La suite Elastic : Beat ElasticSearch Logstash Kibana (BELK)

Documentation fonctionnelle mais non terminée

Elastic Stack : https://www.elastic.co/fr/blog/elastic-stack-5-0-0-released

Retour d'expérience :

Retour d'installation : http://www.alasta.com/bigdata/2016/05/05/elasticstack-alpha-decouverte.html (2016) - http://magieweb.org/2017/04/tutoriel-mise-en-place-dun-serveur-de-monitoring-avec-elastic-stack-elk/ (2017) - http://blog.kinokocorp.com/?p=191 (2017 - Centos7 )

Recensement
Indexation
Correlation
Actions
Securité des données
Service en continue
Authentification
Sauvegarde
Alerte XMPP ? https://elastalert.readthedocs.io/en/latest/

Tableau avec les outils gratuit et payant : https://www.elastic.co/fr/subscriptions#request-info

Les gratuits :

ElasticSearch : Recherchez, analysez et stockez vos données https://www.elastic.co/products/elasticsearch
Logstash : Intégrer les données https://www.elastic.co/fr/products/logstash
- pipeline qui ingère et traite simultanément des données provenant d'une multitude de sources, puis les transforme. On préférera la solution Beats
Kibana : Visualisez los données https://www.elastic.co/fr/products/kibana
Beats : Intégrer les données https://www.elastic.co/fr/products/beats
- Filebeat : log fichier
- Metricbeat : indicateur
- Packetbeat : données réseau
- Winlogbeat : logs windows
- Heartbeat : heartbeat
X-pack : Search Profiler
X-pack : Monitoring

Palier au module LDAP uniquement payant, faire une authentification apache/nginx : https://mapr.com/blog/how-secure-elasticsearch-and-kibana/

Beats de la communautée : https://www.elastic.co/guide/en/beats/libbeat/current/community-beats.html
systemd.journald / http / apache / mysql /ping / openconfig / nagios
Exemple de journald : https://github.com/mheese/journalbeat

Kafka en complément d’ElasticSearch, afin de faire tampon entre ElasticSearch et ceux qui envoient les messages. Cela permet par exemple d’arrêter ElasticSearch le temps d’une mise à jour, Kafka se chargeant de stocker les messages et de les transmettre une fois que le serveur ElasticSearch est à nouveau disponible
https://kafka.apache.org/
Kafka-manager : Afin d'avoir une interface web de gestion du cluster

elasticsearch-HQ est un outil web permettant l'administration d'un cluster ElasticSearch. Il permet de voir l'état des nœuds, voir les différents documents.
https://github.com/royrusso/elasticsearch-HQ

kafka1.domaine.fr
kafka2.domaine.fr
elasticstack.domaine.fr elasticsearch.domaine.fr kibana.domaine.fr ( même machine )
clientweb1.domaine.fr ( filebeat )
clientdns1.domaine.fr ( logstash )

4vcpu; 6Go
selinux : disable
firewalld : disable

openJDK

Il faut installer openjdk sur les noeuds elasticsearch et les noeuds kafka
```
yum install java-1.8.0-openjdk
```

Clés RPM & folder

rpm --import https://artifacts.elastic.co/GPG-KEY-elasticsearch
mkdir -p /local/rpm
cd /local/rpm

https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-5.5.2.rpm
rpm --install elasticsearch-5.2.1.rpm

Configuration

(https://www.elastic.co/guide/en/elasticsearch/reference/5.5/important-settings.html )

vi /etc/elasticsearch/elasticsearch.yml

cluster.name: cluster-test
node.name: ${HOSTNAME}
bootstrap.memory_lock: true
path.data: /local/elasticsearch/data
path.logs: /local/elasticsearch/logs
network.host: localhost
http.port: 9200
#memory_lock = Désactiver le swap pour Elasticsearch : ( pour la gestion java des gros traitement)

vi /usr/lib/systemd/system/elasticsearch.service

#Décommenter la ligne suivante :
LimitMEMLOCK=infinity
#Supprimer l'option --quiet du paramètre ExecStart pour voir les évènements elasticsearch dans journalctl :
--quiet

vi /etc/sysconfig/elasticsearch

#Décommente la ligne suivante :
MAX_LOCKED_MEMORY=unlimited

Start :

systemctl daemon-reload
systemctl enable elasticsearch
systemctl start elasticsearch

Check

netstat -ltpn
tcp6       0      0 127.0.0.1:9200          :::*                    LISTEN      2344/java
 
curl -XGET 'localhost:9200/_nodes?filter_path=**.mlockall&pretty'
nodes :{......}
 
curl -XGET 'localhost:9200/?pretty'
{
  "name" : "8Y5O47R",
  "cluster_name" : "elasticsearch",
  "cluster_uuid" : "2tt8eL_2TKuUsHVzflH6xQ",
  "version" : {
    "number" : "5.5.2",
    "build_hash" : "b2f0c09",
    "build_date" : "2017-08-14T12:33:14.154Z",
    "build_snapshot" : false,
    "lucene_version" : "6.6.0"
  },
  "tagline" : "You Know, for Search"
}

mkdir /local/kafka 
cd /local/kafka 
wget http://apache.crihan.fr/dist/kafka/0.11.0.0/kafka_2.12-0.11.0.0.tgz
tar -xvf kafka_2.12-0.11.0.0.tgz
cd kafka_2.12-0.11.0.0
groupadd kafka
useradd kafka -d "/local/kafka/" -s "/bin/sh" -g "kafka" -M

Configuration

Import des certificats :

# Conversion au format pkcs12
openssl pkcs12 -export -in /etc/pki/certs/cert.crt -inkey /etc/pki/certs/cert.key -chain -CAfile /etc/pki/certs/certCA.crt -name "elasticstack" -out elasticstack.p12
# import dans le keystore
keytool -importkeystore -deststorepass hhjjkk -destkeystore server.keystore.jks -srckeystore elasticstack.p12 -srcstoretype PKCS12
# Lister le keystore: 
keytool -list -keystore server.keystore.jks
# Autorité de certification : 
keytool -keystore server.truststore.jks -alias CARoot -import -file /etc/pki/certs/certCA.crt

vim config/server.properties

# Ecoute du port + fix problème fqdn/certificat
listeners=PLAINTEXT://:9092,SSL://:9093
advertised.host.name=kafka1.domaine.fr
advertised.listeners=PLAINTEXT://kafka1.domaine.fr:9092,SSL://kafka1.domaine.fr:9093
 
# Replications sur les deux noeud
offsets.topic.replication.factor=2
transaction.state.log.replication.factor=2
transaction.state.log.min.isr=2
default.replication.factor=2
offsets.topic.replication.factor=3
 
# SSL
ssl.keystore.location=/local/kafka/kafka_2.12-0.11.0.0/server.keystore.jks
ssl.keystore.password=hhjjkk
ssl.key.password=hhjjkk
ssl.truststore.location=/local/kafka/kafka_2.12-0.11.0.0/server.truststore.jks
ssl.truststore.password=hhjjkk

vim config/server.properties

dataDir=/tmp/zookeeper
clientPort=2181
tickTime=2000
initLimit=10
syncLimit=5
server.1=kafka1.domaine.fr:2888:3888
server.2=kafka2.domaine.fr:2888:3888

Creation des services systemd

vim /etc/systemd/system/kafka-zookeeper.service

[Unit]
Description=Apache Zookeeper server (Kafka)
Documentation=http://zookeeper.apache.org
Requires=network.target remote-fs.target
After=network.target remote-fs.target
 
[Service]
Type=simple
User=kafka
Group=kafka
Environment=JAVA_HOME=/usr/lib/jvm/jre-1.8.0-openjdk
ExecStart=/local/kafka/kafka_2.12-0.11.0.0/bin/zookeeper-server-start.sh /local/kafka/kafka_2.12-0.11.0.0/config/zookeeper.properties
ExecStop=/local/kafka/kafka_2.12-0.11.0.0/bin/zookeeper-server-stop.sh
 
[Install]
WantedBy=multi-user.target

vi /etc/systemd/system/kafka.service

[Unit]
Description=Apache Kafka server (broker)
Documentation=http://kafka.apache.org/documentation.html
Requires=network.target remote-fs.target
After=network.target remote-fs.target kafka-zookeeper.service
 
[Service]
Type=simple
User=kafka
Group=kafka
Environment=JAVA_HOME=/usr/lib/jvm/jre-1.8.0-openjdk
ExecStart=/local/kafka/kafka_2.12-0.11.0.0/bin/kafka-server-start.sh /local/kafka/kafka_2.12-0.11.0.0/config/server.properties
ExecStop=/local/kafka/kafka_2.12-0.11.0.0/bin/kafka-server-stop.sh
 
[Install]
WantedBy=multi-user.target

Start

systemctl daemon-reload
systemctl start kafka-zookeeper.service
systemctl start kafka.service

Monitoring du cluster

git clone https://github.com/yahoo/kafka-manager.git
cd kafka-manager/
./sbt clean dist
cd target/universal/
unzip kafka-manager-1.3.3.13.zip 
cd kafka-manager-1.3.3.13
ZK_HOSTS=localhost:2181 ./bin/kafka-manager

Go http://kafka1:9000/

—-

wget https://artifacts.elastic.co/downloads/logstash/logstash-5.5.2.rpm
rpm -ivh logstash-5.5.2.rpm

Configuration

Import du certification dans /etc/pki/TERENA

Convertir la clé pour que logstash puisse l'utiliser :

erreur : 
[2017-09-04T15:44:06,011][ERROR][logstash.inputs.beats    ] Looks like you either have an invalid key or your private key was not in PKCS8 format. {:exception=>java.lang.IllegalArgumentException: File does not contain valid private key: /etc/pki/certs/cert.key}  
solution : 
15:45:45 root@elasticstack:/local/rpm# openssl pkcs8 -topk8 -inform PEM -outform PEM -in /etc/pki/certs/cert.key -out /etc/pki/certs/cert.pem -nocrypt

```
cd /etc/logstash/conf.d/
```

webtest.conf (log apache)

input {
	kafka {
                bootstrap_servers => 'kafka1.domaine.fr:9092,kafka2.domaine.fr:9092'
                topics => ["WEB-TEST_APACHE"]
                auto_offset_reset => "earliest"   /* pour que logstash recupère les logs manquant */ 
		codec => json {}
	}
}
filter
{
 
	grok {
		match => { "message" => ["%{IPORHOST:[apache2][access][remote_ip]} - %{DATA:[apache2][access][user_name]} \[%{HTTPDATE:[apache2][access][time]}\] \"%{WORD:[apache2][access][method]} %{DATA:[apache2][access][url]} HTTP/%{NUMBER:[apache2][access][http_version]}\" %{NUMBER:[apache2][access][response_code]} %{NUMBER:[apache2][access][body_sent][bytes]}( \"%{DATA:[apache2][access][referrer]}\")?( \"%{DATA:[apache2][access][agent]}\")?",
			"%{IPORHOST:[apache2][access][remote_ip]} - %{DATA:[apache2][access][user_name]} \\[%{HTTPDATE:[apache2][access][time]}\\] \"-\" %{NUMBER:[apache2][access][response_code]} -" ] }
		remove_field => "message"
	}
	mutate {
		add_field => { "read_timestamp" => "%{@timestamp}" }
	}
	date {
		match => [ "[apache2][access][time]", "dd/MMM/YYYY:H:m:s Z" ]
			remove_field => "[apache2][access][time]"
	}
	useragent {
		source => "[apache2][access][agent]"
			target => "[apache2][access][user_agent]"
			remove_field => "[apache2][access][agent]"
	}
	geoip {
		source => "[apache2][access][remote_ip]"
			target => "[apache2][access][geoip]"
	}
 
 
 
}
output {
	elasticsearch {
		index  => "webtest-logs-%{+YYYY.MM.dd}"
			hosts => ["localhost:9200"]
			sniffing => false
	}
 
	stdout
	{
		codec => rubydebug
	}
}

Tester la configuration et la syntax

/usr/share/logstash/bin/logstash -f /etc/logstash/conf.d/web-test.conf -t

Start

systemctl enable logstash
systemctl start logstash

wget https://artifacts.elastic.co/downloads/beats/filebeat/filebeat-5.5.2-x86_64.rpm
rpm -vi filebeat-5.5.2-x86_64.rpm

Configuration

Import du certification dans /etc/pki/TERENA

filebeat.prospectors:
- input_type: log
  paths:
  - /var/log/httpd/*log
  document_type: apache
 
 
- input_type: log
  paths:
    - /var/log/*.log
 
.............

Exemple vers logstash

output.logstash:
  # The Logstash hosts
  hosts: ["elasticstack.domaine.fr:5443"]
 
  # Optional SSL. By default is off.
  # List of root certificates for HTTPS server verifications
  ssl.certificate_authorities: ["/etc/pki/certs/certCA.crt"]
 
  # Certificate for SSL client authentication
  #ssl.certificate: "/etc/pki/client/cert.pem"
 
  # Client Certificate Key
  #ssl.key: "/etc/pki/client/cert.key"
 
  template.name: "filebeat"
  template.path: "filebeat.template.json"
  template.overwrite: false

Exemple vers kafka

output.kafka: 
output.kafka:
   # initial brokers for reading cluster metadata
   #hosts: ["kafka1.domaine.fr:9092","kafka2.domaine.fr:9092"]
   hosts: ["kafka1.domaine.fr:9093","kafka2.domaine.fr:9093"]
 
   # message topic selection + partitioning
   topic: WEB-TEST_APACHE
   #topic: '%{[type]}'
   partition.round_robin:
     reachable_only: false
 
   required_acks: 1
   compression: gzip
   max_message_bytes: 1000000
 
   ssl.certificate_authorities: ["/etc/pki/certs/certCA.crt"]
   ssl.certificate: "/etc/pki/certs/cert.crt"
   ssl.key: "/etc/pki/certs/cert.key"

Start

systemctl enable filebeat
systemctl start filebeat

wget https://artifacts.elastic.co/downloads/kibana/kibana-5.5.2-x86_64.rpm
rpm -ivh kibana-5.5.2-x86_64.rpm

Configuration

vim /etc/kibana/kibana.yml

server.port: 5601
server.host: "localhost"
elasticsearch.url: "http://localhost:9200"

Start

systemctl enable kibana
systemctl start kibana

ProxyPass

yum install httpd
vim /etc/httpd/conf.d/kibana.conf
<Location "/">
  ProxyPass "http://localhost:5601/"
  ProxyPassReverse "http://localhost:5601/"
  # Ajouter authentification de votre choix (htpasswd, ldap, ... )
</Location>

Utilisation

Selectionner les indexs ( pour le faire après configuration initiale : Management > Inde Patterns )
exemple logstash : filebeat-*
exemple kafka : webtest-logs-*
Time Filter field name : @timestamp

cd /local/
git clone https://github.com/royrusso/elasticsearch-HQ.git

Configuration

vim /etc/elasticsearch/elasticsearch.yml

.....
http.cors.allow-origin: "*" #Mettre ip autorisé à faire l'admin
http.cors.enabled: true

vim /etc/httpd/conf.d/proxypass.conf

<Location "/elasticsearch-HQ">
        ProxyPass "!"
</Location>
<Directory "/var/www/html/elasticsearch-HQ">
    Options Indexes FollowSymLinks
 
    AllowOverride None
    Require all granted
</Directory>

packetbeat

Attention ! Beaucoup de CPU quand beaucoup de requêtes

yum install libpcap
wget https://artifacts.elastic.co/downloads/beats/packetbeat/packetbeat-5.6.0-x86_64.rpm
rpm -vi packetbeat-5.6.0-x86_64.rpm

Importer les dashboard dans kibana

/usr/share/packetbeat/scripts/import_dashboards -es http://elasticsearch.domaine.fr:9200

#################

grok debugger http://grokdebug.herokuapp.com/

Retour

Vous pourriez laisser un commentaire si vous étiez connecté.

La suite Elastic : Beat ElasticSearch Logstash Kibana (BELK)

Projet ELK : La Suite Elastic

Partie 1 : Le Besoin

Partie 2 : Les outils

Tools

LibBeats

Kafka

elasticsearch-HQ

Architecture

Installation

Pré-requis

openJDK

Clés RPM & folder

ELASTICSEARCH

Configuration

Start :

Check

Kafka

Configuration

Creation des services systemd

Start

Monitoring du cluster

LOGSTASH

Configuration

Start

FILEBEAT

Configuration

Start

KIBANA

Configuration

Start

ProxyPass

Utilisation

ELASTICSEARCH-HQ

Configuration

AUTRE

packetbeat

AdminWiki.fr