Install ELK as Centralized Logfile Management Server on CentOS 7
This tutorial explains how to setup a centralized logfile management server using ELK stack on CentOS 7. As anyone who not already know, ELK is the combination of 3 services: ElasticSearch, Logstash, and Kibana. To build a complete centralized log management server using this concept, it would require to have each of this package as it's serve a different purpose and related to each other. Basically it works together like this:
- For every client you want to manage, it will produce it's own log of related services.
- For the server that will be use to manage all logging infomation from each client, it will use LogStash package to collect and transform the data to a relative value. By definition, it's an open source, server-side data processing pipeline that ingests data from a multitude of sources simultaneously, transforms it
- Once data collected and trasform, the management server will use ElasticSearch to help and analyze the data to a relevant value. You can use general query language if want produce a related report as needed
- As related data been verified and analyze, this is where Kibana package come to picture as it can helps to visualize and manage the relevant data to a proper view or combine it into desirable glossy dashboard for easy understanding.
Below picture summarizes the workflow process:
1. Preliminary Note
For this tutorial, I am using CentOS Linux 7.4 in the 64bit version. In this tutorial we will use 3 server: The first one will be used to as the management server and the other 2 will be used as clients. For this exercise, we will use the management server to monitor an existing MySQL service which already has been setup, configured and running under each client. As MySQL is a database service which is used mainly for OLTP purpose, we will make our management server to log 2 logging processes which is the health check of the MySQL service itself and the slow query transaction. By the end of this tutorial, we will see that any information logged from any MySQL service inside dedicate client can be seen,visualize and analyze simultaneously from the management server directly real time.
2. Installation Phase
For the installation phase, we will start with FileBeat installation on both MySQL DB server that act as client. Let's start the process, below are the steps:
[root@mysql_db1 opt]# cd
[root@mysql_db1 ~]# cd /opt/
[root@mysql_db1 opt]# wget https://artifacts.elastic.co/downloads/beats/filebeat/filebeat-6.2.1-x86_64.rpm
--2018-06-09 10:50:46-- https://artifacts.elastic.co/downloads/beats/filebeat/filebeat-6.2.1-x86_64.rpm
Resolving artifacts.elastic.co (artifacts.elastic.co)... 107.21.237.188, 107.21.253.15, 184.73.245.233, ...
Connecting to artifacts.elastic.co (artifacts.elastic.co)|107.21.237.188|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 12697093 (12M) [binary/octet-stream]
Saving to: ‘filebeat-6.2.1-x86_64.rpm’
100%[==============================================================================>] 12,697,093 2.20MB/s in 6.9s
2018-06-09 10:51:00 (1.75 MB/s) - ‘filebeat-6.2.1-x86_64.rpm’ saved [12697093/12697093]
[root@mysql_db1 opt]# yum localinstall -y filebeat-6.2.1-x86_64.rpm
Loaded plugins: fastestmirror, ovl
Examining filebeat-6.2.1-x86_64.rpm: filebeat-6.2.1-1.x86_64
Marking filebeat-6.2.1-x86_64.rpm to be installed
Resolving Dependencies
--> Running transaction check
---> Package filebeat.x86_64 0:6.2.1-1 will be installed
--> Finished Dependency Resolution
Dependencies Resolved
========================================================================================================================
Package Arch Version Repository Size
========================================================================================================================
Installing:
filebeat x86_64 6.2.1-1 /filebeat-6.2.1-x86_64 49 M
Transaction Summary
========================================================================================================================
Install 1 Package
Total size: 49 M
Installed size: 49 M
Downloading packages:
Running transaction check
Running transaction test
Transaction test succeeded
Running transaction
Installing : filebeat-6.2.1-1.x86_64 1/1
Verifying : filebeat-6.2.1-1.x86_64 1/1
Installed:
filebeat.x86_64 0:6.2.1-1
Complete!
Once done, we'll list out the default module that enable by FileBeat package and enable the mysql module which is needed for our cases here. Below are the steps:
[root@mysql_db1 opt]# filebeat modules list
Enabled:
Disabled:
apache2
auditd
icinga
kafka
logstash
mysql
nginx
osquery
postgresql
redis
system
traefik
[root@mysql_db1 opt]# filebeat modules enable mysql
Enabled mysql
Done, now let's edit the configuration needed for mysql module that we've enable just now. By default, once we've enable the mysql module from filebeat package, it will automaticallt created a yaml file inside modules.d directory. Yet if the file was not created, feel free to create a new yaml file inside same location. Below are the steps:
[root@mysql_db1 opt]# vi /etc/filebeat/modules.d/mysql.yml
- module: mysql
error:
enabled: true
var.paths: ["/var/lib/mysql/mysql-error.log*"]
slowlog:
enabled: true
var.paths: ["/var/lib/mysql/log-slow-queries.log*"]
As shown above, we've decided to log 2 logging process from MySQL service which is the healthcheck of the database itself and the slow query log.
Now once everything done, let's make some configuration inside the main config file for filebeat under filebeat.yml file. Below are the configuration set:
[root@mysql_db1 opt]# vi /etc/filebeat/filebeat.yml
#=========================== Filebeat prospectors =============================
filebeat.prospectors:
- type: log
enabled: false
paths:
- /var/lib/mysql/mysql-error.log
- /var/lib/mysql/log-slow-queries.log
#============================= Filebeat modules ===============================
filebeat.config.modules:
path: ${path.config}/modules.d/*.yml
reload.enabled: false
#==================== Elasticsearch template setting ==========================
setup.template.settings:
index.number_of_shards: 3
#================================ General =====================================
setup.kibana:
#----------------------------- Logstash output --------------------------------
output.logstash:
hosts: ["172.17.0.6:5044"]
Notice on the above that we've set an IP address for the logstash host which is 172.17.0.6 .This IP is the address for our centralized management server which will crawl directly to collect the logging data. I've set the hardcoded IP as I didn't make any alternative changes under /etc/hosts file and didn't use any DNS server for this tutorial. Yet, feel free to use the hostname of the management server if you've did make the alternative changes.
As all has been setup as per plan, let's start the filebeat services. Below are the steps:
[root@mysql_db1 opt]# filebeat setup -e
2018-06-09T11:04:37.277Z INFO instance/beat.go:468 Home path: [/usr/share/filebeat] Config path: [/etc/filebeat] Data path: [/var/lib/filebeat] Logs path: [/var/log/filebeat]
2018-06-09T11:04:37.277Z INFO instance/beat.go:475 Beat UUID: 98503460-035e-4476-8e4d-10470433dba5
2018-06-09T11:04:37.277Z INFO instance/beat.go:213 Setup Beat: filebeat; Version: 6.2.1
2018-06-09T11:04:37.277Z INFO pipeline/module.go:76 Beat name: lara
2018-06-09T11:04:37.278Z ERROR instance/beat.go:667 Exiting: Template loading requested but the Elasticsearch output is not configured/enabled
Exiting: Template loading requested but the Elasticsearch output is not configured/enabled
[root@mysql_db1 opt]# filebeat -e &
[1] 22010
[root@mysql_db1 opt]# 2018-06-09T12:45:18.812Z INFO instance/beat.go:468 Home path: [/usr/share/filebeat] Config path: [/etc/filebeat] Data path: [/var/lib/filebeat] Logs path: [/var/log/filebeat]
2018-06-09T12:45:18.813Z INFO instance/beat.go:475 Beat UUID: 98503460-035e-4476-8e4d-10470433dba5
2018-06-09T12:45:18.813Z INFO instance/beat.go:213 Setup Beat: filebeat; Version: 6.2.1
2018-06-09T12:45:18.813Z INFO pipeline/module.go:76 Beat name: lara
2018-06-09T12:45:18.813Z INFO [monitoring] log/log.go:97 Starting metrics logging every 30s
2018-06-09T12:45:18.813Z INFO instance/beat.go:301 filebeat start running.
2018-06-09T12:45:18.814Z INFO registrar/registrar.go:71 No registry file found under: /var/lib/filebeat/registry. Creating a new registry file.
2018-06-09T12:45:18.819Z INFO registrar/registrar.go:108 Loading registrar data from /var/lib/filebeat/registry
2018-06-09T12:45:18.819Z INFO registrar/registrar.go:119 States Loaded from registrar: 0
2018-06-09T12:45:18.819Z WARN beater/filebeat.go:261 Filebeat is unable to load the Ingest Node pipelines for the configured modules because the Elasticsearch output is not configured/enabled. If you have already loaded the Ingest Node pipelines or are using Logstash pipelines, you can ignore this warning.
2018-06-09T12:45:18.820Z INFO crawler/crawler.go:48 Loading Prospectors: 1
2018-06-09T12:45:18.821Z INFO log/prospector.go:111 Configured paths: [/var/lib/mysql/log-slow-queries.log*]
2018-06-09T12:45:18.822Z INFO log/prospector.go:111 Configured paths: [/var/lib/mysql/mysql-error.log*]
2018-06-09T12:45:18.822Z INFO crawler/crawler.go:82 Loading and starting Prospectors completed. Enabled prospectors: 0
2018-06-09T12:45:18.822Z INFO cfgfile/reload.go:127 Config reloader started
2018-06-09T12:45:18.840Z INFO log/prospector.go:111 Configured paths: [/var/lib/mysql/log-slow-queries.log*]
2018-06-09T12:45:18.840Z INFO log/prospector.go:111 Configured paths: [/var/lib/mysql/mysql-error.log*]
2018-06-09T12:45:18.840Z INFO cfgfile/reload.go:258 Starting 1 runners ...
2018-06-09T12:45:18.840Z INFO cfgfile/reload.go:219 Loading of config files completed.
2018-06-09T12:45:18.841Z INFO log/harvester.go:216 Harvester started for file: /var/lib/mysql/mysql-error.log
2018-06-09T12:45:18.841Z INFO log/harvester.go:216 Harvester started for file: /var/lib/mysql/log-slow-queries.log
2018-06-09T12:45:20.841Z ERROR pipeline/output.go:74 Failed to connect: dial tcp 172.17.0.6:5044: getsockopt: connection refused
2018-06-09T12:45:22.842Z ERROR pipeline/output.go:74 Failed to connect: dial tcp 172.17.0.6:5044: getsockopt: connection refused
2018-06-09T12:45:26.842Z ERROR pipeline/output.go:74 Failed to connect: dial tcp 172.17.0.6:5044: getsockopt: connection refused
[root@mysql_db1 ~]# tail -f /var/log/filebeat/filebeat
2018-06-09T10:53:28.853Z INFO instance/beat.go:468 Home path: [/usr/share/filebeat] Config path: [/etc/filebeat] Data path: [/var/lib/filebeat] Logs path: [/var/log/filebeat]
2018-06-09T10:53:28.853Z INFO instance/beat.go:475 Beat UUID: 98503460-035e-4476-8e4d-10470433dba5
Notice that once you startup the filebeat service, there's an error shown under the log. This was due to the management server that been assign was not setup yet. For initial phase you can ignore the error log as it will automatically recovered once our management server has been setup and started to crawl.
As configuration for client base are done, you can continue replicate the steps on the other MySQL server that also act as client.
Moving forward, we will continue with setting up the management server itself.
3. Installation Phase (Centralized Management Server Side)
Now as we've done setting up for client side readiness, let's startup the configuration needed for the management server itself. As per brief, there are 3 core packages that needed to be install and configured for management server which is ElasticSearch , LogStash and Kibana.
For this phase, we will start the installation and configuration needed for ElasticSearch first, below are the steps:
[root@elk_master ~]# cd /opt/
[root@elk_master opt]# ls
[root@elk_master opt]# wget https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-6.2.1.tar.gz
--2018-06-09 12:47:59-- https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-6.2.1.tar.gz
Resolving artifacts.elastic.co (artifacts.elastic.co)... 107.21.237.188, 54.235.82.130, 107.21.253.15, ...
Connecting to artifacts.elastic.co (artifacts.elastic.co)|107.21.237.188|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 29049089 (28M) [binary/octet-stream]
Saving to: ‘elasticsearch-6.2.1.tar.gz’
100%[==============================================================================>] 29,049,089 2.47MB/s in 16s
2018-06-09 12:48:21 (1.76 MB/s) - ‘elasticsearch-6.2.1.tar.gz’ saved [29049089/29049089]
[root@elk_master opt]#
[root@elk_master opt]#
[root@elk_master opt]# tar -zxvf elasticsearch-6.2.1.tar.gz
[root@elk_master opt]# ln -s /opt/elasticsearch-6.2.1 /opt/elasticsearch
[root@elk_master opt]# ll
total 28372
lrwxrwxrwx 1 root root 24 Jun 9 12:49 elasticsearch -> /opt/elasticsearch-6.2.1
drwxr-xr-x 8 root root 143 Feb 7 19:36 elasticsearch-6.2.1
-rw-r--r-- 1 root root 29049089 May 15 04:56 elasticsearch-6.2.1.tar.gz
As the installation for elasticsearch are done, let's continue the configuration part. For the configuration side, we will assign directory /data/data to store the collected logging data that been analyze . The directory itself also will be use to store index which will be use by elasticSearch itself for faster query. For directory /data/logs will be used by elasticSearch itself for its own logging purpose. Below are the steps:
[root@elk_master opt]# mkdir -p /data/data
[root@elk_master opt]# mkdir -p /data/logs
[root@elk_master opt]#
[root@elk_master opt]# cd elasticsearch
[root@elk_master elasticsearch]# ls
bin config lib LICENSE.txt logs modules NOTICE.txt plugins README.textile
[root@elk_master elasticsearch]# cd config/
[root@elk_master config]# vi elasticsearch.yml
# ---------------------------------- Cluster -----------------------------------
cluster.name: log_cluster
#
# ------------------------------------ Node ------------------------------------
#
node.name: elk_master
#
# ----------------------------------- Paths ------------------------------------
#
path.data: /data/data
path.logs: /data/logs
#
network.host: 172.17.0.6
Done, in order for ElasticSearch to work, it require Java to be setup. Below are the steps on installing and configuring Java into the server.
[root@elk_master config]# wget --no-cookies --no-check-certificate --header "Cookie: gpw_e24=http%3A%2F%2Fwww.oracle.com%2F; oraclelicense=accept-securebackup-cookie" "http://download.oracle.com/otn-pub/java/jdk/8u131-b11/d54c1d3a095b4ff2b6607d096fa80163/jdk-8u131-linux-x64.rpm"
--2018-06-09 12:57:05-- http://download.oracle.com/otn-pub/java/jdk/8u131-b11/d54c1d3a095b4ff2b6607d096fa80163/jdk-8u131-linux-x64.rpm
Resolving download.oracle.com (download.oracle.com)... 23.49.16.62
Connecting to download.oracle.com (download.oracle.com)|23.49.16.62|:80... connected.
HTTP request sent, awaiting response... 302 Moved Temporarily
Location: https://edelivery.oracle.com/otn-pub/java/jdk/8u131-b11/d54c1d3a095b4ff2b6607d096fa80163/jdk-8u131-linux-x64.rpm [following]
--2018-06-09 12:57:10-- https://edelivery.oracle.com/otn-pub/java/jdk/8u131-b11/d54c1d3a095b4ff2b6607d096fa80163/jdk-8u131-linux-x64.rpm
Resolving edelivery.oracle.com (edelivery.oracle.com)... 104.103.48.174, 2600:1417:58:181::2d3e, 2600:1417:58:188::2d3e
Connecting to edelivery.oracle.com (edelivery.oracle.com)|104.103.48.174|:443... connected.
HTTP request sent, awaiting response... 302 Moved Temporarily
Location: http://download.oracle.com/otn-pub/java/jdk/8u131-b11/d54c1d3a095b4ff2b6607d096fa80163/jdk-8u131-linux-x64.rpm?AuthParam=1528549151_b1fd01d854bc0423600a83c36240028e [following]
--2018-06-09 12:57:11-- http://download.oracle.com/otn-pub/java/jdk/8u131-b11/d54c1d3a095b4ff2b6607d096fa80163/jdk-8u131-linux-x64.rpm?AuthParam=1528549151_b1fd01d854bc0423600a83c36240028e
Connecting to download.oracle.com (download.oracle.com)|23.49.16.62|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 169983496 (162M) [application/x-redhat-package-manager]
Saving to: ‘jdk-8u131-linux-x64.rpm’
100%[==============================================================================>] 169,983,496 2.56MB/s in 64s
2018-06-09 12:58:15 (2.54 MB/s) - ‘jdk-8u131-linux-x64.rpm’ saved [169983496/169983496]
[root@elk_master config]# yum localinstall -y jdk-8u131-linux-x64.rpm
[root@elk_master config]# vi /root/.bash_profile
export JAVA_HOME=/usr/java/jdk1.8.0_131
PATH=$JAVA_HOME/bin:$PATH:$HOME/bin
export PATH
[root@elk_master config]# . /root/.bash_profile
[root@elk_master config]# java -version
java version "1.8.0_131"
Java(TM) SE Runtime Environment (build 1.8.0_131-b11)
Java HotSpot(TM) 64-Bit Server VM (build 25.131-b11, mixed mode)
Done, now elasticSearch have been install and configured in the server. Yet, due to some security policies, elasticSearch are forbid to be run by user root, therefore we will create an additional user to be an owner for elasticSearch service and run it. Below are the steps on creating the dedicated user for it:
[root@elk_master config]# useradd -s /bin/bash shahril
[root@elk_master config]# passwd shahril
Changing password for user shahril.
New password:
BAD PASSWORD: The password fails the dictionary check - it is too simplistic/systematic
Retype new password:
passwd: all authentication tokens updated successfully.
[root@elk_master config]# chown -R shahril:shahril /data/
[root@elk_master config]# sysctl -w vm.max_map_count=262144
vm.max_map_count = 262144
Once done, log in as the user and you may start the elasticSearch services.
