How to Install ELK Stack on CentOS 7 / RHEL 7

Logs analysis has always been an important part system administration but it is one the most tedious and tiresome task, especially when dealing with a number of systems. Fortunately, ELK stack has eased up the task, ELK stack is now used for log inspection/analysis & it’s a combination of following three open source products

•  ElasticSearch -It is a No-SQL database that indexes and stores information
•  Logstash –  It is a log pipeline tool that collects & parses logs
•  Kibana – It provides GUI(Graphical User Interface) and used for visualization of the data & works on top of elasticsearch.

ELK stack receives logs from client through beats protocol, sent by using a beats client.In this tutorial, we are going to create an ELK stack on a Centos 7 machine & will also install beat client named ‘File Beat’ on the Client Machine.

• 192.168.0.180    elk-stack (CentOS 7)
• 192.168.0.70     client (CentOS 7)

Pre-requisite

Machine on which we will install ELK should have Java version 8 installed on it as . So make sure that java open-jdk version 1.8.0_* is installed and running and in case it is not installed, then run the beneath yum command to install

[[email protected] ~]# yum install java-1.8.0-openjdk

[[email protected] ~]# java -version
openjdk version "1.8.0_131"
OpenJDK Runtime Environment (build 1.8.0_131-b12)
OpenJDK 64-Bit Server VM (build 25.131-b12, mixed mode)
[[email protected] ~]#

Set the Hostname and update /etc/hosts file

[[email protected] ~]# hostnamectl set-hostname "elk-stack.example.com"

Update /etc/hosts file

192.168.0.180  elk-stack.example.com elk-stack

Installation Steps of ELK Stack Elasticsearch

We will start by importing the GPG keys for elasticsearch, this key will also be shared with logstash & kibana. To install elasticsearch, run

[[email protected] ~]# rpm –import https://packages.elastic.co/GPG-KEY-elasticsearch

[[email protected] ~]#

Now we will create a repo for the elastic-search repository,

[[email protected] ~]# vi /etc/yum.repos.d/elasticsearch.repo

[elasticsearch]
name=Elasticsearch repository
baseurl=http://packages.elastic.co/elasticsearch/2.x/centos
gpgcheck=1
gpgkey=http://packages.elastic.co/GPG-KEY-elasticsearch
enabled=1

Once the repository has been added, install elasticsearch using yum,

[[email protected] ~]# yum install elasticsearch -y

Now we start the service & will also set to start at boot time

[[email protected] ~]# systemctl daemon-reload
[[email protected] ~]# systemctl start elasticsearch
[[email protected] ~]# systemctl enable elasticsearch

Allow the 9200 tcp port in the OS firewall. In case Firewall is running

[[email protected] ~]# firewall-cmd --permanent --add-port 9200/tcp

We will now test elasticsearch to make sure that its responding to queries

[[email protected] ~]# curl -X GET http://localhost:9200

Output of above command should be something like below:

Logstash

We will now add logstash repository,

[[email protected] ~]# vi /etc/yum.repos.d/logstash.repo

[logstash]
name=Logstash
baseurl=http://packages.elasticsearch.org/logstash/2.2/centos
gpgcheck=1
gpgkey=http://packages.elasticsearch.org/GPG-KEY-elasticsearch
enabled=1

Now install logstash,

[[email protected] ~]# yum install logstash -y

Kibana

We will now create a repository for kibana,

[[email protected] ~]# vi /etc/yum.repos.d/kibana.repo

[kibana-4.5]
name=Kibana repository for 4.5.x packages
baseurl=http://packages.elastic.co/kibana/4.5/centos
gpgcheck=1
gpgkey=http://packages.elastic.co/GPG-KEY-elasticsearch
enabled=1

Now install kibana using yum,

[[email protected] ~]# yum install kibana -y

After installation, start service & enable it at boot time

[[email protected] ~]# systemctl start kibana
[[email protected] ~]# systemctl enable kibana

Allow 5601 port in OS Firewall

[[email protected] ~]# firewall-cmd --permanent --add-port 5601/tcp
[[email protected] ~]# firewall-cmd --reload

Next, we will access the webpage for kibana to make sure it’s working. To do that, open web-browser & enter the following url

http://IP-Address:5601/

Installation of ELK stack is now complete & we will make the necessary configurations.

Configuration SSL certificate for logstash

After the logstash installation, we will now create a SSL certificate for securing communication between logstash & filebeat (clients). Since we will be using IP address to connect to server, we will create SSL certificate for IP SAN.

Before creating a SSL certificate, we will make an entry of our IP in openssl.cnf,

[[email protected] ~]# vi /etc/pki/tls/openssl.cnf

and look for parameter with ‘subjectAltName’ under [ v3_ca ] section  & add your server IP to it,

subjectAltName = IP: 192.168.0.180

Now change the directory to /etc/ssl & create SSL certificate with 365 days validity,

[[email protected] ~]# cd /etc/ssl/
[[email protected] ssl]# openssl req -x509 -days 365 -batch -nodes -newkey rsa:2048 -keyout logstash-forwarder.key -out logstash_frwrd.crt
Generating a 2048 bit RSA private key
.....+++
...........+++
writing new private key to 'logstash-forwarder.key'
-----
[[email protected] ssl]#

Once the certificate is ready, this should be copied to all the clients using scp command.

Configuring Logstash

We will now create a configuration file for logstash under the folder ‘/etc/logstash/conf.d‘. This file will be divided into three sections i.e. input, filter & output section

[[email protected] ~]# vi /etc/logstash/conf.d/logstash.conf

# input section
input {
 beats {
   port => 5044
   ssl => true
   ssl_certificate => "/etc/ssl/logstash_frwrd.crt"
   ssl_key => "/etc/ssl/logstash-forwarder.key"
   congestion_threshold => "40"
  }
}

This section makes logstash to listen on port 5044 for incoming logs & also provides SSL certificate details for secure connection.

Next section i.e. ‘filter section’ will parse the logs before sending them to elasticsearch ,

# Filter section
filter {
if [type] == "syslog" {
    grok {
      match => { "message" => "%{SYSLOGLINE}" }
    }
    date {
match => [ "timestamp", "MMM  d HH:mm:ss", "MMM dd HH:mm:ss" ]

}
  }
}

Last section is ‘output section’  & it defines the location for the storage of logs,

# output section
output {
 elasticsearch {
  hosts => localhost
    index => "%{[@metadata][beat]}-%{+YYYY.MM.dd}"
       }
stdout {
    codec => rubydebug
       }
}

Now save the file & exit. All these sections can also be divided into three separate files but we have used them in single file for ease of configuration. We will now start the logstash service & enable it at boot time,

[[email protected] ~]# systemctl daemon-reload
[[email protected] ~]# systemctl start logstash
[[email protected] ~]# systemctl enable logstash

Allow 5044 tcp port in the OS firewall with following command so that Logstash get logs from Clients

[[email protected] conf.d]# firewall-cmd --permanent --add-port=5044/tcp
success
[[email protected] conf.d]# firewall-cmd --reload
success
[[email protected] conf.d]#

Installing Filebeat on Clients

Filebeat needs to installed on every system for which we need to analyse logs. Let’s first Copy certificate file from elk-stack server to the client

[[email protected] ~]# scp /etc/ssl/logstash_frwrd.crt [email protected]:/etc/ssl

To install filebeat, we will first add the repo for it,

[[email protected] ~]# vi /etc/yum.repos.d/filebeat.repo
[beats]
name=Elastic Beats Repository
baseurl=https://packages.elastic.co/beats/yum/el/$basearch
enabled=1
gpgkey=https://packages.elastic.co/GPG-KEY-elasticsearch
gpgcheck=1

Now install filebeat by running beneath command

[[email protected] ~]# yum install filebeat

We will now make changes to the configuration file to connect the filebeat clients to our ELK stack but before we do that make sure that the certificate we created during logstash configuration is copied on the client under ‘/etc/ssl’ directory. Once that’s done, we will start the filebeat configuration,

[[email protected] ~]# vi /etc/filebeat/filebeat.yml

We need to make changes to the following three sections,

.  .  .
       paths:
        - /var/log/*.log
        - /var/log/secure
        - /var/log/messages
.  .  .

Under this section, we can allow the logs that needs to be analysed. I have left it at default i.e. all logs but you can modify it to send only one or two log files. For the next section, change the document_type to ‘syslog’,

.  .  .
      document_type: syslog
.  .  .

& in the last section i.e ‘output’, we will define our ELK stack server IP address & location for the ssl certificate,

.   .   .
output:
  logstash:
    hosts: ["192.168.0.180:5044"]
    tls:
      certificate_authorities: ["/etc/ssl/logstash_frwrd.crt"]
.   .   .

 

Note: Disable elasticsearch output, comment out the entry 「hosts: [「localhost:9200″]」 in case it is enable.

Now start the service & enable it at boot time,

[[email protected] ~]# systemctl restart filebeat
[[email protected] ~]# systemctl enable filebeat

That’s it, configurations on both server end & client end are now complete. We can now login to the  kibana web interface to look for analysed logs.

Create Index Pattern, change logstash-* to filebeat-*

Click on Create

Click on Discover and then search, we will get logs something like below

That’s all from this article, Please share your valuable feedback and comments.