream elasticsearch {
        zone elasticsearch 64K;
        server elk-node1:9200;
        server elk-node2:9200;
        server elk-node3:9200;
    }

    server {
        listen 80;
        server_name 192.168.112.3;

        location / {
            proxy_pass http://elasticsearch;
            proxy_redirect off;
            proxy_set_header Host $host;
            proxy_set_header X-Real-IP $remote_addr;
            proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        }

        access_log /var/log/es_access.log;
    }


    include /etc/nginx/conf.d/*.conf;
}

重启服务

[root@elk-node1 ~]# nginx -t
nginx: the configuration file /etc/nginx/nginx.conf syntax is ok
nginx: configuration file /etc/nginx/nginx.conf test is successful
[root@elk-node1 ~]# nginx -s reload
[root@elk-node1 ~]# systemctl restart nginx

Kibana配置

[root@elk-node1 ~]# cat /etc/kibana/kibana.yml | grep -v ^#
server.port: 5601
server.host: 192.168.112.3
elasticsearch.url: "http://192.168.112.3:80"

启动服务

[root@elk-node1 ~]# systemctl enable kibana --now
Created symlink from /etc/systemd/system/multi-user.target.wants/kibana.service to /etc/systemd/system/kibana.service.
[root@elk-node1 ~]# ps -ef | grep kibana
kibana    13384      1 32 06:02 ?        00:00:02 /usr/share/kibana/bin/../node/bin/node --no-warnings /usr/share/kibana/bin/../src/cli -c /etc/kibana/kibana.yml
root      13396   1822  0 06:03 pts/0    00:00:00 grep --color=auto kibana

浏览器访问

Zoopeeper集群部署

安装Zoopeeper

[root@elk-node1 ~]# tar -zxvf apache-zookeeper-3.8.3-bin.tar.gz -C /usr/local/
[root@elk-node1 ~]# mv /usr/local/apache-zookeeper-3.8.3-bin/ /usr/local/zookeeper
[root@elk-node1 ~]# cp /usr/local/zookeeper/conf/zoo_sample.cfg /usr/local/zookeeper/conf/zoo.cfg

配置环境变量

[root@elk-node1 ~]# cat >> /etc/profile << EOF
export ZOOKEEPER_HOME=/usr/local/zookeeper
export PATH=$ZOOKEEPER_HOME/bin:$PATH
EOF

[root@elk-node1 ~]# source /etc/profile

[root@elk-node1 ~]# scp /etc/profile 192.168.112.4:/etc/profile
[root@elk-node1 ~]# scp /etc/profile 192.168.112.5:/etc/profile

[root@elk-node2 ~]# source /etc/profile
[root@elk-node3 ~]# source /etc/profile

配置zoopeeper

[root@elk-node1 ~]# cat /usr/local/zookeeper/conf/zoo.cfg
# The number of milliseconds of each tick
tickTime=2000   #### zookeeper 之间心跳间隔2秒
# The number of ticks that the initial 
# synchronization phase can take
initLimit=10    ### LF初始通信时限
# The number of ticks that can pass between 
# sending a request and getting an acknowledgement
syncLimit=5     ### LF同步通信时限
# the directory where the snapshot is stored.
# do not use /tmp for storage, /tmp here is just 
# example sakes.
dataDir=/tmp/zookeeper  ### zookeeper保存数据的目录
dataLogDir=/usr/local/zookeeper/logs ### zookeeper保存日志文件的目录
# the port at which the clients will connect
clientPort=2181         ### 客户端连接 zookeeper 服务器的端口
# the maximum number of client connections.
# increase this if you need to handle more clients
#maxClientCnxns=60
maxClientCnxns=60
#
# Be sure to read the maintenance section of the 
# administrator guide before turning on autopurge.
#
# https://zookeeper.apache.org/doc/current/zookeeperAdmin.html#sc_maintenance
#
# The number of snapshots to retain in dataDir
#autopurge.snapRetainCount=3
autopurge.snapRetainCount=3
# Purge task interval in hours
# Set to "0" to disable auto purge feature
#autopurge.purgeInterval=1
autopurge.purgeInterval=1

server.1=elk-node1:2888:3888
server.2=elk-node2:2888:3888
server.3=elk-node3:2888:3888

## Metrics Providers
#
# https://prometheus.io Metrics Exporter
#metricsProvider.className=org.apache.zookeeper.metrics.prometheus.PrometheusMetricsProvider
#metricsProvider.httpHost=0.0.0.0
#metricsProvider.httpPort=7000
#metricsProvider.exportJvmInfo=true

配置节点标识

[root@elk-node1 ~]# scp /usr/local/zookeeper/conf/zoo.cfg 192.168.112.4:/usr/local/zookeeper/conf/zoo.cfg
[root@elk-node1 ~]# scp /usr/local/zookeeper/conf/zoo.cfg 192.168.112.5:/usr/local/zookeeper/conf/zoo.cfg

[root@elk-node1 ~]# mkdir /tmp/zookeeper
[root@elk-node1 ~]# echo "1" > /tmp/zookeeper/myid

[root@elk-node2 ~]# mkdir /tmp/zookeeper
[root@elk-node2 ~]# echo "2" > /tmp/zookeeper/myid

[root@elk-node3 ~]# mkdir /tmp/zookeeper
[root@elk-node3 ~]# echo "3" > /tmp/zookeeper/myid

启动服务

三个节点都需要启动否测报错

[root@elk-node1 ~]# zkServer.sh start
/bin/java
ZooKeeper JMX enabled by default
Using config: /usr/local/zookeeper/bin/../conf/zoo.cfg
Starting zookeeper ... STARTED

查看服务状态

[root@elk-node1 ~]# zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /usr/local/zookeeper/bin/../conf/zoo.cfg
Client port found: 2181. Client address: localhost. Client SSL: false.
Mode: follower

Kafka集群部署

安装Kafka

[root@elk-node1 ~]# tar -zxvf kafka_2.12-3.6.1.tgz -C /usr/local/
[root@elk-node1 ~]# mv /usr/local/kafka_2.12-3.6.1/ /usr/local/kafka
[root@elk-node1 ~]# cp /usr/local/kafka/config/server.properties{,.bak}
[root@elk-node1 ~]# scp kafka_2.12-3.6.1.tgz 192.168.112.4:/root
[root@elk-node1 ~]# scp kafka_2.12-3.6.1.tgz 192.168.112.5:/root

配置环境变量

[root@elk-node1 ~]# cat >> /etc/profile << EOF
export KAFKA_HOME=/usr/local/kafka
export PATH=$KAFKA_HOME/bin:$PATH
EOF

[root@elk-node1 ~]# source /etc/profile
[root@elk-node1 ~]# echo $KAFKA_HOME
/usr/local/kafka

[root@elk-node1 ~]# scp /etc/profile 192.168.112.4:/etc/profile
[root@elk-node1 ~]# scp /etc/profile 192.168.112.5:/etc/profile

[root@elk-node2 ~]# source /etc/profile
[root@elk-node3 ~]# source /etc/profile

配置Kafka

[root@elk-node1 ~]# grep -v "^#" /usr/local/kafka/config/server.properties.bak > /usr/local/kafka/config/server.properties
[root@elk-node1 ~]# vim /usr/local/kafka/config/server.properties
# 每一个broker在集群中的唯一表示，要求是正数
broker.id=1

# 监控的kafka端口
listenters=PLAINTEXT://192.168.112.3:9092

# broker处理消息的最大线程数，一般情况下不需要去修改
num.network.threads=3

# broker处理磁盘IO的线程数，数值应该大于你的硬盘数
num.io.threads=8

# socket的发送缓冲区
socket.send.buffer.bytes=102400

# socket的接受缓冲区
socket.receive.buffer.bytes=102400

# socket请求的最大字节数
socket.request.max.bytes=104857600

# kafka数据的存放地址,多个地址用逗号分割,多个目录分布在不同磁盘上可以提高读写性能 /tmp/kafka-log,/tmp/kafka-log2
log.dirs=/usr/local/kafka/kafka-logs

# 设置partitions的个数
num.partitions=1

num.recovery.threads.per.data.dir=1

offsets.topic.replication.factor=1

transaction.state.log.replication.factor=1

transaction.state.log.min.isr=1

# 数据文件保留多长时间,此处为168h,粒度还可设置为分钟,或按照文件大小
log.retention.hours=168

# topic的分区是以一堆segment文件存储的,这个控制每个segment的大小,会被topic创建时的指定参数覆盖
log.retention.check.interval.ms=300000

# zookeeper集群地址
zookeeper.connect=elk-node1:2181,elk-node2:2181,elk-node3:2181

# kafka连接zookeeper的超时时间
zookeeper.connection.timeout.ms=6000

group.initial.rebalance.delay.ms=0

[root@elk-node1 ~]# scp /usr/local/kafka/config/server.properties 192.168.112.4:/usr/local/kafka/config/server.properties
[root@elk-node1 ~]# scp /usr/local/kafka/config/server.properties 192.168.112.5:/usr/local/kafka/config/server.properties

########	修改节点broker.id
# 每一个broker在集群中的唯一表示，要求是正数
broker.id=1
broker.id=2
broker.id=3

启动Kafka

三个节点都需要启动

### 启动
[root@elk-node1 ~]# kafka-server-start.sh -daemon /usr/local/kafka/config/server.properties
### 关闭
[root@elk-node1 ~]# kafka-server-stop.sh

ps:注： kafka节点默认需要的内存为1G，在⼯作中可能会调⼤该参数，可修改kafka-server-start.sh的配置项。找到KAFKA_HEAP_OPTS配置项，例如修改为：export KAFKA_HEAP_OPTS="-Xmx2G -Xms2G"。

测试Kafka

[root@elk-node1 ~]# jps 
24099 QuorumPeerMain
48614 Jps
47384 Kafka
23258 Elasticsearch

创建Topic

`在kf1(Broker)上创建测试Tpoic：test-ken,这⾥我们指定了3个副本Broker、test-ken有2个分区`

[root@elk-node1 ~]# kafka-topics.sh --create --bootstrap-server elk-node1:9092 --replication-factor 3 --partitions 2 --topic test-ken
Created topic test-ken.

在创建Topic时不允许使⽤"_."之类的符号 选项解释：
--create：创建新的Topic
--bootstrap-server：指定要哪台Kafka服务器上创建Topic，主机加端⼝，指定的主机地址⼀ 定要和配置⽂件中的listeners⼀致
--zookeeper：指定要哪台zookeeper服务器上创建Topic，主机加端⼝，指定的主机地址⼀定要 和配置⽂件中的listeners⼀致
--replication-factor：创建Topic中的每个分区(partition)中的复制因⼦数量，即为Topic
的副本数量，建议和Broker节点数量⼀致，如果复制因⼦超出Broker节点将⽆法创建
--partitions：创建该Topic中的分区(partition)数量
--topic：指定Topic名称

查看Topic

Topic在kf1上创建后也会同步到集群中另外两个副本Broker：kf2、kf3,通过以下命令列出指定Broker的topic信息

[root@elk-node1 ~]# kafka-topics.sh --list --bootstrap-server elk-node1:9092 
test-ken

[root@elk-node1 ~]# kafka-topics.sh --list --bootstrap-server elk-node2:9092 __consumer_offsets
__consumer_offsets
test-ken

查看Topic详情

[root@elk-node3 ~]# kafka-topics.sh --describe --bootstrap-server elk-node1:9092 --topic test-ken
Topic: test-ken TopicId: CMsPBF2XQySuUyr9ekEf7Q PartitionCount: 2       ReplicationFactor: 3    Configs: 
        Topic: test-ken Partition: 0    Leader: 3       Replicas: 3,2,1 Isr: 3,2,1
        Topic: test-ken Partition: 1    Leader: 1       Replicas: 1,3,2 Isr: 1,3,2
        
`Topic:kafka_data`：			# topic名称
`PartitionCount: 2`：		# 分⽚数量
`ReplicationFactor: 3`：	# Topic副本数量

发送消息

向Broker(id=1)的Topic=test-ken发送消息

[root@elk-node1 ~]# kafka-console-producer.sh --broker-list elk-node1:9092 --topic test-ken
>this is test   
>bye

--broker-list：指定使⽤哪台broker来⽣产消息
--topic：指定要往哪个Topic中⽣产消息

验证接收消息

### 消费者：
### 从开始位置消费(所有节点均能收到)

### elk-node1测试
[root@elk-node1 ~]# kafka-console-consumer.sh --bootstrap-server elk-node2:9092 --topic test-ken --from-beginning 
this is test
bye

Processed a total of 2 messages

### elk-node2测试
[root@elk-node2 ~]# kafka-console-consumer.sh --bootstrap-server elk-node1:9092 --topic test-ken --from-beginning     
this is test
bye

Processed a total of 2 messages

### 消费者组：
### ⼀个Consumer group,多个consumer进程,数量⼩于等于partition分区的数量
### test-ken只有2个分区，只能有两个消费者consumer进程去轮询消费消息

[root@elk-node1 ~]# kafka-console-consumer.sh --bootstrap-server elk-node1:9092 --topic test-ken --group testgroup_ken

删除Topic

[root@elk-node1 ~]# kafka-topics.sh --delete --bootstrap-server elk-node1:9092 --topic test-ken

查看删除信息

[root@elk-node3 ~]# kafka-topics.sh --describe --bootstrap-server elk-node1:9092 --topic test-ken
Error while executing topic command : Topic 'test-ken' does not exist as expected
[2024-01-13 15:14:10,659] ERROR java.lang.IllegalArgumentException: Topic 'test-ken' does not exist as expected
        at kafka.admin.TopicCommand$.kafka$admin$TopicCommand$$ensureTopicExists(TopicCommand.scala:400)
        at kafka.admin.TopicCommand$TopicService.describeTopic(TopicCommand.scala:312)
        at kafka.admin.TopicCommand$.main(TopicCommand.scala:63)
        at kafka.admin.TopicCommand.main(TopicCommand.scala)
 (kafka.admin.TopicCommand$)

Zookeeper的作用

1、broker在zk中注册

kafka的每个broker（相当于⼀个节点，相当于⼀个机器）在启动时，都会在zk中注册，告诉zk其b
rokerid，在整个的集群中，broker.id/brokers/ids，当节点失效时，zk就会删除该节点，就 很⽅便的监控整个集群broker的变化，及时调整负载均衡。

WatchedEvent state:SyncConnected type:None path:null
[zk: elk-node1:2181(CONNECTED) 0] ls /brokers 
[ids, seqid, topics]
[zk: elk-node1:2181(CONNECTED) 1] ls /brokers/ids
[1, 2, 3]
[zk: elk-node1:2181(CONNECTED) 2]

2、topic在zk中注册

在kafka中可以定义很多个topic，每个topic⼜被分为很多个分区。⼀般情况下，每个分区独⽴在 存在⼀个broker上，所有的这些topic和broker的对应关系都有zk进⾏维护

刚才已经删除了Topic再次创建
[root@elk-node1 ~]# kafka-topics.sh --create --bootstrap-server elk-node1:9092 --replication-factor 3 --partitions 2 --topic test-ken
Created topic test-ken.

WatchedEvent state:SyncConnected type:None path:null
[zk: elk-node1:2181(CONNECTED) 0] ls /brokers/topics/test-ken/partitions
[0, 1]

3、consumer(消费者)在zk中注册

注意：从kafka-0.9版本及以后，kafka的消费者组和offset信息就不存zookeeper了，⽽是存到
broker服务器上。 所以，如果你为某个消费者指定了⼀个消费者组名称（group.id），那么，⼀旦这个消费者启动， 这个消费者组名和它要消费的那个topic的offset信息就会被记录在broker服务器上。，但是zook
eeper其实并不适合进⾏⼤批量的读写操作，尤其是写操作。因此kafka提供了另⼀种解决⽅案：增 加__consumeroffsets topic，将offset信息写⼊这个topic

[zk: elk-node1:2181(CONNECTED) 0] ls /brokers/topics
[__consumer_offsets, test-ken]
[zk: elk-node1:2181(CONNECTED) 1] ls /brokers/topics/__consumer_offsets/partitions
[0, 1, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 2, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 3, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 4, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 5, 6, 7, 8, 9]

Beats采集⽇志部署

安装Beats

[root@elk-node1 ~]# scp filebeat-6.0.0-x86_64.rpm 192.168.112.4:/root
[root@elk-node1 ~]# scp filebeat-6.0.0-x86_64.rpm 192.168.112.5:/root

[root@elk-node1 ~]# rpm -ivh filebeat-6.0.0-x86_64.rpm 
warning: filebeat-6.0.0-x86_64.rpm: Header V4 RSA/SHA512 Signature, key ID d88e42b4: NOKEY
Preparing...                          ################################# [100%]
Updating / installing...
   1:filebeat-6.0.0-1                 ################################# [100%]

Beats配置

elk-node1节点

### 编辑配置⽂件
[root@elk-node1 ~]# > /etc/filebeat/filebeat.yml
[root@elk-node1 ~]# vim vim /etc/filebeat/filebeat.yml
filebeat.prospectors:
- type: log
  enabled: true
  paths:
    - /var/log/es_access.log	### 此处可⾃⾏改为想要监听的⽇志⽂件

output.kafka:
  enabled: true
  hosts: ["elk-node1:9092","elk-node2:9092","elk-node3:9092"]
  topic: "es_access"		### 对应zookeeper⽣成的topic
  keep_alive: 10s

elk-node2节点

[root@elk-node2 ~]# > /etc/filebeat/filebeat.yml
[root@elk-node2 ~]# vim /etc/filebeat/filebeat.yml 
filebeat.prospectors:
- type: log
  enabled: true
  paths:
    - /var/log/vmware-network.log

output.kafka:
  enabled: true
  hosts: ["elk-node1:9092","elk-node2:9092","elk-node3:9092"]
  topic: "vmware-network"
  keep_alive: 10s

elk-node3节点

[root@elk-node3 ~]# > /etc/filebeat/filebeat.yml
[root@elk-node3 ~]# vim /etc/filebeat/filebeat.yml
filebeat.prospectors:
- type: log
  enabled: true
  paths:
    - /var/log/access.log

output.kafka:
  enabled: true
  hosts: ["elk-node1:9092","elk-node2:9092","elk-node3:9092"]
  topic: "access"
  keep_alive: 10s

启动服务

[root@elk-node1 ~]# systemctl enable filebeat --now
Created symlink from /etc/systemd/system/multi-user.target.wants/filebeat.service to /usr/lib/systemd/system/filebeat.service.

[root@elk-node1 ~]# systemctl status filebeat       
● filebeat.service - filebeat
   Loaded: loaded (/usr/lib/systemd/system/filebeat.service; enabled; vendor preset: disabled)
   Active: active (running) since Sat 2024-01-13 15:43:19 CST; 6s ago
     Docs: https://www.elastic.co/guide/en/beats/filebeat/current/index.html
 Main PID: 55537 (filebeat)
   CGroup: /system.slice/filebeat.service
           └─55537 /usr/share/filebeat/bin/filebeat -c /etc/filebeat/filebeat.yml -path.home /usr/share/filebeat -path.config /etc/filebeat -path.data /var/lib/filebeat...

Jan 13 15:43:19 elk-node1 systemd[1]: Started filebeat

Logstash部署

安装Logstash

[root@elk-node1 ~]# wget https://artifacts.elastic.co/downloads/logstash/logstash-6.0.0.rpm

[root@elk-node1 ~]# rpm -ivh logstash-6.0.0.rpm 
warning: logstash-6.0.0.rpm: Header V4 RSA/SHA512 Signature, key ID d88e42b4: NOKEY
Preparing...                          ################################# [100%]
Updating / installing...
   1:logstash-1:6.0.0-1               ################################# [100%]
Using provided startup.options file: /etc/logstash/startup.options
Successfully created system startup script for Logstash

配置Logstash

elk-node1节点

### 配置/etc/logstash/logstash.yml，修改增加如下
[root@elk-node1 ~]# grep -v '^#' /etc/logstash/logstash.yml 
http.host: "192.168.112.3"
path.data: /var/lib/logstash
path.config: /etc/logstash/conf.d/*.conf
path.logs: /var/log/logstash

elk-node2节点

### 配置logstash收集es_access的⽇志
[root@elk-node2 ~]# cat /etc/logstash/conf.d/es_access.conf
# Settings file in YAML
input {
  kafka {
    bootstrap_servers => "elk-node1:9092,elk-node2:9092,elk-node3:9092"
    group_id => "logstash"
    auto_offset_reset => "earliest"
    decorate_events => true
    topics => ["es_access"]
    type => "messages"
  }
}

output {
  if [type] == "messages" {
    elasticsearch {
      hosts => ["elk-node1:9200","elk-node2:9200","elk-node3:9200"]
      index => "es_access-%{+YYYY.MM.dd}"
    }
  }
}

### 配置logstash收集vmware的⽇志
[root@elk-node2 ~]# cat /etc/logstash/conf.d/vmware.conf
# Settings file in YAML
input {
  kafka {
    bootstrap_servers => "elk-node1:9092,elk-node2:9092,elk-node3:9092"
    group_id => "logstash"
    auto_offset_reset => "earliest"
    decorate_events => true
    topics => ["vmware"]
    type => "messages"
  }
}

output {
  if [type] == "messages" {
    elasticsearch {
      hosts => ["elk-node1:9200","elk-node2:9200","elk-node3:9200"]
      index => "vmware-%{+YYYY.MM.dd}"
    }
  }
}

### 配置logstash收集nginx的⽇志
[root@elk-node2 ~]# cat /etc/logstash/conf.d/nginx.conf
# Settings file in YAML
input {
  kafka {
    bootstrap_servers => "elk-node1:9092,elk-node2:9092,elk-node3:9092"
    group_id => "logstash"
    auto_offset_reset => "earliest"
    decorate_events => true
    topics => ["nginx"]
    type => "messages"
  }
}

output {
  if [type] == "messages" {
    elasticsearch {
      hosts => ["elk-node1:9200","elk-node2:9200","elk-node3:9200"]
      index => "nginx-%{+YYYY.MM.dd}"
    }
  }
}

检查配置文件是否有误

[root@elk-node2 ~]# ln -s /usr/share/logstash/bin/logstash /usr/bin/

### 检查es_access
[root@elk-node2 ~]# logstash --path.settings /etc/logstash/ -f /etc/logstash/conf.d/es_access.conf --config.test_and_exit
Sending Logstash's logs to /var/log/logstash which is now configured via log4j2.properties
Configuration OK

### 检查vmware
[root@elk-node2 ~]# logstash --path.settings /etc/logstash/ -f /etc/logstash/conf.d/vmware.conf --config.test_and_exit   
Sending Logstash's logs to /var/log/logstash which is now configured via log4j2.properties
Configuration OK

### 检查nginx
[root@elk-node2 ~]# logstash --path.settings /etc/logstash/ -f /etc/logstash/conf.d/nginx.conf --config.test_and_exit
Sending Logstash's logs to /var/log/logstash which is now configured via log4j2.properties
Configuration OK

### 为ok则代表没问题

### 参数解释：
 --path.settings : ⽤于指定logstash的配置⽂件所在的⽬录
 -f : 指定需要被检测的配置⽂件的路径
 --config.test_and_exit : 指定检测完之后就退出，不然就会直接启动了

启动Logstash

三个节点需要启动

### 检查配置⽂件没有问题后，启动Logstash服务
[root@elk-node2 ~]# systemctl enable logstash --now
Created symlink from /etc/systemd/system/multi-user.target.wants/logstash.service to /etc/systemd/system/logstash.service.

### 查看进程
[root@elk-node2 ~]# ps -ef | grep logstash
logstash  17845      1  0 17:32 ?        00:00:00 /bin/java -XX:+UseParNewGC -XX:+UseConcMarkSweepGC -XX:CMSInitiatingOccupancyFraction=75 -XX:+UseCMSInitiatingOccupancyOnly -XX:+DisableExplicitGC -Djava.awt.headless=true -Dfile.encoding=UTF-8 -XX:+HeapDumpOnOutOfMemoryError -Xmx1g -Xms256m -Xss2048k -Djffi.boot.library.path=/usr/share/logstash/vendor/jruby/lib/jni -Xbootclasspath/a:/usr/share/logstash/vendor/jruby/lib/jruby.jar -classpath : -Djruby.home=/usr/share/logstash/vendor/jruby -Djruby.lib=/usr/share/logstash/vendor/jruby/lib -Djruby.script=jruby -Djruby.shell=/bin/sh org.jruby.Main /usr/share/logstash/lib/bootstrap/environment.rb logstash/runner.rb --path.settings /etc/logstash

### 查看端口
[root@elk-node2 ~]# netstat -ntpl
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address           Foreign Address         State       PID/Program name    
tcp        0      0 127.0.0.1:25            0.0.0.0:*               LISTEN      1151/master         
tcp        0      0 0.0.0.0:22              0.0.0.0:*               LISTEN      1020/sshd           
tcp6       0      0 ::1:25                  :::*                    LISTEN      1151/master         
tcp6       0      0 :::9092                 :::*                    LISTEN      15757/java          
tcp6       0      0 :::2181                 :::*                    LISTEN      14812/java          
tcp6       0      0 :::40039                :::*                    LISTEN      15757/java          
tcp6       0      0 :::42696                :::*                    LISTEN      14812/java          
tcp6       0      0 192.168.112.4:3888      :::*                    LISTEN      14812/java          
tcp6       0      0 :::8080                 :::*                    LISTEN      14812/java          
tcp6       0      0 192.168.112.4:9200      :::*                    LISTEN      13070/java          
tcp6       0      0 192.168.112.4:9300      :::*                    LISTEN      13070/java          
tcp6       0      0 :::22                   :::*                    LISTEN      1020/sshd

启动报错解决

[root@elk-node2 ~]# systemctl start logstash
Failed to start logstash.service: Unit not found.

[root@elk-node2 ~]# sudo /usr/share/logstash/bin/system-install /etc/logstash/startup.options systemd
which: no java in (/sbin:/bin:/usr/sbin:/usr/bin)
could not find java; set JAVA_HOME or ensure java is in PATH

[root@elk-node2 ~]# ln -s /opt/jdk1.8.0_391/bin/java /usr/bin/java

[root@elk-node2 ~]# sudo /usr/share/logstash/bin/system-install /etc/logstash/startup.options systemd
Using provided startup.options file: /etc/logstash/startup.options
Manually creating startup for specified platform: systemd
Successfully created system startup script for Logstash

如果启动服务后，有进程但是没有9600端口，是因为权限问题，之前我们以root的身份在终端启动过logstash，所以产生的相关文件的属组属主都是root，解决方法如下

[root@elk-node2 ~]# cat /var/log/logstash/logstash-plain.log | grep que
[2024-01-13T17:23:56,589][INFO ][logstash.setting.writabledirectory] Creating directory {:setting=>"path.queue", :path=>"/var/lib/logstash/queue"}
[2024-01-13T17:23:56,589][INFO ][logstash.setting.writabledirectory] Creating directory {:setting=>"path.dead_letter_queue", :path=>"/var/lib/logstash/dead_letter_queue"}

[root@elk-node2 ~]# ll /var/lib/logstash/
total 0
drwxr-xr-x. 2 root root 6 Jan 13 17:23 dead_letter_queue
drwxr-xr-x. 2 root root 6 Jan 13 17:23 queue

### 修改/var/lib/logstash/⽬录的所属组为logstash，并重启服务
[root@elk-node2 ~]# chown -R logstash /var/lib/logstash/
[root@elk-node2 ~]# ll /var/lib/logstash/               
total 0
drwxr-xr-x. 2 logstash root 6 Jan 13 17:23 dead_letter_queue
drwxr-xr-x. 2 logstash root 6 Jan 13 17:23 queue

[root@elk-node2 ~]# systemctl restart logstash

[root@elk-node2 ~]# netstat -ntpl
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address           Foreign Address         State       PID/Program name    
tcp        0      0 127.0.0.1:25            0.0.0.0:*               LISTEN      1151/master         
tcp        0      0 0.0.0.0:22              0.0.0.0:*               LISTEN      1020/sshd           
tcp6       0      0 ::1:25                  :::*                    LISTEN      1151/master         
tcp6       0      0 127.0.0.1:9600          :::*                    LISTEN      18707/java          
tcp6       0      0 :::9092                 :::*                    LISTEN      15757/java          
tcp6       0      0 :::2181                 :::*                    LISTEN      14812/java          
tcp6       0      0 :::40039                :::*                    LISTEN      15757/java          
tcp6       0      0 :::42696                :::*                    LISTEN      14812/java          
tcp6       0      0 192.168.112.4:3888      :::*                    LISTEN      14812/java          
tcp6       0      0 :::8080                 :::*                    LISTEN      14812/java          
tcp6       0      0 192.168.112.4:9200      :::*                    LISTEN      13070/java          
tcp6       0      0 192.168.112.4:9300      :::*                    LISTEN      13070/java          
tcp6       0      0 :::22                   :::*                    LISTEN      1020/sshd

Kibana查看日志

[root@elk-node1 ~]# curl 'elk-node1:9200/_cat/indices?v'
health status index                uuid                   pri rep docs.count docs.deleted store.size pri.store.size
green  open   .kibana              sQtNJsqNQ3mW4Bs62m5hpQ   1   1          1            0     26.1kb           13kb
green  open   nginx-2024.01.13     KVTsisxoRGKs60LYwdlbVA   5   1        424            0    517.9kb        258.9kb
green  open   vmware-2024.01.13    S_uEeLq6TluD4fajPGAz-g   5   1        424            0    549.8kb        274.9kb
green  open   es_access-2024.01.13 -743RqwoQMOBhBOlkOdVWg   5   1        424            0    540.5kb        270.2kb

Web界⾯配置

浏览器访问192.168.112.3:5601，到Kibana上配置索引

此处的 Index pattern 使用 curl 'elk-node1:9200/_cat/indices?v'获取index

⽣产部署⽅案

在⼀个⽣产集群中我们可以对这些节点进⾏划分。

建议集群中设置3台以上的节点作为master节点【

node.master: true node.data: false】

这些节点只负责成为主节点，维护整个集群的状态。

再根据数据量设置⼀批data节点【

node.master: false node.data: true】

这些节点只负责存储数据，后期提供建⽴索引和查询索引的服务，这样的话如果⽤户请求⽐较频繁，这

些节点的压⼒也会⽐较⼤

所以在集群中建议再设置⼀批client节点【

node.master: false node.data:false】

这些节点只负责处理⽤户请求，实现请求转发，负载均衡等功能。

master节点：普通服务器即可(CPU 内存消耗⼀般)

data节点：主要消耗磁盘，内存

client节点：普通服务器即可(如果要进⾏分组聚合操作的话，建议这个节点内存也分配多⼀点)