CentOS 7.6 部署ELK日志分析系统步骤

188次阅读

共计 14632 个字符，预计需要花费 37 分钟才能阅读完成。

记录在 CentOS 7.6 下部署 ELK 日志分析系统的过程步骤，希望对大家有所帮助。

下载 elasticsearch

创建 elk 用户并授权
useradd elk
chown -R elk:elk /home/elk/elasticsearch
chown -R elk:elk /home/elk/elasticsearch1
chown -R elk:elk /home/elk/elasticsearch2
mkdir -p /home/eladata
mkdir -p /var/log/elk
chown -R elk:elk /home/eladata
chown -R elk:elk /var/log/elk

主节点 master

elasticsearch 解压, 修改配置文件
/home/elk/elasticsearch/config
[root@localhost config]# grep -v “^#” elasticsearch.yml
cluster.name: my-application
node.name: node0
node.master: true
node.attr.rack: r1
node.max_local_storage_nodes: 3
path.data: /home/eladata
path.logs: /var/log/elk
http.cors.enabled: true
http.cors.allow-origin: “*”
network.host: 192.168.1.70
http.port: 9200
transport.tcp.port: 9301
discovery.zen.minimum_master_nodes: 1
cluster.initial_master_nodes: [“node0”]

手动启动命令
su elk -l -c ‘/home/elk/elasticsearch/bin/elasticsearch -d’

启动文件 elasticsearch.service
[root@localhost system]# pwd
/lib/systemd/system
[root@localhost system]# cat elasticsearch.service
[Unit]
Description=Elasticsearch
Documentation=http://www.elastic.co
Wants=network-online.target
After=network-online.target
[Service]
RuntimeDirectory=elasticsearch
PrivateTmp=true
Environment=ES_HOME=/home/elk/elasticsearch
Environment=ES_PATH_CONF=/home/elk/elasticsearch/config
Environment=PID_DIR=/var/run/elasticsearch
EnvironmentFile=-/etc/sysconfig/elasticsearch
WorkingDirectory=/home/elk/elasticsearch
User=elk
Group=elk
ExecStart=/home/elk/elasticsearch/bin/elasticsearch -p ${PID_DIR}/elasticsearch.pid –quiet
StandardOutput=journal
StandardError=inherit
LimitNOFILE=65536
LimitNPROC=4096
LimitAS=infinity
LimitFSIZE=infinity
TimeoutStopSec=0
KillSignal=SIGTERM
KillMode=process
SendSIGKILL=no
SuccessExitStatus=143
[Install]
WantedBy=multi-user.target

[root@localhost system]#

Node1 节点
/home/elk/elasticsearch1/config
[root@localhost config]# grep -v “^#” elasticsearch.yml
cluster.name: my-application
node.name: node1
node.master: false
node.attr.rack: r1
node.max_local_storage_nodes: 3
path.data: /home/eladata
path.logs: /var/log/elk
http.cors.enabled: true
http.cors.allow-origin: “*”
network.host: 192.168.1.70
transport.tcp.port: 9303
http.port: 9302
discovery.zen.ping.unicast.hosts: [“192.168.1.70:9301”]
[root@localhost config]#

手动启动命令
su elk -l -c ‘/home/elk/elasticsearch1/bin/elasticsearch1 -d’

启动文件 elasticsearch1.service
[root@localhost system]# pwd
/lib/systemd/system
[root@localhost system]# cat elasticsearch1.service
[Unit]
Description=Elasticsearch
Documentation=http://www.elastic.co
Wants=network-online.target
After=network-online.target
[Service]
RuntimeDirectory=elasticsearch1
PrivateTmp=true
Environment=ES_HOME=/home/elk/elasticsearch1
Environment=ES_PATH_CONF=/home/elk/elasticsearch1/config
Environment=PID_DIR=/var/run/elasticsearch
EnvironmentFile=-/etc/sysconfig/elasticsearch
WorkingDirectory=/home/elk/elasticsearch
User=elk
Group=elk
ExecStart=/home/elk/elasticsearch1/bin/elasticsearch -p ${PID_DIR}/elasticsearch.pid –quiet
StandardOutput=journal
StandardError=inherit
LimitNOFILE=65536
LimitNPROC=4096
LimitAS=infinity
LimitFSIZE=infinity
TimeoutStopSec=0
KillSignal=SIGTERM
KillMode=process
SendSIGKILL=no
SuccessExitStatus=143
[Install]
WantedBy=multi-user.target

[root@localhost system]#

Node2 节点
/home/elk/elasticsearch2/config
[root@localhost config]# grep -v “^#” elasticsearch.yml
cluster.name: my-application
node.name: node2
node.attr.rack: r1
node.master: false
node.max_local_storage_nodes: 3
path.data: /home/eladata
path.logs: /var/log/elk
http.cors.enabled: true
http.cors.allow-origin: “*”
network.host: 192.168.1.70
http.port: 9203
transport.tcp.port: 9304
discovery.zen.ping.unicast.hosts: [“192.168.1.70:9301”]
discovery.zen.minimum_master_nodes: 1
[root@localhost config]#

手动启动命令
su elk -l -c ‘/home/elk/elasticsearch2/bin/elasticsearch2 -d’

启动文件 elasticsearch2.service
[root@localhost system]# pwd
/lib/systemd/system
[root@localhost system]# cat elasticsearch2.service
[Unit]
Description=Elasticsearch
Documentation=http://www.elastic.co
Wants=network-online.target
After=network-online.target
[Service]
RuntimeDirectory=elasticsearch2
PrivateTmp=true
Environment=ES_HOME=/home/elk/elasticsearch2
Environment=ES_PATH_CONF=/home/elk/elasticsearch2/config
Environment=PID_DIR=/var/run/elasticsearch
EnvironmentFile=-/etc/sysconfig/elasticsearch
WorkingDirectory=/home/elk/elasticsearch2
User=elk
Group=elk
ExecStart=/home/elk/elasticsearch2/bin/elasticsearch -p ${PID_DIR}/elasticsearch.pid –quiet
StandardOutput=journal
StandardError=inherit
LimitNOFILE=65536
LimitNPROC=4096
LimitAS=infinity
LimitFSIZE=infinity
TimeoutStopSec=0
KillSignal=SIGTERM
KillMode=process
SendSIGKILL=no
SuccessExitStatus=143
[Install]
WantedBy=multi-user.target

[root@localhost system]#

下载 logstash

目录如下，默认配置即可
[root@localhost logstash]# pwd
/home/elk/logstash
[root@localhost logstash]#

手动启动命令
./logstash -f ../dev.conf
nohup ./logstash -f ../dev.conf &

下载 kibana

配置文件如下
[root@localhost config]# pwd
/home/elk/kibana/config
[root@localhost config]# grep -v “^#” kibana.yml
server.host: “192.168.1.70”
elasticsearch.hosts: [“http://192.168.1.70:9200”]
kibana.index: “.kibana”
i18n.locale: “zh-CN”

手动启动命令
./kibana
nohup ./kibana &

kibana 启动文件
[root@localhost system]# pwd
/lib/systemd/system
[root@localhost system]# cat kibana.service
[Unit]
Description=Kibana Server Manager
[Service]
ExecStart=/home/elk/kibana/bin/kibana
[Install]
WantedBy=multi-user.target
[root@localhost system]#

端口为：5601 访问：192.168.1.70:5601

安装 Elasticsearch -head
yum install git npm
git clone https://github.com/mobz/elasticsearch-head.git
[root@localhost elasticsearch-head]# pwd
/home/elk/elasticsearch-head
[root@localhost elasticsearch-head]#

启动
npm install
npm run start
nohup npm run start &

curl -XPUT ‘192.168.2.67:9100/book’

访问 192.168.2.67：9100 即可访问

下载 kafka

修改配置文件如下
[root@localhost config]# pwd
/home/elk/kafka/config
[root@localhost config]# grep -v “^#” server.properties
broker.id=0
listeners=PLAINTEXT://192.168.1.70:9092
num.network.threads=3
num.io.threads=8
socket.send.buffer.bytes=102400
socket.receive.buffer.bytes=102400
socket.request.max.bytes=104857600
log.dirs=/var/log/kafka-logs
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
log.retention.hours=168
log.segment.bytes=1073741824
log.retention.check.interval.ms=300000
zookeeper.connect=localhost:2181
zookeeper.connection.timeout.ms=6000
group.initial.rebalance.delay.ms=0
delete.topic.enable=true
[root@localhost config]#

kafka 配置启动 zookeeper

手动启动方式
[root@localhost bin]# pwd
/home/elk/kafka/bin
[root@localhost bin]#
./zookeeper-server-start.sh ../config/zookeeper.properties

systemctl 启动 zookeeper
[root@localhost system]# pwd
/lib/systemd/system
[root@localhost system]# cat zookeeper.service
[Service]
Type=forking
SyslogIdentifier=zookeeper
Restart=always
RestartSec=0s
ExecStart=/home/elk/kafka/bin/zookeeper-server-start.sh -daemon /home/elk/kafka/config/zookeeper.properties
ExecStop=/home/elk/kafka/bin/zookeeper-server-stop.sh
[root@localhost system]#

启动 kafka 服务

手动启动方式
./kafka-server-start.sh ../config/server.properties

systemctl 启动 kafka
[root@localhost system]# pwd
/lib/systemd/system
[root@localhost system]# cat kafka.service
[Unit]
Description=Apache kafka
After=network.target
[Service]
Type=simple
Restart=always
RestartSec=0s
ExecStart=/home/elk/kafka/bin/kafka-server-start.sh /home/elk/kafka/config/server.properties
ExecStop=/home/elk/kafka/bin/kafka-server-stop.sh
[root@localhost system]#

测试 kafka

新建一个名字为 test 的 topic
/kafka-topics.sh –create –zookeeper 192.168.1.70:2181 –replication-factor 1 –partitions 1 –topic test

查看 kafka 中的 topic
./kafka-topics.sh –list –zookeeper 192.168.1.70:2181

往 kafka topic 为 test 中生产消息
./kafka-console-producer.sh –broker-list 192.168.1.70:9092 –topic test

在 kafka topic 为 test 中消费消息
bin/kafka-console-consumer.sh –bootstrap-server 192.168.1.70:9092 –topic test –from-beginning

生产的消息，消费那边接受到即是 ok 的

目标机器安装 filebeat

安装 6.5 版本的
[root@localhost filebeat]# pwd
/usr/local/filebeat
[root@localhost filebeat]# cat filebeat.yml
filebeat.prospectors:
– type: log
paths:
– /opt/logs/workphone-tcp/catalina.out
fields:
tag: 54_tcp_catalina_out
– type: log
paths:
– /opt/logs/workphone-webservice/catalina.out
fields:
tag: 54_web_catalina_out
name: 192.168.1.54
filebeat.config.modules:
path: ${path.config}/modules.d/*.yml
reload.enabled: false
setup.template.settings:
index.number_of_shards: 3
output.kafka:
hosts: [“192.168.1.70:9092”]
topic: “filebeat-log”
partition.hash:
reachable_only: true
compression: gzip
max_message_bytes: 1000000
required_acks: 1

[root@localhost filebeat]#

安装完成后去 logstash 编辑配置文件

logstash 操作
[root@localhost logstash]# pwd
/home/elk/logstash
[root@localhost logstash]# cat dev.conf
input {
kafka{
bootstrap_servers => “192.168.1.70:9092”
topics => [“filebeat-log”]
codec => “json”
}
}
filter {
if [fields][tag]==”jpwebmap” {
json{
source => “message”
remove_field => “message”
}
geoip {
source => “client”
target => “geoip”
add_field => [“[geoip][coordinates]”, “%{[geoip][longitude]}” ]
add_field => [“[geoip][coordinates]”, “%{[geoip][latitude]}” ]
}
mutate {
convert => [“[geoip][coordinates]”, “float”]
}
}
if [fields][tag] == “54_tcp_catalina_out”{
grok {
match => [“message”, “%{TIMESTAMP_ISO8601:logdate}”]
}
date {
match => [“logdate”, “ISO8601”]
}
mutate {
remove_field => [“logdate”]
}
}
if [fields][tag] == “54_web_catalina_out”{
grok {
match => [“message”, “%{TIMESTAMP_ISO8601:logdate}”]
}
date {
match => [“logdate”, “ISO8601”]
}
mutate {
remove_field => [“logdate”]
}
}
if [fields][tag] == “55_tcp_catalina_out”{
grok {
match => [“message”, “%{TIMESTAMP_ISO8601:logdate}”]
}
date {
match => [“logdate”, “ISO8601”]
}
mutate {
remove_field => [“logdate”]
}
}
if [fields][tag] == “55_web_catalina_out”{
grok {
match => [“message”, “%{TIMESTAMP_ISO8601:logdate}”]
}
date {
match => [“logdate”, “ISO8601”]
}
mutate {
remove_field => [“logdate”]
}
}
if [fields][tag] == “51_nginx80_access_log” {
mutate {
add_field => {“spstr” => “%{[log][file][path]}” }
}
mutate {
split => [“spstr” , “/”]
# save the last element of the array as the api_method.
add_field => [“src”, “%{[spstr][-1]}” ]
}
mutate{
remove_field => [“friends”, “ecs”, “agent” , “spstr”]
}
grok {
match => {“message” => “%{IPORHOST:remote_addr} – %{DATA:remote_user} \[%{HTTPDATE:time}\] \”%{WORD:method} %{DATA:url} HTTP/%{NUMBER:http_version}\” %{NUMBER:response_code} %{NUMBER:body_sent:bytes} \”%{DATA:referrer}\” \”%{DATA:agent}\” \”%{DATA:x_forwarded_for}\” \”%{NUMBER:request_time}\” \”%{DATA:upstream_addr}\” \”%{DATA:upstream_status}\”” }
remove_field => “message”
}
date {
match => [“time”, “dd/MMM/yyyy:HH:mm:ss Z”]
target => “@timestamp”
}
geoip {
source => “x_forwarded_for”
target => “geoip”
database => “/home/elk/logstash/GeoLite2-City.mmdb”
add_field => [“[geoip][coordinates]”, “%{[geoip][longitude]}” ]
add_field => [“[geoip][coordinates]”, “%{[geoip][latitude]}” ]
}
mutate {
convert => [“[geoip][coordinates]”, “float”]
}
}
}
output {
if [fields][tag] == “wori”{
elasticsearch {
hosts => [“192.168.1.70:9200”]
index => “zabbix”
}
}
if [fields][tag] == “54_tcp_catalina_out”{
elasticsearch {
hosts => [“192.168.1.70:9200”]
index => “54_tcp_catalina_out”
}
}
if [fields][tag] == “54_web_catalina_out”{
elasticsearch {
hosts => [“192.168.1.70:9200”]
index => “54_web_catalina_out”
}
}
if [fields][tag] == “55_tcp_catalina_out”{
elasticsearch {
hosts => [“192.168.1.70:9200”]
index => “55_tcp_catalina_out”
}
}
if [fields][tag] == “55_web_catalina_out”{
elasticsearch {
hosts => [“192.168.1.70:9200”]
index => “55_web_catalina_out”
}
}
if [fields][tag] == “51_nginx80_access_log” {
stdout{}
elasticsearch {
hosts => [“192.168.1.70:9200”]
index => “51_nginx80_access_log”
}
}
}

其他的配置文件

index.conf
filter {
mutate {
add_field => {“spstr” => “%{[log][file][path]}” }
}
mutate {
split => [“spstr” , “/”]
# save the last element of the array as the api_method.
add_field => [“src”, “%{[spstr][-1]}” ]
}
mutate{
remove_field => [“friends”, “ecs”, “agent” , “spstr”]
}
}

Java.conf
filter {
if [fields][tag] == “java”{
grok {
match => [“message”, “%{TIMESTAMP_ISO8601:logdate}”]
}
date {
match => [“logdate”, “ISO8601”]
}
mutate {
remove_field => [“logdate”]
}
} #End if
}

kafkainput.conf
input {
kafka{
bootstrap_servers => “172.16.11.68:9092”
#topics => [“ql-prod-tomcat”]
topics => [“ql-prod-dubbo”,”ql-prod-nginx”,”ql-prod-tomcat”]
codec => “json”
consumer_threads => 5
decorate_events => true
#auto_offset_reset => “latest”
group_id => “logstash”
#client_id => “”
############################# HELK Optimizing Latency #############################
fetch_min_bytes => “1”
request_timeout_ms => “305000”
############################# HELK Optimizing Availability #############################
session_timeout_ms => “10000”
max_poll_records => “550”
max_poll_interval_ms => “300000”
}

}
#input {
# kafka{
# bootstrap_servers => “172.16.11.68:9092”
# topics => [“ql-prod-java-dubbo”,”ql-prod”,”ql-prod-java”]
# codec => “json”
# consumer_threads => 15
# decorate_events => true
# auto_offset_reset => “latest”
# group_id => “logstash-1”
# ############################# HELK Optimizing Latency #############################
# fetch_min_bytes => “1”
# request_timeout_ms => “305000”
# ############################# HELK Optimizing Availability #############################
# session_timeout_ms => “10000”
# max_poll_records => “550”
# max_poll_interval_ms => “300000”
# }

#}

nginx.conf
filter {
if [fields][tag] == “nginx-access” {
mutate {
add_field => {“spstr” => “%{[log][file][path]}” }
}
mutate {
split => [“spstr” , “/”]
# save the last element of the array as the api_method.
add_field => [“src”, “%{[spstr][-1]}” ]
}
mutate{
remove_field => [“friends”, “ecs”, “agent” , “spstr”]
}

grok {
match => {“message” => “%{IPORHOST:remote_addr} – %{DATA:remote_user} \[%{HTTPDATE:time}\] \”%{WORD:method} %{DATA:url} HTTP/%{NUMBER:http_version}\” %{NUMBER:response_code} %{NUMBER:body_sent:bytes} \”%{DATA:referrer}\” \”%{DATA:agent}\” \”%{DATA:x_forwarded_for}\” \”%{NUMBER:request_time}\” \”%{DATA:upstream_addr}\” \”%{DATA:upstream_status}\”” }
remove_field => “message”
}
date {
match => [“time”, “dd/MMM/yyyy:HH:mm:ss Z”]
target => “@timestamp”
}
geoip {
source => “x_forwarded_for”
target => “geoip”
database => “/opt/logstash-6.2.4/GeoLite2-City.mmdb”
add_field => [“[geoip][coordinates]”, “%{[geoip][longitude]}” ]
add_field => [“[geoip][coordinates]”, “%{[geoip][latitude]}” ]

}
mutate {
convert => [“[geoip][coordinates]”, “float”]
}

} #endif
}

ouput.conf
output{
if [fields][tag] == “nginx-access” {
stdout{}
elasticsearch {
user => elastic
password => WR141bp2sveJuGFaD4oR
hosts => [“172.16.11.67:9200”]
index => “logstash-%{[fields][proname]}-%{+YYYY.MM.dd}”
}
}
#stdout{}
if [fields][tag] == “java” {
elasticsearch {
user => elastic
password => WR141bp2sveJuGFaD4oR
hosts => [“172.16.11.66:9200″,”172.16.11.68:9200”]
index => “%{[host][name]}-%{[src]}”
}
}
}